store: return just one filename in walk functions

Various walk functions return `(revlog_type, decoded, encoded)` where
decoded could be None. But no-one cares about `encoded` and expects
`unencoded` to be present, except verify (because this can only happen
with old repo formats).

Simplify all this by either failing outright if a decoding a filename
fails (instead of almost certainly failing with a type error due to
treating None as a bytes), or skipping the filename but providing in
an out argument for hg verify.

Differential Revision: https://phab.mercurial-scm.org/D11248

hgext/narrow/narrowcommands.py

@@ -1,694 +1,694 b''
 # narrowcommands.py - command modifications for narrowhg extension
 #
 # Copyright 2017 Google, 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 itertools
 import os
 
 from mercurial.i18n import _
 from mercurial.node import (
     hex,
     short,
 )
 from mercurial import (
     bundle2,
     cmdutil,
     commands,
     discovery,
     encoding,
     error,
     exchange,
     extensions,
     hg,
     narrowspec,
     pathutil,
     pycompat,
     registrar,
     repair,
     repoview,
     requirements,
     sparse,
     util,
     wireprototypes,
 )
 from mercurial.utils import (
     urlutil,
 )
 
 table = {}
 command = registrar.command(table)
 
 
 def setup():
     """Wraps user-facing mercurial commands with narrow-aware versions."""
 
     entry = extensions.wrapcommand(commands.table, b'clone', clonenarrowcmd)
     entry[1].append(
         (b'', b'narrow', None, _(b"create a narrow clone of select files"))
     )
     entry[1].append(
         (
             b'',
             b'depth',
             b'',
             _(b"limit the history fetched by distance from heads"),
         )
     )
     entry[1].append((b'', b'narrowspec', b'', _(b"read narrowspecs from file")))
     # TODO(durin42): unify sparse/narrow --include/--exclude logic a bit
     if b'sparse' not in extensions.enabled():
         entry[1].append(
             (b'', b'include', [], _(b"specifically fetch this file/directory"))
         )
         entry[1].append(
             (
                 b'',
                 b'exclude',
                 [],
                 _(b"do not fetch this file/directory, even if included"),
             )
         )
 
     entry = extensions.wrapcommand(commands.table, b'pull', pullnarrowcmd)
     entry[1].append(
         (
             b'',
             b'depth',
             b'',
             _(b"limit the history fetched by distance from heads"),
         )
     )
 
     extensions.wrapcommand(commands.table, b'archive', archivenarrowcmd)
 
 
 def clonenarrowcmd(orig, ui, repo, *args, **opts):
     """Wraps clone command, so 'hg clone' first wraps localrepo.clone()."""
     opts = pycompat.byteskwargs(opts)
     wrappedextraprepare = util.nullcontextmanager()
     narrowspecfile = opts[b'narrowspec']
 
     if narrowspecfile:
         filepath = os.path.join(encoding.getcwd(), narrowspecfile)
         ui.status(_(b"reading narrowspec from '%s'\n") % filepath)
         try:
             fdata = util.readfile(filepath)
         except IOError as inst:
             raise error.Abort(
                 _(b"cannot read narrowspecs from '%s': %s")
                 % (filepath, encoding.strtolocal(inst.strerror))
             )
 
         includes, excludes, profiles = sparse.parseconfig(ui, fdata, b'narrow')
         if profiles:
             raise error.ConfigError(
                 _(
                     b"cannot specify other files using '%include' in"
                     b" narrowspec"
                 )
             )
 
         narrowspec.validatepatterns(includes)
         narrowspec.validatepatterns(excludes)
 
         # narrowspec is passed so we should assume that user wants narrow clone
         opts[b'narrow'] = True
         opts[b'include'].extend(includes)
         opts[b'exclude'].extend(excludes)
 
     if opts[b'narrow']:
 
         def pullbundle2extraprepare_widen(orig, pullop, kwargs):
             orig(pullop, kwargs)
 
             if opts.get(b'depth'):
                 kwargs[b'depth'] = opts[b'depth']
 
         wrappedextraprepare = extensions.wrappedfunction(
             exchange, b'_pullbundle2extraprepare', pullbundle2extraprepare_widen
         )
 
     with wrappedextraprepare:
         return orig(ui, repo, *args, **pycompat.strkwargs(opts))
 
 
 def pullnarrowcmd(orig, ui, repo, *args, **opts):
     """Wraps pull command to allow modifying narrow spec."""
     wrappedextraprepare = util.nullcontextmanager()
     if requirements.NARROW_REQUIREMENT in repo.requirements:
 
         def pullbundle2extraprepare_widen(orig, pullop, kwargs):
             orig(pullop, kwargs)
             if opts.get('depth'):
                 kwargs[b'depth'] = opts['depth']
 
         wrappedextraprepare = extensions.wrappedfunction(
             exchange, b'_pullbundle2extraprepare', pullbundle2extraprepare_widen
         )
 
     with wrappedextraprepare:
         return orig(ui, repo, *args, **opts)
 
 
 def archivenarrowcmd(orig, ui, repo, *args, **opts):
     """Wraps archive command to narrow the default includes."""
     if requirements.NARROW_REQUIREMENT in repo.requirements:
         repo_includes, repo_excludes = repo.narrowpats
         includes = set(opts.get('include', []))
         excludes = set(opts.get('exclude', []))
         includes, excludes, unused_invalid = narrowspec.restrictpatterns(
             includes, excludes, repo_includes, repo_excludes
         )
         if includes:
             opts['include'] = includes
         if excludes:
             opts['exclude'] = excludes
     return orig(ui, repo, *args, **opts)
 
 
 def pullbundle2extraprepare(orig, pullop, kwargs):
     repo = pullop.repo
     if requirements.NARROW_REQUIREMENT not in repo.requirements:
         return orig(pullop, kwargs)
 
     if wireprototypes.NARROWCAP not in pullop.remote.capabilities():
         raise error.Abort(_(b"server does not support narrow clones"))
     orig(pullop, kwargs)
     kwargs[b'narrow'] = True
     include, exclude = repo.narrowpats
     kwargs[b'oldincludepats'] = include
     kwargs[b'oldexcludepats'] = exclude
     if include:
         kwargs[b'includepats'] = include
     if exclude:
         kwargs[b'excludepats'] = exclude
     # calculate known nodes only in ellipses cases because in non-ellipses cases
     # we have all the nodes
     if wireprototypes.ELLIPSESCAP1 in pullop.remote.capabilities():
         kwargs[b'known'] = [
             hex(ctx.node())
             for ctx in repo.set(b'::%ln', pullop.common)
             if ctx.node() != repo.nullid
         ]
         if not kwargs[b'known']:
             # Mercurial serializes an empty list as '' and deserializes it as
             # [''], so delete it instead to avoid handling the empty string on
             # the server.
             del kwargs[b'known']
 
 
 extensions.wrapfunction(
     exchange, b'_pullbundle2extraprepare', pullbundle2extraprepare
 )
 
 
 def _narrow(
     ui,
     repo,
     remote,
     commoninc,
     oldincludes,
     oldexcludes,
     newincludes,
     newexcludes,
     force,
     backup,
 ):
     oldmatch = narrowspec.match(repo.root, oldincludes, oldexcludes)
     newmatch = narrowspec.match(repo.root, newincludes, newexcludes)
 
     # This is essentially doing "hg outgoing" to find all local-only
     # commits. We will then check that the local-only commits don't
     # have any changes to files that will be untracked.
     unfi = repo.unfiltered()
     outgoing = discovery.findcommonoutgoing(unfi, remote, commoninc=commoninc)
     ui.status(_(b'looking for local changes to affected paths\n'))
     progress = ui.makeprogress(
         topic=_(b'changesets'),
         unit=_(b'changesets'),
         total=len(outgoing.missing) + len(outgoing.excluded),
     )
     localnodes = []
     with progress:
         for n in itertools.chain(outgoing.missing, outgoing.excluded):
             progress.increment()
             if any(oldmatch(f) and not newmatch(f) for f in unfi[n].files()):
                 localnodes.append(n)
     revstostrip = unfi.revs(b'descendants(%ln)', localnodes)
     hiddenrevs = repoview.filterrevs(repo, b'visible')
     visibletostrip = list(
         repo.changelog.node(r) for r in (revstostrip - hiddenrevs)
     )
     if visibletostrip:
         ui.status(
             _(
                 b'The following changeset(s) or their ancestors have '
                 b'local changes not on the remote:\n'
             )
         )
         maxnodes = 10
         if ui.verbose or len(visibletostrip) <= maxnodes:
             for n in visibletostrip:
                 ui.status(b'%s\n' % short(n))
         else:
             for n in visibletostrip[:maxnodes]:
                 ui.status(b'%s\n' % short(n))
             ui.status(
                 _(b'...and %d more, use --verbose to list all\n')
                 % (len(visibletostrip) - maxnodes)
             )
         if not force:
             raise error.StateError(
                 _(b'local changes found'),
                 hint=_(b'use --force-delete-local-changes to ignore'),
             )
 
     with ui.uninterruptible():
         if revstostrip:
             tostrip = [unfi.changelog.node(r) for r in revstostrip]
             if repo[b'.'].node() in tostrip:
                 # stripping working copy, so move to a different commit first
                 urev = max(
                     repo.revs(
                         b'(::%n) - %ln + null',
                         repo[b'.'].node(),
                         visibletostrip,
                     )
                 )
                 hg.clean(repo, urev)
             overrides = {(b'devel', b'strip-obsmarkers'): False}
             if backup:
                 ui.status(_(b'moving unwanted changesets to backup\n'))
             else:
                 ui.status(_(b'deleting unwanted changesets\n'))
             with ui.configoverride(overrides, b'narrow'):
                 repair.strip(ui, unfi, tostrip, topic=b'narrow', backup=backup)
 
         todelete = []
-        for t, f, f2, size in repo.store.datafiles():
+        for t, f, size in repo.store.datafiles():
             if f.startswith(b'data/'):
                 file = f[5:-2]
                 if not newmatch(file):
                     todelete.append(f)
             elif f.startswith(b'meta/'):
                 dir = f[5:-13]
                 dirs = sorted(pathutil.dirs({dir})) + [dir]
                 include = True
                 for d in dirs:
                     visit = newmatch.visitdir(d)
                     if not visit:
                         include = False
                         break
                     if visit == b'all':
                         break
                 if not include:
                     todelete.append(f)
 
         repo.destroying()
 
         with repo.transaction(b'narrowing'):
             # Update narrowspec before removing revlogs, so repo won't be
             # corrupt in case of crash
             repo.setnarrowpats(newincludes, newexcludes)
 
             for f in todelete:
                 ui.status(_(b'deleting %s\n') % f)
                 util.unlinkpath(repo.svfs.join(f))
                 repo.store.markremoved(f)
 
             ui.status(_(b'deleting unwanted files from working copy\n'))
             with repo.dirstate.parentchange():
                 narrowspec.updateworkingcopy(repo, assumeclean=True)
                 narrowspec.copytoworkingcopy(repo)
 
         repo.destroyed()
 
 
 def _widen(
     ui,
     repo,
     remote,
     commoninc,
     oldincludes,
     oldexcludes,
     newincludes,
     newexcludes,
 ):
     # for now we assume that if a server has ellipses enabled, we will be
     # exchanging ellipses nodes. In future we should add ellipses as a client
     # side requirement (maybe) to distinguish a client is shallow or not and
     # then send that information to server whether we want ellipses or not.
     # Theoretically a non-ellipses repo should be able to use narrow
     # functionality from an ellipses enabled server
     remotecap = remote.capabilities()
     ellipsesremote = any(
         cap in remotecap for cap in wireprototypes.SUPPORTED_ELLIPSESCAP
     )
 
     # check whether we are talking to a server which supports old version of
     # ellipses capabilities
     isoldellipses = (
         ellipsesremote
         and wireprototypes.ELLIPSESCAP1 in remotecap
         and wireprototypes.ELLIPSESCAP not in remotecap
     )
 
     def pullbundle2extraprepare_widen(orig, pullop, kwargs):
         orig(pullop, kwargs)
         # The old{in,ex}cludepats have already been set by orig()
         kwargs[b'includepats'] = newincludes
         kwargs[b'excludepats'] = newexcludes
 
     wrappedextraprepare = extensions.wrappedfunction(
         exchange, b'_pullbundle2extraprepare', pullbundle2extraprepare_widen
     )
 
