scmutil: allowing different files to be prefetched per revision

The old API takes a list of revision separate from the file matcher, and thus
provides no way to fetch different sets of files from each revision. In
preparation for adding one such usage, I'm changing the API to take a list of
(revision, file matcher) tuples instead.

Differential Revision: https://phab.mercurial-scm.org/D8721

hgext/lfs/wrapper.py

@@ -1,546 +1,546 b''
 # wrapper.py - methods wrapping core mercurial logic
 #
 # Copyright 2017 Facebook, Inc.
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import hashlib
 
 from mercurial.i18n import _
 from mercurial.node import bin, hex, nullid, short
 from mercurial.pycompat import (
     getattr,
     setattr,
 )
 
 from mercurial import (
     bundle2,
     changegroup,
     cmdutil,
     context,
     error,
     exchange,
     exthelper,
     localrepo,
     pycompat,
     revlog,
     scmutil,
     upgrade,
     util,
     vfs as vfsmod,
     wireprotov1server,
 )
 
 from mercurial.interfaces import repository
 
 from mercurial.utils import (
     storageutil,
     stringutil,
 )
 
 from ..largefiles import lfutil
 
 from . import (
     blobstore,
     pointer,
 )
 
 eh = exthelper.exthelper()
 
 
 @eh.wrapfunction(localrepo, b'makefilestorage')
 def localrepomakefilestorage(orig, requirements, features, **kwargs):
     if b'lfs' in requirements:
         features.add(repository.REPO_FEATURE_LFS)
 
     return orig(requirements=requirements, features=features, **kwargs)
 
 
 @eh.wrapfunction(changegroup, b'allsupportedversions')
 def allsupportedversions(orig, ui):
     versions = orig(ui)
     versions.add(b'03')
     return versions
 
 
 @eh.wrapfunction(wireprotov1server, b'_capabilities')
 def _capabilities(orig, repo, proto):
     '''Wrap server command to announce lfs server capability'''
     caps = orig(repo, proto)
     if util.safehasattr(repo.svfs, b'lfslocalblobstore'):
         # Advertise a slightly different capability when lfs is *required*, so
         # that the client knows it MUST load the extension.  If lfs is not
         # required on the server, there's no reason to autoload the extension
         # on the client.
         if b'lfs' in repo.requirements:
             caps.append(b'lfs-serve')
 
         caps.append(b'lfs')
     return caps
 
 
 def bypasscheckhash(self, text):
     return False
 
 
 def readfromstore(self, text):
     """Read filelog content from local blobstore transform for flagprocessor.
 
     Default tranform for flagprocessor, returning contents from blobstore.
     Returns a 2-typle (text, validatehash) where validatehash is True as the
     contents of the blobstore should be checked using checkhash.
     """
     p = pointer.deserialize(text)
     oid = p.oid()
     store = self.opener.lfslocalblobstore
     if not store.has(oid):
         p.filename = self.filename
         self.opener.lfsremoteblobstore.readbatch([p], store)
 
     # The caller will validate the content
     text = store.read(oid, verify=False)
 
     # pack hg filelog metadata
     hgmeta = {}
     for k in p.keys():
         if k.startswith(b'x-hg-'):
             name = k[len(b'x-hg-') :]
             hgmeta[name] = p[k]
     if hgmeta or text.startswith(b'\1\n'):
         text = storageutil.packmeta(hgmeta, text)
 
     return (text, True, {})
 
 
 def writetostore(self, text, sidedata):
     # hg filelog metadata (includes rename, etc)
     hgmeta, offset = storageutil.parsemeta(text)
     if offset and offset > 0:
         # lfs blob does not contain hg filelog metadata
         text = text[offset:]
 
     # git-lfs only supports sha256
     oid = hex(hashlib.sha256(text).digest())
     self.opener.lfslocalblobstore.write(oid, text)
 
     # replace contents with metadata
     longoid = b'sha256:%s' % oid
     metadata = pointer.gitlfspointer(oid=longoid, size=b'%d' % len(text))
 
     # by default, we expect the content to be binary. however, LFS could also
     # be used for non-binary content. add a special entry for non-binary data.
     # this will be used by filectx.isbinary().
     if not stringutil.binary(text):
         # not hg filelog metadata (affecting commit hash), no "x-hg-" prefix
         metadata[b'x-is-binary'] = b'0'
 
     # translate hg filelog metadata to lfs metadata with "x-hg-" prefix
     if hgmeta is not None:
         for k, v in pycompat.iteritems(hgmeta):
             metadata[b'x-hg-%s' % k] = v
 
     rawtext = metadata.serialize()
     return (rawtext, False)
 
 
 def _islfs(rlog, node=None, rev=None):
     if rev is None:
         if node is None:
             # both None - likely working copy content where node is not ready
             return False
         rev = rlog.rev(node)
     else:
         node = rlog.node(rev)
     if node == nullid:
         return False
     flags = rlog.flags(rev)
     return bool(flags & revlog.REVIDX_EXTSTORED)
 
 
 # Wrapping may also be applied by remotefilelog
 def filelogaddrevision(
     orig,
     self,
     text,
     transaction,
     link,
     p1,
     p2,
     cachedelta=None,
     node=None,
     flags=revlog.REVIDX_DEFAULT_FLAGS,
     **kwds
 ):
     # The matcher isn't available if reposetup() wasn't called.
     lfstrack = self._revlog.opener.options.get(b'lfstrack')
 
     if lfstrack:
         textlen = len(text)
         # exclude hg rename meta from file size
         meta, offset = storageutil.parsemeta(text)
         if offset:
             textlen -= offset
 
         if lfstrack(self._revlog.filename, textlen):
             flags |= revlog.REVIDX_EXTSTORED
 
     return orig(
         self,
         text,
         transaction,
         link,
         p1,
         p2,
         cachedelta=cachedelta,
         node=node,
         flags=flags,
         **kwds
     )
 
 
 # Wrapping may also be applied by remotefilelog
 def filelogrenamed(orig, self, node):
     if _islfs(self._revlog, node):
         rawtext = self._revlog.rawdata(node)
         if not rawtext:
             return False
         metadata = pointer.deserialize(rawtext)
         if b'x-hg-copy' in metadata and b'x-hg-copyrev' in metadata:
             return metadata[b'x-hg-copy'], bin(metadata[b'x-hg-copyrev'])
         else:
             return False
     return orig(self, node)
 
 
 # Wrapping may also be applied by remotefilelog
 def filelogsize(orig, self, rev):
     if _islfs(self._revlog, rev=rev):
         # fast path: use lfs metadata to answer size
         rawtext = self._revlog.rawdata(rev)
         metadata = pointer.deserialize(rawtext)
         return int(metadata[b'size'])
     return orig(self, rev)
 
 
 @eh.wrapfunction(revlog, b'_verify_revision')
 def _verify_revision(orig, rl, skipflags, state, node):
     if _islfs(rl, node=node):
         rawtext = rl.rawdata(node)
         metadata = pointer.deserialize(rawtext)
 
         # Don't skip blobs that are stored locally, as local verification is
         # relatively cheap and there's no other way to verify the raw data in
         # the revlog.
         if rl.opener.lfslocalblobstore.has(metadata.oid()):
             skipflags &= ~revlog.REVIDX_EXTSTORED
         elif skipflags & revlog.REVIDX_EXTSTORED:
             # The wrapped method will set `skipread`, but there's enough local
             # info to check renames.
             state[b'safe_renamed'].add(node)
 
     orig(rl, skipflags, state, node)
 
 
 @eh.wrapfunction(context.basefilectx, b'cmp')
 def filectxcmp(orig, self, fctx):
     """returns True if text is different than fctx"""
     # some fctx (ex. hg-git) is not based on basefilectx and do not have islfs
     if self.islfs() and getattr(fctx, 'islfs', lambda: False)():
         # fast path: check LFS oid
         p1 = pointer.deserialize(self.rawdata())
         p2 = pointer.deserialize(fctx.rawdata())
         return p1.oid() != p2.oid()
     return orig(self, fctx)
 
 
 @eh.wrapfunction(context.basefilectx, b'isbinary')
 def filectxisbinary(orig, self):
     if self.islfs():
         # fast path: use lfs metadata to answer isbinary
         metadata = pointer.deserialize(self.rawdata())
         # if lfs metadata says nothing, assume it's binary by default
         return bool(int(metadata.get(b'x-is-binary', 1)))
     return orig(self)
 
 
 def filectxislfs(self):
     return _islfs(self.filelog()._revlog, self.filenode())
 
 
 @eh.wrapfunction(cmdutil, b'_updatecatformatter')
 def _updatecatformatter(orig, fm, ctx, matcher, path, decode):
     orig(fm, ctx, matcher, path, decode)
     fm.data(rawdata=ctx[path].rawdata())
 
 
 @eh.wrapfunction(scmutil, b'wrapconvertsink')
 def convertsink(orig, sink):
     sink = orig(sink)
     if sink.repotype == b'hg':
 
         class lfssink(sink.__class__):
             def putcommit(
                 self,
                 files,
                 copies,
                 parents,
                 commit,
                 source,
                 revmap,
                 full,
                 cleanp2,
             ):
                 pc = super(lfssink, self).putcommit
                 node = pc(
                     files,
                     copies,
                     parents,
                     commit,
                     source,
                     revmap,
                     full,
                     cleanp2,
                 )
 
                 if b'lfs' not in self.repo.requirements:
                     ctx = self.repo[node]
 
                     # The file list may contain removed files, so check for
                     # membership before assuming it is in the context.
                     if any(f in ctx and ctx[f].islfs() for f, n in files):
                         self.repo.requirements.add(b'lfs')
                         self.repo._writerequirements()
 
                 return node
 
         sink.__class__ = lfssink
 
     return sink
 
 
 # bundlerepo uses "vfsmod.readonlyvfs(othervfs)", we need to make sure lfs
 # options and blob stores are passed from othervfs to the new readonlyvfs.
 @eh.wrapfunction(vfsmod.readonlyvfs, b'__init__')
 def vfsinit(orig, self, othervfs):
     orig(self, othervfs)
     # copy lfs related options
     for k, v in othervfs.options.items():
         if k.startswith(b'lfs'):
             self.options[k] = v
     # also copy lfs blobstores. note: this can run before reposetup, so lfs
     # blobstore attributes are not always ready at this time.
     for name in [b'lfslocalblobstore', b'lfsremoteblobstore']:
         if util.safehasattr(othervfs, name):
             setattr(self, name, getattr(othervfs, name))
 
 
-def _prefetchfiles(repo, revs, match):
+def _prefetchfiles(repo, revmatches):
     """Ensure that required LFS blobs are present, fetching them as a group if
     needed."""
     if not util.safehasattr(repo.svfs, b'lfslocalblobstore'):
         return
 
     pointers = []
     oids = set()
     localstore = repo.svfs.lfslocalblobstore
 
-    for rev in revs:
+    for rev, match in revmatches:
         ctx = repo[rev]
         for f in ctx.walk(match):
             p = pointerfromctx(ctx, f)
             if p and p.oid() not in oids and not localstore.has(p.oid()):
                 p.filename = f
                 pointers.append(p)
                 oids.add(p.oid())
 
     if pointers:
         # Recalculating the repo store here allows 'paths.default' that is set
         # on the repo by a clone command to be used for the update.
         blobstore.remote(repo).readbatch(pointers, localstore)
 
 
 def _canskipupload(repo):
     # Skip if this hasn't been passed to reposetup()
     if not util.safehasattr(repo.svfs, b'lfsremoteblobstore'):
         return True
 
     # if remotestore is a null store, upload is a no-op and can be skipped
     return isinstance(repo.svfs.lfsremoteblobstore, blobstore._nullremote)
 
 
 def candownload(repo):
     # Skip if this hasn't been passed to reposetup()
     if not util.safehasattr(repo.svfs, b'lfsremoteblobstore'):
         return False
 
     # if remotestore is a null store, downloads will lead to nothing
     return not isinstance(repo.svfs.lfsremoteblobstore, blobstore._nullremote)
 
 
 def uploadblobsfromrevs(repo, revs):
     '''upload lfs blobs introduced by revs
 
     Note: also used by other extensions e. g. infinitepush. avoid renaming.
     '''
     if _canskipupload(repo):
         return
     pointers = extractpointers(repo, revs)
     uploadblobs(repo, pointers)
 
 
 def prepush(pushop):
     """Prepush hook.
 
     Read through the revisions to push, looking for filelog entries that can be
     deserialized into metadata so that we can block the push on their upload to
     the remote blobstore.
     """
     return uploadblobsfromrevs(pushop.repo, pushop.outgoing.missing)
 
 
 @eh.wrapfunction(exchange, b'push')
 def push(orig, repo, remote, *args, **kwargs):
     """bail on push if the extension isn't enabled on remote when needed, and
     update the remote store based on the destination path."""
     if b'lfs' in repo.requirements:
         # If the remote peer is for a local repo, the requirement tests in the
         # base class method enforce lfs support.  Otherwise, some revisions in
         # this repo use lfs, and the remote repo needs the extension loaded.
         if not remote.local() and not remote.capable(b'lfs'):
             # This is a copy of the message in exchange.push() when requirements
             # are missing between local repos.
             m = _(b"required features are not supported in the destination: %s")
             raise error.Abort(
                 m % b'lfs', hint=_(b'enable the lfs extension on the server')
             )
 
         # Repositories where this extension is disabled won't have the field.
         # But if there's a requirement, then the extension must be loaded AND
         # there may be blobs to push.
         remotestore = repo.svfs.lfsremoteblobstore
         try:
             repo.svfs.lfsremoteblobstore = blobstore.remote(repo, remote.url())
             return orig(repo, remote, *args, **kwargs)
         finally:
             repo.svfs.lfsremoteblobstore = remotestore
     else:
         return orig(repo, remote, *args, **kwargs)
 
 
 # when writing a bundle via "hg bundle" command, upload related LFS blobs
 @eh.wrapfunction(bundle2, b'writenewbundle')
 def writenewbundle(
     orig, ui, repo, source, filename, bundletype, outgoing, *args, **kwargs
 ):
     """upload LFS blobs added by outgoing revisions on 'hg bundle'"""
     uploadblobsfromrevs(repo, outgoing.missing)
     return orig(
         ui, repo, source, filename, bundletype, outgoing, *args, **kwargs
     )
 
 
 def extractpointers(repo, revs):
     """return a list of lfs pointers added by given revs"""
     repo.ui.debug(b'lfs: computing set of blobs to upload\n')
     pointers = {}
 
     makeprogress = repo.ui.makeprogress
     with makeprogress(
         _(b'lfs search'), _(b'changesets'), len(revs)
     ) as progress:
         for r in revs:
             ctx = repo[r]
             for p in pointersfromctx(ctx).values():
                 pointers[p.oid()] = p
             progress.increment()
         return sorted(pointers.values(), key=lambda p: p.oid())
 
 
 def pointerfromctx(ctx, f, removed=False):
     """return a pointer for the named file from the given changectx, or None if
     the file isn't LFS.
 
     Optionally, the pointer for a file deleted from the context can be returned.
     Since no such pointer is actually stored, and to distinguish from a non LFS
     file, this pointer is represented by an empty dict.
     """
     _ctx = ctx
     if f not in ctx:
         if not removed:
             return None
         if f in ctx.p1():
             _ctx = ctx.p1()
         elif f in ctx.p2():
             _ctx = ctx.p2()
         else:
             return None
     fctx = _ctx[f]
     if not _islfs(fctx.filelog()._revlog, fctx.filenode()):
         return None
     try:
         p = pointer.deserialize(fctx.rawdata())
         if ctx == _ctx:
             return p
         return {}
     except pointer.InvalidPointer as ex:
         raise error.Abort(
             _(b'lfs: corrupted pointer (%s@%s): %s\n')
             % (f, short(_ctx.node()), ex)
         )
 
 
 def pointersfromctx(ctx, removed=False):
     """return a dict {path: pointer} for given single changectx.
 
     If ``removed`` == True and the LFS file was removed from ``ctx``, the value
     stored for the path is an empty dict.
     """
     result = {}
     m = ctx.repo().narrowmatch()
 
     # TODO: consider manifest.fastread() instead
     for f in ctx.files():
         if not m(f):
             continue
         p = pointerfromctx(ctx, f, removed=removed)
         if p is not None:
             result[f] = p
     return result
 
 
 def uploadblobs(repo, pointers):
     """upload given pointers from local blobstore"""
     if not pointers:
         return
 
     remoteblob = repo.svfs.lfsremoteblobstore
     remoteblob.writebatch(pointers, repo.svfs.lfslocalblobstore)
 
 
 @eh.wrapfunction(upgrade, b'_finishdatamigration')
 def upgradefinishdatamigration(orig, ui, srcrepo, dstrepo, requirements):
     orig(ui, srcrepo, dstrepo, requirements)
 
     # Skip if this hasn't been passed to reposetup()
     if util.safehasattr(
         srcrepo.svfs, b'lfslocalblobstore'
     ) and util.safehasattr(dstrepo.svfs, b'lfslocalblobstore'):
         srclfsvfs = srcrepo.svfs.lfslocalblobstore.vfs
         dstlfsvfs = dstrepo.svfs.lfslocalblobstore.vfs
 
         for dirpath, dirs, files in srclfsvfs.walk():
             for oid in files:
                 ui.write(_(b'copying lfs blob %s\n') % oid)
                 lfutil.link(srclfsvfs.join(oid), dstlfsvfs.join(oid))
 
 
 @eh.wrapfunction(upgrade, b'preservedrequirements')
 @eh.wrapfunction(upgrade, b'supporteddestrequirements')
 def upgraderequirements(orig, repo):
     reqs = orig(repo)
     if b'lfs' in repo.requirements:
         reqs.add(b'lfs')
     return reqs

hgext/remotefilelog/__init__.py

@@ -1,1285 +1,1285 b''
 # __init__.py - remotefilelog extension
 #
 # Copyright 2013 Facebook, Inc.
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 """remotefilelog causes Mercurial to lazilly fetch file contents (EXPERIMENTAL)
 
 This extension is HIGHLY EXPERIMENTAL. There are NO BACKWARDS COMPATIBILITY
 GUARANTEES. This means that repositories created with this extension may
 only be usable with the exact version of this extension/Mercurial that was
 used. The extension attempts to enforce this in order to prevent repository
 corruption.
 
 remotefilelog works by fetching file contents lazily and storing them
 in a cache on the client rather than in revlogs. This allows enormous
 histories to be transferred only partially, making them easier to
 operate on.
 
 Configs:
 
     ``packs.maxchainlen`` specifies the maximum delta chain length in pack files
 
     ``packs.maxpacksize`` specifies the maximum pack file size
 
     ``packs.maxpackfilecount`` specifies the maximum number of packs in the
       shared cache (trees only for now)
 
     ``remotefilelog.backgroundprefetch`` runs prefetch in background when True
 
     ``remotefilelog.bgprefetchrevs`` specifies revisions to fetch on commit and
       update, and on other commands that use them. Different from pullprefetch.
 
     ``remotefilelog.gcrepack`` does garbage collection during repack when True
 
     ``remotefilelog.nodettl`` specifies maximum TTL of a node in seconds before
       it is garbage collected
 
     ``remotefilelog.repackonhggc`` runs repack on hg gc when True
 
     ``remotefilelog.prefetchdays`` specifies the maximum age of a commit in
       days after which it is no longer prefetched.
 
     ``remotefilelog.prefetchdelay`` specifies delay between background
       prefetches in seconds after operations that change the working copy parent
 
     ``remotefilelog.data.gencountlimit`` constraints the minimum number of data
       pack files required to be considered part of a generation. In particular,
       minimum number of packs files > gencountlimit.
 
     ``remotefilelog.data.generations`` list for specifying the lower bound of
       each generation of the data pack files. For example, list ['100MB','1MB']
       or ['1MB', '100MB'] will lead to three generations: [0, 1MB), [
       1MB, 100MB) and [100MB, infinity).
 
     ``remotefilelog.data.maxrepackpacks`` the maximum number of pack files to
       include in an incremental data repack.
 
     ``remotefilelog.data.repackmaxpacksize`` the maximum size of a pack file for
       it to be considered for an incremental data repack.
 
     ``remotefilelog.data.repacksizelimit`` the maximum total size of pack files
       to include in an incremental data repack.
 
     ``remotefilelog.history.gencountlimit`` constraints the minimum number of
       history pack files required to be considered part of a generation. In
       particular, minimum number of packs files > gencountlimit.
 
     ``remotefilelog.history.generations`` list for specifying the lower bound of
       each generation of the history pack files. For example, list [
       '100MB', '1MB'] or ['1MB', '100MB'] will lead to three generations: [
       0, 1MB), [1MB, 100MB) and [100MB, infinity).
 
     ``remotefilelog.history.maxrepackpacks`` the maximum number of pack files to
       include in an incremental history repack.
 
     ``remotefilelog.history.repackmaxpacksize`` the maximum size of a pack file
       for it to be considered for an incremental history repack.
 
     ``remotefilelog.history.repacksizelimit`` the maximum total size of pack
       files to include in an incremental history repack.
 
     ``remotefilelog.backgroundrepack`` automatically consolidate packs in the
       background
 
     ``remotefilelog.cachepath`` path to cache
 
     ``remotefilelog.cachegroup`` if set, make cache directory sgid to this
       group
 
     ``remotefilelog.cacheprocess`` binary to invoke for fetching file data
 
     ``remotefilelog.debug`` turn on remotefilelog-specific debug output
 
     ``remotefilelog.excludepattern`` pattern of files to exclude from pulls
 
     ``remotefilelog.includepattern`` pattern of files to include in pulls
 
     ``remotefilelog.fetchwarning``: message to print when too many
       single-file fetches occur
 
     ``remotefilelog.getfilesstep`` number of files to request in a single RPC
 
     ``remotefilelog.getfilestype`` if set to 'threaded' use threads to fetch
       files, otherwise use optimistic fetching
 
     ``remotefilelog.pullprefetch`` revset for selecting files that should be
       eagerly downloaded rather than lazily
 
     ``remotefilelog.reponame`` name of the repo. If set, used to partition
       data from other repos in a shared store.
 
     ``remotefilelog.server`` if true, enable server-side functionality
 
     ``remotefilelog.servercachepath`` path for caching blobs on the server
 
     ``remotefilelog.serverexpiration`` number of days to keep cached server
       blobs
 
     ``remotefilelog.validatecache`` if set, check cache entries for corruption
       before returning blobs
 
     ``remotefilelog.validatecachelog`` if set, check cache entries for
       corruption before returning metadata
 
 """
 from __future__ import absolute_import
 
 import os
 import time
 import traceback
 
 from mercurial.node import hex
 from mercurial.i18n import _
 from mercurial.pycompat import open
 from mercurial import (
     changegroup,
     changelog,
     cmdutil,
     commands,
     configitems,
     context,
     copies,
     debugcommands as hgdebugcommands,
     dispatch,
     error,
     exchange,
     extensions,
     hg,
     localrepo,
     match,
     merge,
     node as nodemod,
     patch,
     pycompat,
     registrar,
     repair,
     repoview,
     revset,
     scmutil,
     smartset,
     streamclone,
     util,
 )
 from . import (
     constants,
     debugcommands,
     fileserverclient,
     remotefilectx,
     remotefilelog,
     remotefilelogserver,
     repack as repackmod,
     shallowbundle,
     shallowrepo,
     shallowstore,
     shallowutil,
     shallowverifier,
 )
 
 # ensures debug commands are registered
 hgdebugcommands.command
 
 cmdtable = {}
 command = registrar.command(cmdtable)
 
 configtable = {}
 configitem = registrar.configitem(configtable)
 
 configitem(b'remotefilelog', b'debug', default=False)
 
 configitem(b'remotefilelog', b'reponame', default=b'')
 configitem(b'remotefilelog', b'cachepath', default=None)
 configitem(b'remotefilelog', b'cachegroup', default=None)
 configitem(b'remotefilelog', b'cacheprocess', default=None)
 configitem(b'remotefilelog', b'cacheprocess.includepath', default=None)
 configitem(b"remotefilelog", b"cachelimit", default=b"1000 GB")
 
 configitem(
     b'remotefilelog',
     b'fallbackpath',
     default=configitems.dynamicdefault,
     alias=[(b'remotefilelog', b'fallbackrepo')],
 )
 
 configitem(b'remotefilelog', b'validatecachelog', default=None)
 configitem(b'remotefilelog', b'validatecache', default=b'on')
 configitem(b'remotefilelog', b'server', default=None)
 configitem(b'remotefilelog', b'servercachepath', default=None)
 configitem(b"remotefilelog", b"serverexpiration", default=30)
 configitem(b'remotefilelog', b'backgroundrepack', default=False)
 configitem(b'remotefilelog', b'bgprefetchrevs', default=None)
 configitem(b'remotefilelog', b'pullprefetch', default=None)
 configitem(b'remotefilelog', b'backgroundprefetch', default=False)
 configitem(b'remotefilelog', b'prefetchdelay', default=120)
 configitem(b'remotefilelog', b'prefetchdays', default=14)
 
 configitem(b'remotefilelog', b'getfilesstep', default=10000)
 configitem(b'remotefilelog', b'getfilestype', default=b'optimistic')
 configitem(b'remotefilelog', b'batchsize', configitems.dynamicdefault)
 configitem(b'remotefilelog', b'fetchwarning', default=b'')
 
 configitem(b'remotefilelog', b'includepattern', default=None)
 configitem(b'remotefilelog', b'excludepattern', default=None)
 
 configitem(b'remotefilelog', b'gcrepack', default=False)
 configitem(b'remotefilelog', b'repackonhggc', default=False)
 configitem(b'repack', b'chainorphansbysize', default=True, experimental=True)
 
 configitem(b'packs', b'maxpacksize', default=0)
 configitem(b'packs', b'maxchainlen', default=1000)
 
 configitem(b'devel', b'remotefilelog.bg-wait', default=False)
 
 #  default TTL limit is 30 days
 _defaultlimit = 60 * 60 * 24 * 30
 configitem(b'remotefilelog', b'nodettl', default=_defaultlimit)
 
 configitem(b'remotefilelog', b'data.gencountlimit', default=2),
 configitem(
     b'remotefilelog', b'data.generations', default=[b'1GB', b'100MB', b'1MB']
 )
 configitem(b'remotefilelog', b'data.maxrepackpacks', default=50)
 configitem(b'remotefilelog', b'data.repackmaxpacksize', default=b'4GB')
 configitem(b'remotefilelog', b'data.repacksizelimit', default=b'100MB')
 
 configitem(b'remotefilelog', b'history.gencountlimit', default=2),
 configitem(b'remotefilelog', b'history.generations', default=[b'100MB'])
 configitem(b'remotefilelog', b'history.maxrepackpacks', default=50)
 configitem(b'remotefilelog', b'history.repackmaxpacksize', default=b'400MB')
 configitem(b'remotefilelog', b'history.repacksizelimit', default=b'100MB')
 
 # Note for extension authors: ONLY specify testedwith = 'ships-with-hg-core' for
 # extensions which SHIP WITH MERCURIAL. Non-mainline extensions should
 # be specifying the version(s) of Mercurial they are tested with, or
 # leave the attribute unspecified.
 testedwith = b'ships-with-hg-core'
 
 repoclass = localrepo.localrepository
 repoclass._basesupported.add(constants.SHALLOWREPO_REQUIREMENT)
 
 isenabled = shallowutil.isenabled
 
 
 def uisetup(ui):
     """Wraps user facing Mercurial commands to swap them out with shallow
     versions.
     """
     hg.wirepeersetupfuncs.append(fileserverclient.peersetup)
 
     entry = extensions.wrapcommand(commands.table, b'clone', cloneshallow)
     entry[1].append(
         (
             b'',
             b'shallow',
             None,
             _(b"create a shallow clone which uses remote file history"),
         )
     )
 
     extensions.wrapcommand(
         commands.table, b'debugindex', debugcommands.debugindex
     )
     extensions.wrapcommand(
         commands.table, b'debugindexdot', debugcommands.debugindexdot
     )
     extensions.wrapcommand(commands.table, b'log', log)
     extensions.wrapcommand(commands.table, b'pull', pull)
 
     # Prevent 'hg manifest --all'
     def _manifest(orig, ui, repo, *args, **opts):
         if isenabled(repo) and opts.get('all'):
             raise error.Abort(_(b"--all is not supported in a shallow repo"))
 
         return orig(ui, repo, *args, **opts)
 
     extensions.wrapcommand(commands.table, b"manifest", _manifest)
 
     # Wrap remotefilelog with lfs code
     def _lfsloaded(loaded=False):
         lfsmod = None
         try:
             lfsmod = extensions.find(b'lfs')
         except KeyError:
             pass
         if lfsmod:
             lfsmod.wrapfilelog(remotefilelog.remotefilelog)
             fileserverclient._lfsmod = lfsmod
 
     extensions.afterloaded(b'lfs', _lfsloaded)
 
     # debugdata needs remotefilelog.len to work
     extensions.wrapcommand(commands.table, b'debugdata', debugdatashallow)
 
     changegroup.cgpacker = shallowbundle.shallowcg1packer
 
     extensions.wrapfunction(
         changegroup, b'_addchangegroupfiles', shallowbundle.addchangegroupfiles
     )
     extensions.wrapfunction(
         changegroup, b'makechangegroup', shallowbundle.makechangegroup
     )
     extensions.wrapfunction(localrepo, b'makestore', storewrapper)
     extensions.wrapfunction(exchange, b'pull', exchangepull)
     extensions.wrapfunction(merge, b'applyupdates', applyupdates)
     extensions.wrapfunction(merge, b'_checkunknownfiles', checkunknownfiles)
     extensions.wrapfunction(context.workingctx, b'_checklookup', checklookup)
     extensions.wrapfunction(scmutil, b'_findrenames', findrenames)
     extensions.wrapfunction(
         copies, b'_computeforwardmissing', computeforwardmissing
     )
     extensions.wrapfunction(dispatch, b'runcommand', runcommand)
     extensions.wrapfunction(repair, b'_collectbrokencsets', _collectbrokencsets)
     extensions.wrapfunction(context.changectx, b'filectx', filectx)
     extensions.wrapfunction(context.workingctx, b'filectx', workingfilectx)
     extensions.wrapfunction(patch, b'trydiff', trydiff)
     extensions.wrapfunction(hg, b'verify', _verify)
     scmutil.fileprefetchhooks.add(b'remotefilelog', _fileprefetchhook)
 
     # disappointing hacks below
     extensions.wrapfunction(scmutil, b'getrenamedfn', getrenamedfn)
     extensions.wrapfunction(revset, b'filelog', filelogrevset)
     revset.symbols[b'filelog'] = revset.filelog
     extensions.wrapfunction(cmdutil, b'walkfilerevs', walkfilerevs)
 
 
 def cloneshallow(orig, ui, repo, *args, **opts):
     if opts.get('shallow'):
         repos = []
 
         def pull_shallow(orig, self, *args, **kwargs):
             if not isenabled(self):
                 repos.append(self.unfiltered())
                 # set up the client hooks so the post-clone update works
                 setupclient(self.ui, self.unfiltered())
 
                 # setupclient fixed the class on the repo itself
                 # but we also need to fix it on the repoview
                 if isinstance(self, repoview.repoview):
                     self.__class__.__bases__ = (
                         self.__class__.__bases__[0],
                         self.unfiltered().__class__,
                     )
                 self.requirements.add(constants.SHALLOWREPO_REQUIREMENT)
                 self._writerequirements()
 
                 # Since setupclient hadn't been called, exchange.pull was not
                 # wrapped. So we need to manually invoke our version of it.
                 return exchangepull(orig, self, *args, **kwargs)
             else:
                 return orig(self, *args, **kwargs)
 
         extensions.wrapfunction(exchange, b'pull', pull_shallow)
 
         # Wrap the stream logic to add requirements and to pass include/exclude
         # patterns around.
         def setup_streamout(repo, remote):
             # Replace remote.stream_out with a version that sends file
             # patterns.
             def stream_out_shallow(orig):
                 caps = remote.capabilities()
                 if constants.NETWORK_CAP_LEGACY_SSH_GETFILES in caps:
                     opts = {}
                     if repo.includepattern:
                         opts['includepattern'] = b'\0'.join(repo.includepattern)
                     if repo.excludepattern:
                         opts['excludepattern'] = b'\0'.join(repo.excludepattern)
                     return remote._callstream(b'stream_out_shallow', **opts)
                 else:
                     return orig()
 
             extensions.wrapfunction(remote, b'stream_out', stream_out_shallow)
 
         def stream_wrap(orig, op):
             setup_streamout(op.repo, op.remote)
             return orig(op)
 
         extensions.wrapfunction(
             streamclone, b'maybeperformlegacystreamclone', stream_wrap
         )
 
         def canperformstreamclone(orig, pullop, bundle2=False):
             # remotefilelog is currently incompatible with the
             # bundle2 flavor of streamclones, so force us to use
             # v1 instead.
             if b'v2' in pullop.remotebundle2caps.get(b'stream', []):
                 pullop.remotebundle2caps[b'stream'] = [
                     c for c in pullop.remotebundle2caps[b'stream'] if c != b'v2'
                 ]
             if bundle2:
                 return False, None
             supported, requirements = orig(pullop, bundle2=bundle2)
             if requirements is not None:
                 requirements.add(constants.SHALLOWREPO_REQUIREMENT)
             return supported, requirements
 
         extensions.wrapfunction(
             streamclone, b'canperformstreamclone', canperformstreamclone
         )
 
     try:
         orig(ui, repo, *args, **opts)
     finally:
         if opts.get('shallow'):
             for r in repos:
                 if util.safehasattr(r, b'fileservice'):
                     r.fileservice.close()
 
 
 def debugdatashallow(orig, *args, **kwds):
     oldlen = remotefilelog.remotefilelog.__len__
     try:
         remotefilelog.remotefilelog.__len__ = lambda x: 1
         return orig(*args, **kwds)
     finally:
         remotefilelog.remotefilelog.__len__ = oldlen
 
 
 def reposetup(ui, repo):
     if not repo.local():
         return
 
     # put here intentionally bc doesnt work in uisetup
     ui.setconfig(b'hooks', b'update.prefetch', wcpprefetch)
     ui.setconfig(b'hooks', b'commit.prefetch', wcpprefetch)
 
     isserverenabled = ui.configbool(b'remotefilelog', b'server')
     isshallowclient = isenabled(repo)
 
     if isserverenabled and isshallowclient:
         raise RuntimeError(b"Cannot be both a server and shallow client.")
 
     if isshallowclient:
         setupclient(ui, repo)
 
     if isserverenabled:
         remotefilelogserver.setupserver(ui, repo)
 
 
 def setupclient(ui, repo):
     if not isinstance(repo, localrepo.localrepository):
         return
 
     # Even clients get the server setup since they need to have the
     # wireprotocol endpoints registered.
     remotefilelogserver.onetimesetup(ui)
     onetimeclientsetup(ui)
 
     shallowrepo.wraprepo(repo)
     repo.store = shallowstore.wrapstore(repo.store)
 
 
 def storewrapper(orig, requirements, path, vfstype):
     s = orig(requirements, path, vfstype)
     if constants.SHALLOWREPO_REQUIREMENT in requirements:
         s = shallowstore.wrapstore(s)
 
     return s
 
 
 # prefetch files before update
 def applyupdates(
     orig, repo, actions, wctx, mctx, overwrite, wantfiledata, labels=None
 ):
     if isenabled(repo):
         manifest = mctx.manifest()
         files = []
         for f, args, msg in actions[b'g']:
             files.append((f, hex(manifest[f])))
         # batch fetch the needed files from the server
         repo.fileservice.prefetch(files)
     return orig(
         repo, actions, wctx, mctx, overwrite, wantfiledata, labels=labels
     )
 
 
 # Prefetch merge checkunknownfiles
 def checkunknownfiles(orig, repo, wctx, mctx, force, actions, *args, **kwargs):
     if isenabled(repo):
         files = []
         sparsematch = repo.maybesparsematch(mctx.rev())
         for f, (m, actionargs, msg) in pycompat.iteritems(actions):
             if sparsematch and not sparsematch(f):
                 continue
             if m in (b'c', b'dc', b'cm'):
                 files.append((f, hex(mctx.filenode(f))))
             elif m == b'dg':
                 f2 = actionargs[0]
                 files.append((f2, hex(mctx.filenode(f2))))
         # batch fetch the needed files from the server
         repo.fileservice.prefetch(files)
     return orig(repo, wctx, mctx, force, actions, *args, **kwargs)
 
 
 # Prefetch files before status attempts to look at their size and contents
 def checklookup(orig, self, files):
     repo = self._repo
     if isenabled(repo):
         prefetchfiles = []
         for parent in self._parents:
             for f in files:
                 if f in parent:
                     prefetchfiles.append((f, hex(parent.filenode(f))))
         # batch fetch the needed files from the server
         repo.fileservice.prefetch(prefetchfiles)
     return orig(self, files)
 
 
 # Prefetch the logic that compares added and removed files for renames
 def findrenames(orig, repo, matcher, added, removed, *args, **kwargs):
     if isenabled(repo):
         files = []
         pmf = repo[b'.'].manifest()
         for f in removed:
             if f in pmf:
                 files.append((f, hex(pmf[f])))
         # batch fetch the needed files from the server
         repo.fileservice.prefetch(files)
     return orig(repo, matcher, added, removed, *args, **kwargs)
 
 
 # prefetch files before pathcopies check
 def computeforwardmissing(orig, a, b, match=None):
     missing = orig(a, b, match=match)
     repo = a._repo
     if isenabled(repo):
         mb = b.manifest()
 
         files = []
         sparsematch = repo.maybesparsematch(b.rev())
         if sparsematch:
             sparsemissing = set()
             for f in missing:
                 if sparsematch(f):
                     files.append((f, hex(mb[f])))
                     sparsemissing.add(f)
             missing = sparsemissing
 
         # batch fetch the needed files from the server
         repo.fileservice.prefetch(files)
     return missing
 
 
 # close cache miss server connection after the command has finished
 def runcommand(orig, lui, repo, *args, **kwargs):
     fileservice = None
     # repo can be None when running in chg:
     # - at startup, reposetup was called because serve is not norepo
     # - a norepo command like "help" is called
     if repo and isenabled(repo):
         fileservice = repo.fileservice
     try:
         return orig(lui, repo, *args, **kwargs)
     finally:
         if fileservice:
             fileservice.close()
 
 
 # prevent strip from stripping remotefilelogs
 def _collectbrokencsets(orig, repo, files, striprev):
     if isenabled(repo):
         files = list([f for f in files if not repo.shallowmatch(f)])
     return orig(repo, files, striprev)
 
 
 # changectx wrappers
 def filectx(orig, self, path, fileid=None, filelog=None):
     if fileid is None:
         fileid = self.filenode(path)
     if isenabled(self._repo) and self._repo.shallowmatch(path):
         return remotefilectx.remotefilectx(
             self._repo, path, fileid=fileid, changectx=self, filelog=filelog
         )
     return orig(self, path, fileid=fileid, filelog=filelog)
 
 
 def workingfilectx(orig, self, path, filelog=None):
     if isenabled(self._repo) and self._repo.shallowmatch(path):
         return remotefilectx.remoteworkingfilectx(
             self._repo, path, workingctx=self, filelog=filelog
         )
     return orig(self, path, filelog=filelog)
 
 
 # prefetch required revisions before a diff
 def trydiff(
     orig,
     repo,
     revs,
     ctx1,
     ctx2,
     modified,
     added,
     removed,
     copy,
     getfilectx,
     *args,
     **kwargs
 ):
     if isenabled(repo):
         prefetch = []
         mf1 = ctx1.manifest()
         for fname in modified + added + removed:
             if fname in mf1:
                 fnode = getfilectx(fname, ctx1).filenode()
                 # fnode can be None if it's a edited working ctx file
                 if fnode:
                     prefetch.append((fname, hex(fnode)))
             if fname not in removed:
                 fnode = getfilectx(fname, ctx2).filenode()
                 if fnode:
                     prefetch.append((fname, hex(fnode)))
 
         repo.fileservice.prefetch(prefetch)
 
     return orig(
         repo,
         revs,
         ctx1,
         ctx2,
         modified,
         added,
         removed,
         copy,
         getfilectx,
         *args,
         **kwargs
     )
 
 
 # Prevent verify from processing files
 # a stub for mercurial.hg.verify()
 def _verify(orig, repo, level=None):
     lock = repo.lock()
     try:
         return shallowverifier.shallowverifier(repo).verify()
     finally:
         lock.release()
 
 
 clientonetime = False
 
 
 def onetimeclientsetup(ui):
     global clientonetime
     if clientonetime:
         return
     clientonetime = True
 
     # Don't commit filelogs until we know the commit hash, since the hash
     # is present in the filelog blob.
     # This violates Mercurial's filelog->manifest->changelog write order,
     # but is generally fine for client repos.
     pendingfilecommits = []
 
     def addrawrevision(
         orig,
         self,
         rawtext,
         transaction,
         link,
         p1,
         p2,
         node,
         flags,
         cachedelta=None,
         _metatuple=None,
     ):
         if isinstance(link, int):
             pendingfilecommits.append(
                 (
                     self,
                     rawtext,
                     transaction,
                     link,
                     p1,
                     p2,
                     node,
                     flags,
                     cachedelta,
                     _metatuple,
                 )
             )
             return node
         else:
             return orig(
                 self,
                 rawtext,
                 transaction,
                 link,
                 p1,
                 p2,
                 node,
                 flags,
                 cachedelta,
                 _metatuple=_metatuple,
             )
 
     extensions.wrapfunction(
         remotefilelog.remotefilelog, b'addrawrevision', addrawrevision
     )
 
     def changelogadd(orig, self, *args, **kwargs):
         oldlen = len(self)
         node = orig(self, *args, **kwargs)
         newlen = len(self)
         if oldlen != newlen:
             for oldargs in pendingfilecommits:
                 log, rt, tr, link, p1, p2, n, fl, c, m = oldargs
                 linknode = self.node(link)
                 if linknode == node:
                     log.addrawrevision(rt, tr, linknode, p1, p2, n, fl, c, m)
                 else:
                     raise error.ProgrammingError(
                         b'pending multiple integer revisions are not supported'
                     )
         else:
             # "link" is actually wrong here (it is set to len(changelog))
             # if changelog remains unchanged, skip writing file revisions
             # but still do a sanity check about pending multiple revisions
             if len({x[3] for x in pendingfilecommits}) > 1:
                 raise error.ProgrammingError(
                     b'pending multiple integer revisions are not supported'
                 )
         del pendingfilecommits[:]
         return node
 
     extensions.wrapfunction(changelog.changelog, b'add', changelogadd)
 
 
 def getrenamedfn(orig, repo, endrev=None):
     if not isenabled(repo) or copies.usechangesetcentricalgo(repo):
         return orig(repo, endrev)
 
     rcache = {}
 
     def getrenamed(fn, rev):
         '''looks up all renames for a file (up to endrev) the first
         time the file is given. It indexes on the changerev and only
         parses the manifest if linkrev != changerev.
         Returns rename info for fn at changerev rev.'''
         if rev in rcache.setdefault(fn, {}):
             return rcache[fn][rev]
 
         try:
             fctx = repo[rev].filectx(fn)
             for ancestor in fctx.ancestors():
                 if ancestor.path() == fn:
                     renamed = ancestor.renamed()
                     rcache[fn][ancestor.rev()] = renamed and renamed[0]
 
             renamed = fctx.renamed()
             return renamed and renamed[0]
         except error.LookupError:
             return None
 
     return getrenamed
 
 
 def walkfilerevs(orig, repo, match, follow, revs, fncache):
     if not isenabled(repo):
         return orig(repo, match, follow, revs, fncache)
 
     # remotefilelog's can't be walked in rev order, so throw.
     # The caller will see the exception and walk the commit tree instead.
     if not follow:
         raise cmdutil.FileWalkError(b"Cannot walk via filelog")
 
     wanted = set()
     minrev, maxrev = min(revs), max(revs)
 
     pctx = repo[b'.']
     for filename in match.files():
         if filename not in pctx:
             raise error.Abort(
                 _(b'cannot follow file not in parent revision: "%s"') % filename
             )
         fctx = pctx[filename]
 
         linkrev = fctx.linkrev()
         if linkrev >= minrev and linkrev <= maxrev:
             fncache.setdefault(linkrev, []).append(filename)
             wanted.add(linkrev)
 
         for ancestor in fctx.ancestors():
             linkrev = ancestor.linkrev()
             if linkrev >= minrev and linkrev <= maxrev:
                 fncache.setdefault(linkrev, []).append(ancestor.path())
                 wanted.add(linkrev)
 
     return wanted
 
 
 def filelogrevset(orig, repo, subset, x):
     """``filelog(pattern)``
     Changesets connected to the specified filelog.
 
     For performance reasons, ``filelog()`` does not show every changeset
     that affects the requested file(s). See :hg:`help log` for details. For
     a slower, more accurate result, use ``file()``.
     """
 
     if not isenabled(repo):
         return orig(repo, subset, x)
 
     # i18n: "filelog" is a keyword
     pat = revset.getstring(x, _(b"filelog requires a pattern"))
     m = match.match(
         repo.root, repo.getcwd(), [pat], default=b'relpath', ctx=repo[None]
     )
     s = set()
 
     if not match.patkind(pat):
         # slow
         for r in subset:
             ctx = repo[r]
             cfiles = ctx.files()
             for f in m.files():
                 if f in cfiles:
                     s.add(ctx.rev())
                     break
     else:
         # partial
         files = (f for f in repo[None] if m(f))
         for f in files:
             fctx = repo[None].filectx(f)
             s.add(fctx.linkrev())
             for actx in fctx.ancestors():
                 s.add(actx.linkrev())
 
     return smartset.baseset([r for r in subset if r in s])
 
 
 @command(b'gc', [], _(b'hg gc [REPO...]'), norepo=True)
 def gc(ui, *args, **opts):
     '''garbage collect the client and server filelog caches
     '''
     cachepaths = set()
 
     # get the system client cache
     systemcache = shallowutil.getcachepath(ui, allowempty=True)
     if systemcache:
         cachepaths.add(systemcache)
 
     # get repo client and server cache
     repopaths = []
     pwd = ui.environ.get(b'PWD')
     if pwd:
         repopaths.append(pwd)
 
     repopaths.extend(args)
     repos = []
     for repopath in repopaths:
         try:
             repo = hg.peer(ui, {}, repopath)
             repos.append(repo)
 
             repocache = shallowutil.getcachepath(repo.ui, allowempty=True)
             if repocache:
                 cachepaths.add(repocache)
         except error.RepoError:
             pass
 
     # gc client cache
     for cachepath in cachepaths:
         gcclient(ui, cachepath)
 
     # gc server cache
     for repo in repos:
         remotefilelogserver.gcserver(ui, repo._repo)
 
 
 def gcclient(ui, cachepath):
     # get list of repos that use this cache
     repospath = os.path.join(cachepath, b'repos')
     if not os.path.exists(repospath):
         ui.warn(_(b"no known cache at %s\n") % cachepath)
         return
 
     reposfile = open(repospath, b'rb')
     repos = {r[:-1] for r in reposfile.readlines()}
     reposfile.close()
 
     # build list of useful files
     validrepos = []
     keepkeys = set()
 
     sharedcache = None
     filesrepacked = False
 
     count = 0
     progress = ui.makeprogress(
         _(b"analyzing repositories"), unit=b"repos", total=len(repos)
     )
     for path in repos:
         progress.update(count)
         count += 1
         try:
             path = ui.expandpath(os.path.normpath(path))
         except TypeError as e:
             ui.warn(_(b"warning: malformed path: %r:%s\n") % (path, e))
             traceback.print_exc()
             continue
         try:
             peer = hg.peer(ui, {}, path)
             repo = peer._repo
         except error.RepoError:
             continue
 
         validrepos.append(path)
 
         # Protect against any repo or config changes that have happened since
         # this repo was added to the repos file. We'd rather this loop succeed
         # and too much be deleted, than the loop fail and nothing gets deleted.
         if not isenabled(repo):
             continue
 
         if not util.safehasattr(repo, b'name'):
             ui.warn(
                 _(b"repo %s is a misconfigured remotefilelog repo\n") % path
             )
             continue
 
         # If garbage collection on repack and repack on hg gc are enabled
         # then loose files are repacked and garbage collected.
         # Otherwise regular garbage collection is performed.
         repackonhggc = repo.ui.configbool(b'remotefilelog', b'repackonhggc')
         gcrepack = repo.ui.configbool(b'remotefilelog', b'gcrepack')
         if repackonhggc and gcrepack:
             try:
                 repackmod.incrementalrepack(repo)
                 filesrepacked = True
                 continue
             except (IOError, repackmod.RepackAlreadyRunning):
                 # If repack cannot be performed due to not enough disk space
                 # continue doing garbage collection of loose files w/o repack
                 pass
 
         reponame = repo.name
         if not sharedcache:
             sharedcache = repo.sharedstore
 
         # Compute a keepset which is not garbage collected
         def keyfn(fname, fnode):
             return fileserverclient.getcachekey(reponame, fname, hex(fnode))
 
         keepkeys = repackmod.keepset(repo, keyfn=keyfn, lastkeepkeys=keepkeys)
 
     progress.complete()
 
     # write list of valid repos back
     oldumask = os.umask(0o002)
     try:
         reposfile = open(repospath, b'wb')
         reposfile.writelines([(b"%s\n" % r) for r in validrepos])
         reposfile.close()
     finally:
         os.umask(oldumask)
 
     # prune cache
     if sharedcache is not None:
         sharedcache.gc(keepkeys)
     elif not filesrepacked:
         ui.warn(_(b"warning: no valid repos in repofile\n"))
 
 
 def log(orig, ui, repo, *pats, **opts):
     if not isenabled(repo):
         return orig(ui, repo, *pats, **opts)
 
     follow = opts.get('follow')
     revs = opts.get('rev')
     if pats:
         # Force slowpath for non-follow patterns and follows that start from
         # non-working-copy-parent revs.
         if not follow or revs:
             # This forces the slowpath
             opts['removed'] = True
 
         # If this is a non-follow log without any revs specified, recommend that
         # the user add -f to speed it up.
         if not follow and not revs:
             match = scmutil.match(repo[b'.'], pats, pycompat.byteskwargs(opts))
             isfile = not match.anypats()
             if isfile:
                 for file in match.files():
                     if not os.path.isfile(repo.wjoin(file)):
                         isfile = False
                         break
 
             if isfile:
                 ui.warn(
                     _(
                         b"warning: file log can be slow on large repos - "
                         + b"use -f to speed it up\n"
                     )
                 )
 
     return orig(ui, repo, *pats, **opts)
 
 
 def revdatelimit(ui, revset):
     """Update revset so that only changesets no older than 'prefetchdays' days
     are included. The default value is set to 14 days. If 'prefetchdays' is set
     to zero or negative value then date restriction is not applied.
     """
     days = ui.configint(b'remotefilelog', b'prefetchdays')
     if days > 0:
         revset = b'(%s) & date(-%s)' % (revset, days)
     return revset
 
 
 def readytofetch(repo):
     """Check that enough time has passed since the last background prefetch.
     This only relates to prefetches after operations that change the working
     copy parent. Default delay between background prefetches is 2 minutes.
     """
     timeout = repo.ui.configint(b'remotefilelog', b'prefetchdelay')
     fname = repo.vfs.join(b'lastprefetch')
 
     ready = False
     with open(fname, b'a'):
         # the with construct above is used to avoid race conditions
         modtime = os.path.getmtime(fname)
         if (time.time() - modtime) > timeout:
             os.utime(fname, None)
             ready = True
 
     return ready
 
 
 def wcpprefetch(ui, repo, **kwargs):
     """Prefetches in background revisions specified by bgprefetchrevs revset.
     Does background repack if backgroundrepack flag is set in config.
     """
     shallow = isenabled(repo)
     bgprefetchrevs = ui.config(b'remotefilelog', b'bgprefetchrevs')
     isready = readytofetch(repo)
 
     if not (shallow and bgprefetchrevs and isready):
         return
 
     bgrepack = repo.ui.configbool(b'remotefilelog', b'backgroundrepack')
     # update a revset with a date limit
     bgprefetchrevs = revdatelimit(ui, bgprefetchrevs)
 
     def anon(unused_success):
         if util.safehasattr(repo, b'ranprefetch') and repo.ranprefetch:
             return
         repo.ranprefetch = True
         repo.backgroundprefetch(bgprefetchrevs, repack=bgrepack)
 
     repo._afterlock(anon)
 
 
 def pull(orig, ui, repo, *pats, **opts):
     result = orig(ui, repo, *pats, **opts)
 
     if isenabled(repo):
         # prefetch if it's configured
         prefetchrevset = ui.config(b'remotefilelog', b'pullprefetch')
         bgrepack = repo.ui.configbool(b'remotefilelog', b'backgroundrepack')
         bgprefetch = repo.ui.configbool(b'remotefilelog', b'backgroundprefetch')
 
         if prefetchrevset:
             ui.status(_(b"prefetching file contents\n"))
             revs = scmutil.revrange(repo, [prefetchrevset])
             base = repo[b'.'].rev()
             if bgprefetch:
                 repo.backgroundprefetch(prefetchrevset, repack=bgrepack)
             else:
                 repo.prefetch(revs, base=base)
                 if bgrepack:
                     repackmod.backgroundrepack(repo, incremental=True)
         elif bgrepack:
             repackmod.backgroundrepack(repo, incremental=True)
 
     return result
 
 
 def exchangepull(orig, repo, remote, *args, **kwargs):
     # Hook into the callstream/getbundle to insert bundle capabilities
     # during a pull.
     def localgetbundle(
         orig, source, heads=None, common=None, bundlecaps=None, **kwargs
     ):
         if not bundlecaps:
             bundlecaps = set()
         bundlecaps.add(constants.BUNDLE2_CAPABLITY)
         return orig(
             source, heads=heads, common=common, bundlecaps=bundlecaps, **kwargs
         )
 
     if util.safehasattr(remote, b'_callstream'):
         remote._localrepo = repo
     elif util.safehasattr(remote, b'getbundle'):
         extensions.wrapfunction(remote, b'getbundle', localgetbundle)
 
     return orig(repo, remote, *args, **kwargs)
 
 
-def _fileprefetchhook(repo, revs, match):
+def _fileprefetchhook(repo, revmatches):
     if isenabled(repo):
         allfiles = []
-        for rev in revs:
+        for rev, match in revmatches:
             if rev == nodemod.wdirrev or rev is None:
                 continue
             ctx = repo[rev]
             mf = ctx.manifest()
             sparsematch = repo.maybesparsematch(ctx.rev())
             for path in ctx.walk(match):
                 if (not sparsematch or sparsematch(path)) and path in mf:
                     allfiles.append((path, hex(mf[path])))
         repo.fileservice.prefetch(allfiles)
 
 
 @command(
     b'debugremotefilelog',
     [(b'd', b'decompress', None, _(b'decompress the filelog first')),],
     _(b'hg debugremotefilelog <path>'),
     norepo=True,
 )
 def debugremotefilelog(ui, path, **opts):
     return debugcommands.debugremotefilelog(ui, path, **opts)
 
 
 @command(
     b'verifyremotefilelog',
     [(b'd', b'decompress', None, _(b'decompress the filelogs first')),],
     _(b'hg verifyremotefilelogs <directory>'),
     norepo=True,
 )
 def verifyremotefilelog(ui, path, **opts):
     return debugcommands.verifyremotefilelog(ui, path, **opts)
 
 
 @command(
     b'debugdatapack',
     [
         (b'', b'long', None, _(b'print the long hashes')),
         (b'', b'node', b'', _(b'dump the contents of node'), b'NODE'),
     ],
     _(b'hg debugdatapack <paths>'),
     norepo=True,
 )
 def debugdatapack(ui, *paths, **opts):
     return debugcommands.debugdatapack(ui, *paths, **opts)
 
 
 @command(b'debughistorypack', [], _(b'hg debughistorypack <path>'), norepo=True)
 def debughistorypack(ui, path, **opts):
     return debugcommands.debughistorypack(ui, path)
 
 
 @command(b'debugkeepset', [], _(b'hg debugkeepset'))
 def debugkeepset(ui, repo, **opts):
     # The command is used to measure keepset computation time
     def keyfn(fname, fnode):
         return fileserverclient.getcachekey(repo.name, fname, hex(fnode))
 
     repackmod.keepset(repo, keyfn)
     return
 
 
 @command(b'debugwaitonrepack', [], _(b'hg debugwaitonrepack'))
 def debugwaitonrepack(ui, repo, **opts):
     return debugcommands.debugwaitonrepack(repo)
 
 
 @command(b'debugwaitonprefetch', [], _(b'hg debugwaitonprefetch'))
 def debugwaitonprefetch(ui, repo, **opts):
     return debugcommands.debugwaitonprefetch(repo)
 
 
 def resolveprefetchopts(ui, opts):
     if not opts.get(b'rev'):
         revset = [b'.', b'draft()']
 
         prefetchrevset = ui.config(b'remotefilelog', b'pullprefetch', None)
         if prefetchrevset:
             revset.append(b'(%s)' % prefetchrevset)
         bgprefetchrevs = ui.config(b'remotefilelog', b'bgprefetchrevs', None)
         if bgprefetchrevs:
             revset.append(b'(%s)' % bgprefetchrevs)
         revset = b'+'.join(revset)
 
         # update a revset with a date limit
         revset = revdatelimit(ui, revset)
 
         opts[b'rev'] = [revset]
 
     if not opts.get(b'base'):
         opts[b'base'] = None
 
     return opts
 
 
 @command(
     b'prefetch',
     [
         (b'r', b'rev', [], _(b'prefetch the specified revisions'), _(b'REV')),
         (b'', b'repack', False, _(b'run repack after prefetch')),
         (b'b', b'base', b'', _(b"rev that is assumed to already be local")),
     ]
     + commands.walkopts,
     _(b'hg prefetch [OPTIONS] [FILE...]'),
     helpcategory=command.CATEGORY_MAINTENANCE,
 )
 def prefetch(ui, repo, *pats, **opts):
     """prefetch file revisions from the server
 
     Prefetchs file revisions for the specified revs and stores them in the
     local remotefilelog cache.  If no rev is specified, the default rev is
     used which is the union of dot, draft, pullprefetch and bgprefetchrev.
     File names or patterns can be used to limit which files are downloaded.
 
     Return 0 on success.
     """
     opts = pycompat.byteskwargs(opts)
     if not isenabled(repo):
         raise error.Abort(_(b"repo is not shallow"))
 
     opts = resolveprefetchopts(ui, opts)
     revs = scmutil.revrange(repo, opts.get(b'rev'))
     repo.prefetch(revs, opts.get(b'base'), pats, opts)
 
     # Run repack in background
     if opts.get(b'repack'):
         repackmod.backgroundrepack(repo, incremental=True)
 
 
 @command(
     b'repack',
     [
         (b'', b'background', None, _(b'run in a background process'), None),
         (b'', b'incremental', None, _(b'do an incremental repack'), None),
         (
             b'',
             b'packsonly',
             None,
             _(b'only repack packs (skip loose objects)'),
             None,
         ),
     ],
     _(b'hg repack [OPTIONS]'),
 )
 def repack_(ui, repo, *pats, **opts):
     if opts.get('background'):
         repackmod.backgroundrepack(
             repo,
             incremental=opts.get('incremental'),
             packsonly=opts.get('packsonly', False),
         )
         return
 
     options = {b'packsonly': opts.get('packsonly')}
 
     try:
         if opts.get('incremental'):
             repackmod.incrementalrepack(repo, options=options)
         else:
             repackmod.fullrepack(repo, options=options)
     except repackmod.RepackAlreadyRunning as ex:
         # Don't propogate the exception if the repack is already in
         # progress, since we want the command to exit 0.
         repo.ui.warn(b'%s\n' % ex)

mercurial/archival.py

@@ -1,390 +1,390 b''
 # archival.py - revision archival for mercurial
 #
 # Copyright 2006 Vadim Gelfer <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import gzip
 import os
 import struct
 import tarfile
 import time
 import zipfile
 import zlib
 
 from .i18n import _
 from .node import nullrev
 from .pycompat import open
 
 from . import (
     error,
     formatter,
     match as matchmod,
     pycompat,
     scmutil,
     util,
     vfs as vfsmod,
 )
 
 stringio = util.stringio
 
 # from unzip source code:
 _UNX_IFREG = 0x8000
 _UNX_IFLNK = 0xA000
 
 
 def tidyprefix(dest, kind, prefix):
     '''choose prefix to use for names in archive.  make sure prefix is
     safe for consumers.'''
 
     if prefix:
         prefix = util.normpath(prefix)
     else:
         if not isinstance(dest, bytes):
             raise ValueError(b'dest must be string if no prefix')
         prefix = os.path.basename(dest)
         lower = prefix.lower()
         for sfx in exts.get(kind, []):
             if lower.endswith(sfx):
                 prefix = prefix[: -len(sfx)]
                 break
     lpfx = os.path.normpath(util.localpath(prefix))
     prefix = util.pconvert(lpfx)
     if not prefix.endswith(b'/'):
         prefix += b'/'
     # Drop the leading '.' path component if present, so Windows can read the
     # zip files (issue4634)
     if prefix.startswith(b'./'):
         prefix = prefix[2:]
     if prefix.startswith(b'../') or os.path.isabs(lpfx) or b'/../' in prefix:
         raise error.Abort(_(b'archive prefix contains illegal components'))
     return prefix
 
 
 exts = {
     b'tar': [b'.tar'],
     b'tbz2': [b'.tbz2', b'.tar.bz2'],
     b'tgz': [b'.tgz', b'.tar.gz'],
     b'zip': [b'.zip'],
     b'txz': [b'.txz', b'.tar.xz'],
 }
 
 
 def guesskind(dest):
     for kind, extensions in pycompat.iteritems(exts):
         if any(dest.endswith(ext) for ext in extensions):
             return kind
     return None
 
 
 def _rootctx(repo):
     # repo[0] may be hidden
     for rev in repo:
         return repo[rev]
     return repo[nullrev]
 
 
 # {tags} on ctx includes local tags and 'tip', with no current way to limit
 # that to global tags.  Therefore, use {latesttag} as a substitute when
 # the distance is 0, since that will be the list of global tags on ctx.
 _defaultmetatemplate = br'''
 repo: {root}
 node: {ifcontains(rev, revset("wdir()"), "{p1node}{dirty}", "{node}")}
 branch: {branch|utf8}
 {ifeq(latesttagdistance, 0, join(latesttag % "tag: {tag}", "\n"),
       separate("\n",
                join(latesttag % "latesttag: {tag}", "\n"),
                "latesttagdistance: {latesttagdistance}",
                "changessincelatesttag: {changessincelatesttag}"))}
 '''[
     1:
 ]  # drop leading '\n'
 
 
 def buildmetadata(ctx):
     '''build content of .hg_archival.txt'''
     repo = ctx.repo()
 
     opts = {
         b'template': repo.ui.config(
             b'experimental', b'archivemetatemplate', _defaultmetatemplate
         )
     }
 
     out = util.stringio()
 
     fm = formatter.formatter(repo.ui, out, b'archive', opts)
     fm.startitem()
     fm.context(ctx=ctx)
     fm.data(root=_rootctx(repo).hex())
 
     if ctx.rev() is None:
         dirty = b''
         if ctx.dirty(missing=True):
             dirty = b'+'
         fm.data(dirty=dirty)
     fm.end()
 
     return out.getvalue()
 
 
 class tarit(object):
     '''write archive to tar file or stream.  can write uncompressed,
     or compress with gzip or bzip2.'''
 
     if pycompat.ispy3:
         GzipFileWithTime = gzip.GzipFile  # camelcase-required
     else:
 
         class GzipFileWithTime(gzip.GzipFile):
             def __init__(self, *args, **kw):
                 timestamp = None
                 if 'mtime' in kw:
                     timestamp = kw.pop('mtime')
                 if timestamp is None:
                     self.timestamp = time.time()
                 else:
                     self.timestamp = timestamp
                 gzip.GzipFile.__init__(self, *args, **kw)
 
             def _write_gzip_header(self):
                 self.fileobj.write(b'\037\213')  # magic header
                 self.fileobj.write(b'\010')  # compression method
                 fname = self.name
                 if fname and fname.endswith(b'.gz'):
                     fname = fname[:-3]
                 flags = 0
                 if fname:
                     flags = gzip.FNAME  # pytype: disable=module-attr
                 self.fileobj.write(pycompat.bytechr(flags))
                 gzip.write32u(  # pytype: disable=module-attr
                     self.fileobj, int(self.timestamp)
                 )
                 self.fileobj.write(b'\002')
                 self.fileobj.write(b'\377')
                 if fname:
                     self.fileobj.write(fname + b'\000')
 
     def __init__(self, dest, mtime, kind=b''):
         self.mtime = mtime
         self.fileobj = None
 
         def taropen(mode, name=b'', fileobj=None):
             if kind == b'gz':
                 mode = mode[0:1]
                 if not fileobj:
                     fileobj = open(name, mode + b'b')
                 gzfileobj = self.GzipFileWithTime(
                     name,
                     pycompat.sysstr(mode + b'b'),
                     zlib.Z_BEST_COMPRESSION,
                     fileobj,
                     mtime=mtime,
                 )
                 self.fileobj = gzfileobj
                 return tarfile.TarFile.taropen(  # pytype: disable=attribute-error
                     name, pycompat.sysstr(mode), gzfileobj
                 )
             else:
                 return tarfile.open(name, pycompat.sysstr(mode + kind), fileobj)
 
         if isinstance(dest, bytes):
             self.z = taropen(b'w:', name=dest)
         else:
             self.z = taropen(b'w|', fileobj=dest)
 
     def addfile(self, name, mode, islink, data):
         name = pycompat.fsdecode(name)
         i = tarfile.TarInfo(name)
         i.mtime = self.mtime
         i.size = len(data)
         if islink:
             i.type = tarfile.SYMTYPE
             i.mode = 0o777
             i.linkname = pycompat.fsdecode(data)
             data = None
             i.size = 0
         else:
             i.mode = mode
             data = stringio(data)
         self.z.addfile(i, data)
 
     def done(self):
         self.z.close()
         if self.fileobj:
             self.fileobj.close()
 
 
 class zipit(object):
     '''write archive to zip file or stream.  can write uncompressed,
     or compressed with deflate.'''
 
     def __init__(self, dest, mtime, compress=True):
         if isinstance(dest, bytes):
             dest = pycompat.fsdecode(dest)
         self.z = zipfile.ZipFile(
             dest, 'w', compress and zipfile.ZIP_DEFLATED or zipfile.ZIP_STORED
         )
 
         # Python's zipfile module emits deprecation warnings if we try
         # to store files with a date before 1980.
         epoch = 315532800  # calendar.timegm((1980, 1, 1, 0, 0, 0, 1, 1, 0))
         if mtime < epoch:
             mtime = epoch
 
         self.mtime = mtime
         self.date_time = time.gmtime(mtime)[:6]
 
     def addfile(self, name, mode, islink, data):
         i = zipfile.ZipInfo(pycompat.fsdecode(name), self.date_time)
         i.compress_type = self.z.compression  # pytype: disable=attribute-error
         # unzip will not honor unix file modes unless file creator is
         # set to unix (id 3).
         i.create_system = 3
         ftype = _UNX_IFREG
         if islink:
             mode = 0o777
             ftype = _UNX_IFLNK
         i.external_attr = (mode | ftype) << 16
         # add "extended-timestamp" extra block, because zip archives
         # without this will be extracted with unexpected timestamp,
         # if TZ is not configured as GMT
         i.extra += struct.pack(
             b'<hhBl',
             0x5455,  # block type: "extended-timestamp"
             1 + 4,  # size of this block
             1,  # "modification time is present"
             int(self.mtime),
         )  # last modification (UTC)
         self.z.writestr(i, data)
 
     def done(self):
         self.z.close()
 
 
 class fileit(object):
     '''write archive as files in directory.'''
 
     def __init__(self, name, mtime):
         self.basedir = name
         self.opener = vfsmod.vfs(self.basedir)
         self.mtime = mtime
 
     def addfile(self, name, mode, islink, data):
         if islink:
             self.opener.symlink(data, name)
             return
         f = self.opener(name, b"w", atomictemp=False)
         f.write(data)
         f.close()
         destfile = os.path.join(self.basedir, name)
         os.chmod(destfile, mode)
         if self.mtime is not None:
             os.utime(destfile, (self.mtime, self.mtime))
 
     def done(self):
         pass
 
 
 archivers = {
     b'files': fileit,
     b'tar': tarit,
     b'tbz2': lambda name, mtime: tarit(name, mtime, b'bz2'),
     b'tgz': lambda name, mtime: tarit(name, mtime, b'gz'),
     b'txz': lambda name, mtime: tarit(name, mtime, b'xz'),
     b'uzip': lambda name, mtime: zipit(name, mtime, False),
     b'zip': zipit,
 }
 
 
 def archive(
     repo,
     dest,
     node,
     kind,
     decode=True,
     match=None,
     prefix=b'',
     mtime=None,
     subrepos=False,
 ):
     '''create archive of repo as it was at node.
 
     dest can be name of directory, name of archive file, or file
     object to write archive to.
 
     kind is type of archive to create.
 
     decode tells whether to put files through decode filters from
     hgrc.
 
     match is a matcher to filter names of files to write to archive.
 
     prefix is name of path to put before every archive member.
 
     mtime is the modified time, in seconds, or None to use the changeset time.
 
     subrepos tells whether to include subrepos.
     '''
 
     if kind == b'txz' and not pycompat.ispy3:
         raise error.Abort(_(b'xz compression is only available in Python 3'))
 
     if kind == b'files':
         if prefix:
             raise error.Abort(_(b'cannot give prefix when archiving to files'))
     else:
         prefix = tidyprefix(dest, kind, prefix)
 
     def write(name, mode, islink, getdata):
         data = getdata()
         if decode:
             data = repo.wwritedata(name, data)
         archiver.addfile(prefix + name, mode, islink, data)
 
     if kind not in archivers:
         raise error.Abort(_(b"unknown archive type '%s'") % kind)
 
     ctx = repo[node]
     archiver = archivers[kind](dest, mtime or ctx.date()[0])
 
     if not match:
         match = scmutil.matchall(repo)
 
     if repo.ui.configbool(b"ui", b"archivemeta"):
         name = b'.hg_archival.txt'
         if match(name):
             write(name, 0o644, False, lambda: buildmetadata(ctx))
 
     files = list(ctx.manifest().walk(match))
     total = len(files)
     if total:
         files.sort()
         scmutil.prefetchfiles(
-            repo, [ctx.rev()], scmutil.matchfiles(repo, files)
+            repo, [(ctx.rev(), scmutil.matchfiles(repo, files))]
         )
         progress = repo.ui.makeprogress(
             _(b'archiving'), unit=_(b'files'), total=total
         )
         progress.update(0)
         for f in files:
             ff = ctx.flags(f)
             write(f, b'x' in ff and 0o755 or 0o644, b'l' in ff, ctx[f].data)
             progress.increment(item=f)
         progress.complete()
 
     if subrepos:
         for subpath in sorted(ctx.substate):
             sub = ctx.workingsub(subpath)
             submatch = matchmod.subdirmatcher(subpath, match)
             subprefix = prefix + subpath + b'/'
             total += sub.archive(archiver, subprefix, submatch, decode)
 
     if total == 0:
         raise error.Abort(_(b'no files match the archive pattern'))
 
     archiver.done()
     return total

mercurial/cmdutil.py

@@ -1,4213 +1,4215 b''
 # cmdutil.py - help for command processing in mercurial
 #
 # Copyright 2005-2007 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import copy as copymod
 import errno
 import os
 import re
 
 from .i18n import _
 from .node import (
     hex,
     nullid,
     nullrev,
     short,
 )
 from .pycompat import (
     getattr,
     open,
     setattr,
 )
 from .thirdparty import attr
 
 from . import (
     bookmarks,
     changelog,
     copies,
     crecord as crecordmod,
     dirstateguard,
     encoding,
     error,
     formatter,
     logcmdutil,
     match as matchmod,
     merge as mergemod,
     mergestate as mergestatemod,
     mergeutil,
     obsolete,
     patch,
     pathutil,
     phases,
     pycompat,
     repair,
     revlog,
     rewriteutil,
     scmutil,
     smartset,
     state as statemod,
     subrepoutil,
     templatekw,
     templater,
     util,
     vfs as vfsmod,
 )
 
 from .utils import (
     dateutil,
     stringutil,
 )
 
 if pycompat.TYPE_CHECKING:
     from typing import (
         Any,
         Dict,
     )
 
     for t in (Any, Dict):
         assert t
 
 stringio = util.stringio
 
 # templates of common command options
 
 dryrunopts = [
     (b'n', b'dry-run', None, _(b'do not perform actions, just print output')),
 ]
 
 confirmopts = [
     (b'', b'confirm', None, _(b'ask before applying actions')),
 ]
 
 remoteopts = [
     (b'e', b'ssh', b'', _(b'specify ssh command to use'), _(b'CMD')),
     (
         b'',
         b'remotecmd',
         b'',
         _(b'specify hg command to run on the remote side'),
         _(b'CMD'),
     ),
     (
         b'',
         b'insecure',
         None,
         _(b'do not verify server certificate (ignoring web.cacerts config)'),
     ),
 ]
 
 walkopts = [
     (
         b'I',
         b'include',
         [],
         _(b'include names matching the given patterns'),
         _(b'PATTERN'),
     ),
     (
         b'X',
         b'exclude',
         [],
         _(b'exclude names matching the given patterns'),
         _(b'PATTERN'),
     ),
 ]
 
 commitopts = [
     (b'm', b'message', b'', _(b'use text as commit message'), _(b'TEXT')),
     (b'l', b'logfile', b'', _(b'read commit message from file'), _(b'FILE')),
 ]
 
 commitopts2 = [
     (
         b'd',
         b'date',
         b'',
         _(b'record the specified date as commit date'),
         _(b'DATE'),
     ),
     (
         b'u',
         b'user',
         b'',
         _(b'record the specified user as committer'),
         _(b'USER'),
     ),
 ]
 
 commitopts3 = [
     (b'D', b'currentdate', None, _(b'record the current date as commit date')),
     (b'U', b'currentuser', None, _(b'record the current user as committer')),
 ]
 
 formatteropts = [
     (b'T', b'template', b'', _(b'display with template'), _(b'TEMPLATE')),
 ]
 
 templateopts = [
     (
         b'',
         b'style',
         b'',
         _(b'display using template map file (DEPRECATED)'),
         _(b'STYLE'),
     ),
     (b'T', b'template', b'', _(b'display with template'), _(b'TEMPLATE')),
 ]
 
 logopts = [
     (b'p', b'patch', None, _(b'show patch')),
     (b'g', b'git', None, _(b'use git extended diff format')),
     (b'l', b'limit', b'', _(b'limit number of changes displayed'), _(b'NUM')),
     (b'M', b'no-merges', None, _(b'do not show merges')),
     (b'', b'stat', None, _(b'output diffstat-style summary of changes')),
     (b'G', b'graph', None, _(b"show the revision DAG")),
 ] + templateopts
 
 diffopts = [
     (b'a', b'text', None, _(b'treat all files as text')),
     (
         b'g',
         b'git',
         None,
         _(b'use git extended diff format (DEFAULT: diff.git)'),
     ),
     (b'', b'binary', None, _(b'generate binary diffs in git mode (default)')),
     (b'', b'nodates', None, _(b'omit dates from diff headers')),
 ]
 
 diffwsopts = [
     (
         b'w',
         b'ignore-all-space',
         None,
         _(b'ignore white space when comparing lines'),
     ),
     (
         b'b',
         b'ignore-space-change',
         None,
         _(b'ignore changes in the amount of white space'),
     ),
     (
         b'B',
         b'ignore-blank-lines',
         None,
         _(b'ignore changes whose lines are all blank'),
     ),
     (
         b'Z',
         b'ignore-space-at-eol',
         None,
         _(b'ignore changes in whitespace at EOL'),
     ),
 ]
 
 diffopts2 = (
     [
         (b'', b'noprefix', None, _(b'omit a/ and b/ prefixes from filenames')),
         (
             b'p',
             b'show-function',
             None,
             _(
                 b'show which function each change is in (DEFAULT: diff.showfunc)'
             ),
         ),
         (b'', b'reverse', None, _(b'produce a diff that undoes the changes')),
     ]
     + diffwsopts
     + [
         (
             b'U',
             b'unified',
             b'',
             _(b'number of lines of context to show'),
             _(b'NUM'),
         ),
         (b'', b'stat', None, _(b'output diffstat-style summary of changes')),
         (
             b'',
             b'root',
             b'',
             _(b'produce diffs relative to subdirectory'),
             _(b'DIR'),
         ),
     ]
 )
 
 mergetoolopts = [
     (b't', b'tool', b'', _(b'specify merge tool'), _(b'TOOL')),
 ]
 
 similarityopts = [
     (
         b's',
         b'similarity',
         b'',
         _(b'guess renamed files by similarity (0<=s<=100)'),
         _(b'SIMILARITY'),
     )
 ]
 
 subrepoopts = [(b'S', b'subrepos', None, _(b'recurse into subrepositories'))]
 
 debugrevlogopts = [
     (b'c', b'changelog', False, _(b'open changelog')),
     (b'm', b'manifest', False, _(b'open manifest')),
     (b'', b'dir', b'', _(b'open directory manifest')),
 ]
 
 # special string such that everything below this line will be ingored in the
 # editor text
 _linebelow = b"^HG: ------------------------ >8 ------------------------$"
 
 
 def check_at_most_one_arg(opts, *args):
     """abort if more than one of the arguments are in opts
 
     Returns the unique argument or None if none of them were specified.
     """
 
     def to_display(name):
         return pycompat.sysbytes(name).replace(b'_', b'-')
 
     previous = None
     for x in args:
         if opts.get(x):
             if previous:
                 raise error.Abort(
                     _(b'cannot specify both --%s and --%s')
                     % (to_display(previous), to_display(x))
                 )
             previous = x
     return previous
 
 
 def check_incompatible_arguments(opts, first, others):
     """abort if the first argument is given along with any of the others
 
     Unlike check_at_most_one_arg(), `others` are not mutually exclusive
     among themselves, and they're passed as a single collection.
     """
     for other in others:
         check_at_most_one_arg(opts, first, other)
 
 
 def resolvecommitoptions(ui, opts):
     """modify commit options dict to handle related options
 
     The return value indicates that ``rewrite.update-timestamp`` is the reason
     the ``date`` option is set.
     """
     check_at_most_one_arg(opts, b'date', b'currentdate')
     check_at_most_one_arg(opts, b'user', b'currentuser')
 
     datemaydiffer = False  # date-only change should be ignored?
 
     if opts.get(b'currentdate'):
         opts[b'date'] = b'%d %d' % dateutil.makedate()
     elif (
         not opts.get(b'date')
         and ui.configbool(b'rewrite', b'update-timestamp')
         and opts.get(b'currentdate') is None
     ):
         opts[b'date'] = b'%d %d' % dateutil.makedate()
         datemaydiffer = True
 
     if opts.get(b'currentuser'):
         opts[b'user'] = ui.username()
 
     return datemaydiffer
 
 
 def checknotesize(ui, opts):
     """ make sure note is of valid format """
 
     note = opts.get(b'note')
     if not note:
         return
 
     if len(note) > 255:
         raise error.Abort(_(b"cannot store a note of more than 255 bytes"))
     if b'\n' in note:
         raise error.Abort(_(b"note cannot contain a newline"))
 
 
 def ishunk(x):
     hunkclasses = (crecordmod.uihunk, patch.recordhunk)
     return isinstance(x, hunkclasses)
 
 
 def newandmodified(chunks, originalchunks):
     newlyaddedandmodifiedfiles = set()
     alsorestore = set()
     for chunk in chunks:
         if (
             ishunk(chunk)
             and chunk.header.isnewfile()
             and chunk not in originalchunks
         ):
             newlyaddedandmodifiedfiles.add(chunk.header.filename())
             alsorestore.update(
                 set(chunk.header.files()) - {chunk.header.filename()}
             )
     return newlyaddedandmodifiedfiles, alsorestore
 
 
 def parsealiases(cmd):
     return cmd.split(b"|")
 
 
 def setupwrapcolorwrite(ui):
     # wrap ui.write so diff output can be labeled/colorized
     def wrapwrite(orig, *args, **kw):
         label = kw.pop('label', b'')
         for chunk, l in patch.difflabel(lambda: args):
             orig(chunk, label=label + l)
 
     oldwrite = ui.write
 
     def wrap(*args, **kwargs):
         return wrapwrite(oldwrite, *args, **kwargs)
 
     setattr(ui, 'write', wrap)
     return oldwrite
 
 
 def filterchunks(ui, originalhunks, usecurses, testfile, match, operation=None):
     try:
         if usecurses:
             if testfile:
                 recordfn = crecordmod.testdecorator(
                     testfile, crecordmod.testchunkselector
                 )
             else:
                 recordfn = crecordmod.chunkselector
 
             return crecordmod.filterpatch(
                 ui, originalhunks, recordfn, operation
             )
     except crecordmod.fallbackerror as e:
         ui.warn(b'%s\n' % e)
         ui.warn(_(b'falling back to text mode\n'))
 
     return patch.filterpatch(ui, originalhunks, match, operation)
 
 
 def recordfilter(ui, originalhunks, match, operation=None):
     """ Prompts the user to filter the originalhunks and return a list of
     selected hunks.
     *operation* is used for to build ui messages to indicate the user what
     kind of filtering they are doing: reverting, committing, shelving, etc.
     (see patch.filterpatch).
     """
     usecurses = crecordmod.checkcurses(ui)
     testfile = ui.config(b'experimental', b'crecordtest')
     oldwrite = setupwrapcolorwrite(ui)
     try:
         newchunks, newopts = filterchunks(
             ui, originalhunks, usecurses, testfile, match, operation
         )
     finally:
         ui.write = oldwrite
     return newchunks, newopts
 
 
 def dorecord(
     ui, repo, commitfunc, cmdsuggest, backupall, filterfn, *pats, **opts
 ):
     opts = pycompat.byteskwargs(opts)
     if not ui.interactive():
         if cmdsuggest:
             msg = _(b'running non-interactively, use %s instead') % cmdsuggest
         else:
             msg = _(b'running non-interactively')
         raise error.Abort(msg)
 
     # make sure username is set before going interactive
     if not opts.get(b'user'):
         ui.username()  # raise exception, username not provided
 
     def recordfunc(ui, repo, message, match, opts):
         """This is generic record driver.
 
         Its job is to interactively filter local changes, and
         accordingly prepare working directory into a state in which the
         job can be delegated to a non-interactive commit command such as
         'commit' or 'qrefresh'.
 
         After the actual job is done by non-interactive command, the
         working directory is restored to its original state.
 
         In the end we'll record interesting changes, and everything else
         will be left in place, so the user can continue working.
         """
         if not opts.get(b'interactive-unshelve'):
             checkunfinished(repo, commit=True)
         wctx = repo[None]
         merge = len(wctx.parents()) > 1
         if merge:
             raise error.Abort(
                 _(
                     b'cannot partially commit a merge '
                     b'(use "hg commit" instead)'
                 )
             )
 
         def fail(f, msg):
             raise error.Abort(b'%s: %s' % (f, msg))
 
         force = opts.get(b'force')
         if not force:
             match = matchmod.badmatch(match, fail)
 
         status = repo.status(match=match)
 
         overrides = {(b'ui', b'commitsubrepos'): True}
 
         with repo.ui.configoverride(overrides, b'record'):
             # subrepoutil.precommit() modifies the status
             tmpstatus = scmutil.status(
                 copymod.copy(status.modified),
                 copymod.copy(status.added),
                 copymod.copy(status.removed),
                 copymod.copy(status.deleted),
                 copymod.copy(status.unknown),
                 copymod.copy(status.ignored),
                 copymod.copy(status.clean),  # pytype: disable=wrong-arg-count
             )
 
             # Force allows -X subrepo to skip the subrepo.
             subs, commitsubs, newstate = subrepoutil.precommit(
                 repo.ui, wctx, tmpstatus, match, force=True
             )
             for s in subs:
                 if s in commitsubs:
                     dirtyreason = wctx.sub(s).dirtyreason(True)
                     raise error.Abort(dirtyreason)
 
         if not force:
             repo.checkcommitpatterns(wctx, match, status, fail)
         diffopts = patch.difffeatureopts(
             ui,
             opts=opts,
             whitespace=True,
             section=b'commands',
             configprefix=b'commit.interactive.',
         )
         diffopts.nodates = True
         diffopts.git = True
         diffopts.showfunc = True
         originaldiff = patch.diff(repo, changes=status, opts=diffopts)
         originalchunks = patch.parsepatch(originaldiff)
         match = scmutil.match(repo[None], pats)
 
         # 1. filter patch, since we are intending to apply subset of it
         try:
             chunks, newopts = filterfn(ui, originalchunks, match)
         except error.PatchError as err:
             raise error.Abort(_(b'error parsing patch: %s') % err)
         opts.update(newopts)
 
         # We need to keep a backup of files that have been newly added and
         # modified during the recording process because there is a previous
         # version without the edit in the workdir. We also will need to restore
         # files that were the sources of renames so that the patch application
         # works.
         newlyaddedandmodifiedfiles, alsorestore = newandmodified(
             chunks, originalchunks
         )
         contenders = set()
         for h in chunks:
             try:
                 contenders.update(set(h.files()))
             except AttributeError:
                 pass
 
         changed = status.modified + status.added + status.removed
         newfiles = [f for f in changed if f in contenders]
         if not newfiles:
             ui.status(_(b'no changes to record\n'))
             return 0
 
         modified = set(status.modified)
 
         # 2. backup changed files, so we can restore them in the end
 
         if backupall:
             tobackup = changed
         else:
             tobackup = [
                 f
                 for f in newfiles
                 if f in modified or f in newlyaddedandmodifiedfiles
             ]
         backups = {}
         if tobackup:
             backupdir = repo.vfs.join(b'record-backups')
             try:
                 os.mkdir(backupdir)
             except OSError as err:
                 if err.errno != errno.EEXIST:
                     raise
         try:
             # backup continues
             for f in tobackup:
                 fd, tmpname = pycompat.mkstemp(
                     prefix=f.replace(b'/', b'_') + b'.', dir=backupdir
                 )
                 os.close(fd)
                 ui.debug(b'backup %r as %r\n' % (f, tmpname))
                 util.copyfile(repo.wjoin(f), tmpname, copystat=True)
                 backups[f] = tmpname
 
             fp = stringio()
             for c in chunks:
                 fname = c.filename()
                 if fname in backups:
                     c.write(fp)
             dopatch = fp.tell()
             fp.seek(0)
 
             # 2.5 optionally review / modify patch in text editor
             if opts.get(b'review', False):
                 patchtext = (
                     crecordmod.diffhelptext
                     + crecordmod.patchhelptext
                     + fp.read()
                 )
                 reviewedpatch = ui.edit(
                     patchtext, b"", action=b"diff", repopath=repo.path
                 )
                 fp.truncate(0)
                 fp.write(reviewedpatch)
                 fp.seek(0)
 
             [os.unlink(repo.wjoin(c)) for c in newlyaddedandmodifiedfiles]
             # 3a. apply filtered patch to clean repo  (clean)
             if backups:
                 m = scmutil.matchfiles(repo, set(backups.keys()) | alsorestore)
                 mergemod.revert_to(repo[b'.'], matcher=m)
 
             # 3b. (apply)
             if dopatch:
                 try:
                     ui.debug(b'applying patch\n')
                     ui.debug(fp.getvalue())
                     patch.internalpatch(ui, repo, fp, 1, eolmode=None)
                 except error.PatchError as err:
                     raise error.Abort(pycompat.bytestr(err))
             del fp
 
             # 4. We prepared working directory according to filtered
             #    patch. Now is the time to delegate the job to
             #    commit/qrefresh or the like!
 
             # Make all of the pathnames absolute.
             newfiles = [repo.wjoin(nf) for nf in newfiles]
             return commitfunc(ui, repo, *newfiles, **pycompat.strkwargs(opts))
         finally:
             # 5. finally restore backed-up files
             try:
                 dirstate = repo.dirstate
                 for realname, tmpname in pycompat.iteritems(backups):
                     ui.debug(b'restoring %r to %r\n' % (tmpname, realname))
 
                     if dirstate[realname] == b'n':
                         # without normallookup, restoring timestamp
                         # may cause partially committed files
                         # to be treated as unmodified
                         dirstate.normallookup(realname)
 
                     # copystat=True here and above are a hack to trick any
                     # editors that have f open that we haven't modified them.
                     #
                     # Also note that this racy as an editor could notice the
                     # file's mtime before we've finished writing it.
                     util.copyfile(tmpname, repo.wjoin(realname), copystat=True)
                     os.unlink(tmpname)
                 if tobackup:
                     os.rmdir(backupdir)
             except OSError:
                 pass
 
     def recordinwlock(ui, repo, message, match, opts):
         with repo.wlock():
             return recordfunc(ui, repo, message, match, opts)
 
     return commit(ui, repo, recordinwlock, pats, opts)
 
 
 class dirnode(object):
     """
     Represent a directory in user working copy with information required for
     the purpose of tersing its status.
 
     path is the path to the directory, without a trailing '/'
 
     statuses is a set of statuses of all files in this directory (this includes
     all the files in all the subdirectories too)
 
     files is a list of files which are direct child of this directory
 
     subdirs is a dictionary of sub-directory name as the key and it's own
     dirnode object as the value
     """
 
     def __init__(self, dirpath):
         self.path = dirpath
         self.statuses = set()
         self.files = []
         self.subdirs = {}
 
     def _addfileindir(self, filename, status):
         """Add a file in this directory as a direct child."""
         self.files.append((filename, status))
 
     def addfile(self, filename, status):
         """
         Add a file to this directory or to its direct parent directory.
 
         If the file is not direct child of this directory, we traverse to the
         directory of which this file is a direct child of and add the file
         there.
         """
 
         # the filename contains a path separator, it means it's not the direct
         # child of this directory
         if b'/' in filename:
             subdir, filep = filename.split(b'/', 1)
 
             # does the dirnode object for subdir exists
             if subdir not in self.subdirs:
                 subdirpath = pathutil.join(self.path, subdir)
                 self.subdirs[subdir] = dirnode(subdirpath)
 
             # try adding the file in subdir
             self.subdirs[subdir].addfile(filep, status)
 
         else:
             self._addfileindir(filename, status)
 
         if status not in self.statuses:
             self.statuses.add(status)
 
     def iterfilepaths(self):
         """Yield (status, path) for files directly under this directory."""
         for f, st in self.files:
             yield st, pathutil.join(self.path, f)
 
     def tersewalk(self, terseargs):
         """
         Yield (status, path) obtained by processing the status of this
         dirnode.
 
         terseargs is the string of arguments passed by the user with `--terse`
         flag.
 
         Following are the cases which can happen:
 
         1) All the files in the directory (including all the files in its
         subdirectories) share the same status and the user has asked us to terse
         that status. -> yield (status, dirpath).  dirpath will end in '/'.
 
         2) Otherwise, we do following:
 
                 a) Yield (status, filepath)  for all the files which are in this
                     directory (only the ones in this directory, not the subdirs)
 
                 b) Recurse the function on all the subdirectories of this
                    directory
         """
 
         if len(self.statuses) == 1:
             onlyst = self.statuses.pop()
 
             # Making sure we terse only when the status abbreviation is
             # passed as terse argument
             if onlyst in terseargs:
                 yield onlyst, self.path + b'/'
                 return
 
         # add the files to status list
         for st, fpath in self.iterfilepaths():
             yield st, fpath
 
         # recurse on the subdirs
         for dirobj in self.subdirs.values():
             for st, fpath in dirobj.tersewalk(terseargs):
                 yield st, fpath
 
 
 def tersedir(statuslist, terseargs):
     """
     Terse the status if all the files in a directory shares the same status.
 
     statuslist is scmutil.status() object which contains a list of files for
     each status.
     terseargs is string which is passed by the user as the argument to `--terse`
     flag.
 
     The function makes a tree of objects of dirnode class, and at each node it
     stores the information required to know whether we can terse a certain
     directory or not.
     """
     # the order matters here as that is used to produce final list
     allst = (b'm', b'a', b'r', b'd', b'u', b'i', b'c')
 
     # checking the argument validity
     for s in pycompat.bytestr(terseargs):
         if s not in allst:
             raise error.Abort(_(b"'%s' not recognized") % s)
 
     # creating a dirnode object for the root of the repo
     rootobj = dirnode(b'')
     pstatus = (
         b'modified',
         b'added',
         b'deleted',
         b'clean',
         b'unknown',
         b'ignored',
         b'removed',
     )
 
     tersedict = {}
     for attrname in pstatus:
         statuschar = attrname[0:1]
         for f in getattr(statuslist, attrname):
             rootobj.addfile(f, statuschar)
         tersedict[statuschar] = []
 
     # we won't be tersing the root dir, so add files in it
     for st, fpath in rootobj.iterfilepaths():
         tersedict[st].append(fpath)
 
     # process each sub-directory and build tersedict
     for subdir in rootobj.subdirs.values():
         for st, f in subdir.tersewalk(terseargs):
             tersedict[st].append(f)
 
     tersedlist = []
     for st in allst:
         tersedict[st].sort()
         tersedlist.append(tersedict[st])
 
     return scmutil.status(*tersedlist)
 
 
 def _commentlines(raw):
     '''Surround lineswith a comment char and a new line'''
     lines = raw.splitlines()
     commentedlines = [b'# %s' % line for line in lines]
     return b'\n'.join(commentedlines) + b'\n'
 
 
 @attr.s(frozen=True)
 class morestatus(object):
     reporoot = attr.ib()
     unfinishedop = attr.ib()
     unfinishedmsg = attr.ib()
     activemerge = attr.ib()
     unresolvedpaths = attr.ib()
     _formattedpaths = attr.ib(init=False, default=set())
     _label = b'status.morestatus'
 
     def formatfile(self, path, fm):
         self._formattedpaths.add(path)
         if self.activemerge and path in self.unresolvedpaths:
             fm.data(unresolved=True)
 
     def formatfooter(self, fm):
         if self.unfinishedop or self.unfinishedmsg:
             fm.startitem()
             fm.data(itemtype=b'morestatus')
 
         if self.unfinishedop:
             fm.data(unfinished=self.unfinishedop)
             statemsg = (
                 _(b'The repository is in an unfinished *%s* state.')
                 % self.unfinishedop
             )
             fm.plain(b'%s\n' % _commentlines(statemsg), label=self._label)
         if self.unfinishedmsg:
             fm.data(unfinishedmsg=self.unfinishedmsg)
 
         # May also start new data items.
         self._formatconflicts(fm)
 
         if self.unfinishedmsg:
             fm.plain(
                 b'%s\n' % _commentlines(self.unfinishedmsg), label=self._label
             )
 
     def _formatconflicts(self, fm):
         if not self.activemerge:
             return
 
         if self.unresolvedpaths:
             mergeliststr = b'\n'.join(
                 [
                     b'    %s'
                     % util.pathto(self.reporoot, encoding.getcwd(), path)
                     for path in self.unresolvedpaths
                 ]
             )
             msg = (
                 _(
                     '''Unresolved merge conflicts:
 
 %s
 
 To mark files as resolved:  hg resolve --mark FILE'''
                 )
                 % mergeliststr
             )
 
             # If any paths with unresolved conflicts were not previously
             # formatted, output them now.
             for f in self.unresolvedpaths:
                 if f in self._formattedpaths:
                     # Already output.
                     continue
                 fm.startitem()
                 # We can't claim to know the status of the file - it may just
                 # have been in one of the states that were not requested for
                 # display, so it could be anything.
                 fm.data(itemtype=b'file', path=f, unresolved=True)
 
         else:
             msg = _(b'No unresolved merge conflicts.')
 
         fm.plain(b'%s\n' % _commentlines(msg), label=self._label)
 
 
 def readmorestatus(repo):
     """Returns a morestatus object if the repo has unfinished state."""
     statetuple = statemod.getrepostate(repo)
     mergestate = mergestatemod.mergestate.read(repo)
     activemerge = mergestate.active()
     if not statetuple and not activemerge:
         return None
 
     unfinishedop = unfinishedmsg = unresolved = None
     if statetuple:
         unfinishedop, unfinishedmsg = statetuple
     if activemerge:
         unresolved = sorted(mergestate.unresolved())
     return morestatus(
         repo.root, unfinishedop, unfinishedmsg, activemerge, unresolved
     )
 
 
 def findpossible(cmd, table, strict=False):
     """
     Return cmd -> (aliases, command table entry)
     for each matching command.
     Return debug commands (or their aliases) only if no normal command matches.
     """
     choice = {}
     debugchoice = {}
 
     if cmd in table:
         # short-circuit exact matches, "log" alias beats "log|history"
         keys = [cmd]
     else:
         keys = table.keys()
 
     allcmds = []
     for e in keys:
         aliases = parsealiases(e)
         allcmds.extend(aliases)
         found = None
         if cmd in aliases:
             found = cmd
         elif not strict:
             for a in aliases:
                 if a.startswith(cmd):
                     found = a
                     break
         if found is not None:
             if aliases[0].startswith(b"debug") or found.startswith(b"debug"):
                 debugchoice[found] = (aliases, table[e])
             else:
                 choice[found] = (aliases, table[e])
 
     if not choice and debugchoice:
         choice = debugchoice
 
     return choice, allcmds
 
 
 def findcmd(cmd, table, strict=True):
     """Return (aliases, command table entry) for command string."""
     choice, allcmds = findpossible(cmd, table, strict)
 
     if cmd in choice:
         return choice[cmd]
 
     if len(choice) > 1:
         clist = sorted(choice)
         raise error.AmbiguousCommand(cmd, clist)
 
     if choice:
         return list(choice.values())[0]
 
     raise error.UnknownCommand(cmd, allcmds)
 
 
 def changebranch(ui, repo, revs, label, opts):
     """ Change the branch name of given revs to label """
 
     with repo.wlock(), repo.lock(), repo.transaction(b'branches'):
         # abort in case of uncommitted merge or dirty wdir
         bailifchanged(repo)
         revs = scmutil.revrange(repo, revs)
         if not revs:
             raise error.Abort(b"empty revision set")
         roots = repo.revs(b'roots(%ld)', revs)
         if len(roots) > 1:
             raise error.Abort(
                 _(b"cannot change branch of non-linear revisions")
             )
         rewriteutil.precheck(repo, revs, b'change branch of')
 
         root = repo[roots.first()]
         rpb = {parent.branch() for parent in root.parents()}
         if (
             not opts.get(b'force')
             and label not in rpb
             and label in repo.branchmap()
         ):
             raise error.Abort(_(b"a branch of the same name already exists"))
 
         if repo.revs(b'obsolete() and %ld', revs):
             raise error.Abort(
                 _(b"cannot change branch of a obsolete changeset")
             )
 
         # make sure only topological heads
         if repo.revs(b'heads(%ld) - head()', revs):
             raise error.Abort(_(b"cannot change branch in middle of a stack"))
 
         replacements = {}
         # avoid import cycle mercurial.cmdutil -> mercurial.context ->
         # mercurial.subrepo -> mercurial.cmdutil
         from . import context
 
         for rev in revs:
             ctx = repo[rev]
             oldbranch = ctx.branch()
             # check if ctx has same branch
             if oldbranch == label:
                 continue
 
             def filectxfn(repo, newctx, path):
                 try:
                     return ctx[path]
                 except error.ManifestLookupError:
                     return None
 
             ui.debug(
                 b"changing branch of '%s' from '%s' to '%s'\n"
                 % (hex(ctx.node()), oldbranch, label)
             )
             extra = ctx.extra()
             extra[b'branch_change'] = hex(ctx.node())
             # While changing branch of set of linear commits, make sure that
             # we base our commits on new parent rather than old parent which
             # was obsoleted while changing the branch
             p1 = ctx.p1().node()
             p2 = ctx.p2().node()
             if p1 in replacements:
                 p1 = replacements[p1][0]
             if p2 in replacements:
                 p2 = replacements[p2][0]
 
             mc = context.memctx(
                 repo,
                 (p1, p2),
                 ctx.description(),
                 ctx.files(),
                 filectxfn,
                 user=ctx.user(),
                 date=ctx.date(),
                 extra=extra,
                 branch=label,
             )
 
             newnode = repo.commitctx(mc)
             replacements[ctx.node()] = (newnode,)
             ui.debug(b'new node id is %s\n' % hex(newnode))
 
         # create obsmarkers and move bookmarks
         scmutil.cleanupnodes(
             repo, replacements, b'branch-change', fixphase=True
         )
 
         # move the working copy too
         wctx = repo[None]
         # in-progress merge is a bit too complex for now.
         if len(wctx.parents()) == 1:
             newid = replacements.get(wctx.p1().node())
             if newid is not None:
                 # avoid import cycle mercurial.cmdutil -> mercurial.hg ->
                 # mercurial.cmdutil
                 from . import hg
 
                 hg.update(repo, newid[0], quietempty=True)
 
         ui.status(_(b"changed branch on %d changesets\n") % len(replacements))
 
 
 def findrepo(p):
     while not os.path.isdir(os.path.join(p, b".hg")):
         oldp, p = p, os.path.dirname(p)
         if p == oldp:
             return None
 
     return p
 
 
 def bailifchanged(repo, merge=True, hint=None):
     """ enforce the precondition that working directory must be clean.
 
     'merge' can be set to false if a pending uncommitted merge should be
     ignored (such as when 'update --check' runs).
 
     'hint' is the usual hint given to Abort exception.
     """
 
     if merge and repo.dirstate.p2() != nullid:
         raise error.Abort(_(b'outstanding uncommitted merge'), hint=hint)
     st = repo.status()
     if st.modified or st.added or st.removed or st.deleted:
         raise error.Abort(_(b'uncommitted changes'), hint=hint)
     ctx = repo[None]
     for s in sorted(ctx.substate):
         ctx.sub(s).bailifchanged(hint=hint)
 
 
 def logmessage(ui, opts):
     """ get the log message according to -m and -l option """
 
     check_at_most_one_arg(opts, b'message', b'logfile')
 
     message = opts.get(b'message')
     logfile = opts.get(b'logfile')
 
     if not message and logfile:
         try:
             if isstdiofilename(logfile):
                 message = ui.fin.read()
             else:
                 message = b'\n'.join(util.readfile(logfile).splitlines())
         except IOError as inst:
             raise error.Abort(
                 _(b"can't read commit message '%s': %s")
                 % (logfile, encoding.strtolocal(inst.strerror))
             )
     return message
 
 
 def mergeeditform(ctxorbool, baseformname):
     """return appropriate editform name (referencing a committemplate)
 
     'ctxorbool' is either a ctx to be committed, or a bool indicating whether
     merging is committed.
 
     This returns baseformname with '.merge' appended if it is a merge,
     otherwise '.normal' is appended.
     """
     if isinstance(ctxorbool, bool):
         if ctxorbool:
             return baseformname + b".merge"
     elif len(ctxorbool.parents()) > 1:
         return baseformname + b".merge"
 
     return baseformname + b".normal"
 
 
 def getcommiteditor(
     edit=False, finishdesc=None, extramsg=None, editform=b'', **opts
 ):
     """get appropriate commit message editor according to '--edit' option
 
     'finishdesc' is a function to be called with edited commit message
     (= 'description' of the new changeset) just after editing, but
     before checking empty-ness. It should return actual text to be
     stored into history. This allows to change description before
     storing.
 
     'extramsg' is a extra message to be shown in the editor instead of
     'Leave message empty to abort commit' line. 'HG: ' prefix and EOL
     is automatically added.
 
     'editform' is a dot-separated list of names, to distinguish
     the purpose of commit text editing.
 
     'getcommiteditor' returns 'commitforceeditor' regardless of
     'edit', if one of 'finishdesc' or 'extramsg' is specified, because
     they are specific for usage in MQ.
     """
     if edit or finishdesc or extramsg:
         return lambda r, c, s: commitforceeditor(
             r, c, s, finishdesc=finishdesc, extramsg=extramsg, editform=editform
         )
     elif editform:
         return lambda r, c, s: commiteditor(r, c, s, editform=editform)
     else:
         return commiteditor
 
 
 def _escapecommandtemplate(tmpl):
     parts = []
     for typ, start, end in templater.scantemplate(tmpl, raw=True):
         if typ == b'string':
             parts.append(stringutil.escapestr(tmpl[start:end]))
         else:
             parts.append(tmpl[start:end])
     return b''.join(parts)
 
 
 def rendercommandtemplate(ui, tmpl, props):
     r"""Expand a literal template 'tmpl' in a way suitable for command line
 
     '\' in outermost string is not taken as an escape character because it
     is a directory separator on Windows.
 
     >>> from . import ui as uimod
     >>> ui = uimod.ui()
     >>> rendercommandtemplate(ui, b'c:\\{path}', {b'path': b'foo'})
     'c:\\foo'
     >>> rendercommandtemplate(ui, b'{"c:\\{path}"}', {'path': b'foo'})
     'c:{path}'
     """
     if not tmpl:
         return tmpl
     t = formatter.maketemplater(ui, _escapecommandtemplate(tmpl))
     return t.renderdefault(props)
 
 
 def rendertemplate(ctx, tmpl, props=None):
     """Expand a literal template 'tmpl' byte-string against one changeset
 
     Each props item must be a stringify-able value or a callable returning
     such value, i.e. no bare list nor dict should be passed.
     """
     repo = ctx.repo()
     tres = formatter.templateresources(repo.ui, repo)
     t = formatter.maketemplater(
         repo.ui, tmpl, defaults=templatekw.keywords, resources=tres
     )
     mapping = {b'ctx': ctx}
     if props:
         mapping.update(props)
     return t.renderdefault(mapping)
 
 
 def _buildfntemplate(pat, total=None, seqno=None, revwidth=None, pathname=None):
     r"""Convert old-style filename format string to template string
 
     >>> _buildfntemplate(b'foo-%b-%n.patch', seqno=0)
     'foo-{reporoot|basename}-{seqno}.patch'
     >>> _buildfntemplate(b'%R{tags % "{tag}"}%H')
     '{rev}{tags % "{tag}"}{node}'
 
     '\' in outermost strings has to be escaped because it is a directory
     separator on Windows:
 
     >>> _buildfntemplate(b'c:\\tmp\\%R\\%n.patch', seqno=0)
     'c:\\\\tmp\\\\{rev}\\\\{seqno}.patch'
     >>> _buildfntemplate(b'\\\\foo\\bar.patch')
     '\\\\\\\\foo\\\\bar.patch'
     >>> _buildfntemplate(b'\\{tags % "{tag}"}')
     '\\\\{tags % "{tag}"}'
 
     but inner strings follow the template rules (i.e. '\' is taken as an
     escape character):
 
     >>> _buildfntemplate(br'{"c:\tmp"}', seqno=0)
     '{"c:\\tmp"}'
     """
     expander = {
         b'H': b'{node}',
         b'R': b'{rev}',
         b'h': b'{node|short}',
         b'm': br'{sub(r"[^\w]", "_", desc|firstline)}',
         b'r': b'{if(revwidth, pad(rev, revwidth, "0", left=True), rev)}',
         b'%': b'%',
         b'b': b'{reporoot|basename}',
     }
     if total is not None:
         expander[b'N'] = b'{total}'
     if seqno is not None:
         expander[b'n'] = b'{seqno}'
     if total is not None and seqno is not None:
         expander[b'n'] = b'{pad(seqno, total|stringify|count, "0", left=True)}'
     if pathname is not None:
         expander[b's'] = b'{pathname|basename}'
         expander[b'd'] = b'{if(pathname|dirname, pathname|dirname, ".")}'
         expander[b'p'] = b'{pathname}'
 
     newname = []
     for typ, start, end in templater.scantemplate(pat, raw=True):
         if typ != b'string':
             newname.append(pat[start:end])
             continue
         i = start
         while i < end:
             n = pat.find(b'%', i, end)
             if n < 0:
                 newname.append(stringutil.escapestr(pat[i:end]))
                 break
             newname.append(stringutil.escapestr(pat[i:n]))
             if n + 2 > end:
                 raise error.Abort(
                     _(b"incomplete format spec in output filename")
                 )
             c = pat[n + 1 : n + 2]
             i = n + 2
             try:
                 newname.append(expander[c])
             except KeyError:
                 raise error.Abort(
                     _(b"invalid format spec '%%%s' in output filename") % c
                 )
     return b''.join(newname)
 
 
 def makefilename(ctx, pat, **props):
     if not pat:
         return pat
     tmpl = _buildfntemplate(pat, **props)
     # BUG: alias expansion shouldn't be made against template fragments
     # rewritten from %-format strings, but we have no easy way to partially
     # disable the expansion.
     return rendertemplate(ctx, tmpl, pycompat.byteskwargs(props))
 
 
 def isstdiofilename(pat):
     """True if the given pat looks like a filename denoting stdin/stdout"""
     return not pat or pat == b'-'
 
 
 class _unclosablefile(object):
     def __init__(self, fp):
         self._fp = fp
 
     def close(self):
         pass
 
     def __iter__(self):
         return iter(self._fp)
 
     def __getattr__(self, attr):
         return getattr(self._fp, attr)
 
     def __enter__(self):
         return self
 
     def __exit__(self, exc_type, exc_value, exc_tb):
         pass
 
 
 def makefileobj(ctx, pat, mode=b'wb', **props):
     writable = mode not in (b'r', b'rb')
 
     if isstdiofilename(pat):
         repo = ctx.repo()
         if writable:
             fp = repo.ui.fout
         else:
             fp = repo.ui.fin
         return _unclosablefile(fp)
     fn = makefilename(ctx, pat, **props)
     return open(fn, mode)
 
 
 def openstorage(repo, cmd, file_, opts, returnrevlog=False):
     """opens the changelog, manifest, a filelog or a given revlog"""
     cl = opts[b'changelog']
     mf = opts[b'manifest']
     dir = opts[b'dir']
     msg = None
     if cl and mf:
         msg = _(b'cannot specify --changelog and --manifest at the same time')
     elif cl and dir:
         msg = _(b'cannot specify --changelog and --dir at the same time')
     elif cl or mf or dir:
         if file_:
             msg = _(b'cannot specify filename with --changelog or --manifest')
         elif not repo:
             msg = _(
                 b'cannot specify --changelog or --manifest or --dir '
                 b'without a repository'
             )
     if msg:
         raise error.Abort(msg)
 
     r = None
     if repo:
         if cl:
             r = repo.unfiltered().changelog
         elif dir:
             if b'treemanifest' not in repo.requirements:
                 raise error.Abort(
                     _(
                         b"--dir can only be used on repos with "
                         b"treemanifest enabled"
                     )
                 )
             if not dir.endswith(b'/'):
                 dir = dir + b'/'
             dirlog = repo.manifestlog.getstorage(dir)
             if len(dirlog):
                 r = dirlog
         elif mf:
             r = repo.manifestlog.getstorage(b'')
         elif file_:
             filelog = repo.file(file_)
             if len(filelog):
                 r = filelog
 
         # Not all storage may be revlogs. If requested, try to return an actual
         # revlog instance.
         if returnrevlog:
             if isinstance(r, revlog.revlog):
                 pass
             elif util.safehasattr(r, b'_revlog'):
                 r = r._revlog  # pytype: disable=attribute-error
             elif r is not None:
                 raise error.Abort(_(b'%r does not appear to be a revlog') % r)
 
     if not r:
         if not returnrevlog:
             raise error.Abort(_(b'cannot give path to non-revlog'))
 
         if not file_:
             raise error.CommandError(cmd, _(b'invalid arguments'))
         if not os.path.isfile(file_):
             raise error.Abort(_(b"revlog '%s' not found") % file_)
         r = revlog.revlog(
             vfsmod.vfs(encoding.getcwd(), audit=False), file_[:-2] + b".i"
         )
     return r
 
 
 def openrevlog(repo, cmd, file_, opts):
     """Obtain a revlog backing storage of an item.
 