     # define a function that narrowbundle2 can call after creating the
     # backup bundle, but before applying the bundle from the server
     def setnewnarrowpats():
         repo.setnarrowpats(newincludes, newexcludes)
 
     repo.setnewnarrowpats = setnewnarrowpats
     # silence the devel-warning of applying an empty changegroup
     overrides = {(b'devel', b'all-warnings'): False}
 
     common = commoninc[0]
     with ui.uninterruptible():
         if ellipsesremote:
             ds = repo.dirstate
             p1, p2 = ds.p1(), ds.p2()
             with ds.parentchange():
                 ds.setparents(repo.nullid, repo.nullid)
         if isoldellipses:
             with wrappedextraprepare:
                 exchange.pull(repo, remote, heads=common)
         else:
             known = []
             if ellipsesremote:
                 known = [
                     ctx.node()
                     for ctx in repo.set(b'::%ln', common)
                     if ctx.node() != repo.nullid
                 ]
             with remote.commandexecutor() as e:
                 bundle = e.callcommand(
                     b'narrow_widen',
                     {
                         b'oldincludes': oldincludes,
                         b'oldexcludes': oldexcludes,
                         b'newincludes': newincludes,
                         b'newexcludes': newexcludes,
                         b'cgversion': b'03',
                         b'commonheads': common,
                         b'known': known,
                         b'ellipses': ellipsesremote,
                     },
                 ).result()
 
             trmanager = exchange.transactionmanager(
                 repo, b'widen', remote.url()
             )
             with trmanager, repo.ui.configoverride(overrides, b'widen'):
                 op = bundle2.bundleoperation(
                     repo, trmanager.transaction, source=b'widen'
                 )
                 # TODO: we should catch error.Abort here
                 bundle2.processbundle(repo, bundle, op=op)
 
         if ellipsesremote:
             with ds.parentchange():
                 ds.setparents(p1, p2)
 
         with repo.transaction(b'widening'), repo.dirstate.parentchange():
             repo.setnewnarrowpats()
             narrowspec.updateworkingcopy(repo)
             narrowspec.copytoworkingcopy(repo)
 
 
 # TODO(rdamazio): Make new matcher format and update description
 @command(
     b'tracked',
     [
         (b'', b'addinclude', [], _(b'new paths to include')),
         (b'', b'removeinclude', [], _(b'old paths to no longer include')),
         (
             b'',
             b'auto-remove-includes',
             False,
             _(b'automatically choose unused includes to remove'),
         ),
         (b'', b'addexclude', [], _(b'new paths to exclude')),
         (b'', b'import-rules', b'', _(b'import narrowspecs from a file')),
         (b'', b'removeexclude', [], _(b'old paths to no longer exclude')),
         (
             b'',
             b'clear',
             False,
             _(b'whether to replace the existing narrowspec'),
         ),
         (
             b'',
             b'force-delete-local-changes',
             False,
             _(b'forces deletion of local changes when narrowing'),
         ),
         (
             b'',
             b'backup',
             True,
             _(b'back up local changes when narrowing'),
         ),
         (
             b'',
             b'update-working-copy',
             False,
             _(b'update working copy when the store has changed'),
         ),
     ]
     + commands.remoteopts,
     _(b'[OPTIONS]... [REMOTE]'),
     inferrepo=True,
     helpcategory=command.CATEGORY_MAINTENANCE,
 )
 def trackedcmd(ui, repo, remotepath=None, *pats, **opts):
     """show or change the current narrowspec
 
     With no argument, shows the current narrowspec entries, one per line. Each
     line will be prefixed with 'I' or 'X' for included or excluded patterns,
     respectively.
 
     The narrowspec is comprised of expressions to match remote files and/or
     directories that should be pulled into your client.
     The narrowspec has *include* and *exclude* expressions, with excludes always
     trumping includes: that is, if a file matches an exclude expression, it will
     be excluded even if it also matches an include expression.
     Excluding files that were never included has no effect.
 
     Each included or excluded entry is in the format described by
     'hg help patterns'.
 
     The options allow you to add or remove included and excluded expressions.
 
     If --clear is specified, then all previous includes and excludes are DROPPED
     and replaced by the new ones specified to --addinclude and --addexclude.
     If --clear is specified without any further options, the narrowspec will be
     empty and will not match any files.
 
     If --auto-remove-includes is specified, then those includes that don't match
     any files modified by currently visible local commits (those not shared by
     the remote) will be added to the set of explicitly specified includes to
     remove.
 
     --import-rules accepts a path to a file containing rules, allowing you to
     add --addinclude, --addexclude rules in bulk. Like the other include and
     exclude switches, the changes are applied immediately.
     """
     opts = pycompat.byteskwargs(opts)
     if requirements.NARROW_REQUIREMENT not in repo.requirements:
         raise error.InputError(
             _(
                 b'the tracked command is only supported on '
                 b'repositories cloned with --narrow'
             )
         )
 
     # Before supporting, decide whether it "hg tracked --clear" should mean
     # tracking no paths or all paths.
     if opts[b'clear']:
         raise error.InputError(_(b'the --clear option is not yet supported'))
 
     # import rules from a file
     newrules = opts.get(b'import_rules')
     if newrules:
         try:
             filepath = os.path.join(encoding.getcwd(), newrules)
             fdata = util.readfile(filepath)
         except IOError as inst:
             raise error.StorageError(
                 _(b"cannot read narrowspecs from '%s': %s")
                 % (filepath, encoding.strtolocal(inst.strerror))
             )
         includepats, excludepats, profiles = sparse.parseconfig(
             ui, fdata, b'narrow'
         )
         if profiles:
             raise error.InputError(
                 _(
                     b"including other spec files using '%include' "
                     b"is not supported in narrowspec"
                 )
             )
         opts[b'addinclude'].extend(includepats)
         opts[b'addexclude'].extend(excludepats)
 
     addedincludes = narrowspec.parsepatterns(opts[b'addinclude'])
     removedincludes = narrowspec.parsepatterns(opts[b'removeinclude'])
     addedexcludes = narrowspec.parsepatterns(opts[b'addexclude'])
     removedexcludes = narrowspec.parsepatterns(opts[b'removeexclude'])
     autoremoveincludes = opts[b'auto_remove_includes']
 
     update_working_copy = opts[b'update_working_copy']
     only_show = not (
         addedincludes
         or removedincludes
         or addedexcludes
         or removedexcludes
         or newrules
         or autoremoveincludes
         or update_working_copy
     )
 
     oldincludes, oldexcludes = repo.narrowpats
 
     # filter the user passed additions and deletions into actual additions and
     # deletions of excludes and includes
     addedincludes -= oldincludes
     removedincludes &= oldincludes
     addedexcludes -= oldexcludes
     removedexcludes &= oldexcludes
 
     widening = addedincludes or removedexcludes
     narrowing = removedincludes or addedexcludes
 
     # Only print the current narrowspec.
     if only_show:
         ui.pager(b'tracked')
         fm = ui.formatter(b'narrow', opts)
         for i in sorted(oldincludes):
             fm.startitem()
             fm.write(b'status', b'%s ', b'I', label=b'narrow.included')
             fm.write(b'pat', b'%s\n', i, label=b'narrow.included')
         for i in sorted(oldexcludes):
             fm.startitem()
             fm.write(b'status', b'%s ', b'X', label=b'narrow.excluded')
             fm.write(b'pat', b'%s\n', i, label=b'narrow.excluded')
         fm.end()
         return 0
 
     if update_working_copy:
         with repo.wlock(), repo.lock(), repo.transaction(
             b'narrow-wc'
         ), repo.dirstate.parentchange():
             narrowspec.updateworkingcopy(repo)
             narrowspec.copytoworkingcopy(repo)
         return 0
 
     if not (widening or narrowing or autoremoveincludes):
         ui.status(_(b"nothing to widen or narrow\n"))
         return 0
 
     with repo.wlock(), repo.lock():
         cmdutil.bailifchanged(repo)
 
         # Find the revisions we have in common with the remote. These will
         # be used for finding local-only changes for narrowing. They will
         # also define the set of revisions to update for widening.
         r = urlutil.get_unique_pull_path(b'tracked', repo, ui, remotepath)
         url, branches = r
         ui.status(_(b'comparing with %s\n') % urlutil.hidepassword(url))
         remote = hg.peer(repo, opts, url)
 
         try:
             # check narrow support before doing anything if widening needs to be
             # performed. In future we should also abort if client is ellipses and
             # server does not support ellipses
             if (
                 widening
                 and wireprototypes.NARROWCAP not in remote.capabilities()
             ):
                 raise error.Abort(_(b"server does not support narrow clones"))
 
             commoninc = discovery.findcommonincoming(repo, remote)
 
             if autoremoveincludes:
                 outgoing = discovery.findcommonoutgoing(
                     repo, remote, commoninc=commoninc
                 )
                 ui.status(_(b'looking for unused includes to remove\n'))
                 localfiles = set()
                 for n in itertools.chain(outgoing.missing, outgoing.excluded):
                     localfiles.update(repo[n].files())
                 suggestedremovals = []
                 for include in sorted(oldincludes):
                     match = narrowspec.match(repo.root, [include], oldexcludes)
                     if not any(match(f) for f in localfiles):
                         suggestedremovals.append(include)
                 if suggestedremovals:
                     for s in suggestedremovals:
                         ui.status(b'%s\n' % s)
                     if (
                         ui.promptchoice(
                             _(
                                 b'remove these unused includes (yn)?'
                                 b'$$ &Yes $$ &No'
                             )
                         )
                         == 0
                     ):
                         removedincludes.update(suggestedremovals)
                         narrowing = True
                 else:
                     ui.status(_(b'found no unused includes\n'))
 
             if narrowing:
                 newincludes = oldincludes - removedincludes
                 newexcludes = oldexcludes | addedexcludes
                 _narrow(
                     ui,
                     repo,
                     remote,
                     commoninc,
                     oldincludes,
                     oldexcludes,
                     newincludes,
                     newexcludes,
                     opts[b'force_delete_local_changes'],
                     opts[b'backup'],
                 )
                 # _narrow() updated the narrowspec and _widen() below needs to
                 # use the updated values as its base (otherwise removed includes
                 # and addedexcludes will be lost in the resulting narrowspec)
                 oldincludes = newincludes
                 oldexcludes = newexcludes
 
             if widening:
                 newincludes = oldincludes | addedincludes
                 newexcludes = oldexcludes - removedexcludes
                 _widen(
                     ui,
                     repo,
                     remote,
                     commoninc,
                     oldincludes,
                     oldexcludes,
                     newincludes,
                     newexcludes,
                 )
         finally:
             remote.close()
 
     return 0

hgext/remotefilelog/contentstore.py

@@ -1,399 +1,399 b''
 from __future__ import absolute_import
 
 import threading
 
 from mercurial.node import (
     hex,
     sha1nodeconstants,
 )
 from mercurial.pycompat import getattr
 from mercurial import (
     mdiff,
     pycompat,
     revlog,
 )
 from . import (
     basestore,
     constants,
     shallowutil,
 )
 
 
 class ChainIndicies(object):
     """A static class for easy reference to the delta chain indicies."""
 
     # The filename of this revision delta
     NAME = 0
     # The mercurial file node for this revision delta
     NODE = 1
     # The filename of the delta base's revision. This is useful when delta
     # between different files (like in the case of a move or copy, we can delta
     # against the original file content).
     BASENAME = 2
     # The mercurial file node for the delta base revision. This is the nullid if
     # this delta is a full text.
     BASENODE = 3
     # The actual delta or full text data.
     DATA = 4
 
 
 class unioncontentstore(basestore.baseunionstore):
     def __init__(self, *args, **kwargs):
         super(unioncontentstore, self).__init__(*args, **kwargs)
 
         self.stores = args
         self.writestore = kwargs.get('writestore')
 
         # If allowincomplete==True then the union store can return partial
         # delta chains, otherwise it will throw a KeyError if a full
         # deltachain can't be found.
         self.allowincomplete = kwargs.get('allowincomplete', False)
 
     def get(self, name, node):
         """Fetches the full text revision contents of the given name+node pair.
         If the full text doesn't exist, throws a KeyError.
 