     This is similar to ``openstorage()`` except it always returns a revlog.
 
     In most cases, a caller cares about the main storage object - not the
     revlog backing it. Therefore, this function should only be used by code
     that needs to examine low-level revlog implementation details. e.g. debug
     commands.
     """
     return openstorage(repo, cmd, file_, opts, returnrevlog=True)
 
 
 def copy(ui, repo, pats, opts, rename=False):
     check_incompatible_arguments(opts, b'forget', [b'dry_run'])
 
     # called with the repo lock held
     #
     # hgsep => pathname that uses "/" to separate directories
     # ossep => pathname that uses os.sep to separate directories
     cwd = repo.getcwd()
     targets = {}
     forget = opts.get(b"forget")
     after = opts.get(b"after")
     dryrun = opts.get(b"dry_run")
     rev = opts.get(b'at_rev')
     if rev:
         if not forget and not after:
             # TODO: Remove this restriction and make it also create the copy
             #       targets (and remove the rename source if rename==True).
             raise error.Abort(_(b'--at-rev requires --after'))
         ctx = scmutil.revsingle(repo, rev)
         if len(ctx.parents()) > 1:
             raise error.Abort(_(b'cannot mark/unmark copy in merge commit'))
     else:
         ctx = repo[None]
 
     pctx = ctx.p1()
 
     uipathfn = scmutil.getuipathfn(repo, legacyrelativevalue=True)
 
     if forget:
         if ctx.rev() is None:
             new_ctx = ctx
         else:
             if len(ctx.parents()) > 1:
                 raise error.Abort(_(b'cannot unmark copy in merge commit'))
             # avoid cycle context -> subrepo -> cmdutil
             from . import context
 
             rewriteutil.precheck(repo, [ctx.rev()], b'uncopy')
             new_ctx = context.overlayworkingctx(repo)
             new_ctx.setbase(ctx.p1())
             mergemod.graft(repo, ctx, wctx=new_ctx)
 
         match = scmutil.match(ctx, pats, opts)
 
         current_copies = ctx.p1copies()
         current_copies.update(ctx.p2copies())
 
         uipathfn = scmutil.getuipathfn(repo)
         for f in ctx.walk(match):
             if f in current_copies:
                 new_ctx[f].markcopied(None)
             elif match.exact(f):
                 ui.warn(
                     _(
                         b'%s: not unmarking as copy - file is not marked as copied\n'
                     )
                     % uipathfn(f)
                 )
 
         if ctx.rev() is not None:
             with repo.lock():
                 mem_ctx = new_ctx.tomemctx_for_amend(ctx)
                 new_node = mem_ctx.commit()
 
                 if repo.dirstate.p1() == ctx.node():
                     with repo.dirstate.parentchange():
                         scmutil.movedirstate(repo, repo[new_node])
                 replacements = {ctx.node(): [new_node]}
                 scmutil.cleanupnodes(
                     repo, replacements, b'uncopy', fixphase=True
                 )
 
         return
 
     pats = scmutil.expandpats(pats)
     if not pats:
         raise error.Abort(_(b'no source or destination specified'))
     if len(pats) == 1:
         raise error.Abort(_(b'no destination specified'))
     dest = pats.pop()
 
     def walkpat(pat):
         srcs = []
         # TODO: Inline and simplify the non-working-copy version of this code
         # since it shares very little with the working-copy version of it.
         ctx_to_walk = ctx if ctx.rev() is None else pctx
         m = scmutil.match(ctx_to_walk, [pat], opts, globbed=True)
         for abs in ctx_to_walk.walk(m):
             rel = uipathfn(abs)
             exact = m.exact(abs)
             if abs not in ctx:
                 if abs in pctx:
                     if not after:
                         if exact:
                             ui.warn(
                                 _(
                                     b'%s: not copying - file has been marked '
                                     b'for remove\n'
                                 )
                                 % rel
                             )
                         continue
                 else:
                     if exact:
                         ui.warn(
                             _(b'%s: not copying - file is not managed\n') % rel
                         )
                     continue
 
             # abs: hgsep
             # rel: ossep
             srcs.append((abs, rel, exact))
         return srcs
 
     if ctx.rev() is not None:
         rewriteutil.precheck(repo, [ctx.rev()], b'uncopy')
         absdest = pathutil.canonpath(repo.root, cwd, dest)
         if ctx.hasdir(absdest):
             raise error.Abort(
                 _(b'%s: --at-rev does not support a directory as destination')
                 % uipathfn(absdest)
             )
         if absdest not in ctx:
             raise error.Abort(
                 _(b'%s: copy destination does not exist in %s')
                 % (uipathfn(absdest), ctx)
             )
 
         # avoid cycle context -> subrepo -> cmdutil
         from . import context
 
         copylist = []
         for pat in pats:
             srcs = walkpat(pat)
             if not srcs:
                 continue
             for abs, rel, exact in srcs:
                 copylist.append(abs)
 
         if not copylist:
             raise error.Abort(_(b'no files to copy'))
         # TODO: Add support for `hg cp --at-rev . foo bar dir` and
         # `hg cp --at-rev . dir1 dir2`, preferably unifying the code with the
         # existing functions below.
         if len(copylist) != 1:
             raise error.Abort(_(b'--at-rev requires a single source'))
 
         new_ctx = context.overlayworkingctx(repo)
         new_ctx.setbase(ctx.p1())
         mergemod.graft(repo, ctx, wctx=new_ctx)
 
         new_ctx.markcopied(absdest, copylist[0])
 
         with repo.lock():
             mem_ctx = new_ctx.tomemctx_for_amend(ctx)
             new_node = mem_ctx.commit()
 
             if repo.dirstate.p1() == ctx.node():
                 with repo.dirstate.parentchange():
                     scmutil.movedirstate(repo, repo[new_node])
             replacements = {ctx.node(): [new_node]}
             scmutil.cleanupnodes(repo, replacements, b'copy', fixphase=True)
 
         return
 
     # abssrc: hgsep
     # relsrc: ossep
     # otarget: ossep
     def copyfile(abssrc, relsrc, otarget, exact):
         abstarget = pathutil.canonpath(repo.root, cwd, otarget)
         if b'/' in abstarget:
             # We cannot normalize abstarget itself, this would prevent
             # case only renames, like a => A.
             abspath, absname = abstarget.rsplit(b'/', 1)
             abstarget = repo.dirstate.normalize(abspath) + b'/' + absname
         reltarget = repo.pathto(abstarget, cwd)
         target = repo.wjoin(abstarget)
         src = repo.wjoin(abssrc)
         state = repo.dirstate[abstarget]
 
         scmutil.checkportable(ui, abstarget)
 
         # check for collisions
         prevsrc = targets.get(abstarget)
         if prevsrc is not None:
             ui.warn(
                 _(b'%s: not overwriting - %s collides with %s\n')
                 % (
                     reltarget,
                     repo.pathto(abssrc, cwd),
                     repo.pathto(prevsrc, cwd),
                 )
             )
             return True  # report a failure
 
         # check for overwrites
         exists = os.path.lexists(target)
         samefile = False
         if exists and abssrc != abstarget:
             if repo.dirstate.normalize(abssrc) == repo.dirstate.normalize(
                 abstarget
             ):
                 if not rename:
                     ui.warn(_(b"%s: can't copy - same file\n") % reltarget)
                     return True  # report a failure
                 exists = False
                 samefile = True
 
         if not after and exists or after and state in b'mn':
             if not opts[b'force']:
                 if state in b'mn':
                     msg = _(b'%s: not overwriting - file already committed\n')
                     if after:
                         flags = b'--after --force'
                     else:
                         flags = b'--force'
                     if rename:
                         hint = (
                             _(
                                 b"('hg rename %s' to replace the file by "
                                 b'recording a rename)\n'
                             )
                             % flags
                         )
                     else:
                         hint = (
                             _(
                                 b"('hg copy %s' to replace the file by "
                                 b'recording a copy)\n'
                             )
                             % flags
                         )
                 else:
                     msg = _(b'%s: not overwriting - file exists\n')
                     if rename:
                         hint = _(
                             b"('hg rename --after' to record the rename)\n"
                         )
                     else:
                         hint = _(b"('hg copy --after' to record the copy)\n")
                 ui.warn(msg % reltarget)
                 ui.warn(hint)
                 return True  # report a failure
 
         if after:
             if not exists:
                 if rename:
                     ui.warn(
                         _(b'%s: not recording move - %s does not exist\n')
                         % (relsrc, reltarget)
                     )
                 else:
                     ui.warn(
                         _(b'%s: not recording copy - %s does not exist\n')
                         % (relsrc, reltarget)
                     )
                 return True  # report a failure
         elif not dryrun:
             try:
                 if exists:
                     os.unlink(target)
                 targetdir = os.path.dirname(target) or b'.'
                 if not os.path.isdir(targetdir):
                     os.makedirs(targetdir)
                 if samefile:
                     tmp = target + b"~hgrename"
                     os.rename(src, tmp)
                     os.rename(tmp, target)
                 else:
                     # Preserve stat info on renames, not on copies; this matches
                     # Linux CLI behavior.
                     util.copyfile(src, target, copystat=rename)
                 srcexists = True
             except IOError as inst:
                 if inst.errno == errno.ENOENT:
                     ui.warn(_(b'%s: deleted in working directory\n') % relsrc)
                     srcexists = False
                 else:
                     ui.warn(
                         _(b'%s: cannot copy - %s\n')
                         % (relsrc, encoding.strtolocal(inst.strerror))
                     )
                     return True  # report a failure
 
         if ui.verbose or not exact:
             if rename:
                 ui.status(_(b'moving %s to %s\n') % (relsrc, reltarget))
             else:
                 ui.status(_(b'copying %s to %s\n') % (relsrc, reltarget))
 
         targets[abstarget] = abssrc
 
         # fix up dirstate
         scmutil.dirstatecopy(
             ui, repo, ctx, abssrc, abstarget, dryrun=dryrun, cwd=cwd
         )
         if rename and not dryrun:
             if not after and srcexists and not samefile:
                 rmdir = repo.ui.configbool(b'experimental', b'removeemptydirs')
                 repo.wvfs.unlinkpath(abssrc, rmdir=rmdir)
             ctx.forget([abssrc])
 
     # pat: ossep
     # dest ossep
     # srcs: list of (hgsep, hgsep, ossep, bool)
     # return: function that takes hgsep and returns ossep
     def targetpathfn(pat, dest, srcs):
         if os.path.isdir(pat):
             abspfx = pathutil.canonpath(repo.root, cwd, pat)
             abspfx = util.localpath(abspfx)
             if destdirexists:
                 striplen = len(os.path.split(abspfx)[0])
             else:
                 striplen = len(abspfx)
             if striplen:
                 striplen += len(pycompat.ossep)
             res = lambda p: os.path.join(dest, util.localpath(p)[striplen:])
         elif destdirexists:
             res = lambda p: os.path.join(
                 dest, os.path.basename(util.localpath(p))
             )
         else:
             res = lambda p: dest
         return res
 
     # pat: ossep
     # dest ossep
     # srcs: list of (hgsep, hgsep, ossep, bool)
     # return: function that takes hgsep and returns ossep
     def targetpathafterfn(pat, dest, srcs):
         if matchmod.patkind(pat):
             # a mercurial pattern
             res = lambda p: os.path.join(
                 dest, os.path.basename(util.localpath(p))
             )
         else:
             abspfx = pathutil.canonpath(repo.root, cwd, pat)
             if len(abspfx) < len(srcs[0][0]):
                 # A directory. Either the target path contains the last
                 # component of the source path or it does not.
                 def evalpath(striplen):
                     score = 0
                     for s in srcs:
                         t = os.path.join(dest, util.localpath(s[0])[striplen:])
                         if os.path.lexists(t):
                             score += 1
                     return score
 
                 abspfx = util.localpath(abspfx)
                 striplen = len(abspfx)
                 if striplen:
                     striplen += len(pycompat.ossep)
                 if os.path.isdir(os.path.join(dest, os.path.split(abspfx)[1])):
                     score = evalpath(striplen)
                     striplen1 = len(os.path.split(abspfx)[0])
                     if striplen1:
                         striplen1 += len(pycompat.ossep)
                     if evalpath(striplen1) > score:
                         striplen = striplen1
                 res = lambda p: os.path.join(dest, util.localpath(p)[striplen:])
             else:
                 # a file
                 if destdirexists:
                     res = lambda p: os.path.join(
                         dest, os.path.basename(util.localpath(p))
                     )
                 else:
                     res = lambda p: dest
         return res
 
     destdirexists = os.path.isdir(dest) and not os.path.islink(dest)
     if not destdirexists:
         if len(pats) > 1 or matchmod.patkind(pats[0]):
             raise error.Abort(
                 _(
                     b'with multiple sources, destination must be an '
                     b'existing directory'
                 )
             )
         if util.endswithsep(dest):
             raise error.Abort(_(b'destination %s is not a directory') % dest)
 
     tfn = targetpathfn
     if after:
         tfn = targetpathafterfn
     copylist = []
     for pat in pats:
         srcs = walkpat(pat)
         if not srcs:
             continue
         copylist.append((tfn(pat, dest, srcs), srcs))
     if not copylist:
         raise error.Abort(_(b'no files to copy'))
 
     errors = 0
     for targetpath, srcs in copylist:
         for abssrc, relsrc, exact in srcs:
             if copyfile(abssrc, relsrc, targetpath(abssrc), exact):
                 errors += 1
 
     return errors != 0
 
 
 ## facility to let extension process additional data into an import patch
 # list of identifier to be executed in order
 extrapreimport = []  # run before commit
 extrapostimport = []  # run after commit
 # mapping from identifier to actual import function
 #
 # 'preimport' are run before the commit is made and are provided the following
 # arguments:
 # - repo: the localrepository instance,
 # - patchdata: data extracted from patch header (cf m.patch.patchheadermap),
 # - extra: the future extra dictionary of the changeset, please mutate it,
 # - opts: the import options.
 # XXX ideally, we would just pass an ctx ready to be computed, that would allow
 # mutation of in memory commit and more. Feel free to rework the code to get
 # there.
 extrapreimportmap = {}
 # 'postimport' are run after the commit is made and are provided the following
 # argument:
 # - ctx: the changectx created by import.
 extrapostimportmap = {}
 
 
 def tryimportone(ui, repo, patchdata, parents, opts, msgs, updatefunc):
     """Utility function used by commands.import to import a single patch
 
     This function is explicitly defined here to help the evolve extension to
     wrap this part of the import logic.
 
     The API is currently a bit ugly because it a simple code translation from
     the import command. Feel free to make it better.
 
     :patchdata: a dictionary containing parsed patch data (such as from
                 ``patch.extract()``)
     :parents: nodes that will be parent of the created commit
     :opts: the full dict of option passed to the import command
     :msgs: list to save commit message to.
            (used in case we need to save it when failing)
     :updatefunc: a function that update a repo to a given node
                  updatefunc(<repo>, <node>)
     """
     # avoid cycle context -> subrepo -> cmdutil
     from . import context
 
     tmpname = patchdata.get(b'filename')
     message = patchdata.get(b'message')
     user = opts.get(b'user') or patchdata.get(b'user')
     date = opts.get(b'date') or patchdata.get(b'date')
     branch = patchdata.get(b'branch')
     nodeid = patchdata.get(b'nodeid')
     p1 = patchdata.get(b'p1')
     p2 = patchdata.get(b'p2')
 
     nocommit = opts.get(b'no_commit')
     importbranch = opts.get(b'import_branch')
     update = not opts.get(b'bypass')
     strip = opts[b"strip"]
     prefix = opts[b"prefix"]
     sim = float(opts.get(b'similarity') or 0)
 
     if not tmpname:
         return None, None, False
 
     rejects = False
 
     cmdline_message = logmessage(ui, opts)
     if cmdline_message:
         # pickup the cmdline msg
         message = cmdline_message
     elif message:
         # pickup the patch msg
         message = message.strip()
     else:
         # launch the editor
         message = None
     ui.debug(b'message:\n%s\n' % (message or b''))
 
     if len(parents) == 1:
         parents.append(repo[nullid])
     if opts.get(b'exact'):
         if not nodeid or not p1:
             raise error.Abort(_(b'not a Mercurial patch'))
         p1 = repo[p1]
         p2 = repo[p2 or nullid]
     elif p2:
         try:
             p1 = repo[p1]
             p2 = repo[p2]
             # Without any options, consider p2 only if the
             # patch is being applied on top of the recorded
             # first parent.
             if p1 != parents[0]:
                 p1 = parents[0]
                 p2 = repo[nullid]
         except error.RepoError:
             p1, p2 = parents
         if p2.node() == nullid:
             ui.warn(
                 _(
                     b"warning: import the patch as a normal revision\n"
                     b"(use --exact to import the patch as a merge)\n"
                 )
             )
     else:
         p1, p2 = parents
 
     n = None
     if update:
         if p1 != parents[0]:
             updatefunc(repo, p1.node())
         if p2 != parents[1]:
             repo.setparents(p1.node(), p2.node())
 
         if opts.get(b'exact') or importbranch:
             repo.dirstate.setbranch(branch or b'default')
 
         partial = opts.get(b'partial', False)
         files = set()
         try:
             patch.patch(
                 ui,
                 repo,
                 tmpname,
                 strip=strip,
                 prefix=prefix,
                 files=files,
                 eolmode=None,
                 similarity=sim / 100.0,
             )
         except error.PatchError as e:
             if not partial:
                 raise error.Abort(pycompat.bytestr(e))
             if partial:
                 rejects = True
 
         files = list(files)
         if nocommit:
             if message:
                 msgs.append(message)
         else:
             if opts.get(b'exact') or p2:
                 # If you got here, you either use --force and know what
                 # you are doing or used --exact or a merge patch while
                 # being updated to its first parent.
                 m = None
             else:
                 m = scmutil.matchfiles(repo, files or [])
             editform = mergeeditform(repo[None], b'import.normal')
             if opts.get(b'exact'):
                 editor = None
             else:
                 editor = getcommiteditor(
                     editform=editform, **pycompat.strkwargs(opts)
                 )
             extra = {}
             for idfunc in extrapreimport:
                 extrapreimportmap[idfunc](repo, patchdata, extra, opts)
             overrides = {}
             if partial:
                 overrides[(b'ui', b'allowemptycommit')] = True
             if opts.get(b'secret'):
                 overrides[(b'phases', b'new-commit')] = b'secret'
             with repo.ui.configoverride(overrides, b'import'):
                 n = repo.commit(
                     message, user, date, match=m, editor=editor, extra=extra
                 )
                 for idfunc in extrapostimport:
                     extrapostimportmap[idfunc](repo[n])
     else:
         if opts.get(b'exact') or importbranch:
             branch = branch or b'default'
         else:
             branch = p1.branch()
         store = patch.filestore()
         try:
             files = set()
             try:
                 patch.patchrepo(
                     ui,
                     repo,
                     p1,
                     store,
                     tmpname,
                     strip,
                     prefix,
                     files,
                     eolmode=None,
                 )
             except error.PatchError as e:
                 raise error.Abort(stringutil.forcebytestr(e))
             if opts.get(b'exact'):
                 editor = None
             else:
                 editor = getcommiteditor(editform=b'import.bypass')
             memctx = context.memctx(
                 repo,
                 (p1.node(), p2.node()),
                 message,
                 files=files,
                 filectxfn=store,
                 user=user,
                 date=date,
                 branch=branch,
                 editor=editor,
             )
 
             overrides = {}
             if opts.get(b'secret'):
                 overrides[(b'phases', b'new-commit')] = b'secret'
             with repo.ui.configoverride(overrides, b'import'):
                 n = memctx.commit()
         finally:
             store.close()
     if opts.get(b'exact') and nocommit:
         # --exact with --no-commit is still useful in that it does merge
         # and branch bits
         ui.warn(_(b"warning: can't check exact import with --no-commit\n"))
     elif opts.get(b'exact') and (not n or hex(n) != nodeid):
         raise error.Abort(_(b'patch is damaged or loses information'))
     msg = _(b'applied to working directory')
     if n:
         # i18n: refers to a short changeset id
         msg = _(b'created %s') % short(n)
     return msg, n, rejects
 
 
 # facility to let extensions include additional data in an exported patch
 # list of identifiers to be executed in order
 extraexport = []
 # mapping from identifier to actual export function
 # function as to return a string to be added to the header or None
 # it is given two arguments (sequencenumber, changectx)
 extraexportmap = {}
 
 
 def _exportsingle(repo, ctx, fm, match, switch_parent, seqno, diffopts):
     node = scmutil.binnode(ctx)
     parents = [p.node() for p in ctx.parents() if p]
     branch = ctx.branch()
     if switch_parent:
         parents.reverse()
 
     if parents:
         prev = parents[0]
     else:
         prev = nullid
 
     fm.context(ctx=ctx)
     fm.plain(b'# HG changeset patch\n')
     fm.write(b'user', b'# User %s\n', ctx.user())
     fm.plain(b'# Date %d %d\n' % ctx.date())
     fm.write(b'date', b'#      %s\n', fm.formatdate(ctx.date()))
     fm.condwrite(
         branch and branch != b'default', b'branch', b'# Branch %s\n', branch
     )
     fm.write(b'node', b'# Node ID %s\n', hex(node))
     fm.plain(b'# Parent  %s\n' % hex(prev))
     if len(parents) > 1:
         fm.plain(b'# Parent  %s\n' % hex(parents[1]))
     fm.data(parents=fm.formatlist(pycompat.maplist(hex, parents), name=b'node'))
 
     # TODO: redesign extraexportmap function to support formatter
     for headerid in extraexport:
         header = extraexportmap[headerid](seqno, ctx)
         if header is not None:
             fm.plain(b'# %s\n' % header)
 
     fm.write(b'desc', b'%s\n', ctx.description().rstrip())
     fm.plain(b'\n')
 
     if fm.isplain():
         chunkiter = patch.diffui(repo, prev, node, match, opts=diffopts)
         for chunk, label in chunkiter:
             fm.plain(chunk, label=label)
     else:
         chunkiter = patch.diff(repo, prev, node, match, opts=diffopts)
         # TODO: make it structured?
         fm.data(diff=b''.join(chunkiter))
 
 
 def _exportfile(repo, revs, fm, dest, switch_parent, diffopts, match):
     """Export changesets to stdout or a single file"""
     for seqno, rev in enumerate(revs, 1):
         ctx = repo[rev]
         if not dest.startswith(b'<'):
             repo.ui.note(b"%s\n" % dest)
         fm.startitem()
         _exportsingle(repo, ctx, fm, match, switch_parent, seqno, diffopts)
 
 
 def _exportfntemplate(
     repo, revs, basefm, fntemplate, switch_parent, diffopts, match
 ):
     """Export changesets to possibly multiple files"""
     total = len(revs)
     revwidth = max(len(str(rev)) for rev in revs)
     filemap = util.sortdict()  # filename: [(seqno, rev), ...]
 
     for seqno, rev in enumerate(revs, 1):
         ctx = repo[rev]
         dest = makefilename(
             ctx, fntemplate, total=total, seqno=seqno, revwidth=revwidth
         )
         filemap.setdefault(dest, []).append((seqno, rev))
 
     for dest in filemap:
         with formatter.maybereopen(basefm, dest) as fm:
             repo.ui.note(b"%s\n" % dest)
             for seqno, rev in filemap[dest]:
                 fm.startitem()
                 ctx = repo[rev]
                 _exportsingle(
                     repo, ctx, fm, match, switch_parent, seqno, diffopts
                 )
 
 
 def _prefetchchangedfiles(repo, revs, match):
     allfiles = set()
     for rev in revs:
         for file in repo[rev].files():
             if not match or match(file):
                 allfiles.add(file)
-    scmutil.prefetchfiles(repo, revs, scmutil.matchfiles(repo, allfiles))
+    match = scmutil.matchfiles(repo, allfiles)
+    revmatches = [(rev, match) for rev in revs]
+    scmutil.prefetchfiles(repo, revmatches)
 
 
 def export(
     repo,
     revs,
     basefm,
     fntemplate=b'hg-%h.patch',
     switch_parent=False,
     opts=None,
     match=None,
 ):
     '''export changesets as hg patches
 
     Args:
       repo: The repository from which we're exporting revisions.
       revs: A list of revisions to export as revision numbers.
       basefm: A formatter to which patches should be written.
       fntemplate: An optional string to use for generating patch file names.
       switch_parent: If True, show diffs against second parent when not nullid.
                      Default is false, which always shows diff against p1.
       opts: diff options to use for generating the patch.
       match: If specified, only export changes to files matching this matcher.
 
     Returns:
       Nothing.
 
     Side Effect:
       "HG Changeset Patch" data is emitted to one of the following
       destinations:
         fntemplate specified: Each rev is written to a unique file named using
                             the given template.
         Otherwise: All revs will be written to basefm.
     '''
     _prefetchchangedfiles(repo, revs, match)
 
     if not fntemplate:
         _exportfile(
             repo, revs, basefm, b'<unnamed>', switch_parent, opts, match
         )
     else:
         _exportfntemplate(
             repo, revs, basefm, fntemplate, switch_parent, opts, match
         )
 
 
 def exportfile(repo, revs, fp, switch_parent=False, opts=None, match=None):
     """Export changesets to the given file stream"""
     _prefetchchangedfiles(repo, revs, match)
 
     dest = getattr(fp, 'name', b'<unnamed>')
     with formatter.formatter(repo.ui, fp, b'export', {}) as fm:
         _exportfile(repo, revs, fm, dest, switch_parent, opts, match)
 
 
 def showmarker(fm, marker, index=None):
     """utility function to display obsolescence marker in a readable way
 
     To be used by debug function."""
     if index is not None:
         fm.write(b'index', b'%i ', index)
     fm.write(b'prednode', b'%s ', hex(marker.prednode()))
     succs = marker.succnodes()
     fm.condwrite(
         succs,
         b'succnodes',
         b'%s ',
         fm.formatlist(map(hex, succs), name=b'node'),
     )
     fm.write(b'flag', b'%X ', marker.flags())
     parents = marker.parentnodes()
     if parents is not None:
         fm.write(
             b'parentnodes',
             b'{%s} ',
             fm.formatlist(map(hex, parents), name=b'node', sep=b', '),
         )
     fm.write(b'date', b'(%s) ', fm.formatdate(marker.date()))
     meta = marker.metadata().copy()
     meta.pop(b'date', None)
     smeta = pycompat.rapply(pycompat.maybebytestr, meta)
     fm.write(
         b'metadata', b'{%s}', fm.formatdict(smeta, fmt=b'%r: %r', sep=b', ')
     )
     fm.plain(b'\n')
 
 
 def finddate(ui, repo, date):
     """Find the tipmost changeset that matches the given date spec"""
 
     df = dateutil.matchdate(date)
     m = scmutil.matchall(repo)
     results = {}
 
     def prep(ctx, fns):
         d = ctx.date()
         if df(d[0]):
             results[ctx.rev()] = d
 
     for ctx in walkchangerevs(repo, m, {b'rev': None}, prep):
         rev = ctx.rev()
         if rev in results:
             ui.status(
                 _(b"found revision %d from %s\n")
                 % (rev, dateutil.datestr(results[rev]))
             )
             return b'%d' % rev
 
     raise error.Abort(_(b"revision matching date not found"))
 
 
 def increasingwindows(windowsize=8, sizelimit=512):
     while True:
         yield windowsize
         if windowsize < sizelimit:
             windowsize *= 2
 
 
 def _walkrevs(repo, opts):
     # Default --rev value depends on --follow but --follow behavior
     # depends on revisions resolved from --rev...
     follow = opts.get(b'follow') or opts.get(b'follow_first')
     if opts.get(b'rev'):
         revs = scmutil.revrange(repo, opts[b'rev'])
     elif follow and repo.dirstate.p1() == nullid:
         revs = smartset.baseset()
     elif follow:
         revs = repo.revs(b'reverse(:.)')
     else:
         revs = smartset.spanset(repo)
         revs.reverse()
     return revs
 
 
 class FileWalkError(Exception):
     pass
 
 
 def walkfilerevs(repo, match, follow, revs, fncache):
     '''Walks the file history for the matched files.
 
     Returns the changeset revs that are involved in the file history.
 
     Throws FileWalkError if the file history can't be walked using
     filelogs alone.
     '''
     wanted = set()
     copies = []
     minrev, maxrev = min(revs), max(revs)
 
     def filerevs(filelog, last):
         """
         Only files, no patterns.  Check the history of each file.
 
         Examines filelog entries within minrev, maxrev linkrev range
         Returns an iterator yielding (linkrev, parentlinkrevs, copied)
         tuples in backwards order
         """
         cl_count = len(repo)
         revs = []
         for j in pycompat.xrange(0, last + 1):
             linkrev = filelog.linkrev(j)
             if linkrev < minrev:
                 continue
             # only yield rev for which we have the changelog, it can
             # happen while doing "hg log" during a pull or commit
             if linkrev >= cl_count:
                 break
 
             parentlinkrevs = []
             for p in filelog.parentrevs(j):
                 if p != nullrev:
                     parentlinkrevs.append(filelog.linkrev(p))
             n = filelog.node(j)
             revs.append(
                 (linkrev, parentlinkrevs, follow and filelog.renamed(n))
             )
 
         return reversed(revs)
 
     def iterfiles():
         pctx = repo[b'.']
         for filename in match.files():
             if follow:
                 if filename not in pctx:
                     raise error.Abort(
                         _(
                             b'cannot follow file not in parent '
                             b'revision: "%s"'
                         )
                         % filename
                     )
                 yield filename, pctx[filename].filenode()
             else:
                 yield filename, None
         for filename_node in copies:
             yield filename_node
 
     for file_, node in iterfiles():
         filelog = repo.file(file_)
         if not len(filelog):
             if node is None:
                 # A zero count may be a directory or deleted file, so
                 # try to find matching entries on the slow path.
                 if follow:
                     raise error.Abort(
                         _(b'cannot follow nonexistent file: "%s"') % file_
                     )
                 raise FileWalkError(b"Cannot walk via filelog")
             else:
                 continue
 
         if node is None:
             last = len(filelog) - 1
         else:
             last = filelog.rev(node)
 
         # keep track of all ancestors of the file
         ancestors = {filelog.linkrev(last)}
 
         # iterate from latest to oldest revision
         for rev, flparentlinkrevs, copied in filerevs(filelog, last):
             if not follow:
                 if rev > maxrev:
                     continue
             else:
                 # Note that last might not be the first interesting
                 # rev to us:
                 # if the file has been changed after maxrev, we'll
                 # have linkrev(last) > maxrev, and we still need
                 # to explore the file graph
                 if rev not in ancestors:
                     continue
                 # XXX insert 1327 fix here
                 if flparentlinkrevs:
                     ancestors.update(flparentlinkrevs)
 
             fncache.setdefault(rev, []).append(file_)
             wanted.add(rev)
             if copied:
                 copies.append(copied)
 
     return wanted
 
 
 class _followfilter(object):
     def __init__(self, repo, onlyfirst=False):
         self.repo = repo
         self.startrev = nullrev
         self.roots = set()
         self.onlyfirst = onlyfirst
 
     def match(self, rev):
         def realparents(rev):
             if self.onlyfirst:
                 return self.repo.changelog.parentrevs(rev)[0:1]
             else:
                 return filter(
                     lambda x: x != nullrev, self.repo.changelog.parentrevs(rev)
                 )
 
         if self.startrev == nullrev:
             self.startrev = rev
             return True
 
         if rev > self.startrev:
             # forward: all descendants
             if not self.roots:
                 self.roots.add(self.startrev)
             for parent in realparents(rev):
                 if parent in self.roots:
                     self.roots.add(rev)
                     return True
         else:
             # backwards: all parents
             if not self.roots:
                 self.roots.update(realparents(self.startrev))
             if rev in self.roots:
                 self.roots.remove(rev)
                 self.roots.update(realparents(rev))
                 return True
 
         return False
 
 
 def walkchangerevs(repo, match, opts, prepare):
     '''Iterate over files and the revs in which they changed.
 
     Callers most commonly need to iterate backwards over the history
     in which they are interested. Doing so has awful (quadratic-looking)
     performance, so we use iterators in a "windowed" way.
 
     We walk a window of revisions in the desired order.  Within the
     window, we first walk forwards to gather data, then in the desired
     order (usually backwards) to display it.
 
     This function returns an iterator yielding contexts. Before
     yielding each context, the iterator will first call the prepare
     function on each context in the window in forward order.'''
 
     allfiles = opts.get(b'all_files')
     follow = opts.get(b'follow') or opts.get(b'follow_first')
     revs = _walkrevs(repo, opts)
     if not revs:
         return []
     wanted = set()
     slowpath = match.anypats() or (not match.always() and opts.get(b'removed'))
     fncache = {}
     change = repo.__getitem__
 
     # First step is to fill wanted, the set of revisions that we want to yield.
     # When it does not induce extra cost, we also fill fncache for revisions in
     # wanted: a cache of filenames that were changed (ctx.files()) and that
     # match the file filtering conditions.
 
     if match.always() or allfiles:
         # No files, no patterns.  Display all revs.
         wanted = revs
     elif not slowpath:
         # We only have to read through the filelog to find wanted revisions
 
         try:
             wanted = walkfilerevs(repo, match, follow, revs, fncache)
         except FileWalkError:
             slowpath = True
 
             # We decided to fall back to the slowpath because at least one
             # of the paths was not a file. Check to see if at least one of them
             # existed in history, otherwise simply return
             for path in match.files():
                 if path == b'.' or path in repo.store:
                     break
             else:
                 return []
 
     if slowpath:
         # We have to read the changelog to match filenames against
         # changed files
 
         if follow:
             raise error.Abort(
                 _(b'can only follow copies/renames for explicit filenames')
             )
 
         # The slow path checks files modified in every changeset.
         # This is really slow on large repos, so compute the set lazily.
         class lazywantedset(object):
             def __init__(self):
                 self.set = set()
                 self.revs = set(revs)
 
             # No need to worry about locality here because it will be accessed
             # in the same order as the increasing window below.
             def __contains__(self, value):
                 if value in self.set:
                     return True
                 elif not value in self.revs:
                     return False
                 else:
                     self.revs.discard(value)
                     ctx = change(value)
                     if allfiles:
                         matches = list(ctx.manifest().walk(match))
                     else:
                         matches = [f for f in ctx.files() if match(f)]
                     if matches:
                         fncache[value] = matches
                         self.set.add(value)
                         return True
                     return False
 
             def discard(self, value):
                 self.revs.discard(value)
                 self.set.discard(value)
 
         wanted = lazywantedset()
 
     # it might be worthwhile to do this in the iterator if the rev range
     # is descending and the prune args are all within that range
     for rev in opts.get(b'prune', ()):
         rev = repo[rev].rev()
         ff = _followfilter(repo)
         stop = min(revs[0], revs[-1])
         for x in pycompat.xrange(rev, stop - 1, -1):
             if ff.match(x):
                 wanted = wanted - [x]
 
     # Now that wanted is correctly initialized, we can iterate over the
     # revision range, yielding only revisions in wanted.
     def iterate():
         if follow and match.always():
             ff = _followfilter(repo, onlyfirst=opts.get(b'follow_first'))
 
             def want(rev):
                 return ff.match(rev) and rev in wanted
 
         else:
 
             def want(rev):
                 return rev in wanted
 
         it = iter(revs)
         stopiteration = False
         for windowsize in increasingwindows():
             nrevs = []
             for i in pycompat.xrange(windowsize):
                 rev = next(it, None)
                 if rev is None:
                     stopiteration = True
                     break
                 elif want(rev):
                     nrevs.append(rev)
             for rev in sorted(nrevs):
                 fns = fncache.get(rev)
                 ctx = change(rev)
                 if not fns:
 
                     def fns_generator():
                         if allfiles:
 
                             def bad(f, msg):
                                 pass
 
                             for f in ctx.matches(matchmod.badmatch(match, bad)):
                                 yield f
                         else:
                             for f in ctx.files():
                                 if match(f):
                                     yield f
 
                     fns = fns_generator()
                 prepare(ctx, fns)
             for rev in nrevs:
                 yield change(rev)
 
             if stopiteration:
                 break
 
     return iterate()
 
 
 def add(ui, repo, match, prefix, uipathfn, explicitonly, **opts):
     bad = []
 
     badfn = lambda x, y: bad.append(x) or match.bad(x, y)
     names = []
     wctx = repo[None]
     cca = None
     abort, warn = scmutil.checkportabilityalert(ui)
     if abort or warn:
         cca = scmutil.casecollisionauditor(ui, abort, repo.dirstate)
 
     match = repo.narrowmatch(match, includeexact=True)
     badmatch = matchmod.badmatch(match, badfn)
     dirstate = repo.dirstate
     # We don't want to just call wctx.walk here, since it would return a lot of
     # clean files, which we aren't interested in and takes time.
     for f in sorted(
         dirstate.walk(
             badmatch,
             subrepos=sorted(wctx.substate),
             unknown=True,
             ignored=False,
             full=False,
         )
     ):
         exact = match.exact(f)
         if exact or not explicitonly and f not in wctx and repo.wvfs.lexists(f):
             if cca:
                 cca(f)
             names.append(f)
             if ui.verbose or not exact:
                 ui.status(
                     _(b'adding %s\n') % uipathfn(f), label=b'ui.addremove.added'
                 )
 
     for subpath in sorted(wctx.substate):
         sub = wctx.sub(subpath)
         try:
             submatch = matchmod.subdirmatcher(subpath, match)
             subprefix = repo.wvfs.reljoin(prefix, subpath)
             subuipathfn = scmutil.subdiruipathfn(subpath, uipathfn)
             if opts.get('subrepos'):
                 bad.extend(
                     sub.add(ui, submatch, subprefix, subuipathfn, False, **opts)
                 )
             else:
                 bad.extend(
                     sub.add(ui, submatch, subprefix, subuipathfn, True, **opts)
                 )
         except error.LookupError:
             ui.status(
                 _(b"skipping missing subrepository: %s\n") % uipathfn(subpath)
             )
 
     if not opts.get('dry_run'):
         rejected = wctx.add(names, prefix)
         bad.extend(f for f in rejected if f in match.files())
     return bad
 
 
 def addwebdirpath(repo, serverpath, webconf):
     webconf[serverpath] = repo.root
     repo.ui.debug(b'adding %s = %s\n' % (serverpath, repo.root))
 
     for r in repo.revs(b'filelog("path:.hgsub")'):
         ctx = repo[r]
         for subpath in ctx.substate:
             ctx.sub(subpath).addwebdirpath(serverpath, webconf)
 
 
 def forget(
     ui, repo, match, prefix, uipathfn, explicitonly, dryrun, interactive
 ):
     if dryrun and interactive:
         raise error.Abort(_(b"cannot specify both --dry-run and --interactive"))
     bad = []
     badfn = lambda x, y: bad.append(x) or match.bad(x, y)
     wctx = repo[None]
     forgot = []
 
     s = repo.status(match=matchmod.badmatch(match, badfn), clean=True)
     forget = sorted(s.modified + s.added + s.deleted + s.clean)
     if explicitonly:
         forget = [f for f in forget if match.exact(f)]
 
     for subpath in sorted(wctx.substate):
         sub = wctx.sub(subpath)
         submatch = matchmod.subdirmatcher(subpath, match)
         subprefix = repo.wvfs.reljoin(prefix, subpath)
         subuipathfn = scmutil.subdiruipathfn(subpath, uipathfn)
         try:
             subbad, subforgot = sub.forget(
                 submatch,
                 subprefix,
                 subuipathfn,
                 dryrun=dryrun,
                 interactive=interactive,
             )
             bad.extend([subpath + b'/' + f for f in subbad])
             forgot.extend([subpath + b'/' + f for f in subforgot])
         except error.LookupError:
             ui.status(
                 _(b"skipping missing subrepository: %s\n") % uipathfn(subpath)
             )
 
     if not explicitonly:
         for f in match.files():
             if f not in repo.dirstate and not repo.wvfs.isdir(f):
                 if f not in forgot:
                     if repo.wvfs.exists(f):
                         # Don't complain if the exact case match wasn't given.
                         # But don't do this until after checking 'forgot', so
                         # that subrepo files aren't normalized, and this op is
                         # purely from data cached by the status walk above.
                         if repo.dirstate.normalize(f) in repo.dirstate:
                             continue
                         ui.warn(
                             _(
                                 b'not removing %s: '
                                 b'file is already untracked\n'
                             )
                             % uipathfn(f)
                         )
                     bad.append(f)
 
     if interactive:
         responses = _(
             b'[Ynsa?]'
             b'$$ &Yes, forget this file'
             b'$$ &No, skip this file'
             b'$$ &Skip remaining files'
             b'$$ Include &all remaining files'
             b'$$ &? (display help)'
         )
         for filename in forget[:]:
             r = ui.promptchoice(
                 _(b'forget %s %s') % (uipathfn(filename), responses)
             )
             if r == 4:  # ?
                 while r == 4:
                     for c, t in ui.extractchoices(responses)[1]:
                         ui.write(b'%s - %s\n' % (c, encoding.lower(t)))
                     r = ui.promptchoice(
                         _(b'forget %s %s') % (uipathfn(filename), responses)
                     )
             if r == 0:  # yes
                 continue
             elif r == 1:  # no
                 forget.remove(filename)
             elif r == 2:  # Skip
                 fnindex = forget.index(filename)
                 del forget[fnindex:]
                 break
             elif r == 3:  # All
                 break
 
     for f in forget:
         if ui.verbose or not match.exact(f) or interactive:
             ui.status(
                 _(b'removing %s\n') % uipathfn(f), label=b'ui.addremove.removed'
             )
 
     if not dryrun:
         rejected = wctx.forget(forget, prefix)
         bad.extend(f for f in rejected if f in match.files())
         forgot.extend(f for f in forget if f not in rejected)
     return bad, forgot
 
 
 def files(ui, ctx, m, uipathfn, fm, fmt, subrepos):
     ret = 1
 
     needsfctx = ui.verbose or {b'size', b'flags'} & fm.datahint()
     if fm.isplain() and not needsfctx:
         # Fast path. The speed-up comes from skipping the formatter, and batching
         # calls to ui.write.
         buf = []
         for f in ctx.matches(m):
             buf.append(fmt % uipathfn(f))
             if len(buf) > 100:
                 ui.write(b''.join(buf))
                 del buf[:]
             ret = 0
         if buf:
             ui.write(b''.join(buf))
     else:
         for f in ctx.matches(m):
             fm.startitem()
             fm.context(ctx=ctx)
             if needsfctx:
                 fc = ctx[f]
                 fm.write(b'size flags', b'% 10d % 1s ', fc.size(), fc.flags())
             fm.data(path=f)
             fm.plain(fmt % uipathfn(f))
             ret = 0
 
     for subpath in sorted(ctx.substate):
         submatch = matchmod.subdirmatcher(subpath, m)
         subuipathfn = scmutil.subdiruipathfn(subpath, uipathfn)
         if subrepos or m.exact(subpath) or any(submatch.files()):
             sub = ctx.sub(subpath)
             try:
                 recurse = m.exact(subpath) or subrepos
                 if (
                     sub.printfiles(ui, submatch, subuipathfn, fm, fmt, recurse)
                     == 0
                 ):
                     ret = 0
             except error.LookupError:
                 ui.status(
                     _(b"skipping missing subrepository: %s\n")
                     % uipathfn(subpath)
                 )
 
     return ret
 
 
 def remove(
     ui, repo, m, prefix, uipathfn, after, force, subrepos, dryrun, warnings=None
 ):
     ret = 0
     s = repo.status(match=m, clean=True)
     modified, added, deleted, clean = s.modified, s.added, s.deleted, s.clean
 
     wctx = repo[None]
 
     if warnings is None:
         warnings = []
         warn = True
     else:
         warn = False
 
     subs = sorted(wctx.substate)
     progress = ui.makeprogress(
         _(b'searching'), total=len(subs), unit=_(b'subrepos')
     )
     for subpath in subs:
         submatch = matchmod.subdirmatcher(subpath, m)
         subprefix = repo.wvfs.reljoin(prefix, subpath)
         subuipathfn = scmutil.subdiruipathfn(subpath, uipathfn)
         if subrepos or m.exact(subpath) or any(submatch.files()):
             progress.increment()
             sub = wctx.sub(subpath)
             try:
                 if sub.removefiles(
                     submatch,
                     subprefix,
                     subuipathfn,
                     after,
                     force,
                     subrepos,
                     dryrun,
                     warnings,
                 ):
                     ret = 1
             except error.LookupError:
                 warnings.append(
                     _(b"skipping missing subrepository: %s\n")
                     % uipathfn(subpath)
                 )
     progress.complete()
 
     # warn about failure to delete explicit files/dirs
     deleteddirs = pathutil.dirs(deleted)
     files = m.files()
     progress = ui.makeprogress(
         _(b'deleting'), total=len(files), unit=_(b'files')
     )
     for f in files:
 
         def insubrepo():
             for subpath in wctx.substate:
                 if f.startswith(subpath + b'/'):
                     return True
             return False
 
         progress.increment()
         isdir = f in deleteddirs or wctx.hasdir(f)
         if f in repo.dirstate or isdir or f == b'.' or insubrepo() or f in subs:
             continue
 
         if repo.wvfs.exists(f):
             if repo.wvfs.isdir(f):
                 warnings.append(
                     _(b'not removing %s: no tracked files\n') % uipathfn(f)
                 )
             else:
                 warnings.append(
                     _(b'not removing %s: file is untracked\n') % uipathfn(f)
                 )
         # missing files will generate a warning elsewhere
         ret = 1
     progress.complete()
 
     if force:
         list = modified + deleted + clean + added
     elif after:
         list = deleted
         remaining = modified + added + clean
         progress = ui.makeprogress(
             _(b'skipping'), total=len(remaining), unit=_(b'files')
         )
         for f in remaining:
             progress.increment()
             if ui.verbose or (f in files):
                 warnings.append(
                     _(b'not removing %s: file still exists\n') % uipathfn(f)
                 )
             ret = 1
         progress.complete()
     else:
         list = deleted + clean
         progress = ui.makeprogress(
             _(b'skipping'), total=(len(modified) + len(added)), unit=_(b'files')
         )
         for f in modified:
             progress.increment()
             warnings.append(
                 _(
                     b'not removing %s: file is modified (use -f'
                     b' to force removal)\n'
                 )
                 % uipathfn(f)
             )
             ret = 1
         for f in added:
             progress.increment()
             warnings.append(
                 _(
                     b"not removing %s: file has been marked for add"
                     b" (use 'hg forget' to undo add)\n"
                 )
                 % uipathfn(f)
             )
             ret = 1
         progress.complete()
 
     list = sorted(list)
     progress = ui.makeprogress(
         _(b'deleting'), total=len(list), unit=_(b'files')
     )
     for f in list:
         if ui.verbose or not m.exact(f):
             progress.increment()
             ui.status(
                 _(b'removing %s\n') % uipathfn(f), label=b'ui.addremove.removed'
             )
     progress.complete()
 
     if not dryrun:
         with repo.wlock():
             if not after:
                 for f in list:
                     if f in added:
                         continue  # we never unlink added files on remove
                     rmdir = repo.ui.configbool(
                         b'experimental', b'removeemptydirs'
                     )
                     repo.wvfs.unlinkpath(f, ignoremissing=True, rmdir=rmdir)
             repo[None].forget(list)
 
     if warn:
         for warning in warnings:
             ui.warn(warning)
 
     return ret
 
 
 def _catfmtneedsdata(fm):
     return not fm.datahint() or b'data' in fm.datahint()
 
 
 def _updatecatformatter(fm, ctx, matcher, path, decode):
     """Hook for adding data to the formatter used by ``hg cat``.
 
     Extensions (e.g., lfs) can wrap this to inject keywords/data, but must call
     this method first."""
 