         Under the hood, this uses getdeltachain() across all the stores to build
         up a full chain to produce the full text.
         """
         chain = self.getdeltachain(name, node)
 
         if chain[-1][ChainIndicies.BASENODE] != sha1nodeconstants.nullid:
             # If we didn't receive a full chain, throw
             raise KeyError((name, hex(node)))
 
         # The last entry in the chain is a full text, so we start our delta
         # applies with that.
         fulltext = chain.pop()[ChainIndicies.DATA]
 
         text = fulltext
         while chain:
             delta = chain.pop()[ChainIndicies.DATA]
             text = mdiff.patches(text, [delta])
 
         return text
 
     @basestore.baseunionstore.retriable
     def getdelta(self, name, node):
         """Return the single delta entry for the given name/node pair."""
         for store in self.stores:
             try:
                 return store.getdelta(name, node)
             except KeyError:
                 pass
 
         raise KeyError((name, hex(node)))
 
     def getdeltachain(self, name, node):
         """Returns the deltachain for the given name/node pair.
 
         Returns an ordered list of:
 
           [(name, node, deltabasename, deltabasenode, deltacontent),...]
 
         where the chain is terminated by a full text entry with a nullid
         deltabasenode.
         """
         chain = self._getpartialchain(name, node)
         while chain[-1][ChainIndicies.BASENODE] != sha1nodeconstants.nullid:
             x, x, deltabasename, deltabasenode, x = chain[-1]
             try:
                 morechain = self._getpartialchain(deltabasename, deltabasenode)
                 chain.extend(morechain)
             except KeyError:
                 # If we allow incomplete chains, don't throw.
                 if not self.allowincomplete:
                     raise
                 break
 
         return chain
 
     @basestore.baseunionstore.retriable
     def getmeta(self, name, node):
         """Returns the metadata dict for given node."""
         for store in self.stores:
             try:
                 return store.getmeta(name, node)
             except KeyError:
                 pass
         raise KeyError((name, hex(node)))
 
     def getmetrics(self):
         metrics = [s.getmetrics() for s in self.stores]
         return shallowutil.sumdicts(*metrics)
 
     @basestore.baseunionstore.retriable
     def _getpartialchain(self, name, node):
         """Returns a partial delta chain for the given name/node pair.
 
         A partial chain is a chain that may not be terminated in a full-text.
         """
         for store in self.stores:
             try:
                 return store.getdeltachain(name, node)
             except KeyError:
                 pass
 
         raise KeyError((name, hex(node)))
 
     def add(self, name, node, data):
         raise RuntimeError(
             b"cannot add content only to remotefilelog contentstore"
         )
 
     def getmissing(self, keys):
         missing = keys
         for store in self.stores:
             if missing:
                 missing = store.getmissing(missing)
         return missing
 
     def addremotefilelognode(self, name, node, data):
         if self.writestore:
             self.writestore.addremotefilelognode(name, node, data)
         else:
             raise RuntimeError(b"no writable store configured")
 
     def markledger(self, ledger, options=None):
         for store in self.stores:
             store.markledger(ledger, options)
 
 
 class remotefilelogcontentstore(basestore.basestore):
     def __init__(self, *args, **kwargs):
         super(remotefilelogcontentstore, self).__init__(*args, **kwargs)
         self._threaddata = threading.local()
 
     def get(self, name, node):
         # return raw revision text
         data = self._getdata(name, node)
 
         offset, size, flags = shallowutil.parsesizeflags(data)
         content = data[offset : offset + size]
 
         ancestormap = shallowutil.ancestormap(data)
         p1, p2, linknode, copyfrom = ancestormap[node]
         copyrev = None
         if copyfrom:
             copyrev = hex(p1)
 
         self._updatemetacache(node, size, flags)
 
         # lfs tracks renames in its own metadata, remove hg copy metadata,
         # because copy metadata will be re-added by lfs flag processor.
         if flags & revlog.REVIDX_EXTSTORED:
             copyrev = copyfrom = None
         revision = shallowutil.createrevlogtext(content, copyfrom, copyrev)
         return revision
 
     def getdelta(self, name, node):
         # Since remotefilelog content stores only contain full texts, just
         # return that.
         revision = self.get(name, node)
         return (
             revision,
             name,
             sha1nodeconstants.nullid,
             self.getmeta(name, node),
         )
 
     def getdeltachain(self, name, node):
         # Since remotefilelog content stores just contain full texts, we return
         # a fake delta chain that just consists of a single full text revision.
         # The nullid in the deltabasenode slot indicates that the revision is a
         # fulltext.
         revision = self.get(name, node)
         return [(name, node, None, sha1nodeconstants.nullid, revision)]
 
     def getmeta(self, name, node):
         self._sanitizemetacache()
         if node != self._threaddata.metacache[0]:
             data = self._getdata(name, node)
             offset, size, flags = shallowutil.parsesizeflags(data)
             self._updatemetacache(node, size, flags)
         return self._threaddata.metacache[1]
 
     def add(self, name, node, data):
         raise RuntimeError(
             b"cannot add content only to remotefilelog contentstore"
         )
 
     def _sanitizemetacache(self):
         metacache = getattr(self._threaddata, 'metacache', None)
         if metacache is None:
             self._threaddata.metacache = (None, None)  # (node, meta)
 
     def _updatemetacache(self, node, size, flags):
         self._sanitizemetacache()
         if node == self._threaddata.metacache[0]:
             return
         meta = {constants.METAKEYFLAG: flags, constants.METAKEYSIZE: size}
         self._threaddata.metacache = (node, meta)
 
 
 class remotecontentstore(object):
     def __init__(self, ui, fileservice, shared):
         self._fileservice = fileservice
         # type(shared) is usually remotefilelogcontentstore
         self._shared = shared
 
     def get(self, name, node):
         self._fileservice.prefetch(
             [(name, hex(node))], force=True, fetchdata=True
         )
         return self._shared.get(name, node)
 
     def getdelta(self, name, node):
         revision = self.get(name, node)
         return (
             revision,
             name,
             sha1nodeconstants.nullid,
             self._shared.getmeta(name, node),
         )
 
     def getdeltachain(self, name, node):
         # Since our remote content stores just contain full texts, we return a
         # fake delta chain that just consists of a single full text revision.
         # The nullid in the deltabasenode slot indicates that the revision is a
         # fulltext.
         revision = self.get(name, node)
         return [(name, node, None, sha1nodeconstants.nullid, revision)]
 
     def getmeta(self, name, node):
         self._fileservice.prefetch(
             [(name, hex(node))], force=True, fetchdata=True
         )
         return self._shared.getmeta(name, node)
 
     def add(self, name, node, data):
         raise RuntimeError(b"cannot add to a remote store")
 
     def getmissing(self, keys):
         return keys
 
     def markledger(self, ledger, options=None):
         pass
 
 
 class manifestrevlogstore(object):
     def __init__(self, repo):
         self._store = repo.store
         self._svfs = repo.svfs
         self._revlogs = dict()
         self._cl = revlog.revlog(self._svfs, radix=b'00changelog.i')
         self._repackstartlinkrev = 0
 
     def get(self, name, node):
         return self._revlog(name).rawdata(node)
 
     def getdelta(self, name, node):
         revision = self.get(name, node)
         return revision, name, self._cl.nullid, self.getmeta(name, node)
 
     def getdeltachain(self, name, node):
         revision = self.get(name, node)
         return [(name, node, None, self._cl.nullid, revision)]
 
     def getmeta(self, name, node):
         rl = self._revlog(name)
         rev = rl.rev(node)
         return {
             constants.METAKEYFLAG: rl.flags(rev),
             constants.METAKEYSIZE: rl.rawsize(rev),
         }
 
     def getancestors(self, name, node, known=None):
         if known is None:
             known = set()
         if node in known:
             return []
 
         rl = self._revlog(name)
         ancestors = {}
         missing = {node}
         for ancrev in rl.ancestors([rl.rev(node)], inclusive=True):
             ancnode = rl.node(ancrev)
             missing.discard(ancnode)
 
             p1, p2 = rl.parents(ancnode)
             if p1 != self._cl.nullid and p1 not in known:
                 missing.add(p1)
             if p2 != self._cl.nullid and p2 not in known:
                 missing.add(p2)
 
             linknode = self._cl.node(rl.linkrev(ancrev))
             ancestors[rl.node(ancrev)] = (p1, p2, linknode, b'')
             if not missing:
                 break
         return ancestors
 
     def getnodeinfo(self, name, node):
         cl = self._cl
         rl = self._revlog(name)
         parents = rl.parents(node)
         linkrev = rl.linkrev(rl.rev(node))
         return (parents[0], parents[1], cl.node(linkrev), None)
 
     def add(self, *args):
         raise RuntimeError(b"cannot add to a revlog store")
 
     def _revlog(self, name):
         rl = self._revlogs.get(name)
         if rl is None:
             revlogname = b'00manifesttree'
             if name != b'':
                 revlogname = b'meta/%s/00manifest' % name
             rl = revlog.revlog(self._svfs, radix=revlogname)
             self._revlogs[name] = rl
         return rl
 
     def getmissing(self, keys):
         missing = []
         for name, node in keys:
             mfrevlog = self._revlog(name)
             if node not in mfrevlog.nodemap:
                 missing.append((name, node))
 
         return missing
 
     def setrepacklinkrevrange(self, startrev, endrev):
         self._repackstartlinkrev = startrev
         self._repackendlinkrev = endrev
 
     def markledger(self, ledger, options=None):
         if options and options.get(constants.OPTION_PACKSONLY):
             return
         treename = b''
         rl = revlog.revlog(self._svfs, radix=b'00manifesttree')
         startlinkrev = self._repackstartlinkrev
         endlinkrev = self._repackendlinkrev
         for rev in pycompat.xrange(len(rl) - 1, -1, -1):
             linkrev = rl.linkrev(rev)
             if linkrev < startlinkrev:
                 break
             if linkrev > endlinkrev:
                 continue
             node = rl.node(rev)
             ledger.markdataentry(self, treename, node)
             ledger.markhistoryentry(self, treename, node)
 
-        for t, path, encoded, size in self._store.datafiles():
+        for t, path, size in self._store.datafiles():
             if path[:5] != b'meta/' or path[-2:] != b'.i':
                 continue
 
             treename = path[5 : -len(b'/00manifest')]
 
             rl = revlog.revlog(self._svfs, indexfile=path[:-2])
             for rev in pycompat.xrange(len(rl) - 1, -1, -1):
                 linkrev = rl.linkrev(rev)
                 if linkrev < startlinkrev:
                     break
                 if linkrev > endlinkrev:
                     continue
                 node = rl.node(rev)
                 ledger.markdataentry(self, treename, node)
                 ledger.markhistoryentry(self, treename, node)
 
     def cleanup(self, ledger):
         pass

hgext/remotefilelog/remotefilelogserver.py

@@ -1,441 +1,441 b''
 # remotefilelogserver.py - server logic for a remotefilelog server
 #
 # 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.
 from __future__ import absolute_import
 
 import errno
 import os
 import stat
 import time
 import zlib
 
 from mercurial.i18n import _
 from mercurial.node import bin, hex
 from mercurial.pycompat import open
 from mercurial import (
     changegroup,
     changelog,
     context,
     error,
     extensions,
     match,
     pycompat,
     scmutil,
     store,
     streamclone,
     util,
     wireprotoserver,
     wireprototypes,
     wireprotov1server,
 )
 from . import (
     constants,
     shallowutil,
 )
 
 _sshv1server = wireprotoserver.sshv1protocolhandler
 
 
 def setupserver(ui, repo):
     """Sets up a normal Mercurial repo so it can serve files to shallow repos."""
     onetimesetup(ui)
 
     # don't send files to shallow clients during pulls
     def generatefiles(
         orig, self, changedfiles, linknodes, commonrevs, source, *args, **kwargs
     ):
         caps = self._bundlecaps or []
         if constants.BUNDLE2_CAPABLITY in caps:
             # only send files that don't match the specified patterns
             includepattern = None
             excludepattern = None
             for cap in self._bundlecaps or []:
                 if cap.startswith(b"includepattern="):
                     includepattern = cap[len(b"includepattern=") :].split(b'\0')
                 elif cap.startswith(b"excludepattern="):
                     excludepattern = cap[len(b"excludepattern=") :].split(b'\0')
 
             m = match.always()
             if includepattern or excludepattern:
                 m = match.match(
                     repo.root, b'', None, includepattern, excludepattern
                 )
 
             changedfiles = list([f for f in changedfiles if not m(f)])
         return orig(
             self, changedfiles, linknodes, commonrevs, source, *args, **kwargs
         )
 
     extensions.wrapfunction(
         changegroup.cgpacker, b'generatefiles', generatefiles
     )
 
 
 onetime = False
 
 
 def onetimesetup(ui):
     """Configures the wireprotocol for both clients and servers."""
     global onetime
     if onetime:
         return
     onetime = True
 
     # support file content requests
     wireprotov1server.wireprotocommand(
         b'x_rfl_getflogheads', b'path', permission=b'pull'
     )(getflogheads)
     wireprotov1server.wireprotocommand(
         b'x_rfl_getfiles', b'', permission=b'pull'
     )(getfiles)
     wireprotov1server.wireprotocommand(
         b'x_rfl_getfile', b'file node', permission=b'pull'
     )(getfile)
 
     class streamstate(object):
         match = None
         shallowremote = False
         noflatmf = False
 
     state = streamstate()
 
     def stream_out_shallow(repo, proto, other):
         includepattern = None
         excludepattern = None
         raw = other.get(b'includepattern')
         if raw:
             includepattern = raw.split(b'\0')
         raw = other.get(b'excludepattern')
         if raw:
             excludepattern = raw.split(b'\0')
 
         oldshallow = state.shallowremote
         oldmatch = state.match
         oldnoflatmf = state.noflatmf
         try:
             state.shallowremote = True
             state.match = match.always()
             state.noflatmf = other.get(b'noflatmanifest') == b'True'
             if includepattern or excludepattern:
                 state.match = match.match(
                     repo.root, b'', None, includepattern, excludepattern
                 )
             streamres = wireprotov1server.stream(repo, proto)
 
             # Force the first value to execute, so the file list is computed
             # within the try/finally scope
             first = next(streamres.gen)
             second = next(streamres.gen)
 
             def gen():
                 yield first
                 yield second
                 for value in streamres.gen:
                     yield value
 
             return wireprototypes.streamres(gen())
         finally:
             state.shallowremote = oldshallow
             state.match = oldmatch
             state.noflatmf = oldnoflatmf
 
     wireprotov1server.commands[b'stream_out_shallow'] = (
         stream_out_shallow,
         b'*',
     )
 
     # don't clone filelogs to shallow clients
     def _walkstreamfiles(orig, repo, matcher=None):
         if state.shallowremote:
             # if we are shallow ourselves, stream our local commits
             if shallowutil.isenabled(repo):
                 striplen = len(repo.store.path) + 1
                 readdir = repo.store.rawvfs.readdir
                 visit = [os.path.join(repo.store.path, b'data')]
                 while visit:
                     p = visit.pop()
                     for f, kind, st in readdir(p, stat=True):
                         fp = p + b'/' + f
                         if kind == stat.S_IFREG:
                             if not fp.endswith(b'.i') and not fp.endswith(
                                 b'.d'
                             ):
                                 n = util.pconvert(fp[striplen:])
                                 d = store.decodedir(n)
                                 t = store.FILETYPE_OTHER
-                                yield (t, d, n, st.st_size)
+                                yield (t, d, st.st_size)
                         if kind == stat.S_IFDIR:
                             visit.append(fp)
 
             if scmutil.istreemanifest(repo):
-                for (t, u, e, s) in repo.store.datafiles():
+                for (t, u, s) in repo.store.datafiles():
                     if u.startswith(b'meta/') and (
                         u.endswith(b'.i') or u.endswith(b'.d')
                     ):
-                        yield (t, u, e, s)
+                        yield (t, u, s)
 
             # Return .d and .i files that do not match the shallow pattern
             match = state.match
             if match and not match.always():
-                for (t, u, e, s) in repo.store.datafiles():
+                for (t, u, s) in repo.store.datafiles():
                     f = u[5:-2]  # trim data/...  and .i/.d
                     if not state.match(f):
-                        yield (t, u, e, s)
+                        yield (t, u, s)
 
             for x in repo.store.topfiles():
                 if state.noflatmf and x[1][:11] == b'00manifest.':
                     continue
                 yield x
 
         elif shallowutil.isenabled(repo):
             # don't allow cloning from a shallow repo to a full repo
             # since it would require fetching every version of every
             # file in order to create the revlogs.
             raise error.Abort(
                 _(b"Cannot clone from a shallow repo to a full repo.")
             )
         else:
             for x in orig(repo, matcher):
                 yield x
 
     extensions.wrapfunction(streamclone, b'_walkstreamfiles', _walkstreamfiles)
 
     # expose remotefilelog capabilities
     def _capabilities(orig, repo, proto):
         caps = orig(repo, proto)
         if shallowutil.isenabled(repo) or ui.configbool(
             b'remotefilelog', b'server'
         ):
             if isinstance(proto, _sshv1server):
                 # legacy getfiles method which only works over ssh
                 caps.append(constants.NETWORK_CAP_LEGACY_SSH_GETFILES)
             caps.append(b'x_rfl_getflogheads')
             caps.append(b'x_rfl_getfile')
         return caps
 
     extensions.wrapfunction(wireprotov1server, b'_capabilities', _capabilities)
 
     def _adjustlinkrev(orig, self, *args, **kwargs):
         # When generating file blobs, taking the real path is too slow on large
         # repos, so force it to just return the linkrev directly.
         repo = self._repo
         if util.safehasattr(repo, b'forcelinkrev') and repo.forcelinkrev:
             return self._filelog.linkrev(self._filelog.rev(self._filenode))
         return orig(self, *args, **kwargs)
 
     extensions.wrapfunction(
         context.basefilectx, b'_adjustlinkrev', _adjustlinkrev
     )
 
     def _iscmd(orig, cmd):
         if cmd == b'x_rfl_getfiles':
             return False
         return orig(cmd)
 
     extensions.wrapfunction(wireprotoserver, b'iscmd', _iscmd)
 
 
 def _loadfileblob(repo, cachepath, path, node):
     filecachepath = os.path.join(cachepath, path, hex(node))
     if not os.path.exists(filecachepath) or os.path.getsize(filecachepath) == 0:
         filectx = repo.filectx(path, fileid=node)
         if filectx.node() == repo.nullid:
             repo.changelog = changelog.changelog(repo.svfs)
             filectx = repo.filectx(path, fileid=node)
 
         text = createfileblob(filectx)
         # TODO configurable compression engines
         text = zlib.compress(text)
 
         # everything should be user & group read/writable
         oldumask = os.umask(0o002)
         try:
             dirname = os.path.dirname(filecachepath)
             if not os.path.exists(dirname):
                 try:
                     os.makedirs(dirname)
                 except OSError as ex:
                     if ex.errno != errno.EEXIST:
                         raise
 
             f = None
             try:
                 f = util.atomictempfile(filecachepath, b"wb")
                 f.write(text)
             except (IOError, OSError):
                 # Don't abort if the user only has permission to read,
                 # and not write.
                 pass
             finally:
                 if f:
                     f.close()
         finally:
             os.umask(oldumask)
     else:
         with open(filecachepath, b"rb") as f:
             text = f.read()
     return text
 
 
 def getflogheads(repo, proto, path):
     """A server api for requesting a filelog's heads"""
     flog = repo.file(path)
     heads = flog.heads()
     return b'\n'.join((hex(head) for head in heads if head != repo.nullid))
 
 
 def getfile(repo, proto, file, node):
     """A server api for requesting a particular version of a file. Can be used
     in batches to request many files at once. The return protocol is:
     <errorcode>\0<data/errormsg> where <errorcode> is 0 for success or
     non-zero for an error.
 
     data is a compressed blob with revlog flag and ancestors information. See
     createfileblob for its content.
     """
     if shallowutil.isenabled(repo):
         return b'1\0' + _(b'cannot fetch remote files from shallow repo')
     cachepath = repo.ui.config(b"remotefilelog", b"servercachepath")
     if not cachepath:
         cachepath = os.path.join(repo.path, b"remotefilelogcache")
     node = bin(node.strip())
     if node == repo.nullid:
         return b'0\0'
     return b'0\0' + _loadfileblob(repo, cachepath, file, node)
 
 
 def getfiles(repo, proto):
     """A server api for requesting particular versions of particular files."""
     if shallowutil.isenabled(repo):
         raise error.Abort(_(b'cannot fetch remote files from shallow repo'))
     if not isinstance(proto, _sshv1server):
         raise error.Abort(_(b'cannot fetch remote files over non-ssh protocol'))
 
     def streamer():
         fin = proto._fin
 
         cachepath = repo.ui.config(b"remotefilelog", b"servercachepath")
         if not cachepath:
             cachepath = os.path.join(repo.path, b"remotefilelogcache")
 
         while True:
             request = fin.readline()[:-1]
             if not request:
                 break
 
             node = bin(request[:40])
             if node == repo.nullid:
                 yield b'0\n'
                 continue
 
             path = request[40:]
 
             text = _loadfileblob(repo, cachepath, path, node)
 
             yield b'%d\n%s' % (len(text), text)
 
             # it would be better to only flush after processing a whole batch
             # but currently we don't know if there are more requests coming
             proto._fout.flush()
 
     return wireprototypes.streamres(streamer())
 
 
 def createfileblob(filectx):
     """
     format:
         v0:
             str(len(rawtext)) + '\0' + rawtext + ancestortext
         v1:
             'v1' + '\n' + metalist + '\0' + rawtext + ancestortext
             metalist := metalist + '\n' + meta | meta
             meta := sizemeta | flagmeta
             sizemeta := METAKEYSIZE + str(len(rawtext))
             flagmeta := METAKEYFLAG + str(flag)
 
             note: sizemeta must exist. METAKEYFLAG and METAKEYSIZE must have a
             length of 1.
     """
     flog = filectx.filelog()
     frev = filectx.filerev()
     revlogflags = flog._revlog.flags(frev)
     if revlogflags == 0:
         # normal files
         text = filectx.data()
     else:
         # lfs, read raw revision data
         text = flog.rawdata(frev)
 
     repo = filectx._repo
 
     ancestors = [filectx]
 
     try:
         repo.forcelinkrev = True
         ancestors.extend([f for f in filectx.ancestors()])
 
         ancestortext = b""
         for ancestorctx in ancestors:
             parents = ancestorctx.parents()
             p1 = repo.nullid
             p2 = repo.nullid
             if len(parents) > 0:
                 p1 = parents[0].filenode()
             if len(parents) > 1:
                 p2 = parents[1].filenode()
 
             copyname = b""
             rename = ancestorctx.renamed()
             if rename:
                 copyname = rename[0]
             linknode = ancestorctx.node()
             ancestortext += b"%s%s%s%s%s\0" % (
                 ancestorctx.filenode(),
                 p1,
                 p2,
                 linknode,
                 copyname,
             )
     finally:
         repo.forcelinkrev = False
 
     header = shallowutil.buildfileblobheader(len(text), revlogflags)
 
     return b"%s\0%s%s" % (header, text, ancestortext)
 
 
 def gcserver(ui, repo):
     if not repo.ui.configbool(b"remotefilelog", b"server"):
         return
 
     neededfiles = set()
     heads = repo.revs(b"heads(tip~25000:) - null")
 
     cachepath = repo.vfs.join(b"remotefilelogcache")
     for head in heads:
         mf = repo[head].manifest()
         for filename, filenode in pycompat.iteritems(mf):
             filecachepath = os.path.join(cachepath, filename, hex(filenode))
             neededfiles.add(filecachepath)
 