     # data() can be expensive to fetch (e.g. lfs), so don't fetch it if it
     # wasn't requested.
     data = b''
     if _catfmtneedsdata(fm):
         data = ctx[path].data()
         if decode:
             data = ctx.repo().wwritedata(path, data)
     fm.startitem()
     fm.context(ctx=ctx)
     fm.write(b'data', b'%s', data)
     fm.data(path=path)
 
 
 def cat(ui, repo, ctx, matcher, basefm, fntemplate, prefix, **opts):
     err = 1
     opts = pycompat.byteskwargs(opts)
 
     def write(path):
         filename = None
         if fntemplate:
             filename = makefilename(
                 ctx, fntemplate, pathname=os.path.join(prefix, path)
             )
             # attempt to create the directory if it does not already exist
             try:
                 os.makedirs(os.path.dirname(filename))
             except OSError:
                 pass
         with formatter.maybereopen(basefm, filename) as fm:
             _updatecatformatter(fm, ctx, matcher, path, opts.get(b'decode'))
 
     # Automation often uses hg cat on single files, so special case it
     # for performance to avoid the cost of parsing the manifest.
     if len(matcher.files()) == 1 and not matcher.anypats():
         file = matcher.files()[0]
         mfl = repo.manifestlog
         mfnode = ctx.manifestnode()
         try:
             if mfnode and mfl[mfnode].find(file)[0]:
                 if _catfmtneedsdata(basefm):
-                    scmutil.prefetchfiles(repo, [ctx.rev()], matcher)
+                    scmutil.prefetchfiles(repo, [(ctx.rev(), matcher)])
                 write(file)
                 return 0
         except KeyError:
             pass
 
     if _catfmtneedsdata(basefm):
-        scmutil.prefetchfiles(repo, [ctx.rev()], matcher)
+        scmutil.prefetchfiles(repo, [(ctx.rev(), matcher)])
 
     for abs in ctx.walk(matcher):
         write(abs)
         err = 0
 
     uipathfn = scmutil.getuipathfn(repo, legacyrelativevalue=True)
     for subpath in sorted(ctx.substate):
         sub = ctx.sub(subpath)
         try:
             submatch = matchmod.subdirmatcher(subpath, matcher)
             subprefix = os.path.join(prefix, subpath)
             if not sub.cat(
                 submatch,
                 basefm,
                 fntemplate,
                 subprefix,
                 **pycompat.strkwargs(opts)
             ):
                 err = 0
         except error.RepoLookupError:
             ui.status(
                 _(b"skipping missing subrepository: %s\n") % uipathfn(subpath)
             )
 
     return err
 
 
 def commit(ui, repo, commitfunc, pats, opts):
     '''commit the specified files or all outstanding changes'''
     date = opts.get(b'date')
     if date:
         opts[b'date'] = dateutil.parsedate(date)
     message = logmessage(ui, opts)
     matcher = scmutil.match(repo[None], pats, opts)
 
     dsguard = None
     # extract addremove carefully -- this function can be called from a command
     # that doesn't support addremove
     if opts.get(b'addremove'):
         dsguard = dirstateguard.dirstateguard(repo, b'commit')
     with dsguard or util.nullcontextmanager():
         if dsguard:
             relative = scmutil.anypats(pats, opts)
             uipathfn = scmutil.getuipathfn(repo, legacyrelativevalue=relative)
             if scmutil.addremove(repo, matcher, b"", uipathfn, opts) != 0:
                 raise error.Abort(
                     _(b"failed to mark all new/missing files as added/removed")
                 )
 
         return commitfunc(ui, repo, message, matcher, opts)
 
 
 def samefile(f, ctx1, ctx2):
     if f in ctx1.manifest():
         a = ctx1.filectx(f)
         if f in ctx2.manifest():
             b = ctx2.filectx(f)
             return not a.cmp(b) and a.flags() == b.flags()
         else:
             return False
     else:
         return f not in ctx2.manifest()
 
 
 def amend(ui, repo, old, extra, pats, opts):
     # avoid cycle context -> subrepo -> cmdutil
     from . import context
 
     # amend will reuse the existing user if not specified, but the obsolete
     # marker creation requires that the current user's name is specified.
     if obsolete.isenabled(repo, obsolete.createmarkersopt):
         ui.username()  # raise exception if username not set
 
     ui.note(_(b'amending changeset %s\n') % old)
     base = old.p1()
 
     with repo.wlock(), repo.lock(), repo.transaction(b'amend'):
         # Participating changesets:
         #
         # wctx     o - workingctx that contains changes from working copy
         #          |   to go into amending commit
         #          |
         # old      o - changeset to amend
         #          |
         # base     o - first parent of the changeset to amend
         wctx = repo[None]
 
         # Copy to avoid mutating input
         extra = extra.copy()
         # Update extra dict from amended commit (e.g. to preserve graft
         # source)
         extra.update(old.extra())
 
         # Also update it from the from the wctx
         extra.update(wctx.extra())
 
         # date-only change should be ignored?
         datemaydiffer = resolvecommitoptions(ui, opts)
 
         date = old.date()
         if opts.get(b'date'):
             date = dateutil.parsedate(opts.get(b'date'))
         user = opts.get(b'user') or old.user()
 
         if len(old.parents()) > 1:
             # ctx.files() isn't reliable for merges, so fall back to the
             # slower repo.status() method
             st = base.status(old)
             files = set(st.modified) | set(st.added) | set(st.removed)
         else:
             files = set(old.files())
 
         # add/remove the files to the working copy if the "addremove" option
         # was specified.
         matcher = scmutil.match(wctx, pats, opts)
         relative = scmutil.anypats(pats, opts)
         uipathfn = scmutil.getuipathfn(repo, legacyrelativevalue=relative)
         if opts.get(b'addremove') and scmutil.addremove(
             repo, matcher, b"", uipathfn, opts
         ):
             raise error.Abort(
                 _(b"failed to mark all new/missing files as added/removed")
             )
 
         # Check subrepos. This depends on in-place wctx._status update in
         # subrepo.precommit(). To minimize the risk of this hack, we do
         # nothing if .hgsub does not exist.
         if b'.hgsub' in wctx or b'.hgsub' in old:
             subs, commitsubs, newsubstate = subrepoutil.precommit(
                 ui, wctx, wctx._status, matcher
             )
             # amend should abort if commitsubrepos is enabled
             assert not commitsubs
             if subs:
                 subrepoutil.writestate(repo, newsubstate)
 
         ms = mergestatemod.mergestate.read(repo)
         mergeutil.checkunresolved(ms)
 
         filestoamend = {f for f in wctx.files() if matcher(f)}
 
         changes = len(filestoamend) > 0
         if changes:
             # Recompute copies (avoid recording a -> b -> a)
             copied = copies.pathcopies(base, wctx, matcher)
             if old.p2:
                 copied.update(copies.pathcopies(old.p2(), wctx, matcher))
 
             # Prune files which were reverted by the updates: if old
             # introduced file X and the file was renamed in the working
             # copy, then those two files are the same and
             # we can discard X from our list of files. Likewise if X
             # was removed, it's no longer relevant. If X is missing (aka
             # deleted), old X must be preserved.
             files.update(filestoamend)
             files = [
                 f
                 for f in files
                 if (f not in filestoamend or not samefile(f, wctx, base))
             ]
 
             def filectxfn(repo, ctx_, path):
                 try:
                     # If the file being considered is not amongst the files
                     # to be amended, we should return the file context from the
                     # old changeset. This avoids issues when only some files in
                     # the working copy are being amended but there are also
                     # changes to other files from the old changeset.
                     if path not in filestoamend:
                         return old.filectx(path)
 
                     # Return None for removed files.
                     if path in wctx.removed():
                         return None
 
                     fctx = wctx[path]
                     flags = fctx.flags()
                     mctx = context.memfilectx(
                         repo,
                         ctx_,
                         fctx.path(),
                         fctx.data(),
                         islink=b'l' in flags,
                         isexec=b'x' in flags,
                         copysource=copied.get(path),
                     )
                     return mctx
                 except KeyError:
                     return None
 
         else:
             ui.note(_(b'copying changeset %s to %s\n') % (old, base))
 
             # Use version of files as in the old cset
             def filectxfn(repo, ctx_, path):
                 try:
                     return old.filectx(path)
                 except KeyError:
                     return None
 
         # See if we got a message from -m or -l, if not, open the editor with
         # the message of the changeset to amend.
         message = logmessage(ui, opts)
 
         editform = mergeeditform(old, b'commit.amend')
 
         if not message:
             message = old.description()
             # Default if message isn't provided and --edit is not passed is to
             # invoke editor, but allow --no-edit. If somehow we don't have any
             # description, let's always start the editor.
             doedit = not message or opts.get(b'edit') in [True, None]
         else:
             # Default if message is provided is to not invoke editor, but allow
             # --edit.
             doedit = opts.get(b'edit') is True
         editor = getcommiteditor(edit=doedit, editform=editform)
 
         pureextra = extra.copy()
         extra[b'amend_source'] = old.hex()
 
         new = context.memctx(
             repo,
             parents=[base.node(), old.p2().node()],
             text=message,
             files=files,
             filectxfn=filectxfn,
             user=user,
             date=date,
             extra=extra,
             editor=editor,
         )
 
         newdesc = changelog.stripdesc(new.description())
         if (
             (not changes)
             and newdesc == old.description()
             and user == old.user()
             and (date == old.date() or datemaydiffer)
             and pureextra == old.extra()
         ):
             # nothing changed. continuing here would create a new node
             # anyway because of the amend_source noise.
             #
             # This not what we expect from amend.
             return old.node()
 
         commitphase = None
         if opts.get(b'secret'):
             commitphase = phases.secret
         newid = repo.commitctx(new)
 
         # Reroute the working copy parent to the new changeset
         repo.setparents(newid, nullid)
         mapping = {old.node(): (newid,)}
         obsmetadata = None
         if opts.get(b'note'):
             obsmetadata = {b'note': encoding.fromlocal(opts[b'note'])}
         backup = ui.configbool(b'rewrite', b'backup-bundle')
         scmutil.cleanupnodes(
             repo,
             mapping,
             b'amend',
             metadata=obsmetadata,
             fixphase=True,
             targetphase=commitphase,
             backup=backup,
         )
 
         # Fixing the dirstate because localrepo.commitctx does not update
         # it. This is rather convenient because we did not need to update
         # the dirstate for all the files in the new commit which commitctx
         # could have done if it updated the dirstate. Now, we can
         # selectively update the dirstate only for the amended files.
         dirstate = repo.dirstate
 
         # Update the state of the files which were added and modified in the
         # amend to "normal" in the dirstate. We need to use "normallookup" since
         # the files may have changed since the command started; using "normal"
         # would mark them as clean but with uncommitted contents.
         normalfiles = set(wctx.modified() + wctx.added()) & filestoamend
         for f in normalfiles:
             dirstate.normallookup(f)
 
         # Update the state of files which were removed in the amend
         # to "removed" in the dirstate.
         removedfiles = set(wctx.removed()) & filestoamend
         for f in removedfiles:
             dirstate.drop(f)
 
     return newid
 
 
 def commiteditor(repo, ctx, subs, editform=b''):
     if ctx.description():
         return ctx.description()
     return commitforceeditor(
         repo, ctx, subs, editform=editform, unchangedmessagedetection=True
     )
 
 
 def commitforceeditor(
     repo,
     ctx,
     subs,
     finishdesc=None,
     extramsg=None,
     editform=b'',
     unchangedmessagedetection=False,
 ):
     if not extramsg:
         extramsg = _(b"Leave message empty to abort commit.")
 
     forms = [e for e in editform.split(b'.') if e]
     forms.insert(0, b'changeset')
     templatetext = None
     while forms:
         ref = b'.'.join(forms)
         if repo.ui.config(b'committemplate', ref):
             templatetext = committext = buildcommittemplate(
                 repo, ctx, subs, extramsg, ref
             )
             break
         forms.pop()
     else:
         committext = buildcommittext(repo, ctx, subs, extramsg)
 
     # run editor in the repository root
     olddir = encoding.getcwd()
     os.chdir(repo.root)
 
     # make in-memory changes visible to external process
     tr = repo.currenttransaction()
     repo.dirstate.write(tr)
     pending = tr and tr.writepending() and repo.root
 
     editortext = repo.ui.edit(
         committext,
         ctx.user(),
         ctx.extra(),
         editform=editform,
         pending=pending,
         repopath=repo.path,
         action=b'commit',
     )
     text = editortext
 
     # strip away anything below this special string (used for editors that want
     # to display the diff)
     stripbelow = re.search(_linebelow, text, flags=re.MULTILINE)
     if stripbelow:
         text = text[: stripbelow.start()]
 
     text = re.sub(b"(?m)^HG:.*(\n|$)", b"", text)
     os.chdir(olddir)
 
     if finishdesc:
         text = finishdesc(text)
     if not text.strip():
         raise error.Abort(_(b"empty commit message"))
     if unchangedmessagedetection and editortext == templatetext:
         raise error.Abort(_(b"commit message unchanged"))
 
     return text
 
 
 def buildcommittemplate(repo, ctx, subs, extramsg, ref):
     ui = repo.ui
     spec = formatter.templatespec(ref, None, None)
     t = logcmdutil.changesettemplater(ui, repo, spec)
     t.t.cache.update(
         (k, templater.unquotestring(v))
         for k, v in repo.ui.configitems(b'committemplate')
     )
 
     if not extramsg:
         extramsg = b''  # ensure that extramsg is string
 
     ui.pushbuffer()
     t.show(ctx, extramsg=extramsg)
     return ui.popbuffer()
 
 
 def hgprefix(msg):
     return b"\n".join([b"HG: %s" % a for a in msg.split(b"\n") if a])
 
 
 def buildcommittext(repo, ctx, subs, extramsg):
     edittext = []
     modified, added, removed = ctx.modified(), ctx.added(), ctx.removed()
     if ctx.description():
         edittext.append(ctx.description())
     edittext.append(b"")
     edittext.append(b"")  # Empty line between message and comments.
     edittext.append(
         hgprefix(
             _(
                 b"Enter commit message."
                 b"  Lines beginning with 'HG:' are removed."
             )
         )
     )
     edittext.append(hgprefix(extramsg))
     edittext.append(b"HG: --")
     edittext.append(hgprefix(_(b"user: %s") % ctx.user()))
     if ctx.p2():
         edittext.append(hgprefix(_(b"branch merge")))
     if ctx.branch():
         edittext.append(hgprefix(_(b"branch '%s'") % ctx.branch()))
     if bookmarks.isactivewdirparent(repo):
         edittext.append(hgprefix(_(b"bookmark '%s'") % repo._activebookmark))
     edittext.extend([hgprefix(_(b"subrepo %s") % s) for s in subs])
     edittext.extend([hgprefix(_(b"added %s") % f) for f in added])
     edittext.extend([hgprefix(_(b"changed %s") % f) for f in modified])
     edittext.extend([hgprefix(_(b"removed %s") % f) for f in removed])
     if not added and not modified and not removed:
         edittext.append(hgprefix(_(b"no files changed")))
     edittext.append(b"")
 
     return b"\n".join(edittext)
 
 
 def commitstatus(repo, node, branch, bheads=None, opts=None):
     if opts is None:
         opts = {}
     ctx = repo[node]
     parents = ctx.parents()
 
     if (
         not opts.get(b'amend')
         and bheads
         and node not in bheads
         and not any(
             p.node() in bheads and p.branch() == branch for p in parents
         )
     ):
         repo.ui.status(_(b'created new head\n'))
         # The message is not printed for initial roots. For the other
         # changesets, it is printed in the following situations:
         #
         # Par column: for the 2 parents with ...
         #   N: null or no parent
         #   B: parent is on another named branch
         #   C: parent is a regular non head changeset
         #   H: parent was a branch head of the current branch
         # Msg column: whether we print "created new head" message
         # In the following, it is assumed that there already exists some
         # initial branch heads of the current branch, otherwise nothing is
         # printed anyway.
         #
         # Par Msg Comment
         # N N  y  additional topo root
         #
         # B N  y  additional branch root
         # C N  y  additional topo head
         # H N  n  usual case
         #
         # B B  y  weird additional branch root
         # C B  y  branch merge
         # H B  n  merge with named branch
         #
         # C C  y  additional head from merge
         # C H  n  merge with a head
         #
         # H H  n  head merge: head count decreases
 
     if not opts.get(b'close_branch'):
         for r in parents:
             if r.closesbranch() and r.branch() == branch:
                 repo.ui.status(
                     _(b'reopening closed branch head %d\n') % r.rev()
                 )
 
     if repo.ui.debugflag:
         repo.ui.write(
             _(b'committed changeset %d:%s\n') % (ctx.rev(), ctx.hex())
         )
     elif repo.ui.verbose:
         repo.ui.write(_(b'committed changeset %d:%s\n') % (ctx.rev(), ctx))
 
 
 def postcommitstatus(repo, pats, opts):
     return repo.status(match=scmutil.match(repo[None], pats, opts))
 
 
 def revert(ui, repo, ctx, parents, *pats, **opts):
     opts = pycompat.byteskwargs(opts)
     parent, p2 = parents
     node = ctx.node()
 
     mf = ctx.manifest()
     if node == p2:
         parent = p2
 
     # need all matching names in dirstate and manifest of target rev,
     # so have to walk both. do not print errors if files exist in one
     # but not other. in both cases, filesets should be evaluated against
     # workingctx to get consistent result (issue4497). this means 'set:**'
     # cannot be used to select missing files from target rev.
 
     # `names` is a mapping for all elements in working copy and target revision
     # The mapping is in the form:
     #   <abs path in repo> -> (<path from CWD>, <exactly specified by matcher?>)
     names = {}
     uipathfn = scmutil.getuipathfn(repo, legacyrelativevalue=True)
 
     with repo.wlock():
         ## filling of the `names` mapping
         # walk dirstate to fill `names`
 
         interactive = opts.get(b'interactive', False)
         wctx = repo[None]
         m = scmutil.match(wctx, pats, opts)
 
         # we'll need this later
         targetsubs = sorted(s for s in wctx.substate if m(s))
 
         if not m.always():
             matcher = matchmod.badmatch(m, lambda x, y: False)
             for abs in wctx.walk(matcher):
                 names[abs] = m.exact(abs)
 
             # walk target manifest to fill `names`
 
             def badfn(path, msg):
                 if path in names:
                     return
                 if path in ctx.substate:
                     return
                 path_ = path + b'/'
                 for f in names:
                     if f.startswith(path_):
                         return
                 ui.warn(b"%s: %s\n" % (uipathfn(path), msg))
 
             for abs in ctx.walk(matchmod.badmatch(m, badfn)):
                 if abs not in names:
                     names[abs] = m.exact(abs)
 
             # Find status of all file in `names`.
             m = scmutil.matchfiles(repo, names)
 
             changes = repo.status(
                 node1=node, match=m, unknown=True, ignored=True, clean=True
             )
         else:
             changes = repo.status(node1=node, match=m)
             for kind in changes:
                 for abs in kind:
                     names[abs] = m.exact(abs)
 
             m = scmutil.matchfiles(repo, names)
 
         modified = set(changes.modified)
         added = set(changes.added)
         removed = set(changes.removed)
         _deleted = set(changes.deleted)
         unknown = set(changes.unknown)
         unknown.update(changes.ignored)
         clean = set(changes.clean)
         modadded = set()
 
         # We need to account for the state of the file in the dirstate,
         # even when we revert against something else than parent. This will
         # slightly alter the behavior of revert (doing back up or not, delete
         # or just forget etc).
         if parent == node:
             dsmodified = modified
             dsadded = added
             dsremoved = removed
             # store all local modifications, useful later for rename detection
             localchanges = dsmodified | dsadded
             modified, added, removed = set(), set(), set()
         else:
             changes = repo.status(node1=parent, match=m)
             dsmodified = set(changes.modified)
             dsadded = set(changes.added)
             dsremoved = set(changes.removed)
             # store all local modifications, useful later for rename detection
             localchanges = dsmodified | dsadded
 
             # only take into account for removes between wc and target
             clean |= dsremoved - removed
             dsremoved &= removed
             # distinct between dirstate remove and other
             removed -= dsremoved
 
             modadded = added & dsmodified
             added -= modadded
 
             # tell newly modified apart.
             dsmodified &= modified
             dsmodified |= modified & dsadded  # dirstate added may need backup
             modified -= dsmodified
 
             # We need to wait for some post-processing to update this set
             # before making the distinction. The dirstate will be used for
             # that purpose.
             dsadded = added
 
         # in case of merge, files that are actually added can be reported as
         # modified, we need to post process the result
         if p2 != nullid:
             mergeadd = set(dsmodified)
             for path in dsmodified:
                 if path in mf:
                     mergeadd.remove(path)
             dsadded |= mergeadd
             dsmodified -= mergeadd
 
         # if f is a rename, update `names` to also revert the source
         for f in localchanges:
             src = repo.dirstate.copied(f)
             # XXX should we check for rename down to target node?
             if src and src not in names and repo.dirstate[src] == b'r':
                 dsremoved.add(src)
                 names[src] = True
 
         # determine the exact nature of the deleted changesets
         deladded = set(_deleted)
         for path in _deleted:
             if path in mf:
                 deladded.remove(path)
         deleted = _deleted - deladded
 
         # distinguish between file to forget and the other
         added = set()
         for abs in dsadded:
             if repo.dirstate[abs] != b'a':
                 added.add(abs)
         dsadded -= added
 
         for abs in deladded:
             if repo.dirstate[abs] == b'a':
                 dsadded.add(abs)
         deladded -= dsadded
 
         # For files marked as removed, we check if an unknown file is present at
         # the same path. If a such file exists it may need to be backed up.
         # Making the distinction at this stage helps have simpler backup
         # logic.
         removunk = set()
         for abs in removed:
             target = repo.wjoin(abs)
             if os.path.lexists(target):
                 removunk.add(abs)
         removed -= removunk
 
         dsremovunk = set()
         for abs in dsremoved:
             target = repo.wjoin(abs)
             if os.path.lexists(target):
                 dsremovunk.add(abs)
         dsremoved -= dsremovunk
 
         # action to be actually performed by revert
         # (<list of file>, message>) tuple
         actions = {
             b'revert': ([], _(b'reverting %s\n')),
             b'add': ([], _(b'adding %s\n')),
             b'remove': ([], _(b'removing %s\n')),
             b'drop': ([], _(b'removing %s\n')),
             b'forget': ([], _(b'forgetting %s\n')),
             b'undelete': ([], _(b'undeleting %s\n')),
             b'noop': (None, _(b'no changes needed to %s\n')),
             b'unknown': (None, _(b'file not managed: %s\n')),
         }
 
         # "constant" that convey the backup strategy.
         # All set to `discard` if `no-backup` is set do avoid checking
         # no_backup lower in the code.
         # These values are ordered for comparison purposes
         backupinteractive = 3  # do backup if interactively modified
         backup = 2  # unconditionally do backup
         check = 1  # check if the existing file differs from target
         discard = 0  # never do backup
         if opts.get(b'no_backup'):
             backupinteractive = backup = check = discard
         if interactive:
             dsmodifiedbackup = backupinteractive
         else:
             dsmodifiedbackup = backup
         tobackup = set()
 
         backupanddel = actions[b'remove']
         if not opts.get(b'no_backup'):
             backupanddel = actions[b'drop']
 
         disptable = (
             # dispatch table:
             #   file state
             #   action
             #   make backup
             ## Sets that results that will change file on disk
             # Modified compared to target, no local change
             (modified, actions[b'revert'], discard),
             # Modified compared to target, but local file is deleted
             (deleted, actions[b'revert'], discard),
             # Modified compared to target, local change
             (dsmodified, actions[b'revert'], dsmodifiedbackup),
             # Added since target
             (added, actions[b'remove'], discard),
             # Added in working directory
             (dsadded, actions[b'forget'], discard),
             # Added since target, have local modification
             (modadded, backupanddel, backup),
             # Added since target but file is missing in working directory
             (deladded, actions[b'drop'], discard),
             # Removed since  target, before working copy parent
             (removed, actions[b'add'], discard),
             # Same as `removed` but an unknown file exists at the same path
             (removunk, actions[b'add'], check),
             # Removed since targe, marked as such in working copy parent
             (dsremoved, actions[b'undelete'], discard),
             # Same as `dsremoved` but an unknown file exists at the same path
             (dsremovunk, actions[b'undelete'], check),
             ## the following sets does not result in any file changes
             # File with no modification
             (clean, actions[b'noop'], discard),
             # Existing file, not tracked anywhere
             (unknown, actions[b'unknown'], discard),
         )
 
         for abs, exact in sorted(names.items()):
             # target file to be touch on disk (relative to cwd)
             target = repo.wjoin(abs)
             # search the entry in the dispatch table.
             # if the file is in any of these sets, it was touched in the working
             # directory parent and we are sure it needs to be reverted.
             for table, (xlist, msg), dobackup in disptable:
                 if abs not in table:
                     continue
                 if xlist is not None:
                     xlist.append(abs)
                     if dobackup:
                         # If in interactive mode, don't automatically create
                         # .orig files (issue4793)
                         if dobackup == backupinteractive:
                             tobackup.add(abs)
                         elif backup <= dobackup or wctx[abs].cmp(ctx[abs]):
                             absbakname = scmutil.backuppath(ui, repo, abs)
                             bakname = os.path.relpath(
                                 absbakname, start=repo.root
                             )
                             ui.note(
                                 _(b'saving current version of %s as %s\n')
                                 % (uipathfn(abs), uipathfn(bakname))
                             )
                             if not opts.get(b'dry_run'):
                                 if interactive:
                                     util.copyfile(target, absbakname)
                                 else:
                                     util.rename(target, absbakname)
                     if opts.get(b'dry_run'):
                         if ui.verbose or not exact:
                             ui.status(msg % uipathfn(abs))
                 elif exact:
                     ui.warn(msg % uipathfn(abs))
                 break
 
         if not opts.get(b'dry_run'):
             needdata = (b'revert', b'add', b'undelete')
             oplist = [actions[name][0] for name in needdata]
             prefetch = scmutil.prefetchfiles
-            matchfiles = scmutil.matchfiles
+            matchfiles = scmutil.matchfiles(
+                repo, [f for sublist in oplist for f in sublist])
             prefetch(
                 repo,
-                [ctx.rev()],
-                matchfiles(repo, [f for sublist in oplist for f in sublist]),
+                [(ctx.rev(), matchfiles)],
             )
             match = scmutil.match(repo[None], pats)
             _performrevert(
                 repo,
                 parents,
                 ctx,
                 names,
                 uipathfn,
                 actions,
                 match,
                 interactive,
                 tobackup,
             )
 
         if targetsubs:
             # Revert the subrepos on the revert list
             for sub in targetsubs:
                 try:
                     wctx.sub(sub).revert(
                         ctx.substate[sub], *pats, **pycompat.strkwargs(opts)
                     )
                 except KeyError:
                     raise error.Abort(
                         b"subrepository '%s' does not exist in %s!"
                         % (sub, short(ctx.node()))
                     )
 
 
 def _performrevert(
     repo,
     parents,
     ctx,
     names,
     uipathfn,
     actions,
     match,
     interactive=False,
     tobackup=None,
 ):
     """function that actually perform all the actions computed for revert
 
     This is an independent function to let extension to plug in and react to
     the imminent revert.
 
     Make sure you have the working directory locked when calling this function.
     """
     parent, p2 = parents
     node = ctx.node()
     excluded_files = []
 
     def checkout(f):
         fc = ctx[f]
         repo.wwrite(f, fc.data(), fc.flags())
 
     def doremove(f):
         try:
             rmdir = repo.ui.configbool(b'experimental', b'removeemptydirs')
             repo.wvfs.unlinkpath(f, rmdir=rmdir)
         except OSError:
             pass
         repo.dirstate.remove(f)
 
     def prntstatusmsg(action, f):
         exact = names[f]
         if repo.ui.verbose or not exact:
             repo.ui.status(actions[action][1] % uipathfn(f))
 
     audit_path = pathutil.pathauditor(repo.root, cached=True)
     for f in actions[b'forget'][0]:
         if interactive:
             choice = repo.ui.promptchoice(
                 _(b"forget added file %s (Yn)?$$ &Yes $$ &No") % uipathfn(f)
             )
             if choice == 0:
                 prntstatusmsg(b'forget', f)
                 repo.dirstate.drop(f)
             else:
                 excluded_files.append(f)
         else:
             prntstatusmsg(b'forget', f)
             repo.dirstate.drop(f)
     for f in actions[b'remove'][0]:
         audit_path(f)
         if interactive:
             choice = repo.ui.promptchoice(
                 _(b"remove added file %s (Yn)?$$ &Yes $$ &No") % uipathfn(f)
             )
             if choice == 0:
                 prntstatusmsg(b'remove', f)
                 doremove(f)
             else:
                 excluded_files.append(f)
         else:
             prntstatusmsg(b'remove', f)
             doremove(f)
     for f in actions[b'drop'][0]:
         audit_path(f)
         prntstatusmsg(b'drop', f)
         repo.dirstate.remove(f)
 
     normal = None
     if node == parent:
         # We're reverting to our parent. If possible, we'd like status
         # to report the file as clean. We have to use normallookup for
         # merges to avoid losing information about merged/dirty files.
         if p2 != nullid:
             normal = repo.dirstate.normallookup
         else:
             normal = repo.dirstate.normal
 
     newlyaddedandmodifiedfiles = set()
     if interactive:
         # Prompt the user for changes to revert
         torevert = [f for f in actions[b'revert'][0] if f not in excluded_files]
         m = scmutil.matchfiles(repo, torevert)
         diffopts = patch.difffeatureopts(
             repo.ui,
             whitespace=True,
             section=b'commands',
             configprefix=b'revert.interactive.',
         )
         diffopts.nodates = True
         diffopts.git = True
         operation = b'apply'
         if node == parent:
             if repo.ui.configbool(
                 b'experimental', b'revert.interactive.select-to-keep'
             ):
                 operation = b'keep'
             else:
                 operation = b'discard'
 
         if operation == b'apply':
             diff = patch.diff(repo, None, ctx.node(), m, opts=diffopts)
         else:
             diff = patch.diff(repo, ctx.node(), None, m, opts=diffopts)
         originalchunks = patch.parsepatch(diff)
 
         try:
 
             chunks, opts = recordfilter(
                 repo.ui, originalchunks, match, operation=operation
             )
             if operation == b'discard':
                 chunks = patch.reversehunks(chunks)
 
         except error.PatchError as err:
             raise error.Abort(_(b'error parsing patch: %s') % err)
 
         # FIXME: when doing an interactive revert of a copy, there's no way of
         # performing a partial revert of the added file, the only option is
         # "remove added file <name> (Yn)?", so we don't need to worry about the
         # alsorestore value. Ideally we'd be able to partially revert
         # copied/renamed files.
         newlyaddedandmodifiedfiles, unusedalsorestore = newandmodified(
             chunks, originalchunks
         )
         if tobackup is None:
             tobackup = set()
         # Apply changes
         fp = stringio()
         # chunks are serialized per file, but files aren't sorted
         for f in sorted({c.header.filename() for c in chunks if ishunk(c)}):
             prntstatusmsg(b'revert', f)
         files = set()
         for c in chunks:
             if ishunk(c):
                 abs = c.header.filename()
                 # Create a backup file only if this hunk should be backed up
                 if c.header.filename() in tobackup:
                     target = repo.wjoin(abs)
                     bakname = scmutil.backuppath(repo.ui, repo, abs)
                     util.copyfile(target, bakname)
                     tobackup.remove(abs)
                 if abs not in files:
                     files.add(abs)
                     if operation == b'keep':
                         checkout(abs)
             c.write(fp)
         dopatch = fp.tell()
         fp.seek(0)
         if dopatch:
             try:
                 patch.internalpatch(repo.ui, repo, fp, 1, eolmode=None)
             except error.PatchError as err:
                 raise error.Abort(pycompat.bytestr(err))
         del fp
     else:
         for f in actions[b'revert'][0]:
             prntstatusmsg(b'revert', f)
             checkout(f)
             if normal:
                 normal(f)
 
     for f in actions[b'add'][0]:
         # Don't checkout modified files, they are already created by the diff
         if f not in newlyaddedandmodifiedfiles:
             prntstatusmsg(b'add', f)
             checkout(f)
             repo.dirstate.add(f)
 
     normal = repo.dirstate.normallookup
     if node == parent and p2 == nullid:
         normal = repo.dirstate.normal
     for f in actions[b'undelete'][0]:
         if interactive:
             choice = repo.ui.promptchoice(
                 _(b"add back removed file %s (Yn)?$$ &Yes $$ &No") % f
             )
             if choice == 0:
                 prntstatusmsg(b'undelete', f)
                 checkout(f)
                 normal(f)
             else:
                 excluded_files.append(f)
         else:
             prntstatusmsg(b'undelete', f)
             checkout(f)
             normal(f)
 
     copied = copies.pathcopies(repo[parent], ctx)
 
     for f in (
         actions[b'add'][0] + actions[b'undelete'][0] + actions[b'revert'][0]
     ):
         if f in copied:
             repo.dirstate.copy(copied[f], f)
 
 
 # a list of (ui, repo, otherpeer, opts, missing) functions called by
 # commands.outgoing.  "missing" is "missing" of the result of
 # "findcommonoutgoing()"
 outgoinghooks = util.hooks()
 
 # a list of (ui, repo) functions called by commands.summary
 summaryhooks = util.hooks()
 
 # a list of (ui, repo, opts, changes) functions called by commands.summary.
 #
 # functions should return tuple of booleans below, if 'changes' is None:
 #  (whether-incomings-are-needed, whether-outgoings-are-needed)
 #
 # otherwise, 'changes' is a tuple of tuples below:
 #  - (sourceurl, sourcebranch, sourcepeer, incoming)
 #  - (desturl,   destbranch,   destpeer,   outgoing)
 summaryremotehooks = util.hooks()
 
 
 def checkunfinished(repo, commit=False, skipmerge=False):
     '''Look for an unfinished multistep operation, like graft, and abort
     if found. It's probably good to check this right before
     bailifchanged().
     '''
     # Check for non-clearable states first, so things like rebase will take
     # precedence over update.
     for state in statemod._unfinishedstates:
         if (
             state._clearable
             or (commit and state._allowcommit)
             or state._reportonly
         ):
             continue
         if state.isunfinished(repo):
             raise error.Abort(state.msg(), hint=state.hint())
 
     for s in statemod._unfinishedstates:
         if (
             not s._clearable
             or (commit and s._allowcommit)
             or (s._opname == b'merge' and skipmerge)
             or s._reportonly
         ):
             continue
         if s.isunfinished(repo):
             raise error.Abort(s.msg(), hint=s.hint())
 
 
 def clearunfinished(repo):
     '''Check for unfinished operations (as above), and clear the ones
     that are clearable.
     '''
     for state in statemod._unfinishedstates:
         if state._reportonly:
             continue
         if not state._clearable and state.isunfinished(repo):
             raise error.Abort(state.msg(), hint=state.hint())
 
     for s in statemod._unfinishedstates:
         if s._opname == b'merge' or state._reportonly:
             continue
         if s._clearable and s.isunfinished(repo):
             util.unlink(repo.vfs.join(s._fname))
 
 
 def getunfinishedstate(repo):
     ''' Checks for unfinished operations and returns statecheck object
         for it'''
     for state in statemod._unfinishedstates:
         if state.isunfinished(repo):
             return state
     return None
 
 
 def howtocontinue(repo):
     '''Check for an unfinished operation and return the command to finish
     it.
 
     statemod._unfinishedstates list is checked for an unfinished operation
     and the corresponding message to finish it is generated if a method to
     continue is supported by the operation.
 
     Returns a (msg, warning) tuple. 'msg' is a string and 'warning' is
     a boolean.
     '''
     contmsg = _(b"continue: %s")
     for state in statemod._unfinishedstates:
         if not state._continueflag:
             continue
         if state.isunfinished(repo):
             return contmsg % state.continuemsg(), True
     if repo[None].dirty(missing=True, merge=False, branch=False):
         return contmsg % _(b"hg commit"), False
     return None, None
 
 
 def checkafterresolved(repo):
     '''Inform the user about the next action after completing hg resolve
 
     If there's a an unfinished operation that supports continue flag,
     howtocontinue will yield repo.ui.warn as the reporter.
 
     Otherwise, it will yield repo.ui.note.
     '''
     msg, warning = howtocontinue(repo)
     if msg is not None:
         if warning:
             repo.ui.warn(b"%s\n" % msg)
         else:
             repo.ui.note(b"%s\n" % msg)
 
 
 def wrongtooltocontinue(repo, task):
     '''Raise an abort suggesting how to properly continue if there is an
     active task.
 
     Uses howtocontinue() to find the active task.
 
     If there's no task (repo.ui.note for 'hg commit'), it does not offer
     a hint.
     '''
     after = howtocontinue(repo)
     hint = None
     if after[1]:
         hint = after[0]
     raise error.Abort(_(b'no %s in progress') % task, hint=hint)
 
 
 def abortgraft(ui, repo, graftstate):
     """abort the interrupted graft and rollbacks to the state before interrupted
     graft"""
     if not graftstate.exists():
         raise error.Abort(_(b"no interrupted graft to abort"))
     statedata = readgraftstate(repo, graftstate)
     newnodes = statedata.get(b'newnodes')
     if newnodes is None:
         # and old graft state which does not have all the data required to abort
         # the graft
         raise error.Abort(_(b"cannot abort using an old graftstate"))
 
     # changeset from which graft operation was started
     if len(newnodes) > 0:
         startctx = repo[newnodes[0]].p1()
     else:
         startctx = repo[b'.']
     # whether to strip or not
     cleanup = False
     from . import hg
 
     if newnodes:
         newnodes = [repo[r].rev() for r in newnodes]
         cleanup = True
         # checking that none of the newnodes turned public or is public
         immutable = [c for c in newnodes if not repo[c].mutable()]
         if immutable:
             repo.ui.warn(
                 _(b"cannot clean up public changesets %s\n")
                 % b', '.join(bytes(repo[r]) for r in immutable),
                 hint=_(b"see 'hg help phases' for details"),
             )
             cleanup = False
 
         # checking that no new nodes are created on top of grafted revs
         desc = set(repo.changelog.descendants(newnodes))
         if desc - set(newnodes):
             repo.ui.warn(
                 _(
                     b"new changesets detected on destination "
                     b"branch, can't strip\n"
                 )
             )
             cleanup = False
 
         if cleanup:
             with repo.wlock(), repo.lock():
                 hg.updaterepo(repo, startctx.node(), overwrite=True)
                 # stripping the new nodes created
                 strippoints = [
                     c.node() for c in repo.set(b"roots(%ld)", newnodes)
                 ]
                 repair.strip(repo.ui, repo, strippoints, backup=False)
 
     if not cleanup:
         # we don't update to the startnode if we can't strip
         startctx = repo[b'.']
         hg.updaterepo(repo, startctx.node(), overwrite=True)
 
     ui.status(_(b"graft aborted\n"))
     ui.status(_(b"working directory is now at %s\n") % startctx.hex()[:12])
     graftstate.delete()
     return 0
 
 
 def readgraftstate(repo, graftstate):
     # type: (Any, statemod.cmdstate) -> Dict[bytes, Any]
     """read the graft state file and return a dict of the data stored in it"""
     try:
         return graftstate.read()
     except error.CorruptedState:
         nodes = repo.vfs.read(b'graftstate').splitlines()
         return {b'nodes': nodes}
 
 
 def hgabortgraft(ui, repo):
     """ abort logic for aborting graft using 'hg abort'"""
     with repo.wlock():
         graftstate = statemod.cmdstate(repo, b'graftstate')
         return abortgraft(ui, repo, graftstate)

mercurial/context.py

@@ -1,3085 +1,3085 b''
 # context.py - changeset and file context objects for mercurial
 #
 # Copyright 2006, 2007 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import errno
 import filecmp
 import os
 import stat
 
 from .i18n import _
 from .node import (
     addednodeid,
     hex,
     modifiednodeid,
     nullid,
     nullrev,
     short,
     wdirfilenodeids,
     wdirhex,
 )
 from .pycompat import (
     getattr,
     open,
 )
 from . import (
     dagop,
     encoding,
     error,
     fileset,
     match as matchmod,
     mergestate as mergestatemod,
     metadata,
     obsolete as obsmod,
     patch,
     pathutil,
     phases,
     pycompat,
     repoview,
     scmutil,
     sparse,
     subrepo,
     subrepoutil,
     util,
 )
 from .utils import (
     dateutil,
     stringutil,
 )
 
 propertycache = util.propertycache
 
 
 class basectx(object):
     """A basectx object represents the common logic for its children:
     changectx: read-only context that is already present in the repo,
     workingctx: a context that represents the working directory and can
                 be committed,
     memctx: a context that represents changes in-memory and can also
             be committed."""
 
     def __init__(self, repo):
         self._repo = repo
 
     def __bytes__(self):
         return short(self.node())
 
     __str__ = encoding.strmethod(__bytes__)
 
     def __repr__(self):
         return "<%s %s>" % (type(self).__name__, str(self))
 
     def __eq__(self, other):
         try:
             return type(self) == type(other) and self._rev == other._rev
         except AttributeError:
             return False
 
     def __ne__(self, other):
         return not (self == other)
 
     def __contains__(self, key):
         return key in self._manifest
 
     def __getitem__(self, key):
         return self.filectx(key)
 
     def __iter__(self):
         return iter(self._manifest)
 
     def _buildstatusmanifest(self, status):
         """Builds a manifest that includes the given status results, if this is
         a working copy context. For non-working copy contexts, it just returns
         the normal manifest."""
         return self.manifest()
 
     def _matchstatus(self, other, match):
         """This internal method provides a way for child objects to override the
         match operator.
         """
         return match
 
     def _buildstatus(
         self, other, s, match, listignored, listclean, listunknown
     ):
         """build a status with respect to another context"""
         # Load earliest manifest first for caching reasons. More specifically,
         # if you have revisions 1000 and 1001, 1001 is probably stored as a
         # delta against 1000. Thus, if you read 1000 first, we'll reconstruct
         # 1000 and cache it so that when you read 1001, we just need to apply a
         # delta to what's in the cache. So that's one full reconstruction + one
         # delta application.
         mf2 = None
         if self.rev() is not None and self.rev() < other.rev():
             mf2 = self._buildstatusmanifest(s)
         mf1 = other._buildstatusmanifest(s)
         if mf2 is None:
             mf2 = self._buildstatusmanifest(s)
 
         modified, added = [], []
         removed = []
         clean = []
         deleted, unknown, ignored = s.deleted, s.unknown, s.ignored
         deletedset = set(deleted)
         d = mf1.diff(mf2, match=match, clean=listclean)
         for fn, value in pycompat.iteritems(d):
             if fn in deletedset:
                 continue
             if value is None:
                 clean.append(fn)
                 continue
             (node1, flag1), (node2, flag2) = value
             if node1 is None:
                 added.append(fn)
             elif node2 is None:
                 removed.append(fn)
             elif flag1 != flag2:
                 modified.append(fn)
             elif node2 not in wdirfilenodeids:
                 # When comparing files between two commits, we save time by
                 # not comparing the file contents when the nodeids differ.
                 # Note that this means we incorrectly report a reverted change
                 # to a file as a modification.
                 modified.append(fn)
             elif self[fn].cmp(other[fn]):
                 modified.append(fn)
             else:
                 clean.append(fn)
 
         if removed:
             # need to filter files if they are already reported as removed
             unknown = [
                 fn
                 for fn in unknown
                 if fn not in mf1 and (not match or match(fn))
             ]
             ignored = [
                 fn
                 for fn in ignored
                 if fn not in mf1 and (not match or match(fn))
             ]
             # if they're deleted, don't report them as removed
             removed = [fn for fn in removed if fn not in deletedset]
 
         return scmutil.status(
             modified, added, removed, deleted, unknown, ignored, clean
         )
 
     @propertycache
     def substate(self):
         return subrepoutil.state(self, self._repo.ui)
 
     def subrev(self, subpath):
         return self.substate[subpath][1]
 
     def rev(self):
         return self._rev
 
     def node(self):
         return self._node
 
     def hex(self):
         return hex(self.node())
 
     def manifest(self):
         return self._manifest
 
     def manifestctx(self):
         return self._manifestctx
 
     def repo(self):
         return self._repo
 
     def phasestr(self):
         return phases.phasenames[self.phase()]
 
     def mutable(self):
         return self.phase() > phases.public
 
     def matchfileset(self, cwd, expr, badfn=None):
         return fileset.match(self, cwd, expr, badfn=badfn)
 
     def obsolete(self):
         """True if the changeset is obsolete"""
         return self.rev() in obsmod.getrevs(self._repo, b'obsolete')
 
     def extinct(self):
         """True if the changeset is extinct"""
         return self.rev() in obsmod.getrevs(self._repo, b'extinct')
 
     def orphan(self):
         """True if the changeset is not obsolete, but its ancestor is"""
         return self.rev() in obsmod.getrevs(self._repo, b'orphan')
 
     def phasedivergent(self):
         """True if the changeset tries to be a successor of a public changeset
 
         Only non-public and non-obsolete changesets may be phase-divergent.
         """
         return self.rev() in obsmod.getrevs(self._repo, b'phasedivergent')
 
     def contentdivergent(self):
         """Is a successor of a changeset with multiple possible successor sets
 
         Only non-public and non-obsolete changesets may be content-divergent.
         """
         return self.rev() in obsmod.getrevs(self._repo, b'contentdivergent')
 
     def isunstable(self):
         """True if the changeset is either orphan, phase-divergent or
         content-divergent"""
         return self.orphan() or self.phasedivergent() or self.contentdivergent()
 
     def instabilities(self):
         """return the list of instabilities affecting this changeset.
 
         Instabilities are returned as strings. possible values are:
         - orphan,
         - phase-divergent,
         - content-divergent.
         """
         instabilities = []
         if self.orphan():
             instabilities.append(b'orphan')
         if self.phasedivergent():
             instabilities.append(b'phase-divergent')
         if self.contentdivergent():
             instabilities.append(b'content-divergent')
         return instabilities
 
     def parents(self):
         """return contexts for each parent changeset"""
         return self._parents
 
     def p1(self):
         return self._parents[0]
 
     def p2(self):
         parents = self._parents
         if len(parents) == 2:
             return parents[1]
         return self._repo[nullrev]
 
     def _fileinfo(self, path):
         if '_manifest' in self.__dict__:
             try:
                 return self._manifest.find(path)
             except KeyError:
                 raise error.ManifestLookupError(
                     self._node, path, _(b'not found in manifest')
                 )
         if '_manifestdelta' in self.__dict__ or path in self.files():
             if path in self._manifestdelta:
                 return (
                     self._manifestdelta[path],
                     self._manifestdelta.flags(path),
                 )
         mfl = self._repo.manifestlog
         try:
             node, flag = mfl[self._changeset.manifest].find(path)
         except KeyError:
             raise error.ManifestLookupError(
                 self._node, path, _(b'not found in manifest')
             )
 
         return node, flag
 
     def filenode(self, path):
         return self._fileinfo(path)[0]
 
     def flags(self, path):
         try:
             return self._fileinfo(path)[1]
         except error.LookupError:
             return b''
 
     @propertycache
     def _copies(self):
         return metadata.computechangesetcopies(self)
 
     def p1copies(self):
         return self._copies[0]
 
     def p2copies(self):
         return self._copies[1]
 
     def sub(self, path, allowcreate=True):
         '''return a subrepo for the stored revision of path, never wdir()'''
         return subrepo.subrepo(self, path, allowcreate=allowcreate)
 
     def nullsub(self, path, pctx):
         return subrepo.nullsubrepo(self, path, pctx)
 
     def workingsub(self, path):
         '''return a subrepo for the stored revision, or wdir if this is a wdir
         context.
         '''
         return subrepo.subrepo(self, path, allowwdir=True)
 
     def match(
         self,
         pats=None,
         include=None,
         exclude=None,
         default=b'glob',
         listsubrepos=False,
         badfn=None,
         cwd=None,
     ):
         r = self._repo
         if not cwd:
             cwd = r.getcwd()
         return matchmod.match(
             r.root,
             cwd,
             pats,
             include,
             exclude,
             default,
             auditor=r.nofsauditor,
             ctx=self,
             listsubrepos=listsubrepos,
             badfn=badfn,
         )
 
     def diff(
         self,
         ctx2=None,
         match=None,
         changes=None,
         opts=None,
         losedatafn=None,
         pathfn=None,
         copy=None,
         copysourcematch=None,
         hunksfilterfn=None,
     ):
         """Returns a diff generator for the given contexts and matcher"""
         if ctx2 is None:
             ctx2 = self.p1()
         if ctx2 is not None:
             ctx2 = self._repo[ctx2]
         return patch.diff(
             self._repo,
             ctx2,
             self,
             match=match,
             changes=changes,
             opts=opts,
             losedatafn=losedatafn,
             pathfn=pathfn,
             copy=copy,
             copysourcematch=copysourcematch,
             hunksfilterfn=hunksfilterfn,
         )
 
     def dirs(self):
         return self._manifest.dirs()
 
     def hasdir(self, dir):
         return self._manifest.hasdir(dir)
 
     def status(
         self,
         other=None,
         match=None,
         listignored=False,
         listclean=False,
         listunknown=False,
         listsubrepos=False,
     ):
         """return status of files between two nodes or node and working
         directory.
 
         If other is None, compare this node with working directory.
 
         returns (modified, added, removed, deleted, unknown, ignored, clean)
         """
 
         ctx1 = self
         ctx2 = self._repo[other]
 
         # This next code block is, admittedly, fragile logic that tests for
         # reversing the contexts and wouldn't need to exist if it weren't for
         # the fast (and common) code path of comparing the working directory
         # with its first parent.
         #
         # What we're aiming for here is the ability to call:
         #
         # workingctx.status(parentctx)
         #
         # If we always built the manifest for each context and compared those,
         # then we'd be done. But the special case of the above call means we
         # just copy the manifest of the parent.
         reversed = False
         if not isinstance(ctx1, changectx) and isinstance(ctx2, changectx):
             reversed = True
             ctx1, ctx2 = ctx2, ctx1
 
         match = self._repo.narrowmatch(match)
         match = ctx2._matchstatus(ctx1, match)
         r = scmutil.status([], [], [], [], [], [], [])
         r = ctx2._buildstatus(
             ctx1, r, match, listignored, listclean, listunknown
         )
 
         if reversed:
             # Reverse added and removed. Clear deleted, unknown and ignored as
             # these make no sense to reverse.
             r = scmutil.status(
                 r.modified, r.removed, r.added, [], [], [], r.clean
             )
 
         if listsubrepos:
             for subpath, sub in scmutil.itersubrepos(ctx1, ctx2):
                 try:
                     rev2 = ctx2.subrev(subpath)
                 except KeyError:
                     # A subrepo that existed in node1 was deleted between
                     # node1 and node2 (inclusive). Thus, ctx2's substate
                     # won't contain that subpath. The best we can do ignore it.
                     rev2 = None
                 submatch = matchmod.subdirmatcher(subpath, match)
                 s = sub.status(
                     rev2,
                     match=submatch,
                     ignored=listignored,
                     clean=listclean,
                     unknown=listunknown,
                     listsubrepos=True,
                 )
                 for k in (
                     'modified',
                     'added',
                     'removed',
                     'deleted',
                     'unknown',
                     'ignored',
                     'clean',
                 ):
                     rfiles, sfiles = getattr(r, k), getattr(s, k)
                     rfiles.extend(b"%s/%s" % (subpath, f) for f in sfiles)
 
         r.modified.sort()
         r.added.sort()
         r.removed.sort()
         r.deleted.sort()
         r.unknown.sort()
         r.ignored.sort()
         r.clean.sort()
 
         return r
 
     def mergestate(self, clean=False):
         """Get a mergestate object for this context."""
         raise NotImplementedError(
             '%s does not implement mergestate()' % self.__class__
         )
 
 
 class changectx(basectx):
     """A changecontext object makes access to data related to a particular
     changeset convenient. It represents a read-only context already present in
     the repo."""
 
     def __init__(self, repo, rev, node, maybe_filtered=True):
         super(changectx, self).__init__(repo)
         self._rev = rev
         self._node = node
         # When maybe_filtered is True, the revision might be affected by
         # changelog filtering and operation through the filtered changelog must be used.
         #
         # When maybe_filtered is False, the revision has already been checked
         # against filtering and is not filtered. Operation through the
         # unfiltered changelog might be used in some case.
         self._maybe_filtered = maybe_filtered
 
     def __hash__(self):
         try:
             return hash(self._rev)
         except AttributeError:
             return id(self)
 
     def __nonzero__(self):
         return self._rev != nullrev
 
     __bool__ = __nonzero__
 
     @propertycache
     def _changeset(self):
         if self._maybe_filtered:
             repo = self._repo
         else:
             repo = self._repo.unfiltered()
         return repo.changelog.changelogrevision(self.rev())
 
     @propertycache
     def _manifest(self):
         return self._manifestctx.read()
 
     @property
     def _manifestctx(self):
         return self._repo.manifestlog[self._changeset.manifest]
 
     @propertycache
     def _manifestdelta(self):
         return self._manifestctx.readdelta()
 
     @propertycache
     def _parents(self):
         repo = self._repo
         if self._maybe_filtered:
             cl = repo.changelog
         else:
             cl = repo.unfiltered().changelog
 
         p1, p2 = cl.parentrevs(self._rev)
         if p2 == nullrev:
             return [changectx(repo, p1, cl.node(p1), maybe_filtered=False)]
         return [
             changectx(repo, p1, cl.node(p1), maybe_filtered=False),
             changectx(repo, p2, cl.node(p2), maybe_filtered=False),
         ]
 
     def changeset(self):
         c = self._changeset
         return (
             c.manifest,
             c.user,
             c.date,
             c.files,
             c.description,
             c.extra,
         )
 
     def manifestnode(self):
         return self._changeset.manifest
 
     def user(self):
         return self._changeset.user
 
     def date(self):
         return self._changeset.date
 
     def files(self):
         return self._changeset.files
 
     def filesmodified(self):
         modified = set(self.files())
         modified.difference_update(self.filesadded())
         modified.difference_update(self.filesremoved())
         return sorted(modified)
 
     def filesadded(self):
         filesadded = self._changeset.filesadded
         compute_on_none = True
         if self._repo.filecopiesmode == b'changeset-sidedata':
             compute_on_none = False
         else:
             source = self._repo.ui.config(b'experimental', b'copies.read-from')
             if source == b'changeset-only':
                 compute_on_none = False
             elif source != b'compatibility':
                 # filelog mode, ignore any changelog content
                 filesadded = None
         if filesadded is None:
             if compute_on_none:
                 filesadded = metadata.computechangesetfilesadded(self)
             else:
                 filesadded = []
         return filesadded
 
     def filesremoved(self):
         filesremoved = self._changeset.filesremoved
         compute_on_none = True
         if self._repo.filecopiesmode == b'changeset-sidedata':
             compute_on_none = False
         else:
             source = self._repo.ui.config(b'experimental', b'copies.read-from')
             if source == b'changeset-only':
                 compute_on_none = False
             elif source != b'compatibility':
                 # filelog mode, ignore any changelog content
                 filesremoved = None
         if filesremoved is None:
             if compute_on_none:
                 filesremoved = metadata.computechangesetfilesremoved(self)
             else:
                 filesremoved = []
         return filesremoved
 
     @propertycache
     def _copies(self):
         p1copies = self._changeset.p1copies
         p2copies = self._changeset.p2copies
         compute_on_none = True
         if self._repo.filecopiesmode == b'changeset-sidedata':
             compute_on_none = False
         else:
             source = self._repo.ui.config(b'experimental', b'copies.read-from')
             # If config says to get copy metadata only from changeset, then
             # return that, defaulting to {} if there was no copy metadata.  In
             # compatibility mode, we return copy data from the changeset if it
             # was recorded there, and otherwise we fall back to getting it from
             # the filelogs (below).
             #
             # If we are in compatiblity mode and there is not data in the
             # changeset), we get the copy metadata from the filelogs.
             #
             # otherwise, when config said to read only from filelog, we get the
             # copy metadata from the filelogs.
             if source == b'changeset-only':
                 compute_on_none = False
             elif source != b'compatibility':
                 # filelog mode, ignore any changelog content
                 p1copies = p2copies = None
         if p1copies is None:
             if compute_on_none:
                 p1copies, p2copies = super(changectx, self)._copies
             else:
                 if p1copies is None:
                     p1copies = {}
         if p2copies is None:
             p2copies = {}
         return p1copies, p2copies
 
     def description(self):
         return self._changeset.description
 
     def branch(self):
         return encoding.tolocal(self._changeset.extra.get(b"branch"))
 
     def closesbranch(self):
         return b'close' in self._changeset.extra
 
     def extra(self):
         """Return a dict of extra information."""
         return self._changeset.extra
 
     def tags(self):
         """Return a list of byte tag names"""
         return self._repo.nodetags(self._node)
 
     def bookmarks(self):
         """Return a list of byte bookmark names."""
         return self._repo.nodebookmarks(self._node)
 
     def phase(self):
         return self._repo._phasecache.phase(self._repo, self._rev)
 
     def hidden(self):
         return self._rev in repoview.filterrevs(self._repo, b'visible')
 
     def isinmemory(self):
         return False
 
     def children(self):
         """return list of changectx contexts for each child changeset.
 
         This returns only the immediate child changesets. Use descendants() to
         recursively walk children.
         """
         c = self._repo.changelog.children(self._node)
         return [self._repo[x] for x in c]
 
     def ancestors(self):
         for a in self._repo.changelog.ancestors([self._rev]):
             yield self._repo[a]
 
     def descendants(self):
         """Recursively yield all children of the changeset.
 
         For just the immediate children, use children()
         """
         for d in self._repo.changelog.descendants([self._rev]):
             yield self._repo[d]
 
     def filectx(self, path, fileid=None, filelog=None):
         """get a file context from this changeset"""
         if fileid is None:
             fileid = self.filenode(path)
         return filectx(
             self._repo, path, fileid=fileid, changectx=self, filelog=filelog
         )
 
     def ancestor(self, c2, warn=False):
         """return the "best" ancestor context of self and c2
 
         If there are multiple candidates, it will show a message and check
         merge.preferancestor configuration before falling back to the
         revlog ancestor."""
         # deal with workingctxs
         n2 = c2._node
         if n2 is None:
             n2 = c2._parents[0]._node
         cahs = self._repo.changelog.commonancestorsheads(self._node, n2)
         if not cahs:
             anc = nullid
         elif len(cahs) == 1:
             anc = cahs[0]
         else:
             # experimental config: merge.preferancestor
             for r in self._repo.ui.configlist(b'merge', b'preferancestor'):
                 try:
                     ctx = scmutil.revsymbol(self._repo, r)
                 except error.RepoLookupError:
                     continue
                 anc = ctx.node()
                 if anc in cahs:
                     break
             else:
                 anc = self._repo.changelog.ancestor(self._node, n2)
             if warn:
                 self._repo.ui.status(
                     (
                         _(b"note: using %s as ancestor of %s and %s\n")
                         % (short(anc), short(self._node), short(n2))
                     )
                     + b''.join(
                         _(
                             b"      alternatively, use --config "
                             b"merge.preferancestor=%s\n"
                         )
                         % short(n)
                         for n in sorted(cahs)
                         if n != anc
                     )
                 )
         return self._repo[anc]
 
     def isancestorof(self, other):
         """True if this changeset is an ancestor of other"""
         return self._repo.changelog.isancestorrev(self._rev, other._rev)
 
     def walk(self, match):
         '''Generates matching file names.'''
 
         # Wrap match.bad method to have message with nodeid
         def bad(fn, msg):
             # The manifest doesn't know about subrepos, so don't complain about
             # paths into valid subrepos.
             if any(fn == s or fn.startswith(s + b'/') for s in self.substate):
                 return
             match.bad(fn, _(b'no such file in rev %s') % self)
 
         m = matchmod.badmatch(self._repo.narrowmatch(match), bad)
         return self._manifest.walk(m)
 
     def matches(self, match):
         return self.walk(match)
 
 
 class basefilectx(object):
     """A filecontext object represents the common logic for its children:
     filectx: read-only access to a filerevision that is already present
              in the repo,
     workingfilectx: a filecontext that represents files from the working
                     directory,
     memfilectx: a filecontext that represents files in-memory,
     """
 
     @propertycache
     def _filelog(self):
         return self._repo.file(self._path)
 
     @propertycache
     def _changeid(self):
         if '_changectx' in self.__dict__:
             return self._changectx.rev()
         elif '_descendantrev' in self.__dict__:
             # this file context was created from a revision with a known
             # descendant, we can (lazily) correct for linkrev aliases
             return self._adjustlinkrev(self._descendantrev)
         else:
             return self._filelog.linkrev(self._filerev)
 
     @propertycache
     def _filenode(self):
         if '_fileid' in self.__dict__:
             return self._filelog.lookup(self._fileid)
         else:
             return self._changectx.filenode(self._path)
 
     @propertycache
     def _filerev(self):
         return self._filelog.rev(self._filenode)
 
     @propertycache
     def _repopath(self):
         return self._path
 
     def __nonzero__(self):
         try:
             self._filenode
             return True
         except error.LookupError:
             # file is missing
             return False
 
     __bool__ = __nonzero__
 
     def __bytes__(self):
         try:
             return b"%s@%s" % (self.path(), self._changectx)
         except error.LookupError:
             return b"%s@???" % self.path()
 
     __str__ = encoding.strmethod(__bytes__)
 
     def __repr__(self):
         return "<%s %s>" % (type(self).__name__, str(self))
 
     def __hash__(self):
         try:
             return hash((self._path, self._filenode))
         except AttributeError:
             return id(self)
 
     def __eq__(self, other):
         try:
             return (
                 type(self) == type(other)
                 and self._path == other._path
                 and self._filenode == other._filenode
             )
         except AttributeError:
             return False
 
     def __ne__(self, other):
         return not (self == other)
 
     def filerev(self):
         return self._filerev
 
     def filenode(self):
         return self._filenode
 
     @propertycache
     def _flags(self):
         return self._changectx.flags(self._path)
 
     def flags(self):
         return self._flags
 
     def filelog(self):
         return self._filelog
 
     def rev(self):
         return self._changeid
 
     def linkrev(self):
         return self._filelog.linkrev(self._filerev)
 
     def node(self):
         return self._changectx.node()
 
     def hex(self):
         return self._changectx.hex()
 
     def user(self):
         return self._changectx.user()
 
     def date(self):
         return self._changectx.date()
 
     def files(self):
         return self._changectx.files()
 
     def description(self):
         return self._changectx.description()
 
     def branch(self):
         return self._changectx.branch()
 
     def extra(self):
         return self._changectx.extra()
 
     def phase(self):
         return self._changectx.phase()
 
     def phasestr(self):
         return self._changectx.phasestr()
 
     def obsolete(self):
         return self._changectx.obsolete()
 
     def instabilities(self):
         return self._changectx.instabilities()
 
     def manifest(self):
         return self._changectx.manifest()
 
     def changectx(self):
         return self._changectx
 
     def renamed(self):
         return self._copied
 
     def copysource(self):
         return self._copied and self._copied[0]
 
     def repo(self):
         return self._repo
 
     def size(self):
         return len(self.data())
 
     def path(self):
         return self._path
 
     def isbinary(self):
         try:
             return stringutil.binary(self.data())
         except IOError:
             return False
 
     def isexec(self):
         return b'x' in self.flags()
 
     def islink(self):
         return b'l' in self.flags()
 
     def isabsent(self):
         """whether this filectx represents a file not in self._changectx
 
         This is mainly for merge code to detect change/delete conflicts. This is
         expected to be True for all subclasses of basectx."""
         return False
 
     _customcmp = False
 
     def cmp(self, fctx):
         """compare with other file context
 
         returns True if different than fctx.
         """
         if fctx._customcmp:
             return fctx.cmp(self)
 
         if self._filenode is None:
             raise error.ProgrammingError(
                 b'filectx.cmp() must be reimplemented if not backed by revlog'
             )
 
         if fctx._filenode is None:
             if self._repo._encodefilterpats:
                 # can't rely on size() because wdir content may be decoded
                 return self._filelog.cmp(self._filenode, fctx.data())
             if self.size() - 4 == fctx.size():
                 # size() can match:
                 # if file data starts with '\1\n', empty metadata block is
                 # prepended, which adds 4 bytes to filelog.size().
                 return self._filelog.cmp(self._filenode, fctx.data())
         if self.size() == fctx.size():
             # size() matches: need to compare content
             return self._filelog.cmp(self._filenode, fctx.data())
 
         # size() differs
         return True
 
     def _adjustlinkrev(self, srcrev, inclusive=False, stoprev=None):
         """return the first ancestor of <srcrev> introducing <fnode>
 
         If the linkrev of the file revision does not point to an ancestor of
         srcrev, we'll walk down the ancestors until we find one introducing
         this file revision.
 