     # delete unneeded older files
     days = repo.ui.configint(b"remotefilelog", b"serverexpiration")
     expiration = time.time() - (days * 24 * 60 * 60)
 
     progress = ui.makeprogress(_(b"removing old server cache"), unit=b"files")
     progress.update(0)
     for root, dirs, files in os.walk(cachepath):
         for file in files:
             filepath = os.path.join(root, file)
             progress.increment()
             if filepath in neededfiles:
                 continue
 
             stat = os.stat(filepath)
             if stat.st_mtime < expiration:
                 os.remove(filepath)
 
     progress.complete()

mercurial/repair.py

@@ -1,563 +1,563 b''
 # repair.py - functions for repository repair for mercurial
 #
 # Copyright 2005, 2006 Chris Mason <[e-mail]>
 # Copyright 2007 Olivia Mackall
 #
 # 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
 
 from .i18n import _
 from .node import (
     hex,
     short,
 )
 from . import (
     bundle2,
     changegroup,
     discovery,
     error,
     exchange,
     obsolete,
     obsutil,
     pathutil,
     phases,
     pycompat,
     requirements,
     scmutil,
     util,
 )
 from .utils import (
     hashutil,
     stringutil,
     urlutil,
 )
 
 
 def backupbundle(
     repo, bases, heads, node, suffix, compress=True, obsolescence=True
 ):
     """create a bundle with the specified revisions as a backup"""
 
     backupdir = b"strip-backup"
     vfs = repo.vfs
     if not vfs.isdir(backupdir):
         vfs.mkdir(backupdir)
 
     # Include a hash of all the nodes in the filename for uniqueness
     allcommits = repo.set(b'%ln::%ln', bases, heads)
     allhashes = sorted(c.hex() for c in allcommits)
     totalhash = hashutil.sha1(b''.join(allhashes)).digest()
     name = b"%s/%s-%s-%s.hg" % (
         backupdir,
         short(node),
         hex(totalhash[:4]),
         suffix,
     )
 
     cgversion = changegroup.localversion(repo)
     comp = None
     if cgversion != b'01':
         bundletype = b"HG20"
         if compress:
             comp = b'BZ'
     elif compress:
         bundletype = b"HG10BZ"
     else:
         bundletype = b"HG10UN"
 
     outgoing = discovery.outgoing(repo, missingroots=bases, ancestorsof=heads)
     contentopts = {
         b'cg.version': cgversion,
         b'obsolescence': obsolescence,
         b'phases': True,
     }
     return bundle2.writenewbundle(
         repo.ui,
         repo,
         b'strip',
         name,
         bundletype,
         outgoing,
         contentopts,
         vfs,
         compression=comp,
     )
 
 
 def _collectfiles(repo, striprev):
     """find out the filelogs affected by the strip"""
     files = set()
 
     for x in pycompat.xrange(striprev, len(repo)):
         files.update(repo[x].files())
 
     return sorted(files)
 
 
 def _collectrevlog(revlog, striprev):
     _, brokenset = revlog.getstrippoint(striprev)
     return [revlog.linkrev(r) for r in brokenset]
 
 
 def _collectbrokencsets(repo, files, striprev):
     """return the changesets which will be broken by the truncation"""
     s = set()
 
     for revlog in manifestrevlogs(repo):
         s.update(_collectrevlog(revlog, striprev))
     for fname in files:
         s.update(_collectrevlog(repo.file(fname), striprev))
 
     return s
 
 
 def strip(ui, repo, nodelist, backup=True, topic=b'backup'):
     # This function requires the caller to lock the repo, but it operates
     # within a transaction of its own, and thus requires there to be no current
     # transaction when it is called.
     if repo.currenttransaction() is not None:
         raise error.ProgrammingError(b'cannot strip from inside a transaction')
 
     # Simple way to maintain backwards compatibility for this
     # argument.
     if backup in [b'none', b'strip']:
         backup = False
 
     repo = repo.unfiltered()
     repo.destroying()
     vfs = repo.vfs
     # load bookmark before changelog to avoid side effect from outdated
     # changelog (see repo._refreshchangelog)
     repo._bookmarks
     cl = repo.changelog
 
     # TODO handle undo of merge sets
     if isinstance(nodelist, bytes):
         nodelist = [nodelist]
     striplist = [cl.rev(node) for node in nodelist]
     striprev = min(striplist)
 
     files = _collectfiles(repo, striprev)
     saverevs = _collectbrokencsets(repo, files, striprev)
 
     # Some revisions with rev > striprev may not be descendants of striprev.
     # We have to find these revisions and put them in a bundle, so that
     # we can restore them after the truncations.
     # To create the bundle we use repo.changegroupsubset which requires
     # the list of heads and bases of the set of interesting revisions.
     # (head = revision in the set that has no descendant in the set;
     #  base = revision in the set that has no ancestor in the set)
     tostrip = set(striplist)
     saveheads = set(saverevs)
     for r in cl.revs(start=striprev + 1):
         if any(p in tostrip for p in cl.parentrevs(r)):
             tostrip.add(r)
 
         if r not in tostrip:
             saverevs.add(r)
             saveheads.difference_update(cl.parentrevs(r))
             saveheads.add(r)
     saveheads = [cl.node(r) for r in saveheads]
 
     # compute base nodes
     if saverevs:
         descendants = set(cl.descendants(saverevs))
         saverevs.difference_update(descendants)
     savebases = [cl.node(r) for r in saverevs]
     stripbases = [cl.node(r) for r in tostrip]
 
     stripobsidx = obsmarkers = ()
     if repo.ui.configbool(b'devel', b'strip-obsmarkers'):
         obsmarkers = obsutil.exclusivemarkers(repo, stripbases)
     if obsmarkers:
         stripobsidx = [
             i for i, m in enumerate(repo.obsstore) if m in obsmarkers
         ]
 
     newbmtarget, updatebm = _bookmarkmovements(repo, tostrip)
 
     backupfile = None
     node = nodelist[-1]
     if backup:
         backupfile = _createstripbackup(repo, stripbases, node, topic)
     # create a changegroup for all the branches we need to keep
     tmpbundlefile = None
     if saveheads:
         # do not compress temporary bundle if we remove it from disk later
         #
         # We do not include obsolescence, it might re-introduce prune markers
         # we are trying to strip.  This is harmless since the stripped markers
         # are already backed up and we did not touched the markers for the
         # saved changesets.
         tmpbundlefile = backupbundle(
             repo,
             savebases,
             saveheads,
             node,
             b'temp',
             compress=False,
             obsolescence=False,
         )
 
     with ui.uninterruptible():
         try:
             with repo.transaction(b"strip") as tr:
                 # TODO this code violates the interface abstraction of the
                 # transaction and makes assumptions that file storage is
                 # using append-only files. We'll need some kind of storage
                 # API to handle stripping for us.
                 oldfiles = set(tr._offsetmap.keys())
                 oldfiles.update(tr._newfiles)
 
                 tr.startgroup()
                 cl.strip(striprev, tr)
                 stripmanifest(repo, striprev, tr, files)
 
                 for fn in files:
                     repo.file(fn).strip(striprev, tr)
                 tr.endgroup()
 
                 entries = tr.readjournal()
 
                 for file, troffset in entries:
                     if file in oldfiles:
                         continue
                     with repo.svfs(file, b'a', checkambig=True) as fp:
                         fp.truncate(troffset)
                     if troffset == 0:
                         repo.store.markremoved(file)
 
                 deleteobsmarkers(repo.obsstore, stripobsidx)
                 del repo.obsstore
                 repo.invalidatevolatilesets()
                 repo._phasecache.filterunknown(repo)
 
             if tmpbundlefile:
                 ui.note(_(b"adding branch\n"))
                 f = vfs.open(tmpbundlefile, b"rb")
                 gen = exchange.readbundle(ui, f, tmpbundlefile, vfs)
                 # silence internal shuffling chatter
                 maybe_silent = (
                     repo.ui.silent()
                     if not repo.ui.verbose
                     else util.nullcontextmanager()
                 )
                 with maybe_silent:
                     tmpbundleurl = b'bundle:' + vfs.join(tmpbundlefile)
                     txnname = b'strip'
                     if not isinstance(gen, bundle2.unbundle20):
                         txnname = b"strip\n%s" % urlutil.hidepassword(
                             tmpbundleurl
                         )
                     with repo.transaction(txnname) as tr:
                         bundle2.applybundle(
                             repo, gen, tr, source=b'strip', url=tmpbundleurl
                         )
                 f.close()
 
             with repo.transaction(b'repair') as tr:
                 bmchanges = [(m, repo[newbmtarget].node()) for m in updatebm]
                 repo._bookmarks.applychanges(repo, tr, bmchanges)
 
             # remove undo files
             for undovfs, undofile in repo.undofiles():
                 try:
                     undovfs.unlink(undofile)
                 except OSError as e:
                     if e.errno != errno.ENOENT:
                         ui.warn(
                             _(b'error removing %s: %s\n')
                             % (
                                 undovfs.join(undofile),
                                 stringutil.forcebytestr(e),
                             )
                         )
 
         except:  # re-raises
             if backupfile:
                 ui.warn(
                     _(b"strip failed, backup bundle stored in '%s'\n")
                     % vfs.join(backupfile)
                 )
             if tmpbundlefile:
                 ui.warn(
                     _(b"strip failed, unrecovered changes stored in '%s'\n")
                     % vfs.join(tmpbundlefile)
                 )
                 ui.warn(
                     _(
                         b"(fix the problem, then recover the changesets with "
                         b"\"hg unbundle '%s'\")\n"
                     )
                     % vfs.join(tmpbundlefile)
                 )
             raise
         else:
             if tmpbundlefile:
                 # Remove temporary bundle only if there were no exceptions
                 vfs.unlink(tmpbundlefile)
 
     repo.destroyed()
     # return the backup file path (or None if 'backup' was False) so
     # extensions can use it
     return backupfile
 
 
 def softstrip(ui, repo, nodelist, backup=True, topic=b'backup'):
     """perform a "soft" strip using the archived phase"""
     tostrip = [c.node() for c in repo.set(b'sort(%ln::)', nodelist)]
     if not tostrip:
         return None
 
     backupfile = None
     if backup:
         node = tostrip[0]
         backupfile = _createstripbackup(repo, tostrip, node, topic)
 
     newbmtarget, updatebm = _bookmarkmovements(repo, tostrip)
     with repo.transaction(b'strip') as tr:
         phases.retractboundary(repo, tr, phases.archived, tostrip)
         bmchanges = [(m, repo[newbmtarget].node()) for m in updatebm]
         repo._bookmarks.applychanges(repo, tr, bmchanges)
     return backupfile
 
 
 def _bookmarkmovements(repo, tostrip):
     # compute necessary bookmark movement
     bm = repo._bookmarks
     updatebm = []
     for m in bm:
         rev = repo[bm[m]].rev()
         if rev in tostrip:
             updatebm.append(m)
     newbmtarget = None
     # If we need to move bookmarks, compute bookmark
     # targets. Otherwise we can skip doing this logic.
     if updatebm:
         # For a set s, max(parents(s) - s) is the same as max(heads(::s - s)),
         # but is much faster
         newbmtarget = repo.revs(b'max(parents(%ld) - (%ld))', tostrip, tostrip)
         if newbmtarget:
             newbmtarget = repo[newbmtarget.first()].node()
         else:
             newbmtarget = b'.'
     return newbmtarget, updatebm
 
 
 def _createstripbackup(repo, stripbases, node, topic):
     # backup the changeset we are about to strip
     vfs = repo.vfs
     cl = repo.changelog
     backupfile = backupbundle(repo, stripbases, cl.heads(), node, topic)
     repo.ui.status(_(b"saved backup bundle to %s\n") % vfs.join(backupfile))
     repo.ui.log(
         b"backupbundle", b"saved backup bundle to %s\n", vfs.join(backupfile)
     )
     return backupfile
 
 
 def safestriproots(ui, repo, nodes):
     """return list of roots of nodes where descendants are covered by nodes"""
     torev = repo.unfiltered().changelog.rev
     revs = {torev(n) for n in nodes}
     # tostrip = wanted - unsafe = wanted - ancestors(orphaned)
     # orphaned = affected - wanted
     # affected = descendants(roots(wanted))
     # wanted = revs
     revset = b'%ld - ( ::( (roots(%ld):: and not _phase(%s)) -%ld) )'
     tostrip = set(repo.revs(revset, revs, revs, phases.internal, revs))
     notstrip = revs - tostrip
     if notstrip:
         nodestr = b', '.join(sorted(short(repo[n].node()) for n in notstrip))
         ui.warn(
             _(b'warning: orphaned descendants detected, not stripping %s\n')
             % nodestr
         )
     return [c.node() for c in repo.set(b'roots(%ld)', tostrip)]
 
 
 class stripcallback(object):
     """used as a transaction postclose callback"""
 
     def __init__(self, ui, repo, backup, topic):
         self.ui = ui
         self.repo = repo
         self.backup = backup
         self.topic = topic or b'backup'
         self.nodelist = []
 
     def addnodes(self, nodes):
         self.nodelist.extend(nodes)
 
     def __call__(self, tr):
         roots = safestriproots(self.ui, self.repo, self.nodelist)
         if roots:
             strip(self.ui, self.repo, roots, self.backup, self.topic)
 
 
 def delayedstrip(ui, repo, nodelist, topic=None, backup=True):
     """like strip, but works inside transaction and won't strip irreverent revs
 
     nodelist must explicitly contain all descendants. Otherwise a warning will
     be printed that some nodes are not stripped.
 
     Will do a backup if `backup` is True. The last non-None "topic" will be
     used as the backup topic name. The default backup topic name is "backup".
     """
     tr = repo.currenttransaction()
     if not tr:
         nodes = safestriproots(ui, repo, nodelist)
         return strip(ui, repo, nodes, backup=backup, topic=topic)
     # transaction postclose callbacks are called in alphabet order.
     # use '\xff' as prefix so we are likely to be called last.
     callback = tr.getpostclose(b'\xffstrip')
     if callback is None:
         callback = stripcallback(ui, repo, backup=backup, topic=topic)
         tr.addpostclose(b'\xffstrip', callback)
     if topic:
         callback.topic = topic
     callback.addnodes(nodelist)
 
 
 def stripmanifest(repo, striprev, tr, files):
     for revlog in manifestrevlogs(repo):
         revlog.strip(striprev, tr)
 
 
 def manifestrevlogs(repo):
     yield repo.manifestlog.getstorage(b'')
     if scmutil.istreemanifest(repo):
         # This logic is safe if treemanifest isn't enabled, but also
         # pointless, so we skip it if treemanifest isn't enabled.
-        for t, unencoded, encoded, size in repo.store.datafiles():
+        for t, unencoded, size in repo.store.datafiles():
             if unencoded.startswith(b'meta/') and unencoded.endswith(
                 b'00manifest.i'
             ):
                 dir = unencoded[5:-12]
                 yield repo.manifestlog.getstorage(dir)
 
 
 def rebuildfncache(ui, repo, only_data=False):
     """Rebuilds the fncache file from repo history.
 
     Missing entries will be added. Extra entries will be removed.
     """
     repo = repo.unfiltered()
 
     if requirements.FNCACHE_REQUIREMENT not in repo.requirements:
         ui.warn(
             _(
                 b'(not rebuilding fncache because repository does not '
                 b'support fncache)\n'
             )
         )
         return
 
     with repo.lock():
         fnc = repo.store.fncache
         fnc.ensureloaded(warn=ui.warn)
 
         oldentries = set(fnc.entries)
         newentries = set()
         seenfiles = set()
 
         if only_data:
             # Trust the listing of .i from the fncache, but not the .d. This is
             # much faster, because we only need to stat every possible .d files,
             # instead of reading the full changelog
             for f in fnc:
                 if f[:5] == b'data/' and f[-2:] == b'.i':
                     seenfiles.add(f[5:-2])
                     newentries.add(f)
                     dataf = f[:-2] + b'.d'
                     if repo.store._exists(dataf):
                         newentries.add(dataf)
         else:
             progress = ui.makeprogress(
                 _(b'rebuilding'), unit=_(b'changesets'), total=len(repo)
             )
             for rev in repo:
                 progress.update(rev)
 
                 ctx = repo[rev]
                 for f in ctx.files():
                     # This is to minimize I/O.
                     if f in seenfiles:
                         continue
                     seenfiles.add(f)
 
                     i = b'data/%s.i' % f
                     d = b'data/%s.d' % f
 
                     if repo.store._exists(i):
                         newentries.add(i)
                     if repo.store._exists(d):
                         newentries.add(d)
 
             progress.complete()
 
         if requirements.TREEMANIFEST_REQUIREMENT in repo.requirements:
             # This logic is safe if treemanifest isn't enabled, but also
             # pointless, so we skip it if treemanifest isn't enabled.
             for dir in pathutil.dirs(seenfiles):
                 i = b'meta/%s/00manifest.i' % dir
                 d = b'meta/%s/00manifest.d' % dir
 
                 if repo.store._exists(i):
                     newentries.add(i)
                 if repo.store._exists(d):
                     newentries.add(d)
 
         addcount = len(newentries - oldentries)
         removecount = len(oldentries - newentries)
         for p in sorted(oldentries - newentries):
             ui.write(_(b'removing %s\n') % p)
         for p in sorted(newentries - oldentries):
             ui.write(_(b'adding %s\n') % p)
 
         if addcount or removecount:
             ui.write(
                 _(b'%d items added, %d removed from fncache\n')
                 % (addcount, removecount)
             )
             fnc.entries = newentries
             fnc._dirty = True
 
             with repo.transaction(b'fncache') as tr:
                 fnc.write(tr)
         else:
             ui.write(_(b'fncache already up to date\n'))
 
 
 def deleteobsmarkers(obsstore, indices):
     """Delete some obsmarkers from obsstore and return how many were deleted
 
     'indices' is a list of ints which are the indices
     of the markers to be deleted.
 
     Every invocation of this function completely rewrites the obsstore file,
     skipping the markers we want to be removed. The new temporary file is
     created, remaining markers are written there and on .close() this file
     gets atomically renamed to obsstore, thus guaranteeing consistency."""
     if not indices:
         # we don't want to rewrite the obsstore with the same content
         return
 
     left = []
     current = obsstore._all
     n = 0
     for i, m in enumerate(current):
         if i in indices:
             n += 1
             continue
         left.append(m)
 
     newobsstorefile = obsstore.svfs(b'obsstore', b'w', atomictemp=True)
     for bytes in obsolete.encodemarkers(left, True, obsstore._version):
         newobsstorefile.write(bytes)
     newobsstorefile.close()
     return n

mercurial/revlogutils/rewrite.py

@@ -1,886 +1,886 b''
 # censor code related to censoring revision
 # coding: utf8
 #
 # Copyright 2021 Pierre-Yves David <[e-mail]>
 # Copyright 2015 Google, Inc <[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.
 
 import binascii
 import contextlib
 import os
 import struct
 
 from ..node import (
     nullrev,
 )
 from .constants import (
     COMP_MODE_PLAIN,
     ENTRY_DATA_COMPRESSED_LENGTH,
     ENTRY_DATA_COMPRESSION_MODE,
     ENTRY_DATA_OFFSET,
     ENTRY_DATA_UNCOMPRESSED_LENGTH,
     ENTRY_DELTA_BASE,
     ENTRY_LINK_REV,
     ENTRY_NODE_ID,
     ENTRY_PARENT_1,
     ENTRY_PARENT_2,
     ENTRY_SIDEDATA_COMPRESSED_LENGTH,
     ENTRY_SIDEDATA_COMPRESSION_MODE,
     ENTRY_SIDEDATA_OFFSET,
     REVIDX_ISCENSORED,
     REVLOGV0,
     REVLOGV1,
 )
 from ..i18n import _
 
 from .. import (
     error,
     mdiff,
     pycompat,
     revlogutils,
     util,
 )
 from ..utils import (
     storageutil,
 )
 from . import (
     constants,
     deltas,
 )
 
 
 def v1_censor(rl, tr, censornode, tombstone=b''):
     """censors a revision in a "version 1" revlog"""
     assert rl._format_version == constants.REVLOGV1, rl._format_version
 
     # avoid cycle
     from .. import revlog
 
     censorrev = rl.rev(censornode)
     tombstone = storageutil.packmeta({b'censored': tombstone}, b'')
 
     # Rewriting the revlog in place is hard. Our strategy for censoring is
     # to create a new revlog, copy all revisions to it, then replace the
     # revlogs on transaction close.
     #
     # This is a bit dangerous. We could easily have a mismatch of state.
     newrl = revlog.revlog(
         rl.opener,
         target=rl.target,
         radix=rl.radix,
         postfix=b'tmpcensored',
         censorable=True,
     )
     newrl._format_version = rl._format_version
     newrl._format_flags = rl._format_flags
     newrl._generaldelta = rl._generaldelta
     newrl._parse_index = rl._parse_index
 
     for rev in rl.revs():
         node = rl.node(rev)
         p1, p2 = rl.parents(node)
 
         if rev == censorrev:
             newrl.addrawrevision(
                 tombstone,
                 tr,
                 rl.linkrev(censorrev),
                 p1,
                 p2,
                 censornode,
                 constants.REVIDX_ISCENSORED,
             )
 
             if newrl.deltaparent(rev) != nullrev:
                 m = _(b'censored revision stored as delta; cannot censor')
                 h = _(
                     b'censoring of revlogs is not fully implemented;'
                     b' please report this bug'
                 )
                 raise error.Abort(m, hint=h)
             continue
 
         if rl.iscensored(rev):
             if rl.deltaparent(rev) != nullrev:
                 m = _(
                     b'cannot censor due to censored '
                     b'revision having delta stored'
                 )
                 raise error.Abort(m)
             rawtext = rl._chunk(rev)
         else:
             rawtext = rl.rawdata(rev)
 
         newrl.addrawrevision(
             rawtext, tr, rl.linkrev(rev), p1, p2, node, rl.flags(rev)
         )
 
     tr.addbackup(rl._indexfile, location=b'store')
     if not rl._inline:
         tr.addbackup(rl._datafile, location=b'store')
 
     rl.opener.rename(newrl._indexfile, rl._indexfile)
     if not rl._inline:
         rl.opener.rename(newrl._datafile, rl._datafile)
 
     rl.clearcaches()
     rl._loadindex()
 
 
 def v2_censor(revlog, tr, censornode, tombstone=b''):
     """censors a revision in a "version 2" revlog"""
     assert revlog._format_version != REVLOGV0, revlog._format_version
     assert revlog._format_version != REVLOGV1, revlog._format_version
 
     censor_revs = {revlog.rev(censornode)}
     _rewrite_v2(revlog, tr, censor_revs, tombstone)
 
 
 def _rewrite_v2(revlog, tr, censor_revs, tombstone=b''):
     """rewrite a revlog to censor some of its content
 
     General principle
 
     We create new revlog files (index/data/sidedata) to copy the content of
     the existing data without the censored data.
 
     We need to recompute new delta for any revision that used the censored
     revision as delta base. As the cumulative size of the new delta may be
     large, we store them in a temporary file until they are stored in their
     final destination.
 
     All data before the censored data can be blindly copied. The rest needs
     to be copied as we go and the associated index entry needs adjustement.
     """
     assert revlog._format_version != REVLOGV0, revlog._format_version
     assert revlog._format_version != REVLOGV1, revlog._format_version
 
     old_index = revlog.index
     docket = revlog._docket
 
     tombstone = storageutil.packmeta({b'censored': tombstone}, b'')
 
     first_excl_rev = min(censor_revs)
 
     first_excl_entry = revlog.index[first_excl_rev]
     index_cutoff = revlog.index.entry_size * first_excl_rev
     data_cutoff = first_excl_entry[ENTRY_DATA_OFFSET] >> 16
     sidedata_cutoff = revlog.sidedata_cut_off(first_excl_rev)
 
     with pycompat.unnamedtempfile(mode=b"w+b") as tmp_storage:
         # rev → (new_base, data_start, data_end, compression_mode)
         rewritten_entries = _precompute_rewritten_delta(
             revlog,
             old_index,
             censor_revs,
             tmp_storage,
         )
 
         all_files = _setup_new_files(
             revlog,
             index_cutoff,
             data_cutoff,
             sidedata_cutoff,
         )
 
         # we dont need to open the old index file since its content already
         # exist in a usable form in `old_index`.
         with all_files() as open_files:
             (
                 old_data_file,
                 old_sidedata_file,
                 new_index_file,
                 new_data_file,
                 new_sidedata_file,
             ) = open_files
 
             # writing the censored revision
 
             # Writing all subsequent revisions
             for rev in range(first_excl_rev, len(old_index)):
                 if rev in censor_revs:
                     _rewrite_censor(
                         revlog,
                         old_index,
                         open_files,
                         rev,
                         tombstone,
                     )
                 else:
                     _rewrite_simple(
                         revlog,
                         old_index,
                         open_files,
                         rev,
                         rewritten_entries,
                         tmp_storage,
                     )
     docket.write(transaction=None, stripping=True)
 
 
 def _precompute_rewritten_delta(
     revlog,
     old_index,
     excluded_revs,
     tmp_storage,
 ):
     """Compute new delta for revisions whose delta is based on revision that
     will not survive as is.
 