         :srcrev: the changeset revision we search ancestors from
         :inclusive: if true, the src revision will also be checked
         :stoprev: an optional revision to stop the walk at. If no introduction
                   of this file content could be found before this floor
                   revision, the function will returns "None" and stops its
                   iteration.
         """
         repo = self._repo
         cl = repo.unfiltered().changelog
         mfl = repo.manifestlog
         # fetch the linkrev
         lkr = self.linkrev()
         if srcrev == lkr:
             return lkr
         # hack to reuse ancestor computation when searching for renames
         memberanc = getattr(self, '_ancestrycontext', None)
         iteranc = None
         if srcrev is None:
             # wctx case, used by workingfilectx during mergecopy
             revs = [p.rev() for p in self._repo[None].parents()]
             inclusive = True  # we skipped the real (revless) source
         else:
             revs = [srcrev]
         if memberanc is None:
             memberanc = iteranc = cl.ancestors(revs, lkr, inclusive=inclusive)
         # check if this linkrev is an ancestor of srcrev
         if lkr not in memberanc:
             if iteranc is None:
                 iteranc = cl.ancestors(revs, lkr, inclusive=inclusive)
             fnode = self._filenode
             path = self._path
             for a in iteranc:
                 if stoprev is not None and a < stoprev:
                     return None
                 ac = cl.read(a)  # get changeset data (we avoid object creation)
                 if path in ac[3]:  # checking the 'files' field.
                     # The file has been touched, check if the content is
                     # similar to the one we search for.
                     if fnode == mfl[ac[0]].readfast().get(path):
                         return a
             # In theory, we should never get out of that loop without a result.
             # But if manifest uses a buggy file revision (not children of the
             # one it replaces) we could. Such a buggy situation will likely
             # result is crash somewhere else at to some point.
         return lkr
 
     def isintroducedafter(self, changelogrev):
         """True if a filectx has been introduced after a given floor revision
         """
         if self.linkrev() >= changelogrev:
             return True
         introrev = self._introrev(stoprev=changelogrev)
         if introrev is None:
             return False
         return introrev >= changelogrev
 
     def introrev(self):
         """return the rev of the changeset which introduced this file revision
 
         This method is different from linkrev because it take into account the
         changeset the filectx was created from. It ensures the returned
         revision is one of its ancestors. This prevents bugs from
         'linkrev-shadowing' when a file revision is used by multiple
         changesets.
         """
         return self._introrev()
 
     def _introrev(self, stoprev=None):
         """
         Same as `introrev` but, with an extra argument to limit changelog
         iteration range in some internal usecase.
 
         If `stoprev` is set, the `introrev` will not be searched past that
         `stoprev` revision and "None" might be returned. This is useful to
         limit the iteration range.
         """
         toprev = None
         attrs = vars(self)
         if '_changeid' in attrs:
             # We have a cached value already
             toprev = self._changeid
         elif '_changectx' in attrs:
             # We know which changelog entry we are coming from
             toprev = self._changectx.rev()
 
         if toprev is not None:
             return self._adjustlinkrev(toprev, inclusive=True, stoprev=stoprev)
         elif '_descendantrev' in attrs:
             introrev = self._adjustlinkrev(self._descendantrev, stoprev=stoprev)
             # be nice and cache the result of the computation
             if introrev is not None:
                 self._changeid = introrev
             return introrev
         else:
             return self.linkrev()
 
     def introfilectx(self):
         """Return filectx having identical contents, but pointing to the
         changeset revision where this filectx was introduced"""
         introrev = self.introrev()
         if self.rev() == introrev:
             return self
         return self.filectx(self.filenode(), changeid=introrev)
 
     def _parentfilectx(self, path, fileid, filelog):
         """create parent filectx keeping ancestry info for _adjustlinkrev()"""
         fctx = filectx(self._repo, path, fileid=fileid, filelog=filelog)
         if '_changeid' in vars(self) or '_changectx' in vars(self):
             # If self is associated with a changeset (probably explicitly
             # fed), ensure the created filectx is associated with a
             # changeset that is an ancestor of self.changectx.
             # This lets us later use _adjustlinkrev to get a correct link.
             fctx._descendantrev = self.rev()
             fctx._ancestrycontext = getattr(self, '_ancestrycontext', None)
         elif '_descendantrev' in vars(self):
             # Otherwise propagate _descendantrev if we have one associated.
             fctx._descendantrev = self._descendantrev
             fctx._ancestrycontext = getattr(self, '_ancestrycontext', None)
         return fctx
 
     def parents(self):
         _path = self._path
         fl = self._filelog
         parents = self._filelog.parents(self._filenode)
         pl = [(_path, node, fl) for node in parents if node != nullid]
 
         r = fl.renamed(self._filenode)
         if r:
             # - In the simple rename case, both parent are nullid, pl is empty.
             # - In case of merge, only one of the parent is null id and should
             # be replaced with the rename information. This parent is -always-
             # the first one.
             #
             # As null id have always been filtered out in the previous list
             # comprehension, inserting to 0 will always result in "replacing
             # first nullid parent with rename information.
             pl.insert(0, (r[0], r[1], self._repo.file(r[0])))
 
         return [self._parentfilectx(path, fnode, l) for path, fnode, l in pl]
 
     def p1(self):
         return self.parents()[0]
 
     def p2(self):
         p = self.parents()
         if len(p) == 2:
             return p[1]
         return filectx(self._repo, self._path, fileid=-1, filelog=self._filelog)
 
     def annotate(self, follow=False, skiprevs=None, diffopts=None):
         """Returns a list of annotateline objects for each line in the file
 
         - line.fctx is the filectx of the node where that line was last changed
         - line.lineno is the line number at the first appearance in the managed
           file
         - line.text is the data on that line (including newline character)
         """
         getlog = util.lrucachefunc(lambda x: self._repo.file(x))
 
         def parents(f):
             # Cut _descendantrev here to mitigate the penalty of lazy linkrev
             # adjustment. Otherwise, p._adjustlinkrev() would walk changelog
             # from the topmost introrev (= srcrev) down to p.linkrev() if it
             # isn't an ancestor of the srcrev.
             f._changeid
             pl = f.parents()
 
             # Don't return renamed parents if we aren't following.
             if not follow:
                 pl = [p for p in pl if p.path() == f.path()]
 
             # renamed filectx won't have a filelog yet, so set it
             # from the cache to save time
             for p in pl:
                 if not '_filelog' in p.__dict__:
                     p._filelog = getlog(p.path())
 
             return pl
 
         # use linkrev to find the first changeset where self appeared
         base = self.introfilectx()
         if getattr(base, '_ancestrycontext', None) is None:
             # it is safe to use an unfiltered repository here because we are
             # walking ancestors only.
             cl = self._repo.unfiltered().changelog
             if base.rev() is None:
                 # wctx is not inclusive, but works because _ancestrycontext
                 # is used to test filelog revisions
                 ac = cl.ancestors(
                     [p.rev() for p in base.parents()], inclusive=True
                 )
             else:
                 ac = cl.ancestors([base.rev()], inclusive=True)
             base._ancestrycontext = ac
 
         return dagop.annotate(
             base, parents, skiprevs=skiprevs, diffopts=diffopts
         )
 
     def ancestors(self, followfirst=False):
         visit = {}
         c = self
         if followfirst:
             cut = 1
         else:
             cut = None
 
         while True:
             for parent in c.parents()[:cut]:
                 visit[(parent.linkrev(), parent.filenode())] = parent
             if not visit:
                 break
             c = visit.pop(max(visit))
             yield c
 
     def decodeddata(self):
         """Returns `data()` after running repository decoding filters.
 
         This is often equivalent to how the data would be expressed on disk.
         """
         return self._repo.wwritedata(self.path(), self.data())
 
 
 class filectx(basefilectx):
     """A filecontext object makes access to data related to a particular
        filerevision convenient."""
 
     def __init__(
         self,
         repo,
         path,
         changeid=None,
         fileid=None,
         filelog=None,
         changectx=None,
     ):
         """changeid must be a revision number, if specified.
            fileid can be a file revision or node."""
         self._repo = repo
         self._path = path
 
         assert (
             changeid is not None or fileid is not None or changectx is not None
         ), (
             b"bad args: changeid=%r, fileid=%r, changectx=%r"
             % (changeid, fileid, changectx,)
         )
 
         if filelog is not None:
             self._filelog = filelog
 
         if changeid is not None:
             self._changeid = changeid
         if changectx is not None:
             self._changectx = changectx
         if fileid is not None:
             self._fileid = fileid
 
     @propertycache
     def _changectx(self):
         try:
             return self._repo[self._changeid]
         except error.FilteredRepoLookupError:
             # Linkrev may point to any revision in the repository.  When the
             # repository is filtered this may lead to `filectx` trying to build
             # `changectx` for filtered revision. In such case we fallback to
             # creating `changectx` on the unfiltered version of the reposition.
             # This fallback should not be an issue because `changectx` from
             # `filectx` are not used in complex operations that care about
             # filtering.
             #
             # This fallback is a cheap and dirty fix that prevent several
             # crashes. It does not ensure the behavior is correct. However the
             # behavior was not correct before filtering either and "incorrect
             # behavior" is seen as better as "crash"
             #
             # Linkrevs have several serious troubles with filtering that are
             # complicated to solve. Proper handling of the issue here should be
             # considered when solving linkrev issue are on the table.
             return self._repo.unfiltered()[self._changeid]
 
     def filectx(self, fileid, changeid=None):
         '''opens an arbitrary revision of the file without
         opening a new filelog'''
         return filectx(
             self._repo,
             self._path,
             fileid=fileid,
             filelog=self._filelog,
             changeid=changeid,
         )
 
     def rawdata(self):
         return self._filelog.rawdata(self._filenode)
 
     def rawflags(self):
         """low-level revlog flags"""
         return self._filelog.flags(self._filerev)
 
     def data(self):
         try:
             return self._filelog.read(self._filenode)
         except error.CensoredNodeError:
             if self._repo.ui.config(b"censor", b"policy") == b"ignore":
                 return b""
             raise error.Abort(
                 _(b"censored node: %s") % short(self._filenode),
                 hint=_(b"set censor.policy to ignore errors"),
             )
 
     def size(self):
         return self._filelog.size(self._filerev)
 
     @propertycache
     def _copied(self):
         """check if file was actually renamed in this changeset revision
 
         If rename logged in file revision, we report copy for changeset only
         if file revisions linkrev points back to the changeset in question
         or both changeset parents contain different file revisions.
         """
 
         renamed = self._filelog.renamed(self._filenode)
         if not renamed:
             return None
 
         if self.rev() == self.linkrev():
             return renamed
 
         name = self.path()
         fnode = self._filenode
         for p in self._changectx.parents():
             try:
                 if fnode == p.filenode(name):
                     return None
             except error.LookupError:
                 pass
         return renamed
 
     def children(self):
         # hard for renames
         c = self._filelog.children(self._filenode)
         return [
             filectx(self._repo, self._path, fileid=x, filelog=self._filelog)
             for x in c
         ]
 
 
 class committablectx(basectx):
     """A committablectx object provides common functionality for a context that
     wants the ability to commit, e.g. workingctx or memctx."""
 
     def __init__(
         self,
         repo,
         text=b"",
         user=None,
         date=None,
         extra=None,
         changes=None,
         branch=None,
     ):
         super(committablectx, self).__init__(repo)
         self._rev = None
         self._node = None
         self._text = text
         if date:
             self._date = dateutil.parsedate(date)
         if user:
             self._user = user
         if changes:
             self._status = changes
 
         self._extra = {}
         if extra:
             self._extra = extra.copy()
         if branch is not None:
             self._extra[b'branch'] = encoding.fromlocal(branch)
         if not self._extra.get(b'branch'):
             self._extra[b'branch'] = b'default'
 
     def __bytes__(self):
         return bytes(self._parents[0]) + b"+"
 
     __str__ = encoding.strmethod(__bytes__)
 
     def __nonzero__(self):
         return True
 
     __bool__ = __nonzero__
 
     @propertycache
     def _status(self):
         return self._repo.status()
 
     @propertycache
     def _user(self):
         return self._repo.ui.username()
 
     @propertycache
     def _date(self):
         ui = self._repo.ui
         date = ui.configdate(b'devel', b'default-date')
         if date is None:
             date = dateutil.makedate()
         return date
 
     def subrev(self, subpath):
         return None
 
     def manifestnode(self):
         return None
 
     def user(self):
         return self._user or self._repo.ui.username()
 
     def date(self):
         return self._date
 
     def description(self):
         return self._text
 
     def files(self):
         return sorted(
             self._status.modified + self._status.added + self._status.removed
         )
 
     def modified(self):
         return self._status.modified
 
     def added(self):
         return self._status.added
 
     def removed(self):
         return self._status.removed
 
     def deleted(self):
         return self._status.deleted
 
     filesmodified = modified
     filesadded = added
     filesremoved = removed
 
     def branch(self):
         return encoding.tolocal(self._extra[b'branch'])
 
     def closesbranch(self):
         return b'close' in self._extra
 
     def extra(self):
         return self._extra
 
     def isinmemory(self):
         return False
 
     def tags(self):
         return []
 
     def bookmarks(self):
         b = []
         for p in self.parents():
             b.extend(p.bookmarks())
         return b
 
     def phase(self):
         phase = phases.newcommitphase(self._repo.ui)
         for p in self.parents():
             phase = max(phase, p.phase())
         return phase
 
     def hidden(self):
         return False
 
     def children(self):
         return []
 
     def flags(self, path):
         if '_manifest' in self.__dict__:
             try:
                 return self._manifest.flags(path)
             except KeyError:
                 return b''
 
         try:
             return self._flagfunc(path)
         except OSError:
             return b''
 
     def ancestor(self, c2):
         """return the "best" ancestor context of self and c2"""
         return self._parents[0].ancestor(c2)  # punt on two parents for now
 
     def ancestors(self):
         for p in self._parents:
             yield p
         for a in self._repo.changelog.ancestors(
             [p.rev() for p in self._parents]
         ):
             yield self._repo[a]
 
     def markcommitted(self, node):
         """Perform post-commit cleanup necessary after committing this ctx
 
         Specifically, this updates backing stores this working context
         wraps to reflect the fact that the changes reflected by this
         workingctx have been committed.  For example, it marks
         modified and added files as normal in the dirstate.
 
         """
 
     def dirty(self, missing=False, merge=True, branch=True):
         return False
 
 
 class workingctx(committablectx):
     """A workingctx object makes access to data related to
     the current working directory convenient.
     date - any valid date string or (unixtime, offset), or None.
     user - username string, or None.
     extra - a dictionary of extra values, or None.
     changes - a list of file lists as returned by localrepo.status()
                or None to use the repository status.
     """
 
     def __init__(
         self, repo, text=b"", user=None, date=None, extra=None, changes=None
     ):
         branch = None
         if not extra or b'branch' not in extra:
             try:
                 branch = repo.dirstate.branch()
             except UnicodeDecodeError:
                 raise error.Abort(_(b'branch name not in UTF-8!'))
         super(workingctx, self).__init__(
             repo, text, user, date, extra, changes, branch=branch
         )
 
     def __iter__(self):
         d = self._repo.dirstate
         for f in d:
             if d[f] != b'r':
                 yield f
 
     def __contains__(self, key):
         return self._repo.dirstate[key] not in b"?r"
 
     def hex(self):
         return wdirhex
 
     @propertycache
     def _parents(self):
         p = self._repo.dirstate.parents()
         if p[1] == nullid:
             p = p[:-1]
         # use unfiltered repo to delay/avoid loading obsmarkers
         unfi = self._repo.unfiltered()
         return [
             changectx(
                 self._repo, unfi.changelog.rev(n), n, maybe_filtered=False
             )
             for n in p
         ]
 
     def setparents(self, p1node, p2node=nullid):
         dirstate = self._repo.dirstate
         with dirstate.parentchange():
             copies = dirstate.setparents(p1node, p2node)
             pctx = self._repo[p1node]
             if copies:
                 # Adjust copy records, the dirstate cannot do it, it
                 # requires access to parents manifests. Preserve them
                 # only for entries added to first parent.
                 for f in copies:
                     if f not in pctx and copies[f] in pctx:
                         dirstate.copy(copies[f], f)
             if p2node == nullid:
                 for f, s in sorted(dirstate.copies().items()):
                     if f not in pctx and s not in pctx:
                         dirstate.copy(None, f)
 
     def _fileinfo(self, path):
         # populate __dict__['_manifest'] as workingctx has no _manifestdelta
         self._manifest
         return super(workingctx, self)._fileinfo(path)
 
     def _buildflagfunc(self):
         # Create a fallback function for getting file flags when the
         # filesystem doesn't support them
 
         copiesget = self._repo.dirstate.copies().get
         parents = self.parents()
         if len(parents) < 2:
             # when we have one parent, it's easy: copy from parent
             man = parents[0].manifest()
 
             def func(f):
                 f = copiesget(f, f)
                 return man.flags(f)
 
         else:
             # merges are tricky: we try to reconstruct the unstored
             # result from the merge (issue1802)
             p1, p2 = parents
             pa = p1.ancestor(p2)
             m1, m2, ma = p1.manifest(), p2.manifest(), pa.manifest()
 
             def func(f):
                 f = copiesget(f, f)  # may be wrong for merges with copies
                 fl1, fl2, fla = m1.flags(f), m2.flags(f), ma.flags(f)
                 if fl1 == fl2:
                     return fl1
                 if fl1 == fla:
                     return fl2
                 if fl2 == fla:
                     return fl1
                 return b''  # punt for conflicts
 
         return func
 
     @propertycache
     def _flagfunc(self):
         return self._repo.dirstate.flagfunc(self._buildflagfunc)
 
     def flags(self, path):
         try:
             return self._flagfunc(path)
         except OSError:
             return b''
 
     def filectx(self, path, filelog=None):
         """get a file context from the working directory"""
         return workingfilectx(
             self._repo, path, workingctx=self, filelog=filelog
         )
 
     def dirty(self, missing=False, merge=True, branch=True):
         """check whether a working directory is modified"""
         # check subrepos first
         for s in sorted(self.substate):
             if self.sub(s).dirty(missing=missing):
                 return True
         # check current working dir
         return (
             (merge and self.p2())
             or (branch and self.branch() != self.p1().branch())
             or self.modified()
             or self.added()
             or self.removed()
             or (missing and self.deleted())
         )
 
     def add(self, list, prefix=b""):
         with self._repo.wlock():
             ui, ds = self._repo.ui, self._repo.dirstate
             uipath = lambda f: ds.pathto(pathutil.join(prefix, f))
             rejected = []
             lstat = self._repo.wvfs.lstat
             for f in list:
                 # ds.pathto() returns an absolute file when this is invoked from
                 # the keyword extension.  That gets flagged as non-portable on
                 # Windows, since it contains the drive letter and colon.
                 scmutil.checkportable(ui, os.path.join(prefix, f))
                 try:
                     st = lstat(f)
                 except OSError:
                     ui.warn(_(b"%s does not exist!\n") % uipath(f))
                     rejected.append(f)
                     continue
                 limit = ui.configbytes(b'ui', b'large-file-limit')
                 if limit != 0 and st.st_size > limit:
                     ui.warn(
                         _(
                             b"%s: up to %d MB of RAM may be required "
                             b"to manage this file\n"
                             b"(use 'hg revert %s' to cancel the "
                             b"pending addition)\n"
                         )
                         % (f, 3 * st.st_size // 1000000, uipath(f))
                     )
                 if not (stat.S_ISREG(st.st_mode) or stat.S_ISLNK(st.st_mode)):
                     ui.warn(
                         _(
                             b"%s not added: only files and symlinks "
                             b"supported currently\n"
                         )
                         % uipath(f)
                     )
                     rejected.append(f)
                 elif ds[f] in b'amn':
                     ui.warn(_(b"%s already tracked!\n") % uipath(f))
                 elif ds[f] == b'r':
                     ds.normallookup(f)
                 else:
                     ds.add(f)
             return rejected
 
     def forget(self, files, prefix=b""):
         with self._repo.wlock():
             ds = self._repo.dirstate
             uipath = lambda f: ds.pathto(pathutil.join(prefix, f))
             rejected = []
             for f in files:
                 if f not in ds:
                     self._repo.ui.warn(_(b"%s not tracked!\n") % uipath(f))
                     rejected.append(f)
                 elif ds[f] != b'a':
                     ds.remove(f)
                 else:
                     ds.drop(f)
             return rejected
 
     def copy(self, source, dest):
         try:
             st = self._repo.wvfs.lstat(dest)
         except OSError as err:
             if err.errno != errno.ENOENT:
                 raise
             self._repo.ui.warn(
                 _(b"%s does not exist!\n") % self._repo.dirstate.pathto(dest)
             )
             return
         if not (stat.S_ISREG(st.st_mode) or stat.S_ISLNK(st.st_mode)):
             self._repo.ui.warn(
                 _(b"copy failed: %s is not a file or a symbolic link\n")
                 % self._repo.dirstate.pathto(dest)
             )
         else:
             with self._repo.wlock():
                 ds = self._repo.dirstate
                 if ds[dest] in b'?':
                     ds.add(dest)
                 elif ds[dest] in b'r':
                     ds.normallookup(dest)
                 ds.copy(source, dest)
 
     def match(
         self,
         pats=None,
         include=None,
         exclude=None,
         default=b'glob',
         listsubrepos=False,
         badfn=None,
         cwd=None,
     ):
         r = self._repo
         if not cwd:
             cwd = r.getcwd()
 
         # Only a case insensitive filesystem needs magic to translate user input
         # to actual case in the filesystem.
         icasefs = not util.fscasesensitive(r.root)
         return matchmod.match(
             r.root,
             cwd,
             pats,
             include,
             exclude,
             default,
             auditor=r.auditor,
             ctx=self,
             listsubrepos=listsubrepos,
             badfn=badfn,
             icasefs=icasefs,
         )
 
     def _filtersuspectsymlink(self, files):
         if not files or self._repo.dirstate._checklink:
             return files
 
         # Symlink placeholders may get non-symlink-like contents
         # via user error or dereferencing by NFS or Samba servers,
         # so we filter out any placeholders that don't look like a
         # symlink
         sane = []
         for f in files:
             if self.flags(f) == b'l':
                 d = self[f].data()
                 if (
                     d == b''
                     or len(d) >= 1024
                     or b'\n' in d
                     or stringutil.binary(d)
                 ):
                     self._repo.ui.debug(
                         b'ignoring suspect symlink placeholder "%s"\n' % f
                     )
                     continue
             sane.append(f)
         return sane
 
     def _checklookup(self, files):
         # check for any possibly clean files
         if not files:
             return [], [], []
 
         modified = []
         deleted = []
         fixup = []
         pctx = self._parents[0]
         # do a full compare of any files that might have changed
         for f in sorted(files):
             try:
                 # This will return True for a file that got replaced by a
                 # directory in the interim, but fixing that is pretty hard.
                 if (
                     f not in pctx
                     or self.flags(f) != pctx.flags(f)
                     or pctx[f].cmp(self[f])
                 ):
                     modified.append(f)
                 else:
                     fixup.append(f)
             except (IOError, OSError):
                 # A file become inaccessible in between? Mark it as deleted,
                 # matching dirstate behavior (issue5584).
                 # The dirstate has more complex behavior around whether a
                 # missing file matches a directory, etc, but we don't need to
                 # bother with that: if f has made it to this point, we're sure
                 # it's in the dirstate.
                 deleted.append(f)
 
         return modified, deleted, fixup
 
     def _poststatusfixup(self, status, fixup):
         """update dirstate for files that are actually clean"""
         poststatus = self._repo.postdsstatus()
         if fixup or poststatus:
             try:
                 oldid = self._repo.dirstate.identity()
 
                 # updating the dirstate is optional
                 # so we don't wait on the lock
                 # wlock can invalidate the dirstate, so cache normal _after_
                 # taking the lock
                 with self._repo.wlock(False):
                     if self._repo.dirstate.identity() == oldid:
                         if fixup:
                             normal = self._repo.dirstate.normal
                             for f in fixup:
                                 normal(f)
                             # write changes out explicitly, because nesting
                             # wlock at runtime may prevent 'wlock.release()'
                             # after this block from doing so for subsequent
                             # changing files
                             tr = self._repo.currenttransaction()
                             self._repo.dirstate.write(tr)
 
                         if poststatus:
                             for ps in poststatus:
                                 ps(self, status)
                     else:
                         # in this case, writing changes out breaks
                         # consistency, because .hg/dirstate was
                         # already changed simultaneously after last
                         # caching (see also issue5584 for detail)
                         self._repo.ui.debug(
                             b'skip updating dirstate: identity mismatch\n'
                         )
             except error.LockError:
                 pass
             finally:
                 # Even if the wlock couldn't be grabbed, clear out the list.
                 self._repo.clearpostdsstatus()
 
     def _dirstatestatus(self, match, ignored=False, clean=False, unknown=False):
         '''Gets the status from the dirstate -- internal use only.'''
         subrepos = []
         if b'.hgsub' in self:
             subrepos = sorted(self.substate)
         cmp, s = self._repo.dirstate.status(
             match, subrepos, ignored=ignored, clean=clean, unknown=unknown
         )
 
         # check for any possibly clean files
         fixup = []
         if cmp:
             modified2, deleted2, fixup = self._checklookup(cmp)
             s.modified.extend(modified2)
             s.deleted.extend(deleted2)
 
             if fixup and clean:
                 s.clean.extend(fixup)
 
         self._poststatusfixup(s, fixup)
 
         if match.always():
             # cache for performance
             if s.unknown or s.ignored or s.clean:
                 # "_status" is cached with list*=False in the normal route
                 self._status = scmutil.status(
                     s.modified, s.added, s.removed, s.deleted, [], [], []
                 )
             else:
                 self._status = s
 
         return s
 
     @propertycache
     def _copies(self):
         p1copies = {}
         p2copies = {}
         parents = self._repo.dirstate.parents()
         p1manifest = self._repo[parents[0]].manifest()
         p2manifest = self._repo[parents[1]].manifest()
         changedset = set(self.added()) | set(self.modified())
         narrowmatch = self._repo.narrowmatch()
         for dst, src in self._repo.dirstate.copies().items():
             if dst not in changedset or not narrowmatch(dst):
                 continue
             if src in p1manifest:
                 p1copies[dst] = src
             elif src in p2manifest:
                 p2copies[dst] = src
         return p1copies, p2copies
 
     @propertycache
     def _manifest(self):
         """generate a manifest corresponding to the values in self._status
 
         This reuse the file nodeid from parent, but we use special node
         identifiers for added and modified files. This is used by manifests
         merge to see that files are different and by update logic to avoid
         deleting newly added files.
         """
         return self._buildstatusmanifest(self._status)
 
     def _buildstatusmanifest(self, status):
         """Builds a manifest that includes the given status results."""
         parents = self.parents()
 
         man = parents[0].manifest().copy()
 
         ff = self._flagfunc
         for i, l in (
             (addednodeid, status.added),
             (modifiednodeid, status.modified),
         ):
             for f in l:
                 man[f] = i
                 try:
                     man.setflag(f, ff(f))
                 except OSError:
                     pass
 
         for f in status.deleted + status.removed:
             if f in man:
                 del man[f]
 
         return man
 
     def _buildstatus(
         self, other, s, match, listignored, listclean, listunknown
     ):
         """build a status with respect to another context
 
         This includes logic for maintaining the fast path of status when
         comparing the working directory against its parent, which is to skip
         building a new manifest if self (working directory) is not comparing
         against its parent (repo['.']).
         """
         s = self._dirstatestatus(match, listignored, listclean, listunknown)
         # Filter out symlinks that, in the case of FAT32 and NTFS filesystems,
         # might have accidentally ended up with the entire contents of the file
         # they are supposed to be linking to.
         s.modified[:] = self._filtersuspectsymlink(s.modified)
         if other != self._repo[b'.']:
             s = super(workingctx, self)._buildstatus(
                 other, s, match, listignored, listclean, listunknown
             )
         return s
 
     def _matchstatus(self, other, match):
         """override the match method with a filter for directory patterns
 
         We use inheritance to customize the match.bad method only in cases of
         workingctx since it belongs only to the working directory when
         comparing against the parent changeset.
 
         If we aren't comparing against the working directory's parent, then we
         just use the default match object sent to us.
         """
         if other != self._repo[b'.']:
 
             def bad(f, msg):
                 # 'f' may be a directory pattern from 'match.files()',
                 # so 'f not in ctx1' is not enough
                 if f not in other and not other.hasdir(f):
                     self._repo.ui.warn(
                         b'%s: %s\n' % (self._repo.dirstate.pathto(f), msg)
                     )
 
             match.bad = bad
         return match
 
     def walk(self, match):
         '''Generates matching file names.'''
         return sorted(
             self._repo.dirstate.walk(
                 self._repo.narrowmatch(match),
                 subrepos=sorted(self.substate),
                 unknown=True,
                 ignored=False,
             )
         )
 
     def matches(self, match):
         match = self._repo.narrowmatch(match)
         ds = self._repo.dirstate
         return sorted(f for f in ds.matches(match) if ds[f] != b'r')
 
     def markcommitted(self, node):
         with self._repo.dirstate.parentchange():
             for f in self.modified() + self.added():
                 self._repo.dirstate.normal(f)
             for f in self.removed():
                 self._repo.dirstate.drop(f)
             self._repo.dirstate.setparents(node)
             self._repo._quick_access_changeid_invalidate()
 
         # write changes out explicitly, because nesting wlock at
         # runtime may prevent 'wlock.release()' in 'repo.commit()'
         # from immediately doing so for subsequent changing files
         self._repo.dirstate.write(self._repo.currenttransaction())
 
         sparse.aftercommit(self._repo, node)
 
     def mergestate(self, clean=False):
         if clean:
             return mergestatemod.mergestate.clean(self._repo)
         return mergestatemod.mergestate.read(self._repo)
 
 
 class committablefilectx(basefilectx):
     """A committablefilectx provides common functionality for a file context
     that wants the ability to commit, e.g. workingfilectx or memfilectx."""
 
     def __init__(self, repo, path, filelog=None, ctx=None):
         self._repo = repo
         self._path = path
         self._changeid = None
         self._filerev = self._filenode = None
 
         if filelog is not None:
             self._filelog = filelog
         if ctx:
             self._changectx = ctx
 
     def __nonzero__(self):
         return True
 
     __bool__ = __nonzero__
 
     def linkrev(self):
         # linked to self._changectx no matter if file is modified or not
         return self.rev()
 
     def renamed(self):
         path = self.copysource()
         if not path:
             return None
         return path, self._changectx._parents[0]._manifest.get(path, nullid)
 
     def parents(self):
         '''return parent filectxs, following copies if necessary'''
 
         def filenode(ctx, path):
             return ctx._manifest.get(path, nullid)
 
         path = self._path
         fl = self._filelog
         pcl = self._changectx._parents
         renamed = self.renamed()
 
         if renamed:
             pl = [renamed + (None,)]
         else:
             pl = [(path, filenode(pcl[0], path), fl)]
 
         for pc in pcl[1:]:
             pl.append((path, filenode(pc, path), fl))
 
         return [
             self._parentfilectx(p, fileid=n, filelog=l)
             for p, n, l in pl
             if n != nullid
         ]
 
     def children(self):
         return []
 
 
 class workingfilectx(committablefilectx):
     """A workingfilectx object makes access to data related to a particular
        file in the working directory convenient."""
 
     def __init__(self, repo, path, filelog=None, workingctx=None):
         super(workingfilectx, self).__init__(repo, path, filelog, workingctx)
 
     @propertycache
     def _changectx(self):
         return workingctx(self._repo)
 
     def data(self):
         return self._repo.wread(self._path)
 
     def copysource(self):
         return self._repo.dirstate.copied(self._path)
 
     def size(self):
         return self._repo.wvfs.lstat(self._path).st_size
 
     def lstat(self):
         return self._repo.wvfs.lstat(self._path)
 
     def date(self):
         t, tz = self._changectx.date()
         try:
             return (self._repo.wvfs.lstat(self._path)[stat.ST_MTIME], tz)
         except OSError as err:
             if err.errno != errno.ENOENT:
                 raise
             return (t, tz)
 
     def exists(self):
         return self._repo.wvfs.exists(self._path)
 
     def lexists(self):
         return self._repo.wvfs.lexists(self._path)
 
     def audit(self):
         return self._repo.wvfs.audit(self._path)
 
     def cmp(self, fctx):
         """compare with other file context
 
         returns True if different than fctx.
         """
         # fctx should be a filectx (not a workingfilectx)
         # invert comparison to reuse the same code path
         return fctx.cmp(self)
 
     def remove(self, ignoremissing=False):
         """wraps unlink for a repo's working directory"""
         rmdir = self._repo.ui.configbool(b'experimental', b'removeemptydirs')
         self._repo.wvfs.unlinkpath(
             self._path, ignoremissing=ignoremissing, rmdir=rmdir
         )
 
     def write(self, data, flags, backgroundclose=False, **kwargs):
         """wraps repo.wwrite"""
         return self._repo.wwrite(
             self._path, data, flags, backgroundclose=backgroundclose, **kwargs
         )
 
     def markcopied(self, src):
         """marks this file a copy of `src`"""
         self._repo.dirstate.copy(src, self._path)
 
     def clearunknown(self):
         """Removes conflicting items in the working directory so that
         ``write()`` can be called successfully.
         """
         wvfs = self._repo.wvfs
         f = self._path
         wvfs.audit(f)
         if self._repo.ui.configbool(
             b'experimental', b'merge.checkpathconflicts'
         ):
             # remove files under the directory as they should already be
             # warned and backed up
             if wvfs.isdir(f) and not wvfs.islink(f):
                 wvfs.rmtree(f, forcibly=True)
             for p in reversed(list(pathutil.finddirs(f))):
                 if wvfs.isfileorlink(p):
                     wvfs.unlink(p)
                     break
         else:
             # don't remove files if path conflicts are not processed
             if wvfs.isdir(f) and not wvfs.islink(f):
                 wvfs.removedirs(f)
 
     def setflags(self, l, x):
         self._repo.wvfs.setflags(self._path, l, x)
 
 
 class overlayworkingctx(committablectx):
     """Wraps another mutable context with a write-back cache that can be
     converted into a commit context.
 
     self._cache[path] maps to a dict with keys: {
         'exists': bool?
         'date': date?
         'data': str?
         'flags': str?
         'copied': str? (path or None)
     }
     If `exists` is True, `flags` must be non-None and 'date' is non-None. If it
     is `False`, the file was deleted.
     """
 
     def __init__(self, repo):
         super(overlayworkingctx, self).__init__(repo)
         self.clean()
 
     def setbase(self, wrappedctx):
         self._wrappedctx = wrappedctx
         self._parents = [wrappedctx]
         # Drop old manifest cache as it is now out of date.
         # This is necessary when, e.g., rebasing several nodes with one
         # ``overlayworkingctx`` (e.g. with --collapse).
         util.clearcachedproperty(self, b'_manifest')
 
     def setparents(self, p1node, p2node=nullid):
         assert p1node == self._wrappedctx.node()
         self._parents = [self._wrappedctx, self._repo.unfiltered()[p2node]]
 
     def data(self, path):
         if self.isdirty(path):
             if self._cache[path][b'exists']:
                 if self._cache[path][b'data'] is not None:
                     return self._cache[path][b'data']
                 else:
                     # Must fallback here, too, because we only set flags.
                     return self._wrappedctx[path].data()
             else:
                 raise error.ProgrammingError(
                     b"No such file or directory: %s" % path
                 )
         else:
             return self._wrappedctx[path].data()
 
     @propertycache
     def _manifest(self):
         parents = self.parents()
         man = parents[0].manifest().copy()
 
         flag = self._flagfunc
         for path in self.added():
             man[path] = addednodeid
             man.setflag(path, flag(path))
         for path in self.modified():
             man[path] = modifiednodeid
             man.setflag(path, flag(path))
         for path in self.removed():
             del man[path]
         return man
 
     @propertycache
     def _flagfunc(self):
         def f(path):
             return self._cache[path][b'flags']
 
         return f
 
     def files(self):
         return sorted(self.added() + self.modified() + self.removed())
 
     def modified(self):
         return [
             f
             for f in self._cache.keys()
             if self._cache[f][b'exists'] and self._existsinparent(f)
         ]
 
     def added(self):
         return [
             f
             for f in self._cache.keys()
             if self._cache[f][b'exists'] and not self._existsinparent(f)
         ]
 
     def removed(self):
         return [
             f
             for f in self._cache.keys()
             if not self._cache[f][b'exists'] and self._existsinparent(f)
         ]
 
     def p1copies(self):
         copies = {}
         narrowmatch = self._repo.narrowmatch()
         for f in self._cache.keys():
             if not narrowmatch(f):
                 continue
             copies.pop(f, None)  # delete if it exists
             source = self._cache[f][b'copied']
             if source:
                 copies[f] = source
         return copies
 
     def p2copies(self):
         copies = {}
         narrowmatch = self._repo.narrowmatch()
         for f in self._cache.keys():
             if not narrowmatch(f):
                 continue
             copies.pop(f, None)  # delete if it exists
             source = self._cache[f][b'copied']
             if source:
                 copies[f] = source
         return copies
 
     def isinmemory(self):
         return True
 
     def filedate(self, path):
         if self.isdirty(path):
             return self._cache[path][b'date']
         else:
             return self._wrappedctx[path].date()
 
     def markcopied(self, path, origin):
         self._markdirty(
             path,
             exists=True,
             date=self.filedate(path),
             flags=self.flags(path),
             copied=origin,
         )
 
     def copydata(self, path):
         if self.isdirty(path):
             return self._cache[path][b'copied']
         else:
             return None
 
     def flags(self, path):
         if self.isdirty(path):
             if self._cache[path][b'exists']:
                 return self._cache[path][b'flags']
             else:
                 raise error.ProgrammingError(
                     b"No such file or directory: %s" % path
                 )
         else:
             return self._wrappedctx[path].flags()
 
     def __contains__(self, key):
         if key in self._cache:
             return self._cache[key][b'exists']
         return key in self.p1()
 
     def _existsinparent(self, path):
         try:
             # ``commitctx` raises a ``ManifestLookupError`` if a path does not
             # exist, unlike ``workingctx``, which returns a ``workingfilectx``
             # with an ``exists()`` function.
             self._wrappedctx[path]
             return True
         except error.ManifestLookupError:
             return False
 
     def _auditconflicts(self, path):
         """Replicates conflict checks done by wvfs.write().
 
         Since we never write to the filesystem and never call `applyupdates` in
         IMM, we'll never check that a path is actually writable -- e.g., because
         it adds `a/foo`, but `a` is actually a file in the other commit.
         """
 
         def fail(path, component):
             # p1() is the base and we're receiving "writes" for p2()'s
             # files.
             if b'l' in self.p1()[component].flags():
                 raise error.Abort(
                     b"error: %s conflicts with symlink %s "
                     b"in %d." % (path, component, self.p1().rev())
                 )
             else:
                 raise error.Abort(
                     b"error: '%s' conflicts with file '%s' in "
                     b"%d." % (path, component, self.p1().rev())
                 )
 
         # Test that each new directory to be created to write this path from p2
         # is not a file in p1.
         components = path.split(b'/')
         for i in pycompat.xrange(len(components)):
             component = b"/".join(components[0:i])
             if component in self:
                 fail(path, component)
 
         # Test the other direction -- that this path from p2 isn't a directory
         # in p1 (test that p1 doesn't have any paths matching `path/*`).
         match = self.match([path], default=b'path')
         mfiles = list(self.p1().manifest().walk(match))
         if len(mfiles) > 0:
             if len(mfiles) == 1 and mfiles[0] == path:
                 return
             # omit the files which are deleted in current IMM wctx
             mfiles = [m for m in mfiles if m in self]
             if not mfiles:
                 return
             raise error.Abort(
                 b"error: file '%s' cannot be written because "
                 b" '%s/' is a directory in %s (containing %d "
                 b"entries: %s)"
                 % (path, path, self.p1(), len(mfiles), b', '.join(mfiles))
             )
 
     def write(self, path, data, flags=b'', **kwargs):
         if data is None:
             raise error.ProgrammingError(b"data must be non-None")
         self._auditconflicts(path)
         self._markdirty(
             path, exists=True, data=data, date=dateutil.makedate(), flags=flags
         )
 
     def setflags(self, path, l, x):
         flag = b''
         if l:
             flag = b'l'
         elif x:
             flag = b'x'
         self._markdirty(path, exists=True, date=dateutil.makedate(), flags=flag)
 
     def remove(self, path):
         self._markdirty(path, exists=False)
 
     def exists(self, path):
         """exists behaves like `lexists`, but needs to follow symlinks and
         return False if they are broken.
         """
         if self.isdirty(path):
             # If this path exists and is a symlink, "follow" it by calling
             # exists on the destination path.
             if (
                 self._cache[path][b'exists']
                 and b'l' in self._cache[path][b'flags']
             ):
                 return self.exists(self._cache[path][b'data'].strip())
             else:
                 return self._cache[path][b'exists']
 
         return self._existsinparent(path)
 
     def lexists(self, path):
         """lexists returns True if the path exists"""
         if self.isdirty(path):
             return self._cache[path][b'exists']
 
         return self._existsinparent(path)
 
     def size(self, path):
         if self.isdirty(path):
             if self._cache[path][b'exists']:
                 return len(self._cache[path][b'data'])
             else:
                 raise error.ProgrammingError(
                     b"No such file or directory: %s" % path
                 )
         return self._wrappedctx[path].size()
 
     def tomemctx(
         self,
         text,
         branch=None,
         extra=None,
         date=None,
         parents=None,
         user=None,
         editor=None,
     ):
         """Converts this ``overlayworkingctx`` into a ``memctx`` ready to be
         committed.
 
         ``text`` is the commit message.
         ``parents`` (optional) are rev numbers.
         """
         # Default parents to the wrapped context if not passed.
         if parents is None:
             parents = self.parents()
             if len(parents) == 1:
                 parents = (parents[0], None)
 
         # ``parents`` is passed as rev numbers; convert to ``commitctxs``.
         if parents[1] is None:
             parents = (self._repo[parents[0]], None)
         else:
             parents = (self._repo[parents[0]], self._repo[parents[1]])
 
         files = self.files()
 
         def getfile(repo, memctx, path):
             if self._cache[path][b'exists']:
                 return memfilectx(
                     repo,
                     memctx,
                     path,
                     self._cache[path][b'data'],
                     b'l' in self._cache[path][b'flags'],
                     b'x' in self._cache[path][b'flags'],
                     self._cache[path][b'copied'],
                 )
             else:
                 # Returning None, but including the path in `files`, is
                 # necessary for memctx to register a deletion.
                 return None
 
         if branch is None:
             branch = self._wrappedctx.branch()
 
         return memctx(
             self._repo,
             parents,
             text,
             files,
             getfile,
             date=date,
             extra=extra,
             user=user,
             branch=branch,
             editor=editor,
         )
 
     def tomemctx_for_amend(self, precursor):
         extra = precursor.extra().copy()
         extra[b'amend_source'] = precursor.hex()
         return self.tomemctx(
             text=precursor.description(),
             branch=precursor.branch(),
             extra=extra,
             date=precursor.date(),
             user=precursor.user(),
         )
 
     def isdirty(self, path):
         return path in self._cache
 
     def isempty(self):
         # We need to discard any keys that are actually clean before the empty
         # commit check.
         self._compact()
         return len(self._cache) == 0
 
     def clean(self):
         self._cache = {}
 
     def _compact(self):
         """Removes keys from the cache that are actually clean, by comparing
         them with the underlying context.
 
         This can occur during the merge process, e.g. by passing --tool :local
         to resolve a conflict.
         """
         keys = []
         # This won't be perfect, but can help performance significantly when
         # using things like remotefilelog.
         scmutil.prefetchfiles(
             self.repo(),
-            [self.p1().rev()],
-            scmutil.matchfiles(self.repo(), self._cache.keys()),
+            [(self.p1().rev(),
+              scmutil.matchfiles(self.repo(), self._cache.keys()))],
         )
 
         for path in self._cache.keys():
             cache = self._cache[path]
             try:
                 underlying = self._wrappedctx[path]
                 if (
                     underlying.data() == cache[b'data']
                     and underlying.flags() == cache[b'flags']
                 ):
                     keys.append(path)
             except error.ManifestLookupError:
                 # Path not in the underlying manifest (created).
                 continue
 
         for path in keys:
             del self._cache[path]
         return keys
 
     def _markdirty(
         self, path, exists, data=None, date=None, flags=b'', copied=None
     ):
         # data not provided, let's see if we already have some; if not, let's
         # grab it from our underlying context, so that we always have data if
         # the file is marked as existing.
         if exists and data is None:
             oldentry = self._cache.get(path) or {}
             data = oldentry.get(b'data')
             if data is None:
                 data = self._wrappedctx[path].data()
 
         self._cache[path] = {
             b'exists': exists,
             b'data': data,
             b'date': date,
             b'flags': flags,
             b'copied': copied,
         }
 
     def filectx(self, path, filelog=None):
         return overlayworkingfilectx(
             self._repo, path, parent=self, filelog=filelog
         )
 
 
 class overlayworkingfilectx(committablefilectx):
     """Wrap a ``workingfilectx`` but intercepts all writes into an in-memory
     cache, which can be flushed through later by calling ``flush()``."""
 
     def __init__(self, repo, path, filelog=None, parent=None):
         super(overlayworkingfilectx, self).__init__(repo, path, filelog, parent)
         self._repo = repo
         self._parent = parent
         self._path = path
 
     def cmp(self, fctx):
         return self.data() != fctx.data()
 
     def changectx(self):
         return self._parent
 
     def data(self):
         return self._parent.data(self._path)
 
     def date(self):
         return self._parent.filedate(self._path)
 
     def exists(self):
         return self.lexists()
 
     def lexists(self):
         return self._parent.exists(self._path)
 
     def copysource(self):
         return self._parent.copydata(self._path)
 
     def size(self):
         return self._parent.size(self._path)
 
     def markcopied(self, origin):
         self._parent.markcopied(self._path, origin)
 
     def audit(self):
         pass
 
     def flags(self):
         return self._parent.flags(self._path)
 
     def setflags(self, islink, isexec):
         return self._parent.setflags(self._path, islink, isexec)
 
     def write(self, data, flags, backgroundclose=False, **kwargs):
         return self._parent.write(self._path, data, flags, **kwargs)
 
     def remove(self, ignoremissing=False):
         return self._parent.remove(self._path)
 
     def clearunknown(self):
         pass
 
 
 class workingcommitctx(workingctx):
     """A workingcommitctx object makes access to data related to
     the revision being committed convenient.
 
     This hides changes in the working directory, if they aren't
     committed in this context.
     """
 
     def __init__(
         self, repo, changes, text=b"", user=None, date=None, extra=None
     ):
         super(workingcommitctx, self).__init__(
             repo, text, user, date, extra, changes
         )
 
     def _dirstatestatus(self, match, ignored=False, clean=False, unknown=False):
         """Return matched files only in ``self._status``
 
         Uncommitted files appear "clean" via this context, even if
         they aren't actually so in the working directory.
         """
         if clean:
             clean = [f for f in self._manifest if f not in self._changedset]
         else:
             clean = []
         return scmutil.status(
             [f for f in self._status.modified if match(f)],
             [f for f in self._status.added if match(f)],
             [f for f in self._status.removed if match(f)],
             [],
             [],
             [],
             clean,
         )
 
     @propertycache
     def _changedset(self):
         """Return the set of files changed in this context
         """
         changed = set(self._status.modified)
         changed.update(self._status.added)
         changed.update(self._status.removed)
         return changed
 
 
 def makecachingfilectxfn(func):
     """Create a filectxfn that caches based on the path.
 
     We can't use util.cachefunc because it uses all arguments as the cache
     key and this creates a cycle since the arguments include the repo and
     memctx.
     """
     cache = {}
 
     def getfilectx(repo, memctx, path):
         if path not in cache:
             cache[path] = func(repo, memctx, path)
         return cache[path]
 
     return getfilectx
 
 
 def memfilefromctx(ctx):
     """Given a context return a memfilectx for ctx[path]
 
     This is a convenience method for building a memctx based on another
     context.
     """
 
     def getfilectx(repo, memctx, path):
         fctx = ctx[path]
         copysource = fctx.copysource()
         return memfilectx(
             repo,
             memctx,
             path,
             fctx.data(),
             islink=fctx.islink(),
             isexec=fctx.isexec(),
             copysource=copysource,
         )
 
     return getfilectx
 
 
 def memfilefrompatch(patchstore):
     """Given a patch (e.g. patchstore object) return a memfilectx
 
     This is a convenience method for building a memctx based on a patchstore.
     """
 
     def getfilectx(repo, memctx, path):
         data, mode, copysource = patchstore.getfile(path)
         if data is None:
             return None
         islink, isexec = mode
         return memfilectx(
             repo,
             memctx,
             path,
             data,
             islink=islink,
             isexec=isexec,
             copysource=copysource,
         )
 
     return getfilectx
 
 
 class memctx(committablectx):
     """Use memctx to perform in-memory commits via localrepo.commitctx().
 
     Revision information is supplied at initialization time while
     related files data and is made available through a callback
     mechanism.  'repo' is the current localrepo, 'parents' is a
     sequence of two parent revisions identifiers (pass None for every
     missing parent), 'text' is the commit message and 'files' lists
     names of files touched by the revision (normalized and relative to
     repository root).
 
     filectxfn(repo, memctx, path) is a callable receiving the
     repository, the current memctx object and the normalized path of
     requested file, relative to repository root. It is fired by the
     commit function for every file in 'files', but calls order is
     undefined. If the file is available in the revision being
     committed (updated or added), filectxfn returns a memfilectx
     object. If the file was removed, filectxfn return None for recent
     Mercurial. Moved files are represented by marking the source file
     removed and the new file added with copy information (see
     memfilectx).
 
     user receives the committer name and defaults to current
     repository username, date is the commit date in any format
     supported by dateutil.parsedate() and defaults to current date, extra
     is a dictionary of metadata or is left empty.
     """
 
     # Mercurial <= 3.1 expects the filectxfn to raise IOError for missing files.
     # Extensions that need to retain compatibility across Mercurial 3.1 can use
     # this field to determine what to do in filectxfn.
     _returnnoneformissingfiles = True
 
     def __init__(
         self,
         repo,
         parents,
         text,
         files,
         filectxfn,
         user=None,
         date=None,
         extra=None,
         branch=None,
         editor=None,
     ):
         super(memctx, self).__init__(
             repo, text, user, date, extra, branch=branch
         )
         self._rev = None
         self._node = None
         parents = [(p or nullid) for p in parents]
         p1, p2 = parents
         self._parents = [self._repo[p] for p in (p1, p2)]
         files = sorted(set(files))
         self._files = files
         self.substate = {}
 
         if isinstance(filectxfn, patch.filestore):
             filectxfn = memfilefrompatch(filectxfn)
         elif not callable(filectxfn):
             # if store is not callable, wrap it in a function
             filectxfn = memfilefromctx(filectxfn)
 
         # memoizing increases performance for e.g. vcs convert scenarios.
         self._filectxfn = makecachingfilectxfn(filectxfn)
 
         if editor:
             self._text = editor(self._repo, self, [])
             self._repo.savecommitmessage(self._text)
 
     def filectx(self, path, filelog=None):
         """get a file context from the working directory
 
         Returns None if file doesn't exist and should be removed."""
         return self._filectxfn(self._repo, self, path)
 
     def commit(self):
         """commit context to the repo"""
         return self._repo.commitctx(self)
 
     @propertycache
     def _manifest(self):
         """generate a manifest based on the return values of filectxfn"""
 
         # keep this simple for now; just worry about p1
         pctx = self._parents[0]
         man = pctx.manifest().copy()
 
         for f in self._status.modified:
             man[f] = modifiednodeid
 
         for f in self._status.added:
             man[f] = addednodeid
 
         for f in self._status.removed:
             if f in man:
                 del man[f]
 
         return man
 
     @propertycache
     def _status(self):
         """Calculate exact status from ``files`` specified at construction
         """
         man1 = self.p1().manifest()
         p2 = self._parents[1]
         # "1 < len(self._parents)" can't be used for checking
         # existence of the 2nd parent, because "memctx._parents" is
         # explicitly initialized by the list, of which length is 2.
         if p2.node() != nullid:
             man2 = p2.manifest()
             managing = lambda f: f in man1 or f in man2
         else:
             managing = lambda f: f in man1
 
         modified, added, removed = [], [], []
         for f in self._files:
             if not managing(f):
                 added.append(f)
             elif self[f]:
                 modified.append(f)
             else:
                 removed.append(f)
 
         return scmutil.status(modified, added, removed, [], [], [], [])
 
 
 class memfilectx(committablefilectx):
     """memfilectx represents an in-memory file to commit.
 
     See memctx and committablefilectx for more details.
     """
 
     def __init__(
         self,
         repo,
         changectx,
         path,
         data,
         islink=False,
         isexec=False,
         copysource=None,
     ):
         """
         path is the normalized file path relative to repository root.
         data is the file content as a string.
         islink is True if the file is a symbolic link.
         isexec is True if the file is executable.
         copied is the source file path if current file was copied in the
         revision being committed, or None."""
         super(memfilectx, self).__init__(repo, path, None, changectx)
         self._data = data
         if islink:
             self._flags = b'l'
         elif isexec:
             self._flags = b'x'
         else:
             self._flags = b''
         self._copysource = copysource
 
     def copysource(self):
         return self._copysource
 
     def cmp(self, fctx):
         return self.data() != fctx.data()
 
     def data(self):
         return self._data
 
     def remove(self, ignoremissing=False):
         """wraps unlink for a repo's working directory"""
         # need to figure out what to do here
         del self._changectx[self._path]
 
     def write(self, data, flags, **kwargs):
         """wraps repo.wwrite"""
         self._data = data
 
 
 class metadataonlyctx(committablectx):
     """Like memctx but it's reusing the manifest of different commit.
     Intended to be used by lightweight operations that are creating
     metadata-only changes.
 
     Revision information is supplied at initialization time.  'repo' is the
     current localrepo, 'ctx' is original revision which manifest we're reuisng
     'parents' is a sequence of two parent revisions identifiers (pass None for
     every missing parent), 'text' is the commit.
 
     user receives the committer name and defaults to current repository
     username, date is the commit date in any format supported by
     dateutil.parsedate() and defaults to current date, extra is a dictionary of
     metadata or is left empty.
     """
 
     def __init__(
         self,
         repo,
         originalctx,
         parents=None,
         text=None,
         user=None,
         date=None,
         extra=None,
         editor=None,
     ):
         if text is None:
             text = originalctx.description()
         super(metadataonlyctx, self).__init__(repo, text, user, date, extra)
         self._rev = None
         self._node = None
         self._originalctx = originalctx
         self._manifestnode = originalctx.manifestnode()
         if parents is None:
             parents = originalctx.parents()
         else:
             parents = [repo[p] for p in parents if p is not None]
         parents = parents[:]
         while len(parents) < 2:
             parents.append(repo[nullid])
         p1, p2 = self._parents = parents
 
         # sanity check to ensure that the reused manifest parents are
         # manifests of our commit parents
         mp1, mp2 = self.manifestctx().parents
         if p1 != nullid and p1.manifestnode() != mp1:
             raise RuntimeError(
                 r"can't reuse the manifest: its p1 "
                 r"doesn't match the new ctx p1"
             )
         if p2 != nullid and p2.manifestnode() != mp2:
             raise RuntimeError(
                 r"can't reuse the manifest: "
                 r"its p2 doesn't match the new ctx p2"
             )
 
         self._files = originalctx.files()
         self.substate = {}
 
         if editor:
             self._text = editor(self._repo, self, [])
             self._repo.savecommitmessage(self._text)
 
     def manifestnode(self):
         return self._manifestnode
 
     @property
     def _manifestctx(self):
         return self._repo.manifestlog[self._manifestnode]
 
     def filectx(self, path, filelog=None):
         return self._originalctx.filectx(path, filelog=filelog)
 
     def commit(self):
         """commit context to the repo"""
         return self._repo.commitctx(self)
 
     @property
     def _manifest(self):
         return self._originalctx.manifest()
 
     @propertycache
     def _status(self):
         """Calculate exact status from ``files`` specified in the ``origctx``
         and parents manifests.
         """
         man1 = self.p1().manifest()
         p2 = self._parents[1]
         # "1 < len(self._parents)" can't be used for checking
         # existence of the 2nd parent, because "metadataonlyctx._parents" is
         # explicitly initialized by the list, of which length is 2.
         if p2.node() != nullid:
             man2 = p2.manifest()
             managing = lambda f: f in man1 or f in man2
         else:
             managing = lambda f: f in man1
 
         modified, added, removed = [], [], []
         for f in self._files:
             if not managing(f):
                 added.append(f)
             elif f in self:
                 modified.append(f)
             else:
                 removed.append(f)
 
         return scmutil.status(modified, added, removed, [], [], [], [])
 
 
 class arbitraryfilectx(object):
     """Allows you to use filectx-like functions on a file in an arbitrary
     location on disk, possibly not in the working directory.
     """
 
     def __init__(self, path, repo=None):
         # Repo is optional because contrib/simplemerge uses this class.
         self._repo = repo
         self._path = path
 
     def cmp(self, fctx):
         # filecmp follows symlinks whereas `cmp` should not, so skip the fast
         # path if either side is a symlink.
         symlinks = b'l' in self.flags() or b'l' in fctx.flags()
         if not symlinks and isinstance(fctx, workingfilectx) and self._repo:
             # Add a fast-path for merge if both sides are disk-backed.
             # Note that filecmp uses the opposite return values (True if same)
             # from our cmp functions (True if different).
             return not filecmp.cmp(self.path(), self._repo.wjoin(fctx.path()))
         return self.data() != fctx.data()
 
     def path(self):
         return self._path
 
     def flags(self):
         return b''
 
     def data(self):
         return util.readfile(self._path)
 
     def decodeddata(self):
         with open(self._path, b"rb") as f:
             return f.read()
 
     def remove(self):
         util.unlink(self._path)
 
     def write(self, data, flags, **kwargs):
         assert not flags
         with open(self._path, b"wb") as f:
             f.write(data)

mercurial/merge.py

@@ -1,2077 +1,2078 b''
 # merge.py - directory-level update/merge handling for Mercurial
 #
 # Copyright 2006, 2007 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import errno
 import stat
 import struct
 
 from .i18n import _
 from .node import (
     addednodeid,
     modifiednodeid,
     nullid,
     nullrev,
 )
 from .thirdparty import attr
 from . import (
     copies,
     encoding,
     error,
     filemerge,
     match as matchmod,
     mergestate as mergestatemod,
     obsutil,
     pathutil,
     pycompat,
     scmutil,
     subrepoutil,
     util,
     worker,
 )
 
 _pack = struct.pack
 _unpack = struct.unpack
 
 
 def _getcheckunknownconfig(repo, section, name):
     config = repo.ui.config(section, name)
     valid = [b'abort', b'ignore', b'warn']
     if config not in valid:
         validstr = b', '.join([b"'" + v + b"'" for v in valid])
         raise error.ConfigError(
             _(b"%s.%s not valid ('%s' is none of %s)")
             % (section, name, config, validstr)
         )
     return config
 
 
 def _checkunknownfile(repo, wctx, mctx, f, f2=None):
     if wctx.isinmemory():
         # Nothing to do in IMM because nothing in the "working copy" can be an
         # unknown file.
         #
         # Note that we should bail out here, not in ``_checkunknownfiles()``,
         # because that function does other useful work.
         return False
 
     if f2 is None:
         f2 = f
     return (
         repo.wvfs.audit.check(f)
         and repo.wvfs.isfileorlink(f)
         and repo.dirstate.normalize(f) not in repo.dirstate
         and mctx[f2].cmp(wctx[f])
     )
 
 
 class _unknowndirschecker(object):
     """
     Look for any unknown files or directories that may have a path conflict
     with a file.  If any path prefix of the file exists as a file or link,
     then it conflicts.  If the file itself is a directory that contains any
     file that is not tracked, then it conflicts.
 
     Returns the shortest path at which a conflict occurs, or None if there is
     no conflict.
     """
 
     def __init__(self):
         # A set of paths known to be good.  This prevents repeated checking of
         # dirs.  It will be updated with any new dirs that are checked and found
         # to be safe.
         self._unknowndircache = set()
 
         # A set of paths that are known to be absent.  This prevents repeated
         # checking of subdirectories that are known not to exist. It will be
         # updated with any new dirs that are checked and found to be absent.
         self._missingdircache = set()
 
     def __call__(self, repo, wctx, f):
         if wctx.isinmemory():
             # Nothing to do in IMM for the same reason as ``_checkunknownfile``.
             return False
 
         # Check for path prefixes that exist as unknown files.
         for p in reversed(list(pathutil.finddirs(f))):
             if p in self._missingdircache:
                 return
             if p in self._unknowndircache:
                 continue
             if repo.wvfs.audit.check(p):
                 if (
                     repo.wvfs.isfileorlink(p)
                     and repo.dirstate.normalize(p) not in repo.dirstate
                 ):
                     return p
                 if not repo.wvfs.lexists(p):
                     self._missingdircache.add(p)
                     return
                 self._unknowndircache.add(p)
 
         # Check if the file conflicts with a directory containing unknown files.
         if repo.wvfs.audit.check(f) and repo.wvfs.isdir(f):
             # Does the directory contain any files that are not in the dirstate?
             for p, dirs, files in repo.wvfs.walk(f):
                 for fn in files:
                     relf = util.pconvert(repo.wvfs.reljoin(p, fn))
                     relf = repo.dirstate.normalize(relf, isknown=True)
                     if relf not in repo.dirstate:
                         return f
         return None
 
 
 def _checkunknownfiles(repo, wctx, mctx, force, actions, mergeforce):
     """
     Considers any actions that care about the presence of conflicting unknown
     files. For some actions, the result is to abort; for others, it is to
     choose a different action.
     """
     fileconflicts = set()
     pathconflicts = set()
     warnconflicts = set()
     abortconflicts = set()
     unknownconfig = _getcheckunknownconfig(repo, b'merge', b'checkunknown')
     ignoredconfig = _getcheckunknownconfig(repo, b'merge', b'checkignored')
     pathconfig = repo.ui.configbool(
         b'experimental', b'merge.checkpathconflicts'
     )
     if not force:
 
         def collectconflicts(conflicts, config):
             if config == b'abort':
                 abortconflicts.update(conflicts)
             elif config == b'warn':
                 warnconflicts.update(conflicts)
 
         checkunknowndirs = _unknowndirschecker()
         for f, (m, args, msg) in pycompat.iteritems(actions):
             if m in (
                 mergestatemod.ACTION_CREATED,
                 mergestatemod.ACTION_DELETED_CHANGED,
             ):
                 if _checkunknownfile(repo, wctx, mctx, f):
                     fileconflicts.add(f)
                 elif pathconfig and f not in wctx:
                     path = checkunknowndirs(repo, wctx, f)
                     if path is not None:
                         pathconflicts.add(path)
             elif m == mergestatemod.ACTION_LOCAL_DIR_RENAME_GET:
                 if _checkunknownfile(repo, wctx, mctx, f, args[0]):
                     fileconflicts.add(f)
 
         allconflicts = fileconflicts | pathconflicts
         ignoredconflicts = {c for c in allconflicts if repo.dirstate._ignore(c)}
         unknownconflicts = allconflicts - ignoredconflicts
         collectconflicts(ignoredconflicts, ignoredconfig)
         collectconflicts(unknownconflicts, unknownconfig)
     else:
         for f, (m, args, msg) in pycompat.iteritems(actions):
             if m == mergestatemod.ACTION_CREATED_MERGE:
                 fl2, anc = args
                 different = _checkunknownfile(repo, wctx, mctx, f)
                 if repo.dirstate._ignore(f):
                     config = ignoredconfig
                 else:
                     config = unknownconfig
 
                 # The behavior when force is True is described by this table:
                 #  config  different  mergeforce  |    action    backup
                 #    *         n          *       |      get        n
                 #    *         y          y       |     merge       -
                 #   abort      y          n       |     merge       -   (1)
                 #   warn       y          n       |  warn + get     y
                 #  ignore      y          n       |      get        y
                 #
                 # (1) this is probably the wrong behavior here -- we should
                 #     probably abort, but some actions like rebases currently
                 #     don't like an abort happening in the middle of
                 #     merge.update.
                 if not different:
                     actions[f] = (
                         mergestatemod.ACTION_GET,
                         (fl2, False),
                         b'remote created',
                     )
                 elif mergeforce or config == b'abort':
                     actions[f] = (
                         mergestatemod.ACTION_MERGE,
                         (f, f, None, False, anc),
                         b'remote differs from untracked local',
                     )
                 elif config == b'abort':
                     abortconflicts.add(f)
                 else:
                     if config == b'warn':
                         warnconflicts.add(f)
                     actions[f] = (
                         mergestatemod.ACTION_GET,
                         (fl2, True),
                         b'remote created',
                     )
 
     for f in sorted(abortconflicts):
         warn = repo.ui.warn
         if f in pathconflicts:
             if repo.wvfs.isfileorlink(f):
                 warn(_(b"%s: untracked file conflicts with directory\n") % f)
             else:
                 warn(_(b"%s: untracked directory conflicts with file\n") % f)
         else:
             warn(_(b"%s: untracked file differs\n") % f)
     if abortconflicts:
         raise error.Abort(
             _(
                 b"untracked files in working directory "
                 b"differ from files in requested revision"
             )
         )
 
     for f in sorted(warnconflicts):
         if repo.wvfs.isfileorlink(f):
             repo.ui.warn(_(b"%s: replacing untracked file\n") % f)
         else:
             repo.ui.warn(_(b"%s: replacing untracked files in directory\n") % f)
 
     for f, (m, args, msg) in pycompat.iteritems(actions):
         if m == mergestatemod.ACTION_CREATED:
             backup = (
                 f in fileconflicts
                 or f in pathconflicts
                 or any(p in pathconflicts for p in pathutil.finddirs(f))
             )
             (flags,) = args
             actions[f] = (mergestatemod.ACTION_GET, (flags, backup), msg)
 
 
 def _forgetremoved(wctx, mctx, branchmerge):
     """
     Forget removed files
 
     If we're jumping between revisions (as opposed to merging), and if
     neither the working directory nor the target rev has the file,
     then we need to remove it from the dirstate, to prevent the
     dirstate from listing the file when it is no longer in the
     manifest.
 
     If we're merging, and the other revision has removed a file
     that is not present in the working directory, we need to mark it
     as removed.
     """
 
     actions = {}
     m = mergestatemod.ACTION_FORGET
     if branchmerge:
         m = mergestatemod.ACTION_REMOVE
     for f in wctx.deleted():
         if f not in mctx:
             actions[f] = m, None, b"forget deleted"
 
     if not branchmerge:
         for f in wctx.removed():
             if f not in mctx:
                 actions[f] = (
                     mergestatemod.ACTION_FORGET,
                     None,
                     b"forget removed",
                 )
 
     return actions
 
 
 def _checkcollision(repo, wmf, actions):
     """
     Check for case-folding collisions.
     """
     # If the repo is narrowed, filter out files outside the narrowspec.
     narrowmatch = repo.narrowmatch()
     if not narrowmatch.always():
         pmmf = set(wmf.walk(narrowmatch))
         if actions:
             narrowactions = {}
             for m, actionsfortype in pycompat.iteritems(actions):
                 narrowactions[m] = []
                 for (f, args, msg) in actionsfortype:
                     if narrowmatch(f):
                         narrowactions[m].append((f, args, msg))
             actions = narrowactions
     else:
         # build provisional merged manifest up
         pmmf = set(wmf)
 
     if actions:
         # KEEP and EXEC are no-op
         for m in (
             mergestatemod.ACTION_ADD,
             mergestatemod.ACTION_ADD_MODIFIED,
             mergestatemod.ACTION_FORGET,
             mergestatemod.ACTION_GET,
             mergestatemod.ACTION_CHANGED_DELETED,
             mergestatemod.ACTION_DELETED_CHANGED,
         ):
             for f, args, msg in actions[m]:
                 pmmf.add(f)
         for f, args, msg in actions[mergestatemod.ACTION_REMOVE]:
             pmmf.discard(f)
         for f, args, msg in actions[mergestatemod.ACTION_DIR_RENAME_MOVE_LOCAL]:
             f2, flags = args
             pmmf.discard(f2)
             pmmf.add(f)
         for f, args, msg in actions[mergestatemod.ACTION_LOCAL_DIR_RENAME_GET]:
             pmmf.add(f)
         for f, args, msg in actions[mergestatemod.ACTION_MERGE]:
             f1, f2, fa, move, anc = args
             if move:
                 pmmf.discard(f1)
             pmmf.add(f)
 
     # check case-folding collision in provisional merged manifest
     foldmap = {}
     for f in pmmf:
         fold = util.normcase(f)
         if fold in foldmap:
             raise error.Abort(
                 _(b"case-folding collision between %s and %s")
                 % (f, foldmap[fold])
             )
         foldmap[fold] = f
 
     # check case-folding of directories
     foldprefix = unfoldprefix = lastfull = b''
     for fold, f in sorted(foldmap.items()):
         if fold.startswith(foldprefix) and not f.startswith(unfoldprefix):
             # the folded prefix matches but actual casing is different
             raise error.Abort(
                 _(b"case-folding collision between %s and directory of %s")
                 % (lastfull, f)
             )
         foldprefix = fold + b'/'
         unfoldprefix = f + b'/'
         lastfull = f
 
 
 def driverpreprocess(repo, ms, wctx, labels=None):
     """run the preprocess step of the merge driver, if any
 
     This is currently not implemented -- it's an extension point."""
     return True
 
 
 def driverconclude(repo, ms, wctx, labels=None):
     """run the conclude step of the merge driver, if any
 
     This is currently not implemented -- it's an extension point."""
     return True
 
 
 def _filesindirs(repo, manifest, dirs):
     """
     Generator that yields pairs of all the files in the manifest that are found
     inside the directories listed in dirs, and which directory they are found
     in.
     """
     for f in manifest:
         for p in pathutil.finddirs(f):
             if p in dirs:
                 yield f, p
                 break
 
 
 def checkpathconflicts(repo, wctx, mctx, actions):
     """
     Check if any actions introduce path conflicts in the repository, updating
     actions to record or handle the path conflict accordingly.
     """
     mf = wctx.manifest()
 
     # The set of local files that conflict with a remote directory.
     localconflicts = set()
 
     # The set of directories that conflict with a remote file, and so may cause
     # conflicts if they still contain any files after the merge.
     remoteconflicts = set()
 
     # The set of directories that appear as both a file and a directory in the
     # remote manifest.  These indicate an invalid remote manifest, which
     # can't be updated to cleanly.
     invalidconflicts = set()
 
     # The set of directories that contain files that are being created.
     createdfiledirs = set()
 
     # The set of files deleted by all the actions.
     deletedfiles = set()
 
     for f, (m, args, msg) in actions.items():
         if m in (
             mergestatemod.ACTION_CREATED,
             mergestatemod.ACTION_DELETED_CHANGED,
             mergestatemod.ACTION_MERGE,
             mergestatemod.ACTION_CREATED_MERGE,
         ):
             # This action may create a new local file.
             createdfiledirs.update(pathutil.finddirs(f))
             if mf.hasdir(f):
                 # The file aliases a local directory.  This might be ok if all
                 # the files in the local directory are being deleted.  This
                 # will be checked once we know what all the deleted files are.
                 remoteconflicts.add(f)
         # Track the names of all deleted files.
         if m == mergestatemod.ACTION_REMOVE:
             deletedfiles.add(f)
         if m == mergestatemod.ACTION_MERGE:
             f1, f2, fa, move, anc = args
             if move:
                 deletedfiles.add(f1)
         if m == mergestatemod.ACTION_DIR_RENAME_MOVE_LOCAL:
             f2, flags = args
             deletedfiles.add(f2)
 
     # Check all directories that contain created files for path conflicts.
     for p in createdfiledirs:
         if p in mf:
             if p in mctx:
                 # A file is in a directory which aliases both a local
                 # and a remote file.  This is an internal inconsistency
                 # within the remote manifest.
                 invalidconflicts.add(p)
             else:
                 # A file is in a directory which aliases a local file.
                 # We will need to rename the local file.
                 localconflicts.add(p)
         if p in actions and actions[p][0] in (
             mergestatemod.ACTION_CREATED,
             mergestatemod.ACTION_DELETED_CHANGED,
             mergestatemod.ACTION_MERGE,
             mergestatemod.ACTION_CREATED_MERGE,
         ):
             # The file is in a directory which aliases a remote file.
             # This is an internal inconsistency within the remote
             # manifest.
             invalidconflicts.add(p)
 
     # Rename all local conflicting files that have not been deleted.
     for p in localconflicts:
         if p not in deletedfiles:
             ctxname = bytes(wctx).rstrip(b'+')
             pnew = util.safename(p, ctxname, wctx, set(actions.keys()))
             porig = wctx[p].copysource() or p
             actions[pnew] = (
                 mergestatemod.ACTION_PATH_CONFLICT_RESOLVE,
                 (p, porig),
                 b'local path conflict',
             )
             actions[p] = (
                 mergestatemod.ACTION_PATH_CONFLICT,
                 (pnew, b'l'),
                 b'path conflict',
             )
 
     if remoteconflicts:
         # Check if all files in the conflicting directories have been removed.
         ctxname = bytes(mctx).rstrip(b'+')
         for f, p in _filesindirs(repo, mf, remoteconflicts):
             if f not in deletedfiles:
                 m, args, msg = actions[p]
                 pnew = util.safename(p, ctxname, wctx, set(actions.keys()))
                 if m in (
                     mergestatemod.ACTION_DELETED_CHANGED,
                     mergestatemod.ACTION_MERGE,
                 ):
                     # Action was merge, just update target.
                     actions[pnew] = (m, args, msg)
                 else:
                     # Action was create, change to renamed get action.
                     fl = args[0]
                     actions[pnew] = (
                         mergestatemod.ACTION_LOCAL_DIR_RENAME_GET,
                         (p, fl),
                         b'remote path conflict',
                     )
                 actions[p] = (
                     mergestatemod.ACTION_PATH_CONFLICT,
                     (pnew, mergestatemod.ACTION_REMOVE),
                     b'path conflict',
                 )
                 remoteconflicts.remove(p)
                 break
 
     if invalidconflicts:
         for p in invalidconflicts:
             repo.ui.warn(_(b"%s: is both a file and a directory\n") % p)
         raise error.Abort(_(b"destination manifest contains path conflicts"))
 
 
 def _filternarrowactions(narrowmatch, branchmerge, actions):
     """
     Filters out actions that can ignored because the repo is narrowed.
 
     Raise an exception if the merge cannot be completed because the repo is
     narrowed.
     """
     nooptypes = {b'k'}  # TODO: handle with nonconflicttypes
     nonconflicttypes = set(b'a am c cm f g gs r e'.split())
     # We mutate the items in the dict during iteration, so iterate
     # over a copy.
     for f, action in list(actions.items()):
         if narrowmatch(f):
             pass
         elif not branchmerge:
             del actions[f]  # just updating, ignore changes outside clone
         elif action[0] in nooptypes:
             del actions[f]  # merge does not affect file
         elif action[0] in nonconflicttypes:
             raise error.Abort(
                 _(
                     b'merge affects file \'%s\' outside narrow, '
                     b'which is not yet supported'
                 )
                 % f,
                 hint=_(b'merging in the other direction may work'),
             )
         else:
             raise error.Abort(
                 _(b'conflict in file \'%s\' is outside narrow clone') % f
             )
 
 
 def manifestmerge(
     repo,
     wctx,
     p2,
     pa,
     branchmerge,
     force,
     matcher,
     acceptremote,
     followcopies,
     forcefulldiff=False,
 ):
     """
     Merge wctx and p2 with ancestor pa and generate merge action list
 
     branchmerge and force are as passed in to update
     matcher = matcher to filter file lists
     acceptremote = accept the incoming changes without prompting
     """
     if matcher is not None and matcher.always():
         matcher = None
 
     # manifests fetched in order are going to be faster, so prime the caches
     [
         x.manifest()
         for x in sorted(wctx.parents() + [p2, pa], key=scmutil.intrev)
     ]
 
     branch_copies1 = copies.branch_copies()
     branch_copies2 = copies.branch_copies()
     diverge = {}
     if followcopies:
         branch_copies1, branch_copies2, diverge = copies.mergecopies(
             repo, wctx, p2, pa
         )
 
     boolbm = pycompat.bytestr(bool(branchmerge))
     boolf = pycompat.bytestr(bool(force))
     boolm = pycompat.bytestr(bool(matcher))
     repo.ui.note(_(b"resolving manifests\n"))
     repo.ui.debug(
         b" branchmerge: %s, force: %s, partial: %s\n" % (boolbm, boolf, boolm)
     )
     repo.ui.debug(b" ancestor: %s, local: %s, remote: %s\n" % (pa, wctx, p2))
 
     m1, m2, ma = wctx.manifest(), p2.manifest(), pa.manifest()
     copied1 = set(branch_copies1.copy.values())
     copied1.update(branch_copies1.movewithdir.values())
     copied2 = set(branch_copies2.copy.values())
     copied2.update(branch_copies2.movewithdir.values())
 
     if b'.hgsubstate' in m1 and wctx.rev() is None:
         # Check whether sub state is modified, and overwrite the manifest
         # to flag the change. If wctx is a committed revision, we shouldn't
         # care for the dirty state of the working directory.
         if any(wctx.sub(s).dirty() for s in wctx.substate):
             m1[b'.hgsubstate'] = modifiednodeid
 
     # Don't use m2-vs-ma optimization if:
     # - ma is the same as m1 or m2, which we're just going to diff again later
     # - The caller specifically asks for a full diff, which is useful during bid
     #   merge.
     if pa not in ([wctx, p2] + wctx.parents()) and not forcefulldiff:
         # Identify which files are relevant to the merge, so we can limit the
         # total m1-vs-m2 diff to just those files. This has significant
         # performance benefits in large repositories.
         relevantfiles = set(ma.diff(m2).keys())
 
         # For copied and moved files, we need to add the source file too.
         for copykey, copyvalue in pycompat.iteritems(branch_copies1.copy):
             if copyvalue in relevantfiles:
                 relevantfiles.add(copykey)
         for movedirkey in branch_copies1.movewithdir:
             relevantfiles.add(movedirkey)
         filesmatcher = scmutil.matchfiles(repo, relevantfiles)
         matcher = matchmod.intersectmatchers(matcher, filesmatcher)
 
     diff = m1.diff(m2, match=matcher)
 
     actions = {}
     for f, ((n1, fl1), (n2, fl2)) in pycompat.iteritems(diff):
         if n1 and n2:  # file exists on both local and remote side
             if f not in ma:
                 # TODO: what if they're renamed from different sources?
                 fa = branch_copies1.copy.get(
                     f, None
                 ) or branch_copies2.copy.get(f, None)
                 if fa is not None:
                     actions[f] = (
                         mergestatemod.ACTION_MERGE,
                         (f, f, fa, False, pa.node()),
                         b'both renamed from %s' % fa,
                     )
                 else:
                     actions[f] = (
                         mergestatemod.ACTION_MERGE,
                         (f, f, None, False, pa.node()),
                         b'both created',
                     )
             else:
                 a = ma[f]
                 fla = ma.flags(f)
                 nol = b'l' not in fl1 + fl2 + fla
                 if n2 == a and fl2 == fla:
                     actions[f] = (
                         mergestatemod.ACTION_KEEP,
                         (),
                         b'remote unchanged',
                     )
                 elif n1 == a and fl1 == fla:  # local unchanged - use remote
                     if n1 == n2:  # optimization: keep local content
                         actions[f] = (
                             mergestatemod.ACTION_EXEC,
                             (fl2,),
                             b'update permissions',
                         )
                     else:
                         actions[f] = (
                             mergestatemod.ACTION_GET_OTHER_AND_STORE
                             if branchmerge
                             else mergestatemod.ACTION_GET,
                             (fl2, False),
                             b'remote is newer',
                         )
                 elif nol and n2 == a:  # remote only changed 'x'
                     actions[f] = (
                         mergestatemod.ACTION_EXEC,
                         (fl2,),
                         b'update permissions',
                     )
                 elif nol and n1 == a:  # local only changed 'x'
                     actions[f] = (
                         mergestatemod.ACTION_GET_OTHER_AND_STORE
                         if branchmerge
                         else mergestatemod.ACTION_GET,
                         (fl1, False),
                         b'remote is newer',
                     )
                 else:  # both changed something
                     actions[f] = (
                         mergestatemod.ACTION_MERGE,
                         (f, f, f, False, pa.node()),
                         b'versions differ',
                     )
         elif n1:  # file exists only on local side
             if f in copied2:
                 pass  # we'll deal with it on m2 side
             elif (
                 f in branch_copies1.movewithdir
             ):  # directory rename, move local
                 f2 = branch_copies1.movewithdir[f]
                 if f2 in m2:
                     actions[f2] = (
                         mergestatemod.ACTION_MERGE,
                         (f, f2, None, True, pa.node()),
                         b'remote directory rename, both created',
                     )
                 else:
                     actions[f2] = (
                         mergestatemod.ACTION_DIR_RENAME_MOVE_LOCAL,
                         (f, fl1),
                         b'remote directory rename - move from %s' % f,
                     )
             elif f in branch_copies1.copy:
                 f2 = branch_copies1.copy[f]
                 actions[f] = (
                     mergestatemod.ACTION_MERGE,
                     (f, f2, f2, False, pa.node()),
                     b'local copied/moved from %s' % f2,
                 )
             elif f in ma:  # clean, a different, no remote
                 if n1 != ma[f]:
                     if acceptremote:
                         actions[f] = (
                             mergestatemod.ACTION_REMOVE,
                             None,
                             b'remote delete',
                         )
                     else:
                         actions[f] = (
                             mergestatemod.ACTION_CHANGED_DELETED,
                             (f, None, f, False, pa.node()),
                             b'prompt changed/deleted',
                         )
                 elif n1 == addednodeid:
                     # This extra 'a' is added by working copy manifest to mark
                     # the file as locally added. We should forget it instead of
                     # deleting it.
                     actions[f] = (
                         mergestatemod.ACTION_FORGET,
                         None,
                         b'remote deleted',
                     )
                 else:
                     actions[f] = (
                         mergestatemod.ACTION_REMOVE,
                         None,
                         b'other deleted',
                     )
         elif n2:  # file exists only on remote side
             if f in copied1:
                 pass  # we'll deal with it on m1 side
             elif f in branch_copies2.movewithdir:
                 f2 = branch_copies2.movewithdir[f]
                 if f2 in m1:
                     actions[f2] = (
                         mergestatemod.ACTION_MERGE,
                         (f2, f, None, False, pa.node()),
                         b'local directory rename, both created',
                     )
                 else:
                     actions[f2] = (
                         mergestatemod.ACTION_LOCAL_DIR_RENAME_GET,
                         (f, fl2),
                         b'local directory rename - get from %s' % f,
                     )
             elif f in branch_copies2.copy:
                 f2 = branch_copies2.copy[f]
                 if f2 in m2:
                     actions[f] = (
                         mergestatemod.ACTION_MERGE,
                         (f2, f, f2, False, pa.node()),
                         b'remote copied from %s' % f2,
                     )
                 else:
                     actions[f] = (
                         mergestatemod.ACTION_MERGE,
                         (f2, f, f2, True, pa.node()),
                         b'remote moved from %s' % f2,
                     )
             elif f not in ma:
                 # local unknown, remote created: the logic is described by the
                 # following table:
                 #
                 # force  branchmerge  different  |  action
                 #   n         *           *      |   create
                 #   y         n           *      |   create
                 #   y         y           n      |   create
                 #   y         y           y      |   merge
                 #
                 # Checking whether the files are different is expensive, so we
                 # don't do that when we can avoid it.
                 if not force:
                     actions[f] = (
                         mergestatemod.ACTION_CREATED,
                         (fl2,),
                         b'remote created',
                     )
                 elif not branchmerge:
                     actions[f] = (
                         mergestatemod.ACTION_CREATED,
                         (fl2,),
                         b'remote created',
                     )
                 else:
                     actions[f] = (
                         mergestatemod.ACTION_CREATED_MERGE,
                         (fl2, pa.node()),
                         b'remote created, get or merge',
                     )
             elif n2 != ma[f]:
                 df = None
                 for d in branch_copies1.dirmove:
                     if f.startswith(d):
                         # new file added in a directory that was moved
                         df = branch_copies1.dirmove[d] + f[len(d) :]
                         break
                 if df is not None and df in m1:
                     actions[df] = (
                         mergestatemod.ACTION_MERGE,
                         (df, f, f, False, pa.node()),
                         b'local directory rename - respect move '
                         b'from %s' % f,
                     )
                 elif acceptremote:
                     actions[f] = (
                         mergestatemod.ACTION_CREATED,
                         (fl2,),
                         b'remote recreating',
                     )
                 else:
                     actions[f] = (
                         mergestatemod.ACTION_DELETED_CHANGED,
                         (None, f, f, False, pa.node()),
                         b'prompt deleted/changed',
                     )
 
     if repo.ui.configbool(b'experimental', b'merge.checkpathconflicts'):
         # If we are merging, look for path conflicts.
         checkpathconflicts(repo, wctx, p2, actions)
 
     narrowmatch = repo.narrowmatch()
     if not narrowmatch.always():
         # Updates "actions" in place
         _filternarrowactions(narrowmatch, branchmerge, actions)
 
     renamedelete = branch_copies1.renamedelete
     renamedelete.update(branch_copies2.renamedelete)
 
     return actions, diverge, renamedelete
 
 
 def _resolvetrivial(repo, wctx, mctx, ancestor, actions):
     """Resolves false conflicts where the nodeid changed but the content
        remained the same."""
     # We force a copy of actions.items() because we're going to mutate
     # actions as we resolve trivial conflicts.
     for f, (m, args, msg) in list(actions.items()):
         if (
             m == mergestatemod.ACTION_CHANGED_DELETED
             and f in ancestor
             and not wctx[f].cmp(ancestor[f])
         ):
             # local did change but ended up with same content
             actions[f] = mergestatemod.ACTION_REMOVE, None, b'prompt same'
         elif (
             m == mergestatemod.ACTION_DELETED_CHANGED
             and f in ancestor
             and not mctx[f].cmp(ancestor[f])
         ):
             # remote did change but ended up with same content
             del actions[f]  # don't get = keep local deleted
 
 
 def calculateupdates(
     repo,
     wctx,
     mctx,
     ancestors,
     branchmerge,
     force,
     acceptremote,
     followcopies,
     matcher=None,
     mergeforce=False,
 ):
     """Calculate the actions needed to merge mctx into wctx using ancestors"""
     # Avoid cycle.
     from . import sparse
 
     if len(ancestors) == 1:  # default
         actions, diverge, renamedelete = manifestmerge(
             repo,
             wctx,
             mctx,
             ancestors[0],
             branchmerge,
             force,
             matcher,
             acceptremote,
             followcopies,
         )
         _checkunknownfiles(repo, wctx, mctx, force, actions, mergeforce)
 
     else:  # only when merge.preferancestor=* - the default
         repo.ui.note(
             _(b"note: merging %s and %s using bids from ancestors %s\n")
             % (
                 wctx,
                 mctx,
                 _(b' and ').join(pycompat.bytestr(anc) for anc in ancestors),
             )
         )
 
         # Call for bids
         fbids = (
             {}
         )  # mapping filename to bids (action method to list af actions)
         diverge, renamedelete = None, None
         for ancestor in ancestors:
             repo.ui.note(_(b'\ncalculating bids for ancestor %s\n') % ancestor)
             actions, diverge1, renamedelete1 = manifestmerge(
                 repo,
                 wctx,
                 mctx,
                 ancestor,
                 branchmerge,
                 force,
                 matcher,
                 acceptremote,
                 followcopies,
                 forcefulldiff=True,
             )
             _checkunknownfiles(repo, wctx, mctx, force, actions, mergeforce)
 
             # Track the shortest set of warning on the theory that bid
             # merge will correctly incorporate more information
             if diverge is None or len(diverge1) < len(diverge):
                 diverge = diverge1
             if renamedelete is None or len(renamedelete) < len(renamedelete1):
                 renamedelete = renamedelete1
 
             for f, a in sorted(pycompat.iteritems(actions)):
                 m, args, msg = a
                 if m == mergestatemod.ACTION_GET_OTHER_AND_STORE:
                     m = mergestatemod.ACTION_GET
                 repo.ui.debug(b' %s: %s -> %s\n' % (f, msg, m))
                 if f in fbids:
                     d = fbids[f]
                     if m in d:
                         d[m].append(a)
                     else:
                         d[m] = [a]
                 else:
                     fbids[f] = {m: [a]}
 
         # Pick the best bid for each file
         repo.ui.note(_(b'\nauction for merging merge bids\n'))
         actions = {}
         for f, bids in sorted(fbids.items()):
             # bids is a mapping from action method to list af actions
             # Consensus?
             if len(bids) == 1:  # all bids are the same kind of method
                 m, l = list(bids.items())[0]
                 if all(a == l[0] for a in l[1:]):  # len(bids) is > 1
                     repo.ui.note(_(b" %s: consensus for %s\n") % (f, m))
                     actions[f] = l[0]
                     continue
             # If keep is an option, just do it.
             if mergestatemod.ACTION_KEEP in bids:
                 repo.ui.note(_(b" %s: picking 'keep' action\n") % f)
                 actions[f] = bids[mergestatemod.ACTION_KEEP][0]
                 continue
             # If there are gets and they all agree [how could they not?], do it.
             if mergestatemod.ACTION_GET in bids:
                 ga0 = bids[mergestatemod.ACTION_GET][0]
                 if all(a == ga0 for a in bids[mergestatemod.ACTION_GET][1:]):
                     repo.ui.note(_(b" %s: picking 'get' action\n") % f)
                     actions[f] = ga0
                     continue
             # TODO: Consider other simple actions such as mode changes
             # Handle inefficient democrazy.
             repo.ui.note(_(b' %s: multiple bids for merge action:\n') % f)
             for m, l in sorted(bids.items()):
                 for _f, args, msg in l:
                     repo.ui.note(b'  %s -> %s\n' % (msg, m))
             # Pick random action. TODO: Instead, prompt user when resolving
             m, l = list(bids.items())[0]
             repo.ui.warn(
                 _(b' %s: ambiguous merge - picked %s action\n') % (f, m)
             )
             actions[f] = l[0]
             continue
         repo.ui.note(_(b'end of auction\n\n'))
 
     if wctx.rev() is None:
         fractions = _forgetremoved(wctx, mctx, branchmerge)
         actions.update(fractions)
 
     prunedactions = sparse.filterupdatesactions(
         repo, wctx, mctx, branchmerge, actions
     )
     _resolvetrivial(repo, wctx, mctx, ancestors[0], actions)
 
     return prunedactions, diverge, renamedelete
 
 
 def _getcwd():
     try:
         return encoding.getcwd()
     except OSError as err:
         if err.errno == errno.ENOENT:
             return None
         raise
 
 
 def batchremove(repo, wctx, actions):
     """apply removes to the working directory
 
     yields tuples for progress updates
     """
     verbose = repo.ui.verbose
     cwd = _getcwd()
     i = 0
     for f, args, msg in actions:
         repo.ui.debug(b" %s: %s -> r\n" % (f, msg))
         if verbose:
             repo.ui.note(_(b"removing %s\n") % f)
         wctx[f].audit()
         try:
             wctx[f].remove(ignoremissing=True)
         except OSError as inst:
             repo.ui.warn(
                 _(b"update failed to remove %s: %s!\n") % (f, inst.strerror)
             )
         if i == 100:
             yield i, f
             i = 0
         i += 1
     if i > 0:
         yield i, f
 
     if cwd and not _getcwd():
         # cwd was removed in the course of removing files; print a helpful
         # warning.
         repo.ui.warn(
             _(
                 b"current directory was removed\n"
                 b"(consider changing to repo root: %s)\n"
             )
             % repo.root
         )
 
 
 def batchget(repo, mctx, wctx, wantfiledata, actions):
     """apply gets to the working directory
 
     mctx is the context to get from
 
     Yields arbitrarily many (False, tuple) for progress updates, followed by
     exactly one (True, filedata). When wantfiledata is false, filedata is an
     empty dict. When wantfiledata is true, filedata[f] is a triple (mode, size,
     mtime) of the file f written for each action.
     """
     filedata = {}
     verbose = repo.ui.verbose
     fctx = mctx.filectx
     ui = repo.ui
     i = 0
     with repo.wvfs.backgroundclosing(ui, expectedcount=len(actions)):
         for f, (flags, backup), msg in actions:
             repo.ui.debug(b" %s: %s -> g\n" % (f, msg))
             if verbose:
                 repo.ui.note(_(b"getting %s\n") % f)
 
             if backup:
                 # If a file or directory exists with the same name, back that
                 # up.  Otherwise, look to see if there is a file that conflicts
                 # with a directory this file is in, and if so, back that up.
                 conflicting = f
                 if not repo.wvfs.lexists(f):
                     for p in pathutil.finddirs(f):
                         if repo.wvfs.isfileorlink(p):
                             conflicting = p
                             break
                 if repo.wvfs.lexists(conflicting):
                     orig = scmutil.backuppath(ui, repo, conflicting)
                     util.rename(repo.wjoin(conflicting), orig)
             wfctx = wctx[f]
             wfctx.clearunknown()
             atomictemp = ui.configbool(b"experimental", b"update.atomic-file")
             size = wfctx.write(
                 fctx(f).data(),
                 flags,
                 backgroundclose=True,
                 atomictemp=atomictemp,
             )
             if wantfiledata:
                 s = wfctx.lstat()
                 mode = s.st_mode
                 mtime = s[stat.ST_MTIME]
                 filedata[f] = (mode, size, mtime)  # for dirstate.normal
             if i == 100:
                 yield False, (i, f)
                 i = 0
             i += 1
     if i > 0:
         yield False, (i, f)
     yield True, filedata
 
 
 def _prefetchfiles(repo, ctx, actions):
     """Invoke ``scmutil.prefetchfiles()`` for the files relevant to the dict
     of merge actions.  ``ctx`` is the context being merged in."""
 
     # Skipping 'a', 'am', 'f', 'r', 'dm', 'e', 'k', 'p' and 'pr', because they
     # don't touch the context to be merged in.  'cd' is skipped, because
     # changed/deleted never resolves to something from the remote side.
     oplist = [
         actions[a]
         for a in (
             mergestatemod.ACTION_GET,
             mergestatemod.ACTION_DELETED_CHANGED,
             mergestatemod.ACTION_LOCAL_DIR_RENAME_GET,
             mergestatemod.ACTION_MERGE,
         )
     ]
     prefetch = scmutil.prefetchfiles
     matchfiles = scmutil.matchfiles
     prefetch(
         repo,
-        [ctx.rev()],
-        matchfiles(repo, [f for sublist in oplist for f, args, msg in sublist]),
+        [(ctx.rev(),
+          matchfiles(repo,
+                     [f for sublist in oplist for f, args, msg in sublist]))],
     )
 
 
 @attr.s(frozen=True)
 class updateresult(object):
     updatedcount = attr.ib()
     mergedcount = attr.ib()
     removedcount = attr.ib()
     unresolvedcount = attr.ib()
 
     def isempty(self):
         return not (
             self.updatedcount
             or self.mergedcount
             or self.removedcount
             or self.unresolvedcount
         )
 
 
 def emptyactions():
     """create an actions dict, to be populated and passed to applyupdates()"""
     return {
         m: []
         for m in (
             mergestatemod.ACTION_ADD,
             mergestatemod.ACTION_ADD_MODIFIED,
             mergestatemod.ACTION_FORGET,
             mergestatemod.ACTION_GET,
             mergestatemod.ACTION_CHANGED_DELETED,
             mergestatemod.ACTION_DELETED_CHANGED,
             mergestatemod.ACTION_REMOVE,
             mergestatemod.ACTION_DIR_RENAME_MOVE_LOCAL,
             mergestatemod.ACTION_LOCAL_DIR_RENAME_GET,
             mergestatemod.ACTION_MERGE,
             mergestatemod.ACTION_EXEC,
             mergestatemod.ACTION_KEEP,
             mergestatemod.ACTION_PATH_CONFLICT,
             mergestatemod.ACTION_PATH_CONFLICT_RESOLVE,
             mergestatemod.ACTION_GET_OTHER_AND_STORE,
         )
     }
 
 
 def applyupdates(
     repo, actions, wctx, mctx, overwrite, wantfiledata, labels=None
 ):
     """apply the merge action list to the working directory
 
     wctx is the working copy context
     mctx is the context to be merged into the working copy
 
     Return a tuple of (counts, filedata), where counts is a tuple
     (updated, merged, removed, unresolved) that describes how many
     files were affected by the update, and filedata is as described in
     batchget.
     """
 
     _prefetchfiles(repo, mctx, actions)
 
     updated, merged, removed = 0, 0, 0
     ms = mergestatemod.mergestate.clean(
         repo, wctx.p1().node(), mctx.node(), labels
     )
 
     # add ACTION_GET_OTHER_AND_STORE to mergestate
     for e in actions[mergestatemod.ACTION_GET_OTHER_AND_STORE]:
         ms.addmergedother(e[0])
 
     moves = []
     for m, l in actions.items():
         l.sort()
 
     # 'cd' and 'dc' actions are treated like other merge conflicts
     mergeactions = sorted(actions[mergestatemod.ACTION_CHANGED_DELETED])
     mergeactions.extend(sorted(actions[mergestatemod.ACTION_DELETED_CHANGED]))
     mergeactions.extend(actions[mergestatemod.ACTION_MERGE])
     for f, args, msg in mergeactions:
         f1, f2, fa, move, anc = args
         if f == b'.hgsubstate':  # merged internally
             continue
         if f1 is None:
             fcl = filemerge.absentfilectx(wctx, fa)
         else:
             repo.ui.debug(b" preserving %s for resolve of %s\n" % (f1, f))
             fcl = wctx[f1]
         if f2 is None:
             fco = filemerge.absentfilectx(mctx, fa)
         else:
             fco = mctx[f2]
         actx = repo[anc]
         if fa in actx:
             fca = actx[fa]
         else:
             # TODO: move to absentfilectx
             fca = repo.filectx(f1, fileid=nullrev)
         ms.add(fcl, fco, fca, f)
         if f1 != f and move:
             moves.append(f1)
 
     # remove renamed files after safely stored
     for f in moves:
         if wctx[f].lexists():
             repo.ui.debug(b"removing %s\n" % f)
             wctx[f].audit()
             wctx[f].remove()
 
     numupdates = sum(
         len(l) for m, l in actions.items() if m != mergestatemod.ACTION_KEEP
     )
     progress = repo.ui.makeprogress(
         _(b'updating'), unit=_(b'files'), total=numupdates
     )
 
     if [
         a
         for a in actions[mergestatemod.ACTION_REMOVE]
         if a[0] == b'.hgsubstate'
     ]:
         subrepoutil.submerge(repo, wctx, mctx, wctx, overwrite, labels)
 
     # record path conflicts
     for f, args, msg in actions[mergestatemod.ACTION_PATH_CONFLICT]:
         f1, fo = args
         s = repo.ui.status
         s(
             _(
                 b"%s: path conflict - a file or link has the same name as a "
                 b"directory\n"
             )
             % f
         )
         if fo == b'l':
             s(_(b"the local file has been renamed to %s\n") % f1)
         else:
             s(_(b"the remote file has been renamed to %s\n") % f1)
         s(_(b"resolve manually then use 'hg resolve --mark %s'\n") % f)
         ms.addpath(f, f1, fo)
         progress.increment(item=f)
 
     # When merging in-memory, we can't support worker processes, so set the
     # per-item cost at 0 in that case.
     cost = 0 if wctx.isinmemory() else 0.001
 
     # remove in parallel (must come before resolving path conflicts and getting)
     prog = worker.worker(
         repo.ui,
         cost,
         batchremove,
         (repo, wctx),
         actions[mergestatemod.ACTION_REMOVE],
     )
     for i, item in prog:
         progress.increment(step=i, item=item)
     removed = len(actions[mergestatemod.ACTION_REMOVE])
 
     # resolve path conflicts (must come before getting)
     for f, args, msg in actions[mergestatemod.ACTION_PATH_CONFLICT_RESOLVE]:
         repo.ui.debug(b" %s: %s -> pr\n" % (f, msg))
         (f0, origf0) = args
         if wctx[f0].lexists():
             repo.ui.note(_(b"moving %s to %s\n") % (f0, f))
             wctx[f].audit()
             wctx[f].write(wctx.filectx(f0).data(), wctx.filectx(f0).flags())
             wctx[f0].remove()
         progress.increment(item=f)
 
     # get in parallel.
     threadsafe = repo.ui.configbool(
         b'experimental', b'worker.wdir-get-thread-safe'
     )
     prog = worker.worker(
         repo.ui,
         cost,
         batchget,
         (repo, mctx, wctx, wantfiledata),
         actions[mergestatemod.ACTION_GET],
         threadsafe=threadsafe,
         hasretval=True,
     )
     getfiledata = {}
     for final, res in prog:
         if final:
             getfiledata = res
         else:
             i, item = res
             progress.increment(step=i, item=item)
     updated = len(actions[mergestatemod.ACTION_GET])
 
     if [a for a in actions[mergestatemod.ACTION_GET] if a[0] == b'.hgsubstate']:
         subrepoutil.submerge(repo, wctx, mctx, wctx, overwrite, labels)
 
     # forget (manifest only, just log it) (must come first)
     for f, args, msg in actions[mergestatemod.ACTION_FORGET]:
         repo.ui.debug(b" %s: %s -> f\n" % (f, msg))
         progress.increment(item=f)
 
     # re-add (manifest only, just log it)
     for f, args, msg in actions[mergestatemod.ACTION_ADD]:
         repo.ui.debug(b" %s: %s -> a\n" % (f, msg))
         progress.increment(item=f)
 
     # re-add/mark as modified (manifest only, just log it)
     for f, args, msg in actions[mergestatemod.ACTION_ADD_MODIFIED]:
         repo.ui.debug(b" %s: %s -> am\n" % (f, msg))
         progress.increment(item=f)
 
     # keep (noop, just log it)
     for f, args, msg in actions[mergestatemod.ACTION_KEEP]:
         repo.ui.debug(b" %s: %s -> k\n" % (f, msg))
         # no progress
 
     # directory rename, move local
     for f, args, msg in actions[mergestatemod.ACTION_DIR_RENAME_MOVE_LOCAL]:
         repo.ui.debug(b" %s: %s -> dm\n" % (f, msg))
         progress.increment(item=f)
         f0, flags = args
         repo.ui.note(_(b"moving %s to %s\n") % (f0, f))
         wctx[f].audit()
         wctx[f].write(wctx.filectx(f0).data(), flags)
         wctx[f0].remove()
         updated += 1
 
     # local directory rename, get
     for f, args, msg in actions[mergestatemod.ACTION_LOCAL_DIR_RENAME_GET]:
         repo.ui.debug(b" %s: %s -> dg\n" % (f, msg))
         progress.increment(item=f)
         f0, flags = args
         repo.ui.note(_(b"getting %s to %s\n") % (f0, f))
         wctx[f].write(mctx.filectx(f0).data(), flags)
         updated += 1
 
     # exec
     for f, args, msg in actions[mergestatemod.ACTION_EXEC]:
         repo.ui.debug(b" %s: %s -> e\n" % (f, msg))
         progress.increment(item=f)
         (flags,) = args
         wctx[f].audit()
         wctx[f].setflags(b'l' in flags, b'x' in flags)
         updated += 1
 
     # the ordering is important here -- ms.mergedriver will raise if the merge
     # driver has changed, and we want to be able to bypass it when overwrite is
     # True
     usemergedriver = not overwrite and mergeactions and ms.mergedriver
 
     if usemergedriver:
         if wctx.isinmemory():
             raise error.InMemoryMergeConflictsError(
                 b"in-memory merge does not support mergedriver"
             )
         ms.commit()
         proceed = driverpreprocess(repo, ms, wctx, labels=labels)
         # the driver might leave some files unresolved
         unresolvedf = set(ms.unresolved())
         if not proceed:
             # XXX setting unresolved to at least 1 is a hack to make sure we
             # error out
             return updateresult(
                 updated, merged, removed, max(len(unresolvedf), 1)
             )
         newactions = []
         for f, args, msg in mergeactions:
             if f in unresolvedf:
                 newactions.append((f, args, msg))
         mergeactions = newactions
 
     try:
         # premerge
         tocomplete = []
         for f, args, msg in mergeactions:
             repo.ui.debug(b" %s: %s -> m (premerge)\n" % (f, msg))
             progress.increment(item=f)
             if f == b'.hgsubstate':  # subrepo states need updating
                 subrepoutil.submerge(
                     repo, wctx, mctx, wctx.ancestor(mctx), overwrite, labels
                 )
                 continue
             wctx[f].audit()
             complete, r = ms.preresolve(f, wctx)
             if not complete:
                 numupdates += 1
                 tocomplete.append((f, args, msg))
 
         # merge
         for f, args, msg in tocomplete:
             repo.ui.debug(b" %s: %s -> m (merge)\n" % (f, msg))
             progress.increment(item=f, total=numupdates)
             ms.resolve(f, wctx)
 
     finally:
         ms.commit()
 
     unresolved = ms.unresolvedcount()
 
     if (
         usemergedriver
         and not unresolved
         and ms.mdstate() != mergestatemod.MERGE_DRIVER_STATE_SUCCESS
     ):
         if not driverconclude(repo, ms, wctx, labels=labels):
             # XXX setting unresolved to at least 1 is a hack to make sure we
             # error out
             unresolved = max(unresolved, 1)
 
         ms.commit()
 
     msupdated, msmerged, msremoved = ms.counts()
     updated += msupdated
     merged += msmerged
     removed += msremoved
 
     extraactions = ms.actions()
     if extraactions:
         mfiles = {a[0] for a in actions[mergestatemod.ACTION_MERGE]}
         for k, acts in pycompat.iteritems(extraactions):
             actions[k].extend(acts)
             if k == mergestatemod.ACTION_GET and wantfiledata:
                 # no filedata until mergestate is updated to provide it
                 for a in acts:
                     getfiledata[a[0]] = None
             # Remove these files from actions[ACTION_MERGE] as well. This is
             # important because in recordupdates, files in actions[ACTION_MERGE]
             # are processed after files in other actions, and the merge driver
             # might add files to those actions via extraactions above. This can
             # lead to a file being recorded twice, with poor results. This is
             # especially problematic for actions[ACTION_REMOVE] (currently only
             # possible with the merge driver in the initial merge process;
             # interrupted merges don't go through this flow).
             #
             # The real fix here is to have indexes by both file and action so
             # that when the action for a file is changed it is automatically
             # reflected in the other action lists. But that involves a more
             # complex data structure, so this will do for now.
             #
             # We don't need to do the same operation for 'dc' and 'cd' because
             # those lists aren't consulted again.
             mfiles.difference_update(a[0] for a in acts)
 
         actions[mergestatemod.ACTION_MERGE] = [
             a for a in actions[mergestatemod.ACTION_MERGE] if a[0] in mfiles
         ]
 
     progress.complete()
     assert len(getfiledata) == (
         len(actions[mergestatemod.ACTION_GET]) if wantfiledata else 0
     )
     return updateresult(updated, merged, removed, unresolved), getfiledata
 
 
 UPDATECHECK_ABORT = b'abort'  # handled at higher layers
 UPDATECHECK_NONE = b'none'
 UPDATECHECK_LINEAR = b'linear'
 UPDATECHECK_NO_CONFLICT = b'noconflict'
 
 
 def update(
     repo,
     node,
     branchmerge,
     force,
     ancestor=None,
     mergeancestor=False,
     labels=None,
     matcher=None,
     mergeforce=False,
     updatedirstate=True,
     updatecheck=None,
     wc=None,
 ):
     """
     Perform a merge between the working directory and the given node
 
     node = the node to update to
     branchmerge = whether to merge between branches
     force = whether to force branch merging or file overwriting
     matcher = a matcher to filter file lists (dirstate not updated)
     mergeancestor = whether it is merging with an ancestor. If true,
       we should accept the incoming changes for any prompts that occur.
       If false, merging with an ancestor (fast-forward) is only allowed
       between different named branches. This flag is used by rebase extension
       as a temporary fix and should be avoided in general.
     labels = labels to use for base, local and other
     mergeforce = whether the merge was run with 'merge --force' (deprecated): if
       this is True, then 'force' should be True as well.
 
     The table below shows all the behaviors of the update command given the
     -c/--check and -C/--clean or no options, whether the working directory is
     dirty, whether a revision is specified, and the relationship of the parent
     rev to the target rev (linear or not). Match from top first. The -n
     option doesn't exist on the command line, but represents the
     experimental.updatecheck=noconflict option.
 
     This logic is tested by test-update-branches.t.
 