     Return a mapping: {rev → (new_base, data_start, data_end, compression_mode)}
     """
     dc = deltas.deltacomputer(revlog)
     rewritten_entries = {}
     first_excl_rev = min(excluded_revs)
     with revlog._segmentfile._open_read() as dfh:
         for rev in range(first_excl_rev, len(old_index)):
             if rev in excluded_revs:
                 # this revision will be preserved as is, so we don't need to
                 # consider recomputing a delta.
                 continue
             entry = old_index[rev]
             if entry[ENTRY_DELTA_BASE] not in excluded_revs:
                 continue
             # This is a revision that use the censored revision as the base
             # for its delta. We need a need new deltas
             if entry[ENTRY_DATA_UNCOMPRESSED_LENGTH] == 0:
                 # this revision is empty, we can delta against nullrev
                 rewritten_entries[rev] = (nullrev, 0, 0, COMP_MODE_PLAIN)
             else:
 
                 text = revlog.rawdata(rev, _df=dfh)
                 info = revlogutils.revisioninfo(
                     node=entry[ENTRY_NODE_ID],
                     p1=revlog.node(entry[ENTRY_PARENT_1]),
                     p2=revlog.node(entry[ENTRY_PARENT_2]),
                     btext=[text],
                     textlen=len(text),
                     cachedelta=None,
                     flags=entry[ENTRY_DATA_OFFSET] & 0xFFFF,
                 )
                 d = dc.finddeltainfo(
                     info, dfh, excluded_bases=excluded_revs, target_rev=rev
                 )
                 default_comp = revlog._docket.default_compression_header
                 comp_mode, d = deltas.delta_compression(default_comp, d)
                 # using `tell` is a bit lazy, but we are not here for speed
                 start = tmp_storage.tell()
                 tmp_storage.write(d.data[1])
                 end = tmp_storage.tell()
                 rewritten_entries[rev] = (d.base, start, end, comp_mode)
     return rewritten_entries
 
 
 def _setup_new_files(
     revlog,
     index_cutoff,
     data_cutoff,
     sidedata_cutoff,
 ):
     """
 
     return a context manager to open all the relevant files:
     - old_data_file,
     - old_sidedata_file,
     - new_index_file,
     - new_data_file,
     - new_sidedata_file,
 
     The old_index_file is not here because it is accessed through the
     `old_index` object if the caller function.
     """
     docket = revlog._docket
     old_index_filepath = revlog.opener.join(docket.index_filepath())
     old_data_filepath = revlog.opener.join(docket.data_filepath())
     old_sidedata_filepath = revlog.opener.join(docket.sidedata_filepath())
 
     new_index_filepath = revlog.opener.join(docket.new_index_file())
     new_data_filepath = revlog.opener.join(docket.new_data_file())
     new_sidedata_filepath = revlog.opener.join(docket.new_sidedata_file())
 
     util.copyfile(old_index_filepath, new_index_filepath, nb_bytes=index_cutoff)
     util.copyfile(old_data_filepath, new_data_filepath, nb_bytes=data_cutoff)
     util.copyfile(
         old_sidedata_filepath,
         new_sidedata_filepath,
         nb_bytes=sidedata_cutoff,
     )
     revlog.opener.register_file(docket.index_filepath())
     revlog.opener.register_file(docket.data_filepath())
     revlog.opener.register_file(docket.sidedata_filepath())
 
     docket.index_end = index_cutoff
     docket.data_end = data_cutoff
     docket.sidedata_end = sidedata_cutoff
 
     # reload the revlog internal information
     revlog.clearcaches()
     revlog._loadindex(docket=docket)
 
     @contextlib.contextmanager
     def all_files_opener():
         # hide opening in an helper function to please check-code, black
         # and various python version at the same time
         with open(old_data_filepath, 'rb') as old_data_file:
             with open(old_sidedata_filepath, 'rb') as old_sidedata_file:
                 with open(new_index_filepath, 'r+b') as new_index_file:
                     with open(new_data_filepath, 'r+b') as new_data_file:
                         with open(
                             new_sidedata_filepath, 'r+b'
                         ) as new_sidedata_file:
                             new_index_file.seek(0, os.SEEK_END)
                             assert new_index_file.tell() == index_cutoff
                             new_data_file.seek(0, os.SEEK_END)
                             assert new_data_file.tell() == data_cutoff
                             new_sidedata_file.seek(0, os.SEEK_END)
                             assert new_sidedata_file.tell() == sidedata_cutoff
                             yield (
                                 old_data_file,
                                 old_sidedata_file,
                                 new_index_file,
                                 new_data_file,
                                 new_sidedata_file,
                             )
 
     return all_files_opener
 
 
 def _rewrite_simple(
     revlog,
     old_index,
     all_files,
     rev,
     rewritten_entries,
     tmp_storage,
 ):
     """append a normal revision to the index after the rewritten one(s)"""
     (
         old_data_file,
         old_sidedata_file,
         new_index_file,
         new_data_file,
         new_sidedata_file,
     ) = all_files
     entry = old_index[rev]
     flags = entry[ENTRY_DATA_OFFSET] & 0xFFFF
     old_data_offset = entry[ENTRY_DATA_OFFSET] >> 16
 
     if rev not in rewritten_entries:
         old_data_file.seek(old_data_offset)
         new_data_size = entry[ENTRY_DATA_COMPRESSED_LENGTH]
         new_data = old_data_file.read(new_data_size)
         data_delta_base = entry[ENTRY_DELTA_BASE]
         d_comp_mode = entry[ENTRY_DATA_COMPRESSION_MODE]
     else:
         (
             data_delta_base,
             start,
             end,
             d_comp_mode,
         ) = rewritten_entries[rev]
         new_data_size = end - start
         tmp_storage.seek(start)
         new_data = tmp_storage.read(new_data_size)
 
     # It might be faster to group continuous read/write operation,
     # however, this is censor, an operation that is not focussed
     # around stellar performance. So I have not written this
     # optimisation yet.
     new_data_offset = new_data_file.tell()
     new_data_file.write(new_data)
 
     sidedata_size = entry[ENTRY_SIDEDATA_COMPRESSED_LENGTH]
     new_sidedata_offset = new_sidedata_file.tell()
     if 0 < sidedata_size:
         old_sidedata_offset = entry[ENTRY_SIDEDATA_OFFSET]
         old_sidedata_file.seek(old_sidedata_offset)
         new_sidedata = old_sidedata_file.read(sidedata_size)
         new_sidedata_file.write(new_sidedata)
 
     data_uncompressed_length = entry[ENTRY_DATA_UNCOMPRESSED_LENGTH]
     sd_com_mode = entry[ENTRY_SIDEDATA_COMPRESSION_MODE]
     assert data_delta_base <= rev, (data_delta_base, rev)
 
     new_entry = revlogutils.entry(
         flags=flags,
         data_offset=new_data_offset,
         data_compressed_length=new_data_size,
         data_uncompressed_length=data_uncompressed_length,
         data_delta_base=data_delta_base,
         link_rev=entry[ENTRY_LINK_REV],
         parent_rev_1=entry[ENTRY_PARENT_1],
         parent_rev_2=entry[ENTRY_PARENT_2],
         node_id=entry[ENTRY_NODE_ID],
         sidedata_offset=new_sidedata_offset,
         sidedata_compressed_length=sidedata_size,
         data_compression_mode=d_comp_mode,
         sidedata_compression_mode=sd_com_mode,
     )
     revlog.index.append(new_entry)
     entry_bin = revlog.index.entry_binary(rev)
     new_index_file.write(entry_bin)
 
     revlog._docket.index_end = new_index_file.tell()
     revlog._docket.data_end = new_data_file.tell()
     revlog._docket.sidedata_end = new_sidedata_file.tell()
 
 
 def _rewrite_censor(
     revlog,
     old_index,
     all_files,
     rev,
     tombstone,
 ):
     """rewrite and append a censored revision"""
     (
         old_data_file,
         old_sidedata_file,
         new_index_file,
         new_data_file,
         new_sidedata_file,
     ) = all_files
     entry = old_index[rev]
 
     # XXX consider trying the default compression too
     new_data_size = len(tombstone)
     new_data_offset = new_data_file.tell()
     new_data_file.write(tombstone)
 
     # we are not adding any sidedata as they might leak info about the censored version
 
     link_rev = entry[ENTRY_LINK_REV]
 
     p1 = entry[ENTRY_PARENT_1]
     p2 = entry[ENTRY_PARENT_2]
 
     new_entry = revlogutils.entry(
         flags=constants.REVIDX_ISCENSORED,
         data_offset=new_data_offset,
         data_compressed_length=new_data_size,
         data_uncompressed_length=new_data_size,
         data_delta_base=rev,
         link_rev=link_rev,
         parent_rev_1=p1,
         parent_rev_2=p2,
         node_id=entry[ENTRY_NODE_ID],
         sidedata_offset=0,
         sidedata_compressed_length=0,
         data_compression_mode=COMP_MODE_PLAIN,
         sidedata_compression_mode=COMP_MODE_PLAIN,
     )
     revlog.index.append(new_entry)
     entry_bin = revlog.index.entry_binary(rev)
     new_index_file.write(entry_bin)
     revlog._docket.index_end = new_index_file.tell()
     revlog._docket.data_end = new_data_file.tell()
 
 
 def _get_filename_from_filelog_index(path):
     # Drop the extension and the `data/` prefix
     path_part = path.rsplit(b'.', 1)[0].split(b'/', 1)
     if len(path_part) < 2:
         msg = _(b"cannot recognize filelog from filename: '%s'")
         msg %= path
         raise error.Abort(msg)
 
     return path_part[1]
 
 
 def _filelog_from_filename(repo, path):
     """Returns the filelog for the given `path`. Stolen from `engine.py`"""
 
     from .. import filelog  # avoid cycle
 
     fl = filelog.filelog(repo.svfs, path)
     return fl
 
 
 def _write_swapped_parents(repo, rl, rev, offset, fp):
     """Swaps p1 and p2 and overwrites the revlog entry for `rev` in `fp`"""
     from ..pure import parsers  # avoid cycle
 
     if repo._currentlock(repo._lockref) is None:
         # Let's be paranoid about it
         msg = "repo needs to be locked to rewrite parents"
         raise error.ProgrammingError(msg)
 
     index_format = parsers.IndexObject.index_format
     entry = rl.index[rev]
     new_entry = list(entry)
     new_entry[5], new_entry[6] = entry[6], entry[5]
     packed = index_format.pack(*new_entry[:8])
     fp.seek(offset)
     fp.write(packed)
 
 
 def _reorder_filelog_parents(repo, fl, to_fix):
     """
     Swaps p1 and p2 for all `to_fix` revisions of filelog `fl` and writes the
     new version to disk, overwriting the old one with a rename.
     """
     from ..pure import parsers  # avoid cycle
 
     ui = repo.ui
     assert len(to_fix) > 0
     rl = fl._revlog
     if rl._format_version != constants.REVLOGV1:
         msg = "expected version 1 revlog, got version '%d'" % rl._format_version
         raise error.ProgrammingError(msg)
 
     index_file = rl._indexfile
     new_file_path = index_file + b'.tmp-parents-fix'
     repaired_msg = _(b"repaired revision %d of 'filelog %s'\n")
 