     -c  -C  -n  -m  dirty  rev  linear  |  result
      y   y   *   *    *     *     *     |    (1)
      y   *   y   *    *     *     *     |    (1)
      y   *   *   y    *     *     *     |    (1)
      *   y   y   *    *     *     *     |    (1)
      *   y   *   y    *     *     *     |    (1)
      *   *   y   y    *     *     *     |    (1)
      *   *   *   *    *     n     n     |     x
      *   *   *   *    n     *     *     |    ok
      n   n   n   n    y     *     y     |   merge
      n   n   n   n    y     y     n     |    (2)
      n   n   n   y    y     *     *     |   merge
      n   n   y   n    y     *     *     |  merge if no conflict
      n   y   n   n    y     *     *     |  discard
      y   n   n   n    y     *     *     |    (3)
 
     x = can't happen
     * = don't-care
     1 = incompatible options (checked in commands.py)
     2 = abort: uncommitted changes (commit or update --clean to discard changes)
     3 = abort: uncommitted changes (checked in commands.py)
 
     The merge is performed inside ``wc``, a workingctx-like objects. It defaults
     to repo[None] if None is passed.
 
     Return the same tuple as applyupdates().
     """
     # Avoid cycle.
     from . import sparse
 
     # This function used to find the default destination if node was None, but
     # that's now in destutil.py.
     assert node is not None
     if not branchmerge and not force:
         # TODO: remove the default once all callers that pass branchmerge=False
         # and force=False pass a value for updatecheck. We may want to allow
         # updatecheck='abort' to better suppport some of these callers.
         if updatecheck is None:
             updatecheck = UPDATECHECK_LINEAR
         if updatecheck not in (
             UPDATECHECK_NONE,
             UPDATECHECK_LINEAR,
             UPDATECHECK_NO_CONFLICT,
         ):
             raise ValueError(
                 r'Invalid updatecheck %r (can accept %r)'
                 % (
                     updatecheck,
                     (
                         UPDATECHECK_NONE,
                         UPDATECHECK_LINEAR,
                         UPDATECHECK_NO_CONFLICT,
                     ),
                 )
             )
     if wc is not None and wc.isinmemory():
         maybe_wlock = util.nullcontextmanager()
     else:
         maybe_wlock = repo.wlock()
     with maybe_wlock:
         if wc is None:
             wc = repo[None]
         pl = wc.parents()
         p1 = pl[0]
         p2 = repo[node]
         if ancestor is not None:
             pas = [repo[ancestor]]
         else:
             if repo.ui.configlist(b'merge', b'preferancestor') == [b'*']:
                 cahs = repo.changelog.commonancestorsheads(p1.node(), p2.node())
                 pas = [repo[anc] for anc in (sorted(cahs) or [nullid])]
             else:
                 pas = [p1.ancestor(p2, warn=branchmerge)]
 
         fp1, fp2, xp1, xp2 = p1.node(), p2.node(), bytes(p1), bytes(p2)
 
         overwrite = force and not branchmerge
         ### check phase
         if not overwrite:
             if len(pl) > 1:
                 raise error.Abort(_(b"outstanding uncommitted merge"))
             ms = mergestatemod.mergestate.read(repo)
             if list(ms.unresolved()):
                 raise error.Abort(
                     _(b"outstanding merge conflicts"),
                     hint=_(b"use 'hg resolve' to resolve"),
                 )
         if branchmerge:
             if pas == [p2]:
                 raise error.Abort(
                     _(
                         b"merging with a working directory ancestor"
                         b" has no effect"
                     )
                 )
             elif pas == [p1]:
                 if not mergeancestor and wc.branch() == p2.branch():
                     raise error.Abort(
                         _(b"nothing to merge"),
                         hint=_(b"use 'hg update' or check 'hg heads'"),
                     )
             if not force and (wc.files() or wc.deleted()):
                 raise error.Abort(
                     _(b"uncommitted changes"),
                     hint=_(b"use 'hg status' to list changes"),
                 )
             if not wc.isinmemory():
                 for s in sorted(wc.substate):
                     wc.sub(s).bailifchanged()
 
         elif not overwrite:
             if p1 == p2:  # no-op update
                 # call the hooks and exit early
                 repo.hook(b'preupdate', throw=True, parent1=xp2, parent2=b'')
                 repo.hook(b'update', parent1=xp2, parent2=b'', error=0)
                 return updateresult(0, 0, 0, 0)
 
             if updatecheck == UPDATECHECK_LINEAR and pas not in (
                 [p1],
                 [p2],
             ):  # nonlinear
                 dirty = wc.dirty(missing=True)
                 if dirty:
                     # Branching is a bit strange to ensure we do the minimal
                     # amount of call to obsutil.foreground.
                     foreground = obsutil.foreground(repo, [p1.node()])
                     # note: the <node> variable contains a random identifier
                     if repo[node].node() in foreground:
                         pass  # allow updating to successors
                     else:
                         msg = _(b"uncommitted changes")
                         hint = _(b"commit or update --clean to discard changes")
                         raise error.UpdateAbort(msg, hint=hint)
                 else:
                     # Allow jumping branches if clean and specific rev given
                     pass
 
         if overwrite:
             pas = [wc]
         elif not branchmerge:
             pas = [p1]
 
         # deprecated config: merge.followcopies
         followcopies = repo.ui.configbool(b'merge', b'followcopies')
         if overwrite:
             followcopies = False
         elif not pas[0]:
             followcopies = False
         if not branchmerge and not wc.dirty(missing=True):
             followcopies = False
 
         ### calculate phase
         actionbyfile, diverge, renamedelete = calculateupdates(
             repo,
             wc,
             p2,
             pas,
             branchmerge,
             force,
             mergeancestor,
             followcopies,
             matcher=matcher,
             mergeforce=mergeforce,
         )
 
         if updatecheck == UPDATECHECK_NO_CONFLICT:
             for f, (m, args, msg) in pycompat.iteritems(actionbyfile):
                 if m not in (
                     mergestatemod.ACTION_GET,
                     mergestatemod.ACTION_KEEP,
                     mergestatemod.ACTION_EXEC,
                     mergestatemod.ACTION_REMOVE,
                     mergestatemod.ACTION_PATH_CONFLICT_RESOLVE,
                     mergestatemod.ACTION_GET_OTHER_AND_STORE,
                 ):
                     msg = _(b"conflicting changes")
                     hint = _(b"commit or update --clean to discard changes")
                     raise error.Abort(msg, hint=hint)
 
         # Prompt and create actions. Most of this is in the resolve phase
         # already, but we can't handle .hgsubstate in filemerge or
         # subrepoutil.submerge yet so we have to keep prompting for it.
         if b'.hgsubstate' in actionbyfile:
             f = b'.hgsubstate'
             m, args, msg = actionbyfile[f]
             prompts = filemerge.partextras(labels)
             prompts[b'f'] = f
             if m == mergestatemod.ACTION_CHANGED_DELETED:
                 if repo.ui.promptchoice(
                     _(
                         b"local%(l)s changed %(f)s which other%(o)s deleted\n"
                         b"use (c)hanged version or (d)elete?"
                         b"$$ &Changed $$ &Delete"
                     )
                     % prompts,
                     0,
                 ):
                     actionbyfile[f] = (
                         mergestatemod.ACTION_REMOVE,
                         None,
                         b'prompt delete',
                     )
                 elif f in p1:
                     actionbyfile[f] = (
                         mergestatemod.ACTION_ADD_MODIFIED,
                         None,
                         b'prompt keep',
                     )
                 else:
                     actionbyfile[f] = (
                         mergestatemod.ACTION_ADD,
                         None,
                         b'prompt keep',
                     )
             elif m == mergestatemod.ACTION_DELETED_CHANGED:
                 f1, f2, fa, move, anc = args
                 flags = p2[f2].flags()
                 if (
                     repo.ui.promptchoice(
                         _(
                             b"other%(o)s changed %(f)s which local%(l)s deleted\n"
                             b"use (c)hanged version or leave (d)eleted?"
                             b"$$ &Changed $$ &Deleted"
                         )
                         % prompts,
                         0,
                     )
                     == 0
                 ):
                     actionbyfile[f] = (
                         mergestatemod.ACTION_GET,
                         (flags, False),
                         b'prompt recreating',
                     )
                 else:
                     del actionbyfile[f]
 
         # Convert to dictionary-of-lists format
         actions = emptyactions()
         for f, (m, args, msg) in pycompat.iteritems(actionbyfile):
             if m not in actions:
                 actions[m] = []
             actions[m].append((f, args, msg))
 
         # ACTION_GET_OTHER_AND_STORE is a mergestatemod.ACTION_GET + store in mergestate
         for e in actions[mergestatemod.ACTION_GET_OTHER_AND_STORE]:
             actions[mergestatemod.ACTION_GET].append(e)
 
         if not util.fscasesensitive(repo.path):
             # check collision between files only in p2 for clean update
             if not branchmerge and (
                 force or not wc.dirty(missing=True, branch=False)
             ):
                 _checkcollision(repo, p2.manifest(), None)
             else:
                 _checkcollision(repo, wc.manifest(), actions)
 
         # divergent renames
         for f, fl in sorted(pycompat.iteritems(diverge)):
             repo.ui.warn(
                 _(
                     b"note: possible conflict - %s was renamed "
                     b"multiple times to:\n"
                 )
                 % f
             )
             for nf in sorted(fl):
                 repo.ui.warn(b" %s\n" % nf)
 
         # rename and delete
         for f, fl in sorted(pycompat.iteritems(renamedelete)):
             repo.ui.warn(
                 _(
                     b"note: possible conflict - %s was deleted "
                     b"and renamed to:\n"
                 )
                 % f
             )
             for nf in sorted(fl):
                 repo.ui.warn(b" %s\n" % nf)
 
         ### apply phase
         if not branchmerge:  # just jump to the new rev
             fp1, fp2, xp1, xp2 = fp2, nullid, xp2, b''
         # If we're doing a partial update, we need to skip updating
         # the dirstate.
         always = matcher is None or matcher.always()
         updatedirstate = updatedirstate and always and not wc.isinmemory()
         if updatedirstate:
             repo.hook(b'preupdate', throw=True, parent1=xp1, parent2=xp2)
             # note that we're in the middle of an update
             repo.vfs.write(b'updatestate', p2.hex())
 
         # Advertise fsmonitor when its presence could be useful.
         #
         # We only advertise when performing an update from an empty working
         # directory. This typically only occurs during initial clone.
         #
         # We give users a mechanism to disable the warning in case it is
         # annoying.
         #
         # We only allow on Linux and MacOS because that's where fsmonitor is
         # considered stable.
         fsmonitorwarning = repo.ui.configbool(b'fsmonitor', b'warn_when_unused')
         fsmonitorthreshold = repo.ui.configint(
             b'fsmonitor', b'warn_update_file_count'
         )
         try:
             # avoid cycle: extensions -> cmdutil -> merge
             from . import extensions
 
             extensions.find(b'fsmonitor')
             fsmonitorenabled = repo.ui.config(b'fsmonitor', b'mode') != b'off'
             # We intentionally don't look at whether fsmonitor has disabled
             # itself because a) fsmonitor may have already printed a warning
             # b) we only care about the config state here.
         except KeyError:
             fsmonitorenabled = False
 
         if (
             fsmonitorwarning
             and not fsmonitorenabled
             and p1.node() == nullid
             and len(actions[mergestatemod.ACTION_GET]) >= fsmonitorthreshold
             and pycompat.sysplatform.startswith((b'linux', b'darwin'))
         ):
             repo.ui.warn(
                 _(
                     b'(warning: large working directory being used without '
                     b'fsmonitor enabled; enable fsmonitor to improve performance; '
                     b'see "hg help -e fsmonitor")\n'
                 )
             )
 
         wantfiledata = updatedirstate and not branchmerge
         stats, getfiledata = applyupdates(
             repo, actions, wc, p2, overwrite, wantfiledata, labels=labels
         )
 
         if updatedirstate:
             with repo.dirstate.parentchange():
                 repo.setparents(fp1, fp2)
                 mergestatemod.recordupdates(
                     repo, actions, branchmerge, getfiledata
                 )
                 # update completed, clear state
                 util.unlink(repo.vfs.join(b'updatestate'))
 
                 if not branchmerge:
                     repo.dirstate.setbranch(p2.branch())
 
     # If we're updating to a location, clean up any stale temporary includes
     # (ex: this happens during hg rebase --abort).
     if not branchmerge:
         sparse.prunetemporaryincludes(repo)
 
     if updatedirstate:
         repo.hook(
             b'update', parent1=xp1, parent2=xp2, error=stats.unresolvedcount
         )
     return stats
 
 
 def merge(ctx, labels=None, force=False, wc=None):
     """Merge another topological branch into the working copy.
 
     force = whether the merge was run with 'merge --force' (deprecated)
     """
 
     return update(
         ctx.repo(),
         ctx.rev(),
         labels=labels,
         branchmerge=True,
         force=force,
         mergeforce=force,
         wc=wc,
     )
 
 
 def clean_update(ctx, wc=None):
     """Do a clean update to the given commit.
 
     This involves updating to the commit and discarding any changes in the
     working copy.
     """
     return update(ctx.repo(), ctx.rev(), branchmerge=False, force=True, wc=wc)
 
 
 def revert_to(ctx, matcher=None, wc=None):
     """Revert the working copy to the given commit.
 
     The working copy will keep its current parent(s) but its content will
     be the same as in the given commit.
     """
 
     return update(
         ctx.repo(),
         ctx.rev(),
         branchmerge=False,
         force=True,
         updatedirstate=False,
         matcher=matcher,
         wc=wc,
     )
 
 
 def graft(
     repo,
     ctx,
     base=None,
     labels=None,
     keepparent=False,
     keepconflictparent=False,
     wctx=None,
 ):
     """Do a graft-like merge.
 
     This is a merge where the merge ancestor is chosen such that one
     or more changesets are grafted onto the current changeset. In
     addition to the merge, this fixes up the dirstate to include only
     a single parent (if keepparent is False) and tries to duplicate any
     renames/copies appropriately.
 
     ctx - changeset to rebase
     base - merge base, or ctx.p1() if not specified
     labels - merge labels eg ['local', 'graft']
     keepparent - keep second parent if any
     keepconflictparent - if unresolved, keep parent used for the merge
 
     """
     # If we're grafting a descendant onto an ancestor, be sure to pass
     # mergeancestor=True to update. This does two things: 1) allows the merge if
     # the destination is the same as the parent of the ctx (so we can use graft
     # to copy commits), and 2) informs update that the incoming changes are
     # newer than the destination so it doesn't prompt about "remote changed foo
     # which local deleted".
     # We also pass mergeancestor=True when base is the same revision as p1. 2)
     # doesn't matter as there can't possibly be conflicts, but 1) is necessary.
     wctx = wctx or repo[None]
     pctx = wctx.p1()
     base = base or ctx.p1()
     mergeancestor = (
         repo.changelog.isancestor(pctx.node(), ctx.node())
         or pctx.rev() == base.rev()
     )
 
     stats = update(
         repo,
         ctx.node(),
         True,
         True,
         base.node(),
         mergeancestor=mergeancestor,
         labels=labels,
         wc=wctx,
     )
 
     if keepconflictparent and stats.unresolvedcount:
         pother = ctx.node()
     else:
         pother = nullid
         parents = ctx.parents()
         if keepparent and len(parents) == 2 and base in parents:
             parents.remove(base)
             pother = parents[0].node()
     # Never set both parents equal to each other
     if pother == pctx.node():
         pother = nullid
 
     if wctx.isinmemory():
         wctx.setparents(pctx.node(), pother)
         # fix up dirstate for copies and renames
         copies.graftcopies(wctx, ctx, base)
     else:
         with repo.dirstate.parentchange():
             repo.setparents(pctx.node(), pother)
             repo.dirstate.write(repo.currenttransaction())
             # fix up dirstate for copies and renames
             copies.graftcopies(wctx, ctx, base)
     return stats
 
 
 def purge(
     repo,
     matcher,
     unknown=True,
     ignored=False,
     removeemptydirs=True,
     removefiles=True,
     abortonerror=False,
     noop=False,
 ):
     """Purge the working directory of untracked files.
 
     ``matcher`` is a matcher configured to scan the working directory -
     potentially a subset.
 
     ``unknown`` controls whether unknown files should be purged.
 
     ``ignored`` controls whether ignored files should be purged.
 
     ``removeemptydirs`` controls whether empty directories should be removed.
 
     ``removefiles`` controls whether files are removed.
 
     ``abortonerror`` causes an exception to be raised if an error occurs
     deleting a file or directory.
 
     ``noop`` controls whether to actually remove files. If not defined, actions
     will be taken.
 
     Returns an iterable of relative paths in the working directory that were
     or would be removed.
     """
 
     def remove(removefn, path):
         try:
             removefn(path)
         except OSError:
             m = _(b'%s cannot be removed') % path
             if abortonerror:
                 raise error.Abort(m)
             else:
                 repo.ui.warn(_(b'warning: %s\n') % m)
 
     # There's no API to copy a matcher. So mutate the passed matcher and
     # restore it when we're done.
     oldtraversedir = matcher.traversedir
 
     res = []
 
     try:
         if removeemptydirs:
             directories = []
             matcher.traversedir = directories.append
 
         status = repo.status(match=matcher, ignored=ignored, unknown=unknown)
 
         if removefiles:
             for f in sorted(status.unknown + status.ignored):
                 if not noop:
                     repo.ui.note(_(b'removing file %s\n') % f)
                     remove(repo.wvfs.unlink, f)
                 res.append(f)
 
         if removeemptydirs:
             for f in sorted(directories, reverse=True):
                 if matcher(f) and not repo.wvfs.listdir(f):
                     if not noop:
                         repo.ui.note(_(b'removing directory %s\n') % f)
                         remove(repo.wvfs.rmdir, f)
                     res.append(f)
 
         return res
 
     finally:
         matcher.traversedir = oldtraversedir

mercurial/patch.py

@@ -1,3227 +1,3231 b''
 # patch.py - patch file parsing routines
 #
 # Copyright 2006 Brendan Cully <[e-mail]>
 # Copyright 2007 Chris Mason <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import, print_function
 
 import collections
 import contextlib
 import copy
 import errno
 import os
 import re
 import shutil
 import zlib
 
 from .i18n import _
 from .node import (
     hex,
     short,
 )
 from .pycompat import open
 from . import (
     copies,
     diffhelper,
     diffutil,
     encoding,
     error,
     mail,
     mdiff,
     pathutil,
     pycompat,
     scmutil,
     similar,
     util,
     vfs as vfsmod,
 )
 from .utils import (
     dateutil,
     hashutil,
     procutil,
     stringutil,
 )
 
 stringio = util.stringio
 
 gitre = re.compile(br'diff --git a/(.*) b/(.*)')
 tabsplitter = re.compile(br'(\t+|[^\t]+)')
 wordsplitter = re.compile(
     br'(\t+| +|[a-zA-Z0-9_\x80-\xff]+|[^ \ta-zA-Z0-9_\x80-\xff])'
 )
 
 PatchError = error.PatchError
 
 # public functions
 
 
 def split(stream):
     '''return an iterator of individual patches from a stream'''
 
     def isheader(line, inheader):
         if inheader and line.startswith((b' ', b'\t')):
             # continuation
             return True
         if line.startswith((b' ', b'-', b'+')):
             # diff line - don't check for header pattern in there
             return False
         l = line.split(b': ', 1)
         return len(l) == 2 and b' ' not in l[0]
 
     def chunk(lines):
         return stringio(b''.join(lines))
 
     def hgsplit(stream, cur):
         inheader = True
 
         for line in stream:
             if not line.strip():
                 inheader = False
             if not inheader and line.startswith(b'# HG changeset patch'):
                 yield chunk(cur)
                 cur = []
                 inheader = True
 
             cur.append(line)
 
         if cur:
             yield chunk(cur)
 
     def mboxsplit(stream, cur):
         for line in stream:
             if line.startswith(b'From '):
                 for c in split(chunk(cur[1:])):
                     yield c
                 cur = []
 
             cur.append(line)
 
         if cur:
             for c in split(chunk(cur[1:])):
                 yield c
 
     def mimesplit(stream, cur):
         def msgfp(m):
             fp = stringio()
             g = mail.Generator(fp, mangle_from_=False)
             g.flatten(m)
             fp.seek(0)
             return fp
 
         for line in stream:
             cur.append(line)
         c = chunk(cur)
 
         m = mail.parse(c)
         if not m.is_multipart():
             yield msgfp(m)
         else:
             ok_types = (b'text/plain', b'text/x-diff', b'text/x-patch')
             for part in m.walk():
                 ct = part.get_content_type()
                 if ct not in ok_types:
                     continue
                 yield msgfp(part)
 
     def headersplit(stream, cur):
         inheader = False
 
         for line in stream:
             if not inheader and isheader(line, inheader):
                 yield chunk(cur)
                 cur = []
                 inheader = True
             if inheader and not isheader(line, inheader):
                 inheader = False
 
             cur.append(line)
 
         if cur:
             yield chunk(cur)
 
     def remainder(cur):
         yield chunk(cur)
 
     class fiter(object):
         def __init__(self, fp):
             self.fp = fp
 
         def __iter__(self):
             return self
 
         def next(self):
             l = self.fp.readline()
             if not l:
                 raise StopIteration
             return l
 
         __next__ = next
 
     inheader = False
     cur = []
 
     mimeheaders = [b'content-type']
 
     if not util.safehasattr(stream, b'next'):
         # http responses, for example, have readline but not next
         stream = fiter(stream)
 
     for line in stream:
         cur.append(line)
         if line.startswith(b'# HG changeset patch'):
             return hgsplit(stream, cur)
         elif line.startswith(b'From '):
             return mboxsplit(stream, cur)
         elif isheader(line, inheader):
             inheader = True
             if line.split(b':', 1)[0].lower() in mimeheaders:
                 # let email parser handle this
                 return mimesplit(stream, cur)
         elif line.startswith(b'--- ') and inheader:
             # No evil headers seen by diff start, split by hand
             return headersplit(stream, cur)
         # Not enough info, keep reading
 
     # if we are here, we have a very plain patch
     return remainder(cur)
 
 
 ## Some facility for extensible patch parsing:
 # list of pairs ("header to match", "data key")
 patchheadermap = [
     (b'Date', b'date'),
     (b'Branch', b'branch'),
     (b'Node ID', b'nodeid'),
 ]
 
 
 @contextlib.contextmanager
 def extract(ui, fileobj):
     '''extract patch from data read from fileobj.
 
     patch can be a normal patch or contained in an email message.
 
     return a dictionary. Standard keys are:
       - filename,
       - message,
       - user,
       - date,
       - branch,
       - node,
       - p1,
       - p2.
     Any item can be missing from the dictionary. If filename is missing,
     fileobj did not contain a patch. Caller must unlink filename when done.'''
 
     fd, tmpname = pycompat.mkstemp(prefix=b'hg-patch-')
     tmpfp = os.fdopen(fd, 'wb')
     try:
         yield _extract(ui, fileobj, tmpname, tmpfp)
     finally:
         tmpfp.close()
         os.unlink(tmpname)
 
 
 def _extract(ui, fileobj, tmpname, tmpfp):
 
     # attempt to detect the start of a patch
     # (this heuristic is borrowed from quilt)
     diffre = re.compile(
         br'^(?:Index:[ \t]|diff[ \t]-|RCS file: |'
         br'retrieving revision [0-9]+(\.[0-9]+)*$|'
         br'---[ \t].*?^\+\+\+[ \t]|'
         br'\*\*\*[ \t].*?^---[ \t])',
         re.MULTILINE | re.DOTALL,
     )
 
     data = {}
 
     msg = mail.parse(fileobj)
 
     subject = msg['Subject'] and mail.headdecode(msg['Subject'])
     data[b'user'] = msg['From'] and mail.headdecode(msg['From'])
     if not subject and not data[b'user']:
         # Not an email, restore parsed headers if any
         subject = (
             b'\n'.join(
                 b': '.join(map(encoding.strtolocal, h)) for h in msg.items()
             )
             + b'\n'
         )
 
     # should try to parse msg['Date']
     parents = []
 
     nodeid = msg['X-Mercurial-Node']
     if nodeid:
         data[b'nodeid'] = nodeid = mail.headdecode(nodeid)
         ui.debug(b'Node ID: %s\n' % nodeid)
 
     if subject:
         if subject.startswith(b'[PATCH'):
             pend = subject.find(b']')
             if pend >= 0:
                 subject = subject[pend + 1 :].lstrip()
         subject = re.sub(br'\n[ \t]+', b' ', subject)
         ui.debug(b'Subject: %s\n' % subject)
     if data[b'user']:
         ui.debug(b'From: %s\n' % data[b'user'])
     diffs_seen = 0
     ok_types = (b'text/plain', b'text/x-diff', b'text/x-patch')
     message = b''
     for part in msg.walk():
         content_type = pycompat.bytestr(part.get_content_type())
         ui.debug(b'Content-Type: %s\n' % content_type)
         if content_type not in ok_types:
             continue
         payload = part.get_payload(decode=True)
         m = diffre.search(payload)
         if m:
             hgpatch = False
             hgpatchheader = False
             ignoretext = False
 
             ui.debug(b'found patch at byte %d\n' % m.start(0))
             diffs_seen += 1
             cfp = stringio()
             for line in payload[: m.start(0)].splitlines():
                 if line.startswith(b'# HG changeset patch') and not hgpatch:
                     ui.debug(b'patch generated by hg export\n')
                     hgpatch = True
                     hgpatchheader = True
                     # drop earlier commit message content
                     cfp.seek(0)
                     cfp.truncate()
                     subject = None
                 elif hgpatchheader:
                     if line.startswith(b'# User '):
                         data[b'user'] = line[7:]
                         ui.debug(b'From: %s\n' % data[b'user'])
                     elif line.startswith(b"# Parent "):
                         parents.append(line[9:].lstrip())
                     elif line.startswith(b"# "):
                         for header, key in patchheadermap:
                             prefix = b'# %s ' % header
                             if line.startswith(prefix):
                                 data[key] = line[len(prefix) :]
                                 ui.debug(b'%s: %s\n' % (header, data[key]))
                     else:
                         hgpatchheader = False
                 elif line == b'---':
                     ignoretext = True
                 if not hgpatchheader and not ignoretext:
                     cfp.write(line)
                     cfp.write(b'\n')
             message = cfp.getvalue()
             if tmpfp:
                 tmpfp.write(payload)
                 if not payload.endswith(b'\n'):
                     tmpfp.write(b'\n')
         elif not diffs_seen and message and content_type == b'text/plain':
             message += b'\n' + payload
 
     if subject and not message.startswith(subject):
         message = b'%s\n%s' % (subject, message)
     data[b'message'] = message
     tmpfp.close()
     if parents:
         data[b'p1'] = parents.pop(0)
         if parents:
             data[b'p2'] = parents.pop(0)
 
     if diffs_seen:
         data[b'filename'] = tmpname
 
     return data
 
 
 class patchmeta(object):
     """Patched file metadata
 
     'op' is the performed operation within ADD, DELETE, RENAME, MODIFY
     or COPY.  'path' is patched file path. 'oldpath' is set to the
     origin file when 'op' is either COPY or RENAME, None otherwise. If
     file mode is changed, 'mode' is a tuple (islink, isexec) where
     'islink' is True if the file is a symlink and 'isexec' is True if
     the file is executable. Otherwise, 'mode' is None.
     """
 
     def __init__(self, path):
         self.path = path
         self.oldpath = None
         self.mode = None
         self.op = b'MODIFY'
         self.binary = False
 
     def setmode(self, mode):
         islink = mode & 0o20000
         isexec = mode & 0o100
         self.mode = (islink, isexec)
 
     def copy(self):
         other = patchmeta(self.path)
         other.oldpath = self.oldpath
         other.mode = self.mode
         other.op = self.op
         other.binary = self.binary
         return other
 
     def _ispatchinga(self, afile):
         if afile == b'/dev/null':
             return self.op == b'ADD'
         return afile == b'a/' + (self.oldpath or self.path)
 
     def _ispatchingb(self, bfile):
         if bfile == b'/dev/null':
             return self.op == b'DELETE'
         return bfile == b'b/' + self.path
 
     def ispatching(self, afile, bfile):
         return self._ispatchinga(afile) and self._ispatchingb(bfile)
 
     def __repr__(self):
         return "<patchmeta %s %r>" % (self.op, self.path)
 
 
 def readgitpatch(lr):
     """extract git-style metadata about patches from <patchname>"""
 
     # Filter patch for git information
     gp = None
     gitpatches = []
     for line in lr:
         line = line.rstrip(b' \r\n')
         if line.startswith(b'diff --git a/'):
             m = gitre.match(line)
             if m:
                 if gp:
                     gitpatches.append(gp)
                 dst = m.group(2)
                 gp = patchmeta(dst)
         elif gp:
             if line.startswith(b'--- '):
                 gitpatches.append(gp)
                 gp = None
                 continue
             if line.startswith(b'rename from '):
                 gp.op = b'RENAME'
                 gp.oldpath = line[12:]
             elif line.startswith(b'rename to '):
                 gp.path = line[10:]
             elif line.startswith(b'copy from '):
                 gp.op = b'COPY'
                 gp.oldpath = line[10:]
             elif line.startswith(b'copy to '):
                 gp.path = line[8:]
             elif line.startswith(b'deleted file'):
                 gp.op = b'DELETE'
             elif line.startswith(b'new file mode '):
                 gp.op = b'ADD'
                 gp.setmode(int(line[-6:], 8))
             elif line.startswith(b'new mode '):
                 gp.setmode(int(line[-6:], 8))
             elif line.startswith(b'GIT binary patch'):
                 gp.binary = True
     if gp:
         gitpatches.append(gp)
 
     return gitpatches
 
 
 class linereader(object):
     # simple class to allow pushing lines back into the input stream
     def __init__(self, fp):
         self.fp = fp
         self.buf = []
 
     def push(self, line):
         if line is not None:
             self.buf.append(line)
 
     def readline(self):
         if self.buf:
             l = self.buf[0]
             del self.buf[0]
             return l
         return self.fp.readline()
 
     def __iter__(self):
         return iter(self.readline, b'')
 
 
 class abstractbackend(object):
     def __init__(self, ui):
         self.ui = ui
 
     def getfile(self, fname):
         """Return target file data and flags as a (data, (islink,
         isexec)) tuple. Data is None if file is missing/deleted.
         """
         raise NotImplementedError
 
     def setfile(self, fname, data, mode, copysource):
         """Write data to target file fname and set its mode. mode is a
         (islink, isexec) tuple. If data is None, the file content should
         be left unchanged. If the file is modified after being copied,
         copysource is set to the original file name.
         """
         raise NotImplementedError
 
     def unlink(self, fname):
         """Unlink target file."""
         raise NotImplementedError
 
     def writerej(self, fname, failed, total, lines):
         """Write rejected lines for fname. total is the number of hunks
         which failed to apply and total the total number of hunks for this
         files.
         """
 
     def exists(self, fname):
         raise NotImplementedError
 
     def close(self):
         raise NotImplementedError
 
 
 class fsbackend(abstractbackend):
     def __init__(self, ui, basedir):
         super(fsbackend, self).__init__(ui)
         self.opener = vfsmod.vfs(basedir)
 
     def getfile(self, fname):
         if self.opener.islink(fname):
             return (self.opener.readlink(fname), (True, False))
 
         isexec = False
         try:
             isexec = self.opener.lstat(fname).st_mode & 0o100 != 0
         except OSError as e:
             if e.errno != errno.ENOENT:
                 raise
         try:
             return (self.opener.read(fname), (False, isexec))
         except IOError as e:
             if e.errno != errno.ENOENT:
                 raise
             return None, None
 
     def setfile(self, fname, data, mode, copysource):
         islink, isexec = mode
         if data is None:
             self.opener.setflags(fname, islink, isexec)
             return
         if islink:
             self.opener.symlink(data, fname)
         else:
             self.opener.write(fname, data)
             if isexec:
                 self.opener.setflags(fname, False, True)
 
     def unlink(self, fname):
         rmdir = self.ui.configbool(b'experimental', b'removeemptydirs')
         self.opener.unlinkpath(fname, ignoremissing=True, rmdir=rmdir)
 
     def writerej(self, fname, failed, total, lines):
         fname = fname + b".rej"
         self.ui.warn(
             _(b"%d out of %d hunks FAILED -- saving rejects to file %s\n")
             % (failed, total, fname)
         )
         fp = self.opener(fname, b'w')
         fp.writelines(lines)
         fp.close()
 
     def exists(self, fname):
         return self.opener.lexists(fname)
 
 
 class workingbackend(fsbackend):
     def __init__(self, ui, repo, similarity):
         super(workingbackend, self).__init__(ui, repo.root)
         self.repo = repo
         self.similarity = similarity
         self.removed = set()
         self.changed = set()
         self.copied = []
 
     def _checkknown(self, fname):
         if self.repo.dirstate[fname] == b'?' and self.exists(fname):
             raise PatchError(_(b'cannot patch %s: file is not tracked') % fname)
 
     def setfile(self, fname, data, mode, copysource):
         self._checkknown(fname)
         super(workingbackend, self).setfile(fname, data, mode, copysource)
         if copysource is not None:
             self.copied.append((copysource, fname))
         self.changed.add(fname)
 
     def unlink(self, fname):
         self._checkknown(fname)
         super(workingbackend, self).unlink(fname)
         self.removed.add(fname)
         self.changed.add(fname)
 
     def close(self):
         wctx = self.repo[None]
         changed = set(self.changed)
         for src, dst in self.copied:
             scmutil.dirstatecopy(self.ui, self.repo, wctx, src, dst)
         if self.removed:
             wctx.forget(sorted(self.removed))
             for f in self.removed:
                 if f not in self.repo.dirstate:
                     # File was deleted and no longer belongs to the
                     # dirstate, it was probably marked added then
                     # deleted, and should not be considered by
                     # marktouched().
                     changed.discard(f)
         if changed:
             scmutil.marktouched(self.repo, changed, self.similarity)
         return sorted(self.changed)
 
 
 class filestore(object):
     def __init__(self, maxsize=None):
         self.opener = None
         self.files = {}
         self.created = 0
         self.maxsize = maxsize
         if self.maxsize is None:
             self.maxsize = 4 * (2 ** 20)
         self.size = 0
         self.data = {}
 
     def setfile(self, fname, data, mode, copied=None):
         if self.maxsize < 0 or (len(data) + self.size) <= self.maxsize:
             self.data[fname] = (data, mode, copied)
             self.size += len(data)
         else:
             if self.opener is None:
                 root = pycompat.mkdtemp(prefix=b'hg-patch-')
                 self.opener = vfsmod.vfs(root)
             # Avoid filename issues with these simple names
             fn = b'%d' % self.created
             self.opener.write(fn, data)
             self.created += 1
             self.files[fname] = (fn, mode, copied)
 
     def getfile(self, fname):
         if fname in self.data:
             return self.data[fname]
         if not self.opener or fname not in self.files:
             return None, None, None
         fn, mode, copied = self.files[fname]
         return self.opener.read(fn), mode, copied
 
     def close(self):
         if self.opener:
             shutil.rmtree(self.opener.base)
 
 
 class repobackend(abstractbackend):
     def __init__(self, ui, repo, ctx, store):
         super(repobackend, self).__init__(ui)
         self.repo = repo
         self.ctx = ctx
         self.store = store
         self.changed = set()
         self.removed = set()
         self.copied = {}
 
     def _checkknown(self, fname):
         if fname not in self.ctx:
             raise PatchError(_(b'cannot patch %s: file is not tracked') % fname)
 
     def getfile(self, fname):
         try:
             fctx = self.ctx[fname]
         except error.LookupError:
             return None, None
         flags = fctx.flags()
         return fctx.data(), (b'l' in flags, b'x' in flags)
 
     def setfile(self, fname, data, mode, copysource):
         if copysource:
             self._checkknown(copysource)
         if data is None:
             data = self.ctx[fname].data()
         self.store.setfile(fname, data, mode, copysource)
         self.changed.add(fname)
         if copysource:
             self.copied[fname] = copysource
 
     def unlink(self, fname):
         self._checkknown(fname)
         self.removed.add(fname)
 
     def exists(self, fname):
         return fname in self.ctx
 
     def close(self):
         return self.changed | self.removed
 
 
 # @@ -start,len +start,len @@ or @@ -start +start @@ if len is 1
 unidesc = re.compile(br'@@ -(\d+)(?:,(\d+))? \+(\d+)(?:,(\d+))? @@')
 contextdesc = re.compile(br'(?:---|\*\*\*) (\d+)(?:,(\d+))? (?:---|\*\*\*)')
 eolmodes = [b'strict', b'crlf', b'lf', b'auto']
 
 
 class patchfile(object):
     def __init__(self, ui, gp, backend, store, eolmode=b'strict'):
         self.fname = gp.path
         self.eolmode = eolmode
         self.eol = None
         self.backend = backend
         self.ui = ui
         self.lines = []
         self.exists = False
         self.missing = True
         self.mode = gp.mode
         self.copysource = gp.oldpath
         self.create = gp.op in (b'ADD', b'COPY', b'RENAME')
         self.remove = gp.op == b'DELETE'
         if self.copysource is None:
             data, mode = backend.getfile(self.fname)
         else:
             data, mode = store.getfile(self.copysource)[:2]
         if data is not None:
             self.exists = self.copysource is None or backend.exists(self.fname)
             self.missing = False
             if data:
                 self.lines = mdiff.splitnewlines(data)
             if self.mode is None:
                 self.mode = mode
             if self.lines:
                 # Normalize line endings
                 if self.lines[0].endswith(b'\r\n'):
                     self.eol = b'\r\n'
                 elif self.lines[0].endswith(b'\n'):
                     self.eol = b'\n'
                 if eolmode != b'strict':
                     nlines = []
                     for l in self.lines:
                         if l.endswith(b'\r\n'):
                             l = l[:-2] + b'\n'
                         nlines.append(l)
                     self.lines = nlines
         else:
             if self.create:
                 self.missing = False
             if self.mode is None:
                 self.mode = (False, False)
         if self.missing:
             self.ui.warn(_(b"unable to find '%s' for patching\n") % self.fname)
             self.ui.warn(
                 _(
                     b"(use '--prefix' to apply patch relative to the "
                     b"current directory)\n"
                 )
             )
 
         self.hash = {}
         self.dirty = 0
         self.offset = 0
         self.skew = 0
         self.rej = []
         self.fileprinted = False
         self.printfile(False)
         self.hunks = 0
 
     def writelines(self, fname, lines, mode):
         if self.eolmode == b'auto':
             eol = self.eol
         elif self.eolmode == b'crlf':
             eol = b'\r\n'
         else:
             eol = b'\n'
 
         if self.eolmode != b'strict' and eol and eol != b'\n':
             rawlines = []
             for l in lines:
                 if l and l.endswith(b'\n'):
                     l = l[:-1] + eol
                 rawlines.append(l)
             lines = rawlines
 
         self.backend.setfile(fname, b''.join(lines), mode, self.copysource)
 
     def printfile(self, warn):
         if self.fileprinted:
             return
         if warn or self.ui.verbose:
             self.fileprinted = True
         s = _(b"patching file %s\n") % self.fname
         if warn:
             self.ui.warn(s)
         else:
             self.ui.note(s)
 
     def findlines(self, l, linenum):
         # looks through the hash and finds candidate lines.  The
         # result is a list of line numbers sorted based on distance
         # from linenum
 
         cand = self.hash.get(l, [])
         if len(cand) > 1:
             # resort our list of potentials forward then back.
             cand.sort(key=lambda x: abs(x - linenum))
         return cand
 
     def write_rej(self):
         # our rejects are a little different from patch(1).  This always
         # creates rejects in the same form as the original patch.  A file
         # header is inserted so that you can run the reject through patch again
         # without having to type the filename.
         if not self.rej:
             return
         base = os.path.basename(self.fname)
         lines = [b"--- %s\n+++ %s\n" % (base, base)]
         for x in self.rej:
             for l in x.hunk:
                 lines.append(l)
                 if l[-1:] != b'\n':
                     lines.append(b"\n\\ No newline at end of file\n")
         self.backend.writerej(self.fname, len(self.rej), self.hunks, lines)
 
     def apply(self, h):
         if not h.complete():
             raise PatchError(
                 _(b"bad hunk #%d %s (%d %d %d %d)")
                 % (h.number, h.desc, len(h.a), h.lena, len(h.b), h.lenb)
             )
 
         self.hunks += 1
 
         if self.missing:
             self.rej.append(h)
             return -1
 
         if self.exists and self.create:
             if self.copysource:
                 self.ui.warn(
                     _(b"cannot create %s: destination already exists\n")
                     % self.fname
                 )
             else:
                 self.ui.warn(_(b"file %s already exists\n") % self.fname)
             self.rej.append(h)
             return -1
 
         if isinstance(h, binhunk):
             if self.remove:
                 self.backend.unlink(self.fname)
             else:
                 l = h.new(self.lines)
                 self.lines[:] = l
                 self.offset += len(l)
                 self.dirty = True
             return 0
 
         horig = h
         if (
             self.eolmode in (b'crlf', b'lf')
             or self.eolmode == b'auto'
             and self.eol
         ):
             # If new eols are going to be normalized, then normalize
             # hunk data before patching. Otherwise, preserve input
             # line-endings.
             h = h.getnormalized()
 
         # fast case first, no offsets, no fuzz
         old, oldstart, new, newstart = h.fuzzit(0, False)
         oldstart += self.offset
         orig_start = oldstart
         # if there's skew we want to emit the "(offset %d lines)" even
         # when the hunk cleanly applies at start + skew, so skip the
         # fast case code
         if self.skew == 0 and diffhelper.testhunk(old, self.lines, oldstart):
             if self.remove:
                 self.backend.unlink(self.fname)
             else:
                 self.lines[oldstart : oldstart + len(old)] = new
                 self.offset += len(new) - len(old)
                 self.dirty = True
             return 0
 
         # ok, we couldn't match the hunk. Lets look for offsets and fuzz it
         self.hash = {}
         for x, s in enumerate(self.lines):
             self.hash.setdefault(s, []).append(x)
 
         for fuzzlen in pycompat.xrange(
             self.ui.configint(b"patch", b"fuzz") + 1
         ):
             for toponly in [True, False]:
                 old, oldstart, new, newstart = h.fuzzit(fuzzlen, toponly)
                 oldstart = oldstart + self.offset + self.skew
                 oldstart = min(oldstart, len(self.lines))
                 if old:
                     cand = self.findlines(old[0][1:], oldstart)
                 else:
                     # Only adding lines with no or fuzzed context, just
                     # take the skew in account
                     cand = [oldstart]
 
                 for l in cand:
                     if not old or diffhelper.testhunk(old, self.lines, l):
                         self.lines[l : l + len(old)] = new
                         self.offset += len(new) - len(old)
                         self.skew = l - orig_start
                         self.dirty = True
                         offset = l - orig_start - fuzzlen
                         if fuzzlen:
                             msg = _(
                                 b"Hunk #%d succeeded at %d "
                                 b"with fuzz %d "
                                 b"(offset %d lines).\n"
                             )
                             self.printfile(True)
                             self.ui.warn(
                                 msg % (h.number, l + 1, fuzzlen, offset)
                             )
                         else:
                             msg = _(
                                 b"Hunk #%d succeeded at %d "
                                 b"(offset %d lines).\n"
                             )
                             self.ui.note(msg % (h.number, l + 1, offset))
                         return fuzzlen
         self.printfile(True)
         self.ui.warn(_(b"Hunk #%d FAILED at %d\n") % (h.number, orig_start))
         self.rej.append(horig)
         return -1
 
     def close(self):
         if self.dirty:
             self.writelines(self.fname, self.lines, self.mode)
         self.write_rej()
         return len(self.rej)
 
 
 class header(object):
     """patch header
     """
 
     diffgit_re = re.compile(b'diff --git a/(.*) b/(.*)$')
     diff_re = re.compile(b'diff -r .* (.*)$')
     allhunks_re = re.compile(b'(?:index|deleted file) ')
     pretty_re = re.compile(b'(?:new file|deleted file) ')
     special_re = re.compile(b'(?:index|deleted|copy|rename|new mode) ')
     newfile_re = re.compile(b'(?:new file|copy to|rename to)')
 
     def __init__(self, header):
         self.header = header
         self.hunks = []
 
     def binary(self):
         return any(h.startswith(b'index ') for h in self.header)
 
     def pretty(self, fp):
         for h in self.header:
             if h.startswith(b'index '):
                 fp.write(_(b'this modifies a binary file (all or nothing)\n'))
                 break
             if self.pretty_re.match(h):
                 fp.write(h)
                 if self.binary():
                     fp.write(_(b'this is a binary file\n'))
                 break
             if h.startswith(b'---'):
                 fp.write(
                     _(b'%d hunks, %d lines changed\n')
                     % (
                         len(self.hunks),
                         sum([max(h.added, h.removed) for h in self.hunks]),
                     )
                 )
                 break
             fp.write(h)
 
     def write(self, fp):
         fp.write(b''.join(self.header))
 
     def allhunks(self):
         return any(self.allhunks_re.match(h) for h in self.header)
 
     def files(self):
         match = self.diffgit_re.match(self.header[0])
         if match:
             fromfile, tofile = match.groups()
             if fromfile == tofile:
                 return [fromfile]
             return [fromfile, tofile]
         else:
             return self.diff_re.match(self.header[0]).groups()
 
     def filename(self):
         return self.files()[-1]
 
     def __repr__(self):
         return '<header %s>' % (
             ' '.join(pycompat.rapply(pycompat.fsdecode, self.files()))
         )
 
     def isnewfile(self):
         return any(self.newfile_re.match(h) for h in self.header)
 
     def special(self):
         # Special files are shown only at the header level and not at the hunk
         # level for example a file that has been deleted is a special file.
         # The user cannot change the content of the operation, in the case of
         # the deleted file he has to take the deletion or not take it, he
         # cannot take some of it.
         # Newly added files are special if they are empty, they are not special
         # if they have some content as we want to be able to change it
         nocontent = len(self.header) == 2
         emptynewfile = self.isnewfile() and nocontent
         return emptynewfile or any(
             self.special_re.match(h) for h in self.header
         )
 
 
 class recordhunk(object):
     """patch hunk
 
     XXX shouldn't we merge this with the other hunk class?
     """
 
     def __init__(
         self,
         header,
         fromline,
         toline,
         proc,
         before,
         hunk,
         after,
         maxcontext=None,
     ):
         def trimcontext(lines, reverse=False):
             if maxcontext is not None:
                 delta = len(lines) - maxcontext
                 if delta > 0:
                     if reverse:
                         return delta, lines[delta:]
                     else:
                         return delta, lines[:maxcontext]
             return 0, lines
 
         self.header = header
         trimedbefore, self.before = trimcontext(before, True)
         self.fromline = fromline + trimedbefore
         self.toline = toline + trimedbefore
         _trimedafter, self.after = trimcontext(after, False)
         self.proc = proc
         self.hunk = hunk
         self.added, self.removed = self.countchanges(self.hunk)
 
     def __eq__(self, v):
         if not isinstance(v, recordhunk):
             return False
 
         return (
             (v.hunk == self.hunk)
             and (v.proc == self.proc)
             and (self.fromline == v.fromline)
             and (self.header.files() == v.header.files())
         )
 
     def __hash__(self):
         return hash(
             (
                 tuple(self.hunk),
                 tuple(self.header.files()),
                 self.fromline,
                 self.proc,
             )
         )
 
     def countchanges(self, hunk):
         """hunk -> (n+,n-)"""
         add = len([h for h in hunk if h.startswith(b'+')])
         rem = len([h for h in hunk if h.startswith(b'-')])
         return add, rem
 
     def reversehunk(self):
         """return another recordhunk which is the reverse of the hunk
 
         If this hunk is diff(A, B), the returned hunk is diff(B, A). To do
         that, swap fromline/toline and +/- signs while keep other things
         unchanged.
         """
         m = {b'+': b'-', b'-': b'+', b'\\': b'\\'}
         hunk = [b'%s%s' % (m[l[0:1]], l[1:]) for l in self.hunk]
         return recordhunk(
             self.header,
             self.toline,
             self.fromline,
             self.proc,
             self.before,
             hunk,
             self.after,
         )
 
     def write(self, fp):
         delta = len(self.before) + len(self.after)
         if self.after and self.after[-1] == b'\\ No newline at end of file\n':
             delta -= 1
         fromlen = delta + self.removed
         tolen = delta + self.added
         fp.write(
             b'@@ -%d,%d +%d,%d @@%s\n'
             % (
                 self.fromline,
                 fromlen,
                 self.toline,
                 tolen,
                 self.proc and (b' ' + self.proc),
             )
         )
         fp.write(b''.join(self.before + self.hunk + self.after))
 
     pretty = write
 
     def filename(self):
         return self.header.filename()
 
     @encoding.strmethod
     def __repr__(self):
         return b'<hunk %r@%d>' % (self.filename(), self.fromline)
 
 
 def getmessages():
     return {
         b'multiple': {
             b'apply': _(b"apply change %d/%d to '%s'?"),
             b'discard': _(b"discard change %d/%d to '%s'?"),
             b'keep': _(b"keep change %d/%d to '%s'?"),
             b'record': _(b"record change %d/%d to '%s'?"),
         },
         b'single': {
             b'apply': _(b"apply this change to '%s'?"),
             b'discard': _(b"discard this change to '%s'?"),
             b'keep': _(b"keep this change to '%s'?"),
             b'record': _(b"record this change to '%s'?"),
         },
         b'help': {
             b'apply': _(
                 b'[Ynesfdaq?]'
                 b'$$ &Yes, apply this change'
                 b'$$ &No, skip this change'
                 b'$$ &Edit this change manually'
                 b'$$ &Skip remaining changes to this file'
                 b'$$ Apply remaining changes to this &file'
                 b'$$ &Done, skip remaining changes and files'
                 b'$$ Apply &all changes to all remaining files'
                 b'$$ &Quit, applying no changes'
                 b'$$ &? (display help)'
             ),
             b'discard': _(
                 b'[Ynesfdaq?]'
                 b'$$ &Yes, discard this change'
                 b'$$ &No, skip this change'
                 b'$$ &Edit this change manually'
                 b'$$ &Skip remaining changes to this file'
                 b'$$ Discard remaining changes to this &file'
                 b'$$ &Done, skip remaining changes and files'
                 b'$$ Discard &all changes to all remaining files'
                 b'$$ &Quit, discarding no changes'
                 b'$$ &? (display help)'
             ),
             b'keep': _(
                 b'[Ynesfdaq?]'
                 b'$$ &Yes, keep this change'
                 b'$$ &No, skip this change'
                 b'$$ &Edit this change manually'
                 b'$$ &Skip remaining changes to this file'
                 b'$$ Keep remaining changes to this &file'
                 b'$$ &Done, skip remaining changes and files'
                 b'$$ Keep &all changes to all remaining files'
                 b'$$ &Quit, keeping all changes'
                 b'$$ &? (display help)'
             ),
             b'record': _(
                 b'[Ynesfdaq?]'
                 b'$$ &Yes, record this change'
                 b'$$ &No, skip this change'
                 b'$$ &Edit this change manually'
                 b'$$ &Skip remaining changes to this file'
                 b'$$ Record remaining changes to this &file'
                 b'$$ &Done, skip remaining changes and files'
                 b'$$ Record &all changes to all remaining files'
                 b'$$ &Quit, recording no changes'
                 b'$$ &? (display help)'
             ),
         },
     }
 
 
 def filterpatch(ui, headers, match, operation=None):
     """Interactively filter patch chunks into applied-only chunks"""
     messages = getmessages()
 
     if operation is None:
         operation = b'record'
 
     def prompt(skipfile, skipall, query, chunk):
         """prompt query, and process base inputs
 
         - y/n for the rest of file
         - y/n for the rest
         - ? (help)
         - q (quit)
 
         Return True/False and possibly updated skipfile and skipall.
         """
         newpatches = None
         if skipall is not None:
             return skipall, skipfile, skipall, newpatches
         if skipfile is not None:
             return skipfile, skipfile, skipall, newpatches
         while True:
             resps = messages[b'help'][operation]
             # IMPORTANT: keep the last line of this prompt short (<40 english
             # chars is a good target) because of issue6158.
             r = ui.promptchoice(b"%s\n(enter ? for help) %s" % (query, resps))
             ui.write(b"\n")
             if r == 8:  # ?
                 for c, t in ui.extractchoices(resps)[1]:
                     ui.write(b'%s - %s\n' % (c, encoding.lower(t)))
                 continue
             elif r == 0:  # yes
                 ret = True
             elif r == 1:  # no
                 ret = False
             elif r == 2:  # Edit patch
                 if chunk is None:
                     ui.write(_(b'cannot edit patch for whole file'))
                     ui.write(b"\n")
                     continue
                 if chunk.header.binary():
                     ui.write(_(b'cannot edit patch for binary file'))
                     ui.write(b"\n")
                     continue
                 # Patch comment based on the Git one (based on comment at end of
                 # https://mercurial-scm.org/wiki/RecordExtension)
                 phelp = b'---' + _(
                     """
 To remove '-' lines, make them ' ' lines (context).
 To remove '+' lines, delete them.
 Lines starting with # will be removed from the patch.
 
 If the patch applies cleanly, the edited hunk will immediately be
 added to the record list. If it does not apply cleanly, a rejects
 file will be generated: you can use that when you try again. If
 all lines of the hunk are removed, then the edit is aborted and
 the hunk is left unchanged.
 """
                 )
                 (patchfd, patchfn) = pycompat.mkstemp(
                     prefix=b"hg-editor-", suffix=b".diff"
                 )
                 ncpatchfp = None
                 try:
                     # Write the initial patch
                     f = util.nativeeolwriter(os.fdopen(patchfd, 'wb'))
                     chunk.header.write(f)
                     chunk.write(f)
                     f.write(
                         b''.join(
                             [b'# ' + i + b'\n' for i in phelp.splitlines()]
                         )
                     )
                     f.close()
                     # Start the editor and wait for it to complete
                     editor = ui.geteditor()
                     ret = ui.system(
                         b"%s \"%s\"" % (editor, patchfn),
                         environ={b'HGUSER': ui.username()},
                         blockedtag=b'filterpatch',
                     )
                     if ret != 0:
                         ui.warn(_(b"editor exited with exit code %d\n") % ret)
                         continue
                     # Remove comment lines
                     patchfp = open(patchfn, 'rb')
                     ncpatchfp = stringio()
                     for line in util.iterfile(patchfp):
                         line = util.fromnativeeol(line)
                         if not line.startswith(b'#'):
                             ncpatchfp.write(line)
                     patchfp.close()
                     ncpatchfp.seek(0)
                     newpatches = parsepatch(ncpatchfp)
                 finally:
                     os.unlink(patchfn)
                     del ncpatchfp
                 # Signal that the chunk shouldn't be applied as-is, but
                 # provide the new patch to be used instead.
                 ret = False
             elif r == 3:  # Skip
                 ret = skipfile = False
             elif r == 4:  # file (Record remaining)
                 ret = skipfile = True
             elif r == 5:  # done, skip remaining
                 ret = skipall = False
             elif r == 6:  # all
                 ret = skipall = True
             elif r == 7:  # quit
                 raise error.Abort(_(b'user quit'))
             return ret, skipfile, skipall, newpatches
 
     seen = set()
     applied = {}  # 'filename' -> [] of chunks
     skipfile, skipall = None, None
     pos, total = 1, sum(len(h.hunks) for h in headers)
     for h in headers:
         pos += len(h.hunks)
         skipfile = None
         fixoffset = 0
         hdr = b''.join(h.header)
         if hdr in seen:
             continue
         seen.add(hdr)
         if skipall is None:
             h.pretty(ui)
         files = h.files()
         msg = _(b'examine changes to %s?') % _(b' and ').join(
             b"'%s'" % f for f in files
         )
         if all(match.exact(f) for f in files):
             r, skipall, np = True, None, None
         else:
             r, skipfile, skipall, np = prompt(skipfile, skipall, msg, None)
         if not r:
             continue
         applied[h.filename()] = [h]
         if h.allhunks():
             applied[h.filename()] += h.hunks
             continue
         for i, chunk in enumerate(h.hunks):
             if skipfile is None and skipall is None:
                 chunk.pretty(ui)
             if total == 1:
                 msg = messages[b'single'][operation] % chunk.filename()
             else:
                 idx = pos - len(h.hunks) + i
                 msg = messages[b'multiple'][operation] % (
                     idx,
                     total,
                     chunk.filename(),
                 )
             r, skipfile, skipall, newpatches = prompt(
                 skipfile, skipall, msg, chunk
             )
             if r:
                 if fixoffset:
                     chunk = copy.copy(chunk)
                     chunk.toline += fixoffset
                 applied[chunk.filename()].append(chunk)
             elif newpatches is not None:
                 for newpatch in newpatches:
                     for newhunk in newpatch.hunks:
                         if fixoffset:
                             newhunk.toline += fixoffset
                         applied[newhunk.filename()].append(newhunk)
             else:
                 fixoffset += chunk.removed - chunk.added
     return (
         sum(
             [
                 h
                 for h in pycompat.itervalues(applied)
                 if h[0].special() or len(h) > 1
             ],
             [],
         ),
         {},
     )
 
 
 class hunk(object):
     def __init__(self, desc, num, lr, context):
         self.number = num
         self.desc = desc
         self.hunk = [desc]
         self.a = []
         self.b = []
         self.starta = self.lena = None
         self.startb = self.lenb = None
         if lr is not None:
             if context:
                 self.read_context_hunk(lr)
             else:
                 self.read_unified_hunk(lr)
 
     def getnormalized(self):
         """Return a copy with line endings normalized to LF."""
 
         def normalize(lines):
             nlines = []
             for line in lines:
                 if line.endswith(b'\r\n'):
                     line = line[:-2] + b'\n'
                 nlines.append(line)
             return nlines
 
         # Dummy object, it is rebuilt manually
         nh = hunk(self.desc, self.number, None, None)
         nh.number = self.number
         nh.desc = self.desc
         nh.hunk = self.hunk
         nh.a = normalize(self.a)
         nh.b = normalize(self.b)
         nh.starta = self.starta
         nh.startb = self.startb
         nh.lena = self.lena
         nh.lenb = self.lenb
         return nh
 
     def read_unified_hunk(self, lr):
         m = unidesc.match(self.desc)
         if not m:
             raise PatchError(_(b"bad hunk #%d") % self.number)
         self.starta, self.lena, self.startb, self.lenb = m.groups()
         if self.lena is None:
             self.lena = 1
         else:
             self.lena = int(self.lena)
         if self.lenb is None:
             self.lenb = 1
         else:
             self.lenb = int(self.lenb)
         self.starta = int(self.starta)
         self.startb = int(self.startb)
         try:
             diffhelper.addlines(
                 lr, self.hunk, self.lena, self.lenb, self.a, self.b
             )
         except error.ParseError as e:
             raise PatchError(_(b"bad hunk #%d: %s") % (self.number, e))
         # if we hit eof before finishing out the hunk, the last line will
         # be zero length.  Lets try to fix it up.
         while len(self.hunk[-1]) == 0:
             del self.hunk[-1]
             del self.a[-1]
             del self.b[-1]
             self.lena -= 1
             self.lenb -= 1
         self._fixnewline(lr)
 
     def read_context_hunk(self, lr):
         self.desc = lr.readline()
         m = contextdesc.match(self.desc)
         if not m:
             raise PatchError(_(b"bad hunk #%d") % self.number)
         self.starta, aend = m.groups()
         self.starta = int(self.starta)
         if aend is None:
             aend = self.starta
         self.lena = int(aend) - self.starta
         if self.starta:
             self.lena += 1
         for x in pycompat.xrange(self.lena):
             l = lr.readline()
             if l.startswith(b'---'):
                 # lines addition, old block is empty
                 lr.push(l)
                 break
             s = l[2:]
             if l.startswith(b'- ') or l.startswith(b'! '):
                 u = b'-' + s
             elif l.startswith(b'  '):
                 u = b' ' + s
             else:
                 raise PatchError(
                     _(b"bad hunk #%d old text line %d") % (self.number, x)
                 )
             self.a.append(u)
             self.hunk.append(u)
 
         l = lr.readline()
         if l.startswith(br'\ '):
             s = self.a[-1][:-1]
             self.a[-1] = s
             self.hunk[-1] = s
             l = lr.readline()
         m = contextdesc.match(l)
         if not m:
             raise PatchError(_(b"bad hunk #%d") % self.number)
         self.startb, bend = m.groups()
         self.startb = int(self.startb)
         if bend is None:
             bend = self.startb
         self.lenb = int(bend) - self.startb
         if self.startb:
             self.lenb += 1
         hunki = 1
         for x in pycompat.xrange(self.lenb):
             l = lr.readline()
             if l.startswith(br'\ '):
                 # XXX: the only way to hit this is with an invalid line range.
                 # The no-eol marker is not counted in the line range, but I
                 # guess there are diff(1) out there which behave differently.
                 s = self.b[-1][:-1]
                 self.b[-1] = s
                 self.hunk[hunki - 1] = s
                 continue
             if not l:
                 # line deletions, new block is empty and we hit EOF
                 lr.push(l)
                 break
             s = l[2:]
             if l.startswith(b'+ ') or l.startswith(b'! '):
                 u = b'+' + s
             elif l.startswith(b'  '):
                 u = b' ' + s
             elif len(self.b) == 0:
                 # line deletions, new block is empty
                 lr.push(l)
                 break
             else:
                 raise PatchError(
                     _(b"bad hunk #%d old text line %d") % (self.number, x)
                 )
             self.b.append(s)
             while True:
                 if hunki >= len(self.hunk):
                     h = b""
                 else:
                     h = self.hunk[hunki]
                 hunki += 1
                 if h == u:
                     break
                 elif h.startswith(b'-'):
                     continue
                 else:
                     self.hunk.insert(hunki - 1, u)
                     break
 
         if not self.a:
             # this happens when lines were only added to the hunk
             for x in self.hunk:
                 if x.startswith(b'-') or x.startswith(b' '):
                     self.a.append(x)
         if not self.b:
             # this happens when lines were only deleted from the hunk
             for x in self.hunk:
                 if x.startswith(b'+') or x.startswith(b' '):
                     self.b.append(x[1:])
         # @@ -start,len +start,len @@
         self.desc = b"@@ -%d,%d +%d,%d @@\n" % (
             self.starta,
             self.lena,
             self.startb,
             self.lenb,
         )
         self.hunk[0] = self.desc
         self._fixnewline(lr)
 
     def _fixnewline(self, lr):
         l = lr.readline()
         if l.startswith(br'\ '):
             diffhelper.fixnewline(self.hunk, self.a, self.b)
         else:
             lr.push(l)
 
     def complete(self):
         return len(self.a) == self.lena and len(self.b) == self.lenb
 
     def _fuzzit(self, old, new, fuzz, toponly):
         # this removes context lines from the top and bottom of list 'l'.  It
         # checks the hunk to make sure only context lines are removed, and then
         # returns a new shortened list of lines.
         fuzz = min(fuzz, len(old))
         if fuzz:
             top = 0
             bot = 0
             hlen = len(self.hunk)
             for x in pycompat.xrange(hlen - 1):
                 # the hunk starts with the @@ line, so use x+1
                 if self.hunk[x + 1].startswith(b' '):
                     top += 1
                 else:
                     break
             if not toponly:
                 for x in pycompat.xrange(hlen - 1):
                     if self.hunk[hlen - bot - 1].startswith(b' '):
                         bot += 1
                     else:
                         break
 
             bot = min(fuzz, bot)
             top = min(fuzz, top)
             return old[top : len(old) - bot], new[top : len(new) - bot], top
         return old, new, 0
 
     def fuzzit(self, fuzz, toponly):
         old, new, top = self._fuzzit(self.a, self.b, fuzz, toponly)
         oldstart = self.starta + top
         newstart = self.startb + top
         # zero length hunk ranges already have their start decremented
         if self.lena and oldstart > 0:
             oldstart -= 1
         if self.lenb and newstart > 0:
             newstart -= 1
         return old, oldstart, new, newstart
 
 
 class binhunk(object):
     """A binary patch file."""
 
     def __init__(self, lr, fname):
         self.text = None
         self.delta = False
         self.hunk = [b'GIT binary patch\n']
         self._fname = fname
         self._read(lr)
 
     def complete(self):
         return self.text is not None
 
     def new(self, lines):
         if self.delta:
             return [applybindelta(self.text, b''.join(lines))]
         return [self.text]
 
     def _read(self, lr):
         def getline(lr, hunk):
             l = lr.readline()
             hunk.append(l)
             return l.rstrip(b'\r\n')
 
         while True:
             line = getline(lr, self.hunk)
             if not line:
                 raise PatchError(
                     _(b'could not extract "%s" binary data') % self._fname
                 )
             if line.startswith(b'literal '):
                 size = int(line[8:].rstrip())
                 break
             if line.startswith(b'delta '):
                 size = int(line[6:].rstrip())
                 self.delta = True
                 break
         dec = []
         line = getline(lr, self.hunk)
         while len(line) > 1:
             l = line[0:1]
             if l <= b'Z' and l >= b'A':
                 l = ord(l) - ord(b'A') + 1
             else:
                 l = ord(l) - ord(b'a') + 27
             try:
                 dec.append(util.b85decode(line[1:])[:l])
             except ValueError as e:
                 raise PatchError(
                     _(b'could not decode "%s" binary patch: %s')
                     % (self._fname, stringutil.forcebytestr(e))
                 )
             line = getline(lr, self.hunk)
         text = zlib.decompress(b''.join(dec))
         if len(text) != size:
             raise PatchError(
                 _(b'"%s" length is %d bytes, should be %d')
                 % (self._fname, len(text), size)
             )
         self.text = text
 
 
 def parsefilename(str):
     # --- filename \t|space stuff
     s = str[4:].rstrip(b'\r\n')
     i = s.find(b'\t')
     if i < 0:
         i = s.find(b' ')
         if i < 0:
             return s
     return s[:i]
 
 
 def reversehunks(hunks):
     '''reverse the signs in the hunks given as argument
 
     This function operates on hunks coming out of patch.filterpatch, that is
     a list of the form: [header1, hunk1, hunk2, header2...]. Example usage:
 
     >>> rawpatch = b"""diff --git a/folder1/g b/folder1/g
     ... --- a/folder1/g
     ... +++ b/folder1/g
     ... @@ -1,7 +1,7 @@
     ... +firstline
     ...  c
     ...  1
     ...  2
     ... + 3
     ... -4
     ...  5
     ...  d
     ... +lastline"""
     >>> hunks = parsepatch([rawpatch])
     >>> hunkscomingfromfilterpatch = []
     >>> for h in hunks:
     ...     hunkscomingfromfilterpatch.append(h)
     ...     hunkscomingfromfilterpatch.extend(h.hunks)
 
     >>> reversedhunks = reversehunks(hunkscomingfromfilterpatch)
     >>> from . import util
     >>> fp = util.stringio()
     >>> for c in reversedhunks:
     ...      c.write(fp)
     >>> fp.seek(0) or None
     >>> reversedpatch = fp.read()
     >>> print(pycompat.sysstr(reversedpatch))
     diff --git a/folder1/g b/folder1/g
     --- a/folder1/g
     +++ b/folder1/g
     @@ -1,4 +1,3 @@
     -firstline
      c
      1
      2
     @@ -2,6 +1,6 @@
      c
      1
      2
     - 3
     +4
      5
      d
     @@ -6,3 +5,2 @@
      5
      d
     -lastline
 
     '''
 
     newhunks = []
     for c in hunks:
         if util.safehasattr(c, b'reversehunk'):
             c = c.reversehunk()
         newhunks.append(c)
     return newhunks
 
 
 def parsepatch(originalchunks, maxcontext=None):
     """patch -> [] of headers -> [] of hunks
 
     If maxcontext is not None, trim context lines if necessary.
 
     >>> rawpatch = b'''diff --git a/folder1/g b/folder1/g
     ... --- a/folder1/g
     ... +++ b/folder1/g
     ... @@ -1,8 +1,10 @@
     ...  1
     ...  2
     ... -3
     ...  4
     ...  5
     ...  6
     ... +6.1
     ... +6.2
     ...  7
     ...  8
     ... +9'''
     >>> out = util.stringio()
     >>> headers = parsepatch([rawpatch], maxcontext=1)
     >>> for header in headers:
     ...     header.write(out)
     ...     for hunk in header.hunks:
     ...         hunk.write(out)
     >>> print(pycompat.sysstr(out.getvalue()))
     diff --git a/folder1/g b/folder1/g
     --- a/folder1/g
     +++ b/folder1/g
     @@ -2,3 +2,2 @@
      2
     -3
      4
     @@ -6,2 +5,4 @@
      6
     +6.1
     +6.2
      7
     @@ -8,1 +9,2 @@
      8
     +9
     """
 
     class parser(object):
         """patch parsing state machine"""
 
         def __init__(self):
             self.fromline = 0
             self.toline = 0
             self.proc = b''
             self.header = None
             self.context = []
             self.before = []
             self.hunk = []
             self.headers = []
 
         def addrange(self, limits):
             self.addcontext([])
             fromstart, fromend, tostart, toend, proc = limits
             self.fromline = int(fromstart)
             self.toline = int(tostart)
             self.proc = proc
 
         def addcontext(self, context):
             if self.hunk:
                 h = recordhunk(
                     self.header,
                     self.fromline,
                     self.toline,
                     self.proc,
                     self.before,
                     self.hunk,
                     context,
                     maxcontext,
                 )
                 self.header.hunks.append(h)
                 self.fromline += len(self.before) + h.removed
                 self.toline += len(self.before) + h.added
                 self.before = []
                 self.hunk = []
             self.context = context
 
         def addhunk(self, hunk):
             if self.context:
                 self.before = self.context
                 self.context = []
             if self.hunk:
                 self.addcontext([])
             self.hunk = hunk
 
         def newfile(self, hdr):
             self.addcontext([])
             h = header(hdr)
             self.headers.append(h)
             self.header = h
 
         def addother(self, line):
             pass  # 'other' lines are ignored
 
         def finished(self):
             self.addcontext([])
             return self.headers
 
         transitions = {
             b'file': {
                 b'context': addcontext,
                 b'file': newfile,
                 b'hunk': addhunk,
                 b'range': addrange,
             },
             b'context': {
                 b'file': newfile,
                 b'hunk': addhunk,
                 b'range': addrange,
                 b'other': addother,
             },
             b'hunk': {
                 b'context': addcontext,
                 b'file': newfile,
                 b'range': addrange,
             },
             b'range': {b'context': addcontext, b'hunk': addhunk},
             b'other': {b'other': addother},
         }
 
     p = parser()
     fp = stringio()
     fp.write(b''.join(originalchunks))
     fp.seek(0)
 
     state = b'context'
     for newstate, data in scanpatch(fp):
         try:
             p.transitions[state][newstate](p, data)
         except KeyError:
             raise PatchError(
                 b'unhandled transition: %s -> %s' % (state, newstate)
             )
         state = newstate
     del fp
     return p.finished()
 
 
 def pathtransform(path, strip, prefix):
     '''turn a path from a patch into a path suitable for the repository
 
     prefix, if not empty, is expected to be normalized with a / at the end.
 
     Returns (stripped components, path in repository).
 
     >>> pathtransform(b'a/b/c', 0, b'')
     ('', 'a/b/c')
     >>> pathtransform(b'   a/b/c   ', 0, b'')
     ('', '   a/b/c')
     >>> pathtransform(b'   a/b/c   ', 2, b'')
     ('a/b/', 'c')
     >>> pathtransform(b'a/b/c', 0, b'd/e/')
     ('', 'd/e/a/b/c')
     >>> pathtransform(b'   a//b/c   ', 2, b'd/e/')
     ('a//b/', 'd/e/c')
     >>> pathtransform(b'a/b/c', 3, b'')
     Traceback (most recent call last):
     PatchError: unable to strip away 1 of 3 dirs from a/b/c
     '''
     pathlen = len(path)
     i = 0
     if strip == 0:
         return b'', prefix + path.rstrip()
     count = strip
     while count > 0:
         i = path.find(b'/', i)
         if i == -1:
             raise PatchError(
                 _(b"unable to strip away %d of %d dirs from %s")
                 % (count, strip, path)
             )
         i += 1
         # consume '//' in the path
         while i < pathlen - 1 and path[i : i + 1] == b'/':
             i += 1
         count -= 1
     return path[:i].lstrip(), prefix + path[i:].rstrip()
 
 
 def makepatchmeta(backend, afile_orig, bfile_orig, hunk, strip, prefix):
     nulla = afile_orig == b"/dev/null"
     nullb = bfile_orig == b"/dev/null"
     create = nulla and hunk.starta == 0 and hunk.lena == 0
     remove = nullb and hunk.startb == 0 and hunk.lenb == 0
     abase, afile = pathtransform(afile_orig, strip, prefix)
     gooda = not nulla and backend.exists(afile)
     bbase, bfile = pathtransform(bfile_orig, strip, prefix)
     if afile == bfile:
         goodb = gooda
     else:
         goodb = not nullb and backend.exists(bfile)
     missing = not goodb and not gooda and not create
 
     # some diff programs apparently produce patches where the afile is
     # not /dev/null, but afile starts with bfile
     abasedir = afile[: afile.rfind(b'/') + 1]
     bbasedir = bfile[: bfile.rfind(b'/') + 1]
     if (
         missing
         and abasedir == bbasedir
         and afile.startswith(bfile)
         and hunk.starta == 0
         and hunk.lena == 0
     ):
         create = True
         missing = False
 
     # If afile is "a/b/foo" and bfile is "a/b/foo.orig" we assume the
     # diff is between a file and its backup. In this case, the original
     # file should be patched (see original mpatch code).
     isbackup = abase == bbase and bfile.startswith(afile)
     fname = None
     if not missing:
         if gooda and goodb:
             if isbackup:
                 fname = afile
             else:
                 fname = bfile
         elif gooda:
             fname = afile
 
     if not fname:
         if not nullb:
             if isbackup:
                 fname = afile
             else:
                 fname = bfile
         elif not nulla:
             fname = afile
         else:
             raise PatchError(_(b"undefined source and destination files"))
 
     gp = patchmeta(fname)
     if create:
         gp.op = b'ADD'
     elif remove:
         gp.op = b'DELETE'
     return gp
 
 
 def scanpatch(fp):
     """like patch.iterhunks, but yield different events
 
     - ('file',    [header_lines + fromfile + tofile])
     - ('context', [context_lines])
     - ('hunk',    [hunk_lines])
     - ('range',   (-start,len, +start,len, proc))
     """
     lines_re = re.compile(br'@@ -(\d+),(\d+) \+(\d+),(\d+) @@\s*(.*)')
     lr = linereader(fp)
 
     def scanwhile(first, p):
         """scan lr while predicate holds"""
         lines = [first]
         for line in iter(lr.readline, b''):
             if p(line):
                 lines.append(line)
             else:
                 lr.push(line)
                 break
         return lines
 
     for line in iter(lr.readline, b''):
         if line.startswith(b'diff --git a/') or line.startswith(b'diff -r '):
 
             def notheader(line):
                 s = line.split(None, 1)
                 return not s or s[0] not in (b'---', b'diff')
 
             header = scanwhile(line, notheader)
             fromfile = lr.readline()
             if fromfile.startswith(b'---'):
                 tofile = lr.readline()
                 header += [fromfile, tofile]
             else:
                 lr.push(fromfile)
             yield b'file', header
         elif line.startswith(b' '):
             cs = (b' ', b'\\')
             yield b'context', scanwhile(line, lambda l: l.startswith(cs))
         elif line.startswith((b'-', b'+')):
             cs = (b'-', b'+', b'\\')
             yield b'hunk', scanwhile(line, lambda l: l.startswith(cs))
         else:
             m = lines_re.match(line)
             if m:
                 yield b'range', m.groups()
             else:
                 yield b'other', line
 
 
 def scangitpatch(lr, firstline):
     """
     Git patches can emit:
     - rename a to b
     - change b
     - copy a to c
     - change c
 
     We cannot apply this sequence as-is, the renamed 'a' could not be
     found for it would have been renamed already. And we cannot copy
     from 'b' instead because 'b' would have been changed already. So
     we scan the git patch for copy and rename commands so we can
     perform the copies ahead of time.
     """
     pos = 0
     try:
         pos = lr.fp.tell()
         fp = lr.fp
     except IOError:
         fp = stringio(lr.fp.read())
     gitlr = linereader(fp)
     gitlr.push(firstline)
     gitpatches = readgitpatch(gitlr)
     fp.seek(pos)
     return gitpatches
 
 
 def iterhunks(fp):
     """Read a patch and yield the following events:
     - ("file", afile, bfile, firsthunk): select a new target file.
     - ("hunk", hunk): a new hunk is ready to be applied, follows a
     "file" event.
     - ("git", gitchanges): current diff is in git format, gitchanges
     maps filenames to gitpatch records. Unique event.
     """
     afile = b""
     bfile = b""
     state = None
     hunknum = 0
     emitfile = newfile = False
     gitpatches = None
 
     # our states
     BFILE = 1
     context = None
     lr = linereader(fp)
 
     for x in iter(lr.readline, b''):
         if state == BFILE and (
             (not context and x.startswith(b'@'))
             or (context is not False and x.startswith(b'***************'))
             or x.startswith(b'GIT binary patch')
         ):
             gp = None
             if gitpatches and gitpatches[-1].ispatching(afile, bfile):
                 gp = gitpatches.pop()
             if x.startswith(b'GIT binary patch'):
                 h = binhunk(lr, gp.path)
             else:
                 if context is None and x.startswith(b'***************'):
                     context = True
                 h = hunk(x, hunknum + 1, lr, context)
             hunknum += 1
             if emitfile:
                 emitfile = False
                 yield b'file', (afile, bfile, h, gp and gp.copy() or None)
             yield b'hunk', h
         elif x.startswith(b'diff --git a/'):
             m = gitre.match(x.rstrip(b' \r\n'))
             if not m:
                 continue
             if gitpatches is None:
                 # scan whole input for git metadata
                 gitpatches = scangitpatch(lr, x)
                 yield b'git', [
                     g.copy() for g in gitpatches if g.op in (b'COPY', b'RENAME')
                 ]
                 gitpatches.reverse()
             afile = b'a/' + m.group(1)
             bfile = b'b/' + m.group(2)
             while gitpatches and not gitpatches[-1].ispatching(afile, bfile):
                 gp = gitpatches.pop()
                 yield b'file', (
                     b'a/' + gp.path,
                     b'b/' + gp.path,
                     None,
                     gp.copy(),
                 )
             if not gitpatches:
                 raise PatchError(
                     _(b'failed to synchronize metadata for "%s"') % afile[2:]
                 )
             newfile = True
         elif x.startswith(b'---'):
             # check for a unified diff
             l2 = lr.readline()
             if not l2.startswith(b'+++'):
                 lr.push(l2)
                 continue
             newfile = True
             context = False
             afile = parsefilename(x)
             bfile = parsefilename(l2)
         elif x.startswith(b'***'):
             # check for a context diff
             l2 = lr.readline()
             if not l2.startswith(b'---'):
                 lr.push(l2)
                 continue
             l3 = lr.readline()
             lr.push(l3)
             if not l3.startswith(b"***************"):
                 lr.push(l2)
                 continue
             newfile = True
             context = True
             afile = parsefilename(x)
             bfile = parsefilename(l2)
 
         if newfile:
             newfile = False
             emitfile = True
             state = BFILE
             hunknum = 0
 
     while gitpatches:
         gp = gitpatches.pop()
         yield b'file', (b'a/' + gp.path, b'b/' + gp.path, None, gp.copy())
 
 
 def applybindelta(binchunk, data):
     """Apply a binary delta hunk
     The algorithm used is the algorithm from git's patch-delta.c
     """
 
     def deltahead(binchunk):
         i = 0
         for c in pycompat.bytestr(binchunk):
             i += 1
             if not (ord(c) & 0x80):
                 return i
         return i
 
     out = b""
     s = deltahead(binchunk)
     binchunk = binchunk[s:]
     s = deltahead(binchunk)
     binchunk = binchunk[s:]
     i = 0
     while i < len(binchunk):
         cmd = ord(binchunk[i : i + 1])
         i += 1
         if cmd & 0x80:
             offset = 0
             size = 0
             if cmd & 0x01:
                 offset = ord(binchunk[i : i + 1])
                 i += 1
             if cmd & 0x02:
                 offset |= ord(binchunk[i : i + 1]) << 8
                 i += 1
             if cmd & 0x04:
                 offset |= ord(binchunk[i : i + 1]) << 16
                 i += 1
             if cmd & 0x08:
                 offset |= ord(binchunk[i : i + 1]) << 24
                 i += 1
             if cmd & 0x10:
                 size = ord(binchunk[i : i + 1])
                 i += 1
             if cmd & 0x20:
                 size |= ord(binchunk[i : i + 1]) << 8
                 i += 1
             if cmd & 0x40:
                 size |= ord(binchunk[i : i + 1]) << 16
                 i += 1
             if size == 0:
                 size = 0x10000
             offset_end = offset + size
             out += data[offset:offset_end]
         elif cmd != 0:
             offset_end = i + cmd
             out += binchunk[i:offset_end]
             i += cmd
         else:
             raise PatchError(_(b'unexpected delta opcode 0'))
     return out
 
 
 def applydiff(ui, fp, backend, store, strip=1, prefix=b'', eolmode=b'strict'):
     """Reads a patch from fp and tries to apply it.
 
     Returns 0 for a clean patch, -1 if any rejects were found and 1 if
     there was any fuzz.
 
     If 'eolmode' is 'strict', the patch content and patched file are
     read in binary mode. Otherwise, line endings are ignored when
     patching then normalized according to 'eolmode'.
     """
     return _applydiff(
         ui,
         fp,
         patchfile,
         backend,
         store,
         strip=strip,
         prefix=prefix,
         eolmode=eolmode,
     )
 
 
 def _canonprefix(repo, prefix):
     if prefix:
         prefix = pathutil.canonpath(repo.root, repo.getcwd(), prefix)
         if prefix != b'':
             prefix += b'/'
     return prefix
 
 
 def _applydiff(
     ui, fp, patcher, backend, store, strip=1, prefix=b'', eolmode=b'strict'
 ):
     prefix = _canonprefix(backend.repo, prefix)
 
     def pstrip(p):
         return pathtransform(p, strip - 1, prefix)[1]
 
     rejects = 0
     err = 0
     current_file = None
 
     for state, values in iterhunks(fp):
         if state == b'hunk':
             if not current_file:
                 continue
             ret = current_file.apply(values)
             if ret > 0:
                 err = 1
         elif state == b'file':
             if current_file:
                 rejects += current_file.close()
                 current_file = None
             afile, bfile, first_hunk, gp = values
             if gp:
                 gp.path = pstrip(gp.path)
                 if gp.oldpath:
                     gp.oldpath = pstrip(gp.oldpath)
             else:
                 gp = makepatchmeta(
                     backend, afile, bfile, first_hunk, strip, prefix
                 )
             if gp.op == b'RENAME':
                 backend.unlink(gp.oldpath)
             if not first_hunk:
                 if gp.op == b'DELETE':
                     backend.unlink(gp.path)
                     continue
                 data, mode = None, None
                 if gp.op in (b'RENAME', b'COPY'):
                     data, mode = store.getfile(gp.oldpath)[:2]
                     if data is None:
                         # This means that the old path does not exist
                         raise PatchError(
                             _(b"source file '%s' does not exist") % gp.oldpath
                         )
                 if gp.mode:
                     mode = gp.mode
                     if gp.op == b'ADD':
                         # Added files without content have no hunk and
                         # must be created
                         data = b''
                 if data or mode:
                     if gp.op in (b'ADD', b'RENAME', b'COPY') and backend.exists(
                         gp.path
                     ):
                         raise PatchError(
                             _(
                                 b"cannot create %s: destination "
                                 b"already exists"
                             )
                             % gp.path
                         )
                     backend.setfile(gp.path, data, mode, gp.oldpath)
                 continue
             try:
                 current_file = patcher(ui, gp, backend, store, eolmode=eolmode)
             except PatchError as inst:
                 ui.warn(stringutil.forcebytestr(inst) + b'\n')
                 current_file = None
                 rejects += 1
                 continue
         elif state == b'git':
             for gp in values:
                 path = pstrip(gp.oldpath)
                 data, mode = backend.getfile(path)
                 if data is None:
                     # The error ignored here will trigger a getfile()
                     # error in a place more appropriate for error
                     # handling, and will not interrupt the patching
                     # process.
                     pass
                 else:
                     store.setfile(path, data, mode)
         else:
             raise error.Abort(_(b'unsupported parser state: %s') % state)
 
     if current_file:
         rejects += current_file.close()
 
     if rejects:
         return -1
     return err
 
 
 def _externalpatch(ui, repo, patcher, patchname, strip, files, similarity):
     """use <patcher> to apply <patchname> to the working directory.
     returns whether patch was applied with fuzz factor."""
 
     fuzz = False
     args = []
     cwd = repo.root
     if cwd:
         args.append(b'-d %s' % procutil.shellquote(cwd))
     cmd = b'%s %s -p%d < %s' % (
         patcher,
         b' '.join(args),
         strip,
         procutil.shellquote(patchname),
     )
     ui.debug(b'Using external patch tool: %s\n' % cmd)
     fp = procutil.popen(cmd, b'rb')
     try:
         for line in util.iterfile(fp):
             line = line.rstrip()
             ui.note(line + b'\n')
             if line.startswith(b'patching file '):
                 pf = util.parsepatchoutput(line)
                 printed_file = False
                 files.add(pf)
             elif line.find(b'with fuzz') >= 0:
                 fuzz = True
                 if not printed_file:
                     ui.warn(pf + b'\n')
                     printed_file = True
                 ui.warn(line + b'\n')
             elif line.find(b'saving rejects to file') >= 0:
                 ui.warn(line + b'\n')
             elif line.find(b'FAILED') >= 0:
                 if not printed_file:
                     ui.warn(pf + b'\n')
                     printed_file = True
                 ui.warn(line + b'\n')
     finally:
         if files:
             scmutil.marktouched(repo, files, similarity)
     code = fp.close()
     if code:
         raise PatchError(
             _(b"patch command failed: %s") % procutil.explainexit(code)
         )
     return fuzz
 
 
 def patchbackend(
     ui, backend, patchobj, strip, prefix, files=None, eolmode=b'strict'
 ):
     if files is None:
         files = set()
     if eolmode is None:
         eolmode = ui.config(b'patch', b'eol')
     if eolmode.lower() not in eolmodes:
         raise error.Abort(_(b'unsupported line endings type: %s') % eolmode)
     eolmode = eolmode.lower()
 
     store = filestore()
     try:
         fp = open(patchobj, b'rb')
     except TypeError:
         fp = patchobj
     try:
         ret = applydiff(
             ui, fp, backend, store, strip=strip, prefix=prefix, eolmode=eolmode
         )
     finally:
         if fp != patchobj:
             fp.close()
         files.update(backend.close())
         store.close()
     if ret < 0:
         raise PatchError(_(b'patch failed to apply'))
     return ret > 0
 
 
 def internalpatch(
     ui,
     repo,
     patchobj,
     strip,
     prefix=b'',
     files=None,
     eolmode=b'strict',
     similarity=0,
 ):
     """use builtin patch to apply <patchobj> to the working directory.
     returns whether patch was applied with fuzz factor."""
     backend = workingbackend(ui, repo, similarity)
     return patchbackend(ui, backend, patchobj, strip, prefix, files, eolmode)
 
 
 def patchrepo(
     ui, repo, ctx, store, patchobj, strip, prefix, files=None, eolmode=b'strict'
 ):
     backend = repobackend(ui, repo, ctx, store)
     return patchbackend(ui, backend, patchobj, strip, prefix, files, eolmode)
 
 
 def patch(
     ui,
     repo,
     patchname,
     strip=1,
     prefix=b'',
     files=None,
     eolmode=b'strict',
     similarity=0,
 ):
     """Apply <patchname> to the working directory.
 
     'eolmode' specifies how end of lines should be handled. It can be:
     - 'strict': inputs are read in binary mode, EOLs are preserved
     - 'crlf': EOLs are ignored when patching and reset to CRLF
     - 'lf': EOLs are ignored when patching and reset to LF
     - None: get it from user settings, default to 'strict'
     'eolmode' is ignored when using an external patcher program.
 
     Returns whether patch was applied with fuzz factor.
     """
     patcher = ui.config(b'ui', b'patch')
     if files is None:
         files = set()
     if patcher:
         return _externalpatch(
             ui, repo, patcher, patchname, strip, files, similarity
         )
     return internalpatch(
         ui, repo, patchname, strip, prefix, files, eolmode, similarity
     )
 
 
 def changedfiles(ui, repo, patchpath, strip=1, prefix=b''):
     backend = fsbackend(ui, repo.root)
     prefix = _canonprefix(repo, prefix)
     with open(patchpath, b'rb') as fp:
         changed = set()
         for state, values in iterhunks(fp):
             if state == b'file':
                 afile, bfile, first_hunk, gp = values
                 if gp:
                     gp.path = pathtransform(gp.path, strip - 1, prefix)[1]
                     if gp.oldpath:
                         gp.oldpath = pathtransform(
                             gp.oldpath, strip - 1, prefix
                         )[1]
                 else:
                     gp = makepatchmeta(
                         backend, afile, bfile, first_hunk, strip, prefix
                     )
                 changed.add(gp.path)
                 if gp.op == b'RENAME':
                     changed.add(gp.oldpath)
             elif state not in (b'hunk', b'git'):
                 raise error.Abort(_(b'unsupported parser state: %s') % state)
         return changed
 
 
 class GitDiffRequired(Exception):
     pass
 
 
 diffopts = diffutil.diffallopts
 diffallopts = diffutil.diffallopts
 difffeatureopts = diffutil.difffeatureopts
 
 
 def diff(
     repo,
     node1=None,
     node2=None,
     match=None,
     changes=None,
     opts=None,
     losedatafn=None,
     pathfn=None,
     copy=None,
     copysourcematch=None,
     hunksfilterfn=None,
 ):
     '''yields diff of changes to files between two nodes, or node and
     working directory.
 
     if node1 is None, use first dirstate parent instead.
     if node2 is None, compare node1 with working directory.
 
     losedatafn(**kwarg) is a callable run when opts.upgrade=True and
     every time some change cannot be represented with the current
     patch format. Return False to upgrade to git patch format, True to
     accept the loss or raise an exception to abort the diff. It is
     called with the name of current file being diffed as 'fn'. If set
     to None, patches will always be upgraded to git format when
     necessary.
 
     prefix is a filename prefix that is prepended to all filenames on
     display (used for subrepos).
 
     relroot, if not empty, must be normalized with a trailing /. Any match
     patterns that fall outside it will be ignored.
 
     copy, if not empty, should contain mappings {dst@y: src@x} of copy
     information.
 
     if copysourcematch is not None, then copy sources will be filtered by this
     matcher
 
     hunksfilterfn, if not None, should be a function taking a filectx and
     hunks generator that may yield filtered hunks.
     '''
     if not node1 and not node2:
         node1 = repo.dirstate.p1()
 
     ctx1 = repo[node1]
     ctx2 = repo[node2]
 
     for fctx1, fctx2, hdr, hunks in diffhunks(
         repo,
         ctx1=ctx1,
         ctx2=ctx2,
         match=match,
         changes=changes,
         opts=opts,
         losedatafn=losedatafn,
         pathfn=pathfn,
         copy=copy,
         copysourcematch=copysourcematch,
     ):
         if hunksfilterfn is not None:
             # If the file has been removed, fctx2 is None; but this should
             # not occur here since we catch removed files early in
             # logcmdutil.getlinerangerevs() for 'hg log -L'.
             assert (
                 fctx2 is not None
             ), b'fctx2 unexpectly None in diff hunks filtering'
             hunks = hunksfilterfn(fctx2, hunks)
         text = b''.join(b''.join(hlines) for hrange, hlines in hunks)
         if hdr and (text or len(hdr) > 1):
             yield b'\n'.join(hdr) + b'\n'
         if text:
             yield text
 
 
 def diffhunks(
     repo,
     ctx1,
     ctx2,
     match=None,
     changes=None,
     opts=None,
     losedatafn=None,
     pathfn=None,
     copy=None,
     copysourcematch=None,
 ):
     """Yield diff of changes to files in the form of (`header`, `hunks`) tuples
     where `header` is a list of diff headers and `hunks` is an iterable of
     (`hunkrange`, `hunklines`) tuples.
 
     See diff() for the meaning of parameters.
     """
 
     if opts is None:
         opts = mdiff.defaultopts
 
     def lrugetfilectx():
         cache = {}
         order = collections.deque()
 
         def getfilectx(f, ctx):
             fctx = ctx.filectx(f, filelog=cache.get(f))
             if f not in cache:
                 if len(cache) > 20:
                     del cache[order.popleft()]
                 cache[f] = fctx.filelog()
             else:
                 order.remove(f)
             order.append(f)
             return fctx
 
         return getfilectx
 
     getfilectx = lrugetfilectx()
 
     if not changes:
         changes = ctx1.status(ctx2, match=match)
     if isinstance(changes, list):
         modified, added, removed = changes[:3]
     else:
         modified, added, removed = (
             changes.modified,
             changes.added,
             changes.removed,
         )
 
     if not modified and not added and not removed:
         return []
 
     if repo.ui.debugflag:
         hexfunc = hex
     else:
         hexfunc = short
     revs = [hexfunc(node) for node in [ctx1.node(), ctx2.node()] if node]
 
     if copy is None:
         copy = {}
         if opts.git or opts.upgrade:
             copy = copies.pathcopies(ctx1, ctx2, match=match)
 
     if copysourcematch:
         # filter out copies where source side isn't inside the matcher
         # (copies.pathcopies() already filtered out the destination)
         copy = {
             dst: src
             for dst, src in pycompat.iteritems(copy)
             if copysourcematch(src)
         }
 
     modifiedset = set(modified)
     addedset = set(added)
     removedset = set(removed)
     for f in modified:
         if f not in ctx1:
             # Fix up added, since merged-in additions appear as
             # modifications during merges
             modifiedset.remove(f)
             addedset.add(f)
     for f in removed:
         if f not in ctx1:
             # Merged-in additions that are then removed are reported as removed.
             # They are not in ctx1, so We don't want to show them in the diff.
             removedset.remove(f)
     modified = sorted(modifiedset)
     added = sorted(addedset)
     removed = sorted(removedset)
     for dst, src in list(copy.items()):
         if src not in ctx1:
             # Files merged in during a merge and then copied/renamed are
             # reported as copies. We want to show them in the diff as additions.
             del copy[dst]
 
     prefetchmatch = scmutil.matchfiles(
         repo, list(modifiedset | addedset | removedset)
     )
-    scmutil.prefetchfiles(repo, [ctx1.rev(), ctx2.rev()], prefetchmatch)
+    revmatches = [
+        (ctx1.rev(), prefetchmatch),
+        (ctx2.rev(), prefetchmatch),
+    ]
+    scmutil.prefetchfiles(repo, revmatches)
 
     def difffn(opts, losedata):
         return trydiff(
             repo,
             revs,
             ctx1,
             ctx2,
             modified,
             added,
             removed,
             copy,
             getfilectx,
             opts,
             losedata,
             pathfn,
         )
 
     if opts.upgrade and not opts.git:
         try:
 
             def losedata(fn):
                 if not losedatafn or not losedatafn(fn=fn):
                     raise GitDiffRequired
 
             # Buffer the whole output until we are sure it can be generated
             return list(difffn(opts.copy(git=False), losedata))
         except GitDiffRequired:
             return difffn(opts.copy(git=True), None)
     else:
         return difffn(opts, None)
 
 
 def diffsinglehunk(hunklines):
     """yield tokens for a list of lines in a single hunk"""
     for line in hunklines:
         # chomp
         chompline = line.rstrip(b'\r\n')
         # highlight tabs and trailing whitespace
         stripline = chompline.rstrip()
         if line.startswith(b'-'):
             label = b'diff.deleted'
         elif line.startswith(b'+'):
             label = b'diff.inserted'
         else:
             raise error.ProgrammingError(b'unexpected hunk line: %s' % line)
         for token in tabsplitter.findall(stripline):
             if token.startswith(b'\t'):
                 yield (token, b'diff.tab')
             else:
                 yield (token, label)
 
         if chompline != stripline:
             yield (chompline[len(stripline) :], b'diff.trailingwhitespace')
         if chompline != line:
             yield (line[len(chompline) :], b'')
 
 
 def diffsinglehunkinline(hunklines):
     """yield tokens for a list of lines in a single hunk, with inline colors"""
     # prepare deleted, and inserted content
     a = b''
     b = b''
     for line in hunklines:
         if line[0:1] == b'-':
             a += line[1:]
         elif line[0:1] == b'+':
             b += line[1:]
         else:
             raise error.ProgrammingError(b'unexpected hunk line: %s' % line)
     # fast path: if either side is empty, use diffsinglehunk
     if not a or not b:
         for t in diffsinglehunk(hunklines):
             yield t
         return
     # re-split the content into words
     al = wordsplitter.findall(a)
     bl = wordsplitter.findall(b)
     # re-arrange the words to lines since the diff algorithm is line-based
     aln = [s if s == b'\n' else s + b'\n' for s in al]
     bln = [s if s == b'\n' else s + b'\n' for s in bl]
     an = b''.join(aln)
     bn = b''.join(bln)
     # run the diff algorithm, prepare atokens and btokens
     atokens = []
     btokens = []
     blocks = mdiff.allblocks(an, bn, lines1=aln, lines2=bln)
     for (a1, a2, b1, b2), btype in blocks:
         changed = btype == b'!'
         for token in mdiff.splitnewlines(b''.join(al[a1:a2])):
             atokens.append((changed, token))
         for token in mdiff.splitnewlines(b''.join(bl[b1:b2])):
             btokens.append((changed, token))
 
     # yield deleted tokens, then inserted ones
     for prefix, label, tokens in [
         (b'-', b'diff.deleted', atokens),
         (b'+', b'diff.inserted', btokens),
     ]:
         nextisnewline = True
         for changed, token in tokens:
             if nextisnewline:
                 yield (prefix, label)
                 nextisnewline = False
             # special handling line end
             isendofline = token.endswith(b'\n')
             if isendofline:
                 chomp = token[:-1]  # chomp
                 if chomp.endswith(b'\r'):
                     chomp = chomp[:-1]
                 endofline = token[len(chomp) :]
                 token = chomp.rstrip()  # detect spaces at the end
                 endspaces = chomp[len(token) :]
             # scan tabs
             for maybetab in tabsplitter.findall(token):
                 if b'\t' == maybetab[0:1]:
                     currentlabel = b'diff.tab'
                 else:
                     if changed:
                         currentlabel = label + b'.changed'
                     else:
                         currentlabel = label + b'.unchanged'
                 yield (maybetab, currentlabel)
             if isendofline:
                 if endspaces:
                     yield (endspaces, b'diff.trailingwhitespace')
                 yield (endofline, b'')
                 nextisnewline = True
 
 
 def difflabel(func, *args, **kw):
     '''yields 2-tuples of (output, label) based on the output of func()'''
     if kw.get('opts') and kw['opts'].worddiff:
         dodiffhunk = diffsinglehunkinline
     else:
         dodiffhunk = diffsinglehunk
     headprefixes = [
         (b'diff', b'diff.diffline'),
         (b'copy', b'diff.extended'),
         (b'rename', b'diff.extended'),
         (b'old', b'diff.extended'),
         (b'new', b'diff.extended'),
         (b'deleted', b'diff.extended'),
         (b'index', b'diff.extended'),
         (b'similarity', b'diff.extended'),
         (b'---', b'diff.file_a'),
         (b'+++', b'diff.file_b'),
     ]
     textprefixes = [
         (b'@', b'diff.hunk'),
         # - and + are handled by diffsinglehunk
     ]
     head = False
 
     # buffers a hunk, i.e. adjacent "-", "+" lines without other changes.
     hunkbuffer = []
 
     def consumehunkbuffer():
         if hunkbuffer:
             for token in dodiffhunk(hunkbuffer):
                 yield token
             hunkbuffer[:] = []
 
     for chunk in func(*args, **kw):
         lines = chunk.split(b'\n')
         linecount = len(lines)
         for i, line in enumerate(lines):
             if head:
                 if line.startswith(b'@'):
                     head = False
             else:
                 if line and not line.startswith(
                     (b' ', b'+', b'-', b'@', b'\\')
                 ):
                     head = True
             diffline = False
             if not head and line and line.startswith((b'+', b'-')):
                 diffline = True
 
             prefixes = textprefixes
             if head:
                 prefixes = headprefixes
             if diffline:
                 # buffered
                 bufferedline = line
                 if i + 1 < linecount:
                     bufferedline += b"\n"
                 hunkbuffer.append(bufferedline)
             else:
                 # unbuffered
                 for token in consumehunkbuffer():
                     yield token
                 stripline = line.rstrip()
                 for prefix, label in prefixes:
                     if stripline.startswith(prefix):
                         yield (stripline, label)
                         if line != stripline:
                             yield (
                                 line[len(stripline) :],
                                 b'diff.trailingwhitespace',
                             )
                         break
                 else:
                     yield (line, b'')
                 if i + 1 < linecount:
                     yield (b'\n', b'')
         for token in consumehunkbuffer():
             yield token
 
 
 def diffui(*args, **kw):
     '''like diff(), but yields 2-tuples of (output, label) for ui.write()'''
     return difflabel(diff, *args, **kw)
 
 
 def _filepairs(modified, added, removed, copy, opts):
     '''generates tuples (f1, f2, copyop), where f1 is the name of the file
     before and f2 is the the name after. For added files, f1 will be None,
     and for removed files, f2 will be None. copyop may be set to None, 'copy'
     or 'rename' (the latter two only if opts.git is set).'''
     gone = set()
 
     copyto = {v: k for k, v in copy.items()}
 
     addedset, removedset = set(added), set(removed)
 
     for f in sorted(modified + added + removed):
         copyop = None
         f1, f2 = f, f
         if f in addedset:
             f1 = None
             if f in copy:
                 if opts.git:
                     f1 = copy[f]
                     if f1 in removedset and f1 not in gone:
                         copyop = b'rename'
                         gone.add(f1)
                     else:
                         copyop = b'copy'
         elif f in removedset:
             f2 = None
             if opts.git:
                 # have we already reported a copy above?
                 if (
                     f in copyto
                     and copyto[f] in addedset
                     and copy[copyto[f]] == f
                 ):
                     continue
         yield f1, f2, copyop
 
 
 def trydiff(
     repo,
     revs,
     ctx1,
     ctx2,
     modified,
     added,
     removed,
     copy,
     getfilectx,
     opts,
     losedatafn,
     pathfn,
 ):
     '''given input data, generate a diff and yield it in blocks
 
     If generating a diff would lose data like flags or binary data and
     losedatafn is not None, it will be called.
 
     pathfn is applied to every path in the diff output.
     '''
 
     def gitindex(text):
         if not text:
             text = b""
         l = len(text)
         s = hashutil.sha1(b'blob %d\0' % l)
         s.update(text)
         return hex(s.digest())
 
     if opts.noprefix:
         aprefix = bprefix = b''
     else:
         aprefix = b'a/'
         bprefix = b'b/'
 
     def diffline(f, revs):
         revinfo = b' '.join([b"-r %s" % rev for rev in revs])
         return b'diff %s %s' % (revinfo, f)
 
     def isempty(fctx):
         return fctx is None or fctx.size() == 0
 
     date1 = dateutil.datestr(ctx1.date())
     date2 = dateutil.datestr(ctx2.date())
 
     gitmode = {b'l': b'120000', b'x': b'100755', b'': b'100644'}
 
     if not pathfn:
         pathfn = lambda f: f
 
     for f1, f2, copyop in _filepairs(modified, added, removed, copy, opts):
         content1 = None
         content2 = None
         fctx1 = None
         fctx2 = None
         flag1 = None
         flag2 = None
         if f1:
             fctx1 = getfilectx(f1, ctx1)
             if opts.git or losedatafn:
                 flag1 = ctx1.flags(f1)
         if f2:
             fctx2 = getfilectx(f2, ctx2)
             if opts.git or losedatafn:
                 flag2 = ctx2.flags(f2)
         # if binary is True, output "summary" or "base85", but not "text diff"
         if opts.text:
             binary = False
         else:
             binary = any(f.isbinary() for f in [fctx1, fctx2] if f is not None)
 
         if losedatafn and not opts.git:
             if (
                 binary
                 or
                 # copy/rename
                 f2 in copy
                 or
                 # empty file creation
                 (not f1 and isempty(fctx2))
                 or
                 # empty file deletion
                 (isempty(fctx1) and not f2)
                 or
                 # create with flags
                 (not f1 and flag2)
                 or
                 # change flags
                 (f1 and f2 and flag1 != flag2)
             ):
                 losedatafn(f2 or f1)
 
         path1 = pathfn(f1 or f2)
         path2 = pathfn(f2 or f1)
         header = []
         if opts.git:
             header.append(
                 b'diff --git %s%s %s%s' % (aprefix, path1, bprefix, path2)
             )
             if not f1:  # added
                 header.append(b'new file mode %s' % gitmode[flag2])
             elif not f2:  # removed
                 header.append(b'deleted file mode %s' % gitmode[flag1])
             else:  # modified/copied/renamed
                 mode1, mode2 = gitmode[flag1], gitmode[flag2]
                 if mode1 != mode2:
                     header.append(b'old mode %s' % mode1)
                     header.append(b'new mode %s' % mode2)
                 if copyop is not None:
                     if opts.showsimilarity:
                         sim = similar.score(ctx1[path1], ctx2[path2]) * 100
                         header.append(b'similarity index %d%%' % sim)
                     header.append(b'%s from %s' % (copyop, path1))
                     header.append(b'%s to %s' % (copyop, path2))
         elif revs:
             header.append(diffline(path1, revs))
 
         #  fctx.is  | diffopts                | what to   | is fctx.data()
         #  binary() | text nobinary git index | output?   | outputted?
         # ------------------------------------|----------------------------
         #  yes      | no   no       no  *     | summary   | no
         #  yes      | no   no       yes *     | base85    | yes
         #  yes      | no   yes      no  *     | summary   | no
         #  yes      | no   yes      yes 0     | summary   | no
         #  yes      | no   yes      yes >0    | summary   | semi [1]
         #  yes      | yes  *        *   *     | text diff | yes
         #  no       | *    *        *   *     | text diff | yes
         # [1]: hash(fctx.data()) is outputted. so fctx.data() cannot be faked
         if binary and (
             not opts.git or (opts.git and opts.nobinary and not opts.index)
         ):
             # fast path: no binary content will be displayed, content1 and
             # content2 are only used for equivalent test. cmp() could have a
             # fast path.
             if fctx1 is not None:
                 content1 = b'\0'
             if fctx2 is not None:
                 if fctx1 is not None and not fctx1.cmp(fctx2):
                     content2 = b'\0'  # not different
                 else:
                     content2 = b'\0\0'
         else:
             # normal path: load contents
             if fctx1 is not None:
                 content1 = fctx1.data()
             if fctx2 is not None:
                 content2 = fctx2.data()
 
         if binary and opts.git and not opts.nobinary:
             text = mdiff.b85diff(content1, content2)
             if text:
                 header.append(
                     b'index %s..%s' % (gitindex(content1), gitindex(content2))
                 )
             hunks = ((None, [text]),)
         else:
             if opts.git and opts.index > 0:
                 flag = flag1
                 if flag is None:
                     flag = flag2
                 header.append(
                     b'index %s..%s %s'
                     % (
                         gitindex(content1)[0 : opts.index],
                         gitindex(content2)[0 : opts.index],
                         gitmode[flag],
                     )
                 )
 
             uheaders, hunks = mdiff.unidiff(
                 content1,
                 date1,
                 content2,
                 date2,
                 path1,
                 path2,
                 binary=binary,
                 opts=opts,
             )
             header.extend(uheaders)
         yield fctx1, fctx2, header, hunks
 
 
 def diffstatsum(stats):
     maxfile, maxtotal, addtotal, removetotal, binary = 0, 0, 0, 0, False
     for f, a, r, b in stats:
         maxfile = max(maxfile, encoding.colwidth(f))
         maxtotal = max(maxtotal, a + r)
         addtotal += a
         removetotal += r
         binary = binary or b
 
     return maxfile, maxtotal, addtotal, removetotal, binary
 
 
 def diffstatdata(lines):
     diffre = re.compile(br'^diff .*-r [a-z0-9]+\s(.*)$')
 
     results = []
     filename, adds, removes, isbinary = None, 0, 0, False
 
     def addresult():
         if filename:
             results.append((filename, adds, removes, isbinary))
 
     # inheader is used to track if a line is in the
     # header portion of the diff.  This helps properly account
     # for lines that start with '--' or '++'
     inheader = False
 
     for line in lines:
         if line.startswith(b'diff'):
             addresult()
             # starting a new file diff
             # set numbers to 0 and reset inheader
             inheader = True
             adds, removes, isbinary = 0, 0, False
             if line.startswith(b'diff --git a/'):
                 filename = gitre.search(line).group(2)
             elif line.startswith(b'diff -r'):
                 # format: "diff -r ... -r ... filename"
                 filename = diffre.search(line).group(1)
         elif line.startswith(b'@@'):
             inheader = False
         elif line.startswith(b'+') and not inheader:
             adds += 1
         elif line.startswith(b'-') and not inheader:
             removes += 1
         elif line.startswith(b'GIT binary patch') or line.startswith(
             b'Binary file'
         ):
             isbinary = True
         elif line.startswith(b'rename from'):
             filename = line[12:]
         elif line.startswith(b'rename to'):
             filename += b' => %s' % line[10:]
     addresult()
     return results
 
 
 def diffstat(lines, width=80):
     output = []
     stats = diffstatdata(lines)
     maxname, maxtotal, totaladds, totalremoves, hasbinary = diffstatsum(stats)
 
     countwidth = len(str(maxtotal))
     if hasbinary and countwidth < 3:
         countwidth = 3
     graphwidth = width - countwidth - maxname - 6
     if graphwidth < 10:
         graphwidth = 10
 
     def scale(i):
         if maxtotal <= graphwidth:
             return i
         # If diffstat runs out of room it doesn't print anything,
         # which isn't very useful, so always print at least one + or -
         # if there were at least some changes.
         return max(i * graphwidth // maxtotal, int(bool(i)))
 
     for filename, adds, removes, isbinary in stats:
         if isbinary:
             count = b'Bin'
         else:
             count = b'%d' % (adds + removes)
         pluses = b'+' * scale(adds)
         minuses = b'-' * scale(removes)
         output.append(
             b' %s%s |  %*s %s%s\n'
             % (
                 filename,
                 b' ' * (maxname - encoding.colwidth(filename)),
                 countwidth,
                 count,
                 pluses,
                 minuses,
             )
         )
 
     if stats:
         output.append(
             _(b' %d files changed, %d insertions(+), %d deletions(-)\n')
             % (len(stats), totaladds, totalremoves)
         )
 
     return b''.join(output)
 
 
 def diffstatui(*args, **kw):
     '''like diffstat(), but yields 2-tuples of (output, label) for
     ui.write()
     '''
 
     for line in diffstat(*args, **kw).splitlines():
         if line and line[-1] in b'+-':
             name, graph = line.rsplit(b' ', 1)
             yield (name + b' ', b'')
             m = re.search(br'\++', graph)
             if m:
                 yield (m.group(0), b'diffstat.inserted')
             m = re.search(br'-+', graph)
             if m:
                 yield (m.group(0), b'diffstat.deleted')
         else:
             yield (line, b'')
         yield (b'\n', b'')