     with ui.uninterruptible():
         try:
             util.copyfile(
                 rl.opener.join(index_file),
                 rl.opener.join(new_file_path),
                 checkambig=rl._checkambig,
             )
 
             with rl.opener(new_file_path, mode=b"r+") as fp:
                 if rl._inline:
                     index = parsers.InlinedIndexObject(fp.read())
                     for rev in fl.revs():
                         if rev in to_fix:
                             offset = index._calculate_index(rev)
                             _write_swapped_parents(repo, rl, rev, offset, fp)
                             ui.write(repaired_msg % (rev, index_file))
                 else:
                     index_format = parsers.IndexObject.index_format
                     for rev in to_fix:
                         offset = rev * index_format.size
                         _write_swapped_parents(repo, rl, rev, offset, fp)
                         ui.write(repaired_msg % (rev, index_file))
 
             rl.opener.rename(new_file_path, index_file)
             rl.clearcaches()
             rl._loadindex()
         finally:
             util.tryunlink(new_file_path)
 
 
 def _is_revision_affected(fl, filerev, metadata_cache=None):
     full_text = lambda: fl._revlog.rawdata(filerev)
     parent_revs = lambda: fl._revlog.parentrevs(filerev)
     return _is_revision_affected_inner(
         full_text, parent_revs, filerev, metadata_cache
     )
 
 
 def _is_revision_affected_inner(
     full_text,
     parents_revs,
     filerev,
     metadata_cache=None,
 ):
     """Mercurial currently (5.9rc0) uses `p1 == nullrev and p2 != nullrev` as a
     special meaning compared to the reverse in the context of filelog-based
     copytracing. issue6528 exists because new code assumed that parent ordering
     didn't matter, so this detects if the revision contains metadata (since
     it's only used for filelog-based copytracing) and its parents are in the
     "wrong" order."""
     try:
         raw_text = full_text()
     except error.CensoredNodeError:
         # We don't care about censored nodes as they never carry metadata
         return False
     has_meta = raw_text.startswith(b'\x01\n')
     if metadata_cache is not None:
         metadata_cache[filerev] = has_meta
     if has_meta:
         (p1, p2) = parents_revs()
         if p1 != nullrev and p2 == nullrev:
             return True
     return False
 
 
 def _is_revision_affected_fast(repo, fl, filerev, metadata_cache):
     rl = fl._revlog
     is_censored = lambda: rl.iscensored(filerev)
     delta_base = lambda: rl.deltaparent(filerev)
     delta = lambda: rl._chunk(filerev)
     full_text = lambda: rl.rawdata(filerev)
     parent_revs = lambda: rl.parentrevs(filerev)
     return _is_revision_affected_fast_inner(
         is_censored,
         delta_base,
         delta,
         full_text,
         parent_revs,
         filerev,
         metadata_cache,
     )
 
 
 def _is_revision_affected_fast_inner(
     is_censored,
     delta_base,
     delta,
     full_text,
     parent_revs,
     filerev,
     metadata_cache,
 ):
     """Optimization fast-path for `_is_revision_affected`.
 
     `metadata_cache` is a dict of `{rev: has_metadata}` which allows any
     revision to check if its base has metadata, saving computation of the full
     text, instead looking at the current delta.
 
     This optimization only works if the revisions are looked at in order."""
 
     if is_censored():
         # Censored revisions don't contain metadata, so they cannot be affected
         metadata_cache[filerev] = False
         return False
 
     p1, p2 = parent_revs()
     if p1 == nullrev or p2 != nullrev:
         return False
 
     delta_parent = delta_base()
     parent_has_metadata = metadata_cache.get(delta_parent)
     if parent_has_metadata is None:
         return _is_revision_affected_inner(
             full_text,
             parent_revs,
             filerev,
             metadata_cache,
         )
 
     chunk = delta()
     if not len(chunk):
         # No diff for this revision
         return parent_has_metadata
 
     header_length = 12
     if len(chunk) < header_length:
         raise error.Abort(_(b"patch cannot be decoded"))
 
     start, _end, _length = struct.unpack(b">lll", chunk[:header_length])
 
     if start < 2:  # len(b'\x01\n') == 2
         # This delta does *something* to the metadata marker (if any).
         # Check it the slow way
         is_affected = _is_revision_affected_inner(
             full_text,
             parent_revs,
             filerev,
             metadata_cache,
         )
         return is_affected
 
     # The diff did not remove or add the metadata header, it's then in the same
     # situation as its parent
     metadata_cache[filerev] = parent_has_metadata
     return parent_has_metadata
 
 
 def _from_report(ui, repo, context, from_report, dry_run):
     """
     Fix the revisions given in the `from_report` file, but still checks if the
     revisions are indeed affected to prevent an unfortunate cyclic situation
     where we'd swap well-ordered parents again.
 
     See the doc for `debug_fix_issue6528` for the format documentation.
     """
     ui.write(_(b"loading report file '%s'\n") % from_report)
 
     with context(), open(from_report, mode='rb') as f:
         for line in f.read().split(b'\n'):
             if not line:
                 continue
             filenodes, filename = line.split(b' ', 1)
             fl = _filelog_from_filename(repo, filename)
             to_fix = set(
                 fl.rev(binascii.unhexlify(n)) for n in filenodes.split(b',')
             )
             excluded = set()
 
             for filerev in to_fix:
                 if _is_revision_affected(fl, filerev):
                     msg = b"found affected revision %d for filelog '%s'\n"
                     ui.warn(msg % (filerev, filename))
                 else:
                     msg = _(b"revision %s of file '%s' is not affected\n")
                     msg %= (binascii.hexlify(fl.node(filerev)), filename)
                     ui.warn(msg)
                     excluded.add(filerev)
 
             to_fix = to_fix - excluded
             if not to_fix:
                 msg = _(b"no affected revisions were found for '%s'\n")
                 ui.write(msg % filename)
                 continue
             if not dry_run:
                 _reorder_filelog_parents(repo, fl, sorted(to_fix))
 
 
 def filter_delta_issue6528(revlog, deltas_iter):
     """filter incomind deltas to repaire issue 6528 on the fly"""
     metadata_cache = {}
 
     deltacomputer = deltas.deltacomputer(revlog)
 
     for rev, d in enumerate(deltas_iter, len(revlog)):
         (
             node,
             p1_node,
             p2_node,
             linknode,
             deltabase,
             delta,
             flags,
             sidedata,
         ) = d
 
         if not revlog.index.has_node(deltabase):
             raise error.LookupError(
                 deltabase, revlog.radix, _(b'unknown parent')
             )
         base_rev = revlog.rev(deltabase)
         if not revlog.index.has_node(p1_node):
             raise error.LookupError(p1_node, revlog.radix, _(b'unknown parent'))
         p1_rev = revlog.rev(p1_node)
         if not revlog.index.has_node(p2_node):
             raise error.LookupError(p2_node, revlog.radix, _(b'unknown parent'))
         p2_rev = revlog.rev(p2_node)
 
         is_censored = lambda: bool(flags & REVIDX_ISCENSORED)
         delta_base = lambda: revlog.rev(delta_base)
         delta_base = lambda: base_rev
         parent_revs = lambda: (p1_rev, p2_rev)
 
         def full_text():
             # note: being able to reuse the full text computation in the
             # underlying addrevision would be useful however this is a bit too
             # intrusive the for the "quick" issue6528 we are writing before the
             # 5.8 release
             textlen = mdiff.patchedsize(revlog.size(base_rev), delta)
 
             revinfo = revlogutils.revisioninfo(
                 node,
                 p1_node,
                 p2_node,
                 [None],
                 textlen,
                 (base_rev, delta),
                 flags,
             )
             # cached by the global "writing" context
             assert revlog._writinghandles is not None
             if revlog._inline:
                 fh = revlog._writinghandles[0]
             else:
                 fh = revlog._writinghandles[1]
             return deltacomputer.buildtext(revinfo, fh)
 
         is_affected = _is_revision_affected_fast_inner(
             is_censored,
             delta_base,
             lambda: delta,
             full_text,
             parent_revs,
             rev,
             metadata_cache,
         )
         if is_affected:
             d = (
                 node,
                 p2_node,
                 p1_node,
                 linknode,
                 deltabase,
                 delta,
                 flags,
                 sidedata,
             )
         yield d
 
 
 def repair_issue6528(
     ui, repo, dry_run=False, to_report=None, from_report=None, paranoid=False
 ):
     from .. import store  # avoid cycle
 
     @contextlib.contextmanager
     def context():
         if dry_run or to_report:  # No need for locking
             yield
         else:
             with repo.wlock(), repo.lock():
                 yield
 
     if from_report:
         return _from_report(ui, repo, context, from_report, dry_run)
 
     report_entries = []
 
     with context():
         files = list(
             (file_type, path)
-            for (file_type, path, _e, _s) in repo.store.datafiles()
+            for (file_type, path, _s) in repo.store.datafiles()
             if path.endswith(b'.i') and file_type & store.FILEFLAGS_FILELOG
         )
 
         progress = ui.makeprogress(
             _(b"looking for affected revisions"),
             unit=_(b"filelogs"),
             total=len(files),
         )
         found_nothing = True
 
         for file_type, path in files:
             if (
                 not path.endswith(b'.i')
                 or not file_type & store.FILEFLAGS_FILELOG
             ):
                 continue
             progress.increment()
             filename = _get_filename_from_filelog_index(path)
             fl = _filelog_from_filename(repo, filename)
 
             # Set of filerevs (or hex filenodes if `to_report`) that need fixing
             to_fix = set()
             metadata_cache = {}
             for filerev in fl.revs():
                 affected = _is_revision_affected_fast(
                     repo, fl, filerev, metadata_cache
                 )
                 if paranoid:
                     slow = _is_revision_affected(fl, filerev)
                     if slow != affected:
                         msg = _(b"paranoid check failed for '%s' at node %s")
                         node = binascii.hexlify(fl.node(filerev))
                         raise error.Abort(msg % (filename, node))
                 if affected:
                     msg = b"found affected revision %d for filelog '%s'\n"
                     ui.warn(msg % (filerev, path))
                     found_nothing = False
                     if not dry_run:
                         if to_report:
                             to_fix.add(binascii.hexlify(fl.node(filerev)))
                         else:
                             to_fix.add(filerev)
 
             if to_fix:
                 to_fix = sorted(to_fix)
                 if to_report:
                     report_entries.append((filename, to_fix))
                 else:
                     _reorder_filelog_parents(repo, fl, to_fix)
 
         if found_nothing:
             ui.write(_(b"no affected revisions were found\n"))
 
         if to_report and report_entries:
             with open(to_report, mode="wb") as f:
                 for path, to_fix in report_entries:
                     f.write(b"%s %s\n" % (b",".join(to_fix), path))
 
         progress.complete()

mercurial/store.py

@@ -1,839 +1,850 b''
 # store.py - repository store handling for Mercurial
 #
 # Copyright 2008 Olivia 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 functools
 import os
 import re
 import stat
 
 from .i18n import _
 from .pycompat import getattr
 from .node import hex
 from . import (
     changelog,
     error,
     manifest,
     policy,
     pycompat,
     util,
     vfs as vfsmod,
 )
 from .utils import hashutil
 
 parsers = policy.importmod('parsers')
 # how much bytes should be read from fncache in one read
 # It is done to prevent loading large fncache files into memory
 fncache_chunksize = 10 ** 6
 
 
 def _matchtrackedpath(path, matcher):
     """parses a fncache entry and returns whether the entry is tracking a path
     matched by matcher or not.
 
     If matcher is None, returns True"""
 
     if matcher is None:
         return True
     path = decodedir(path)
     if path.startswith(b'data/'):
         return matcher(path[len(b'data/') : -len(b'.i')])
     elif path.startswith(b'meta/'):
         return matcher.visitdir(path[len(b'meta/') : -len(b'/00manifest.i')])
 
     raise error.ProgrammingError(b"cannot decode path %s" % path)
 
 
 # This avoids a collision between a file named foo and a dir named
 # foo.i or foo.d
 def _encodedir(path):
     """
     >>> _encodedir(b'data/foo.i')
     'data/foo.i'
     >>> _encodedir(b'data/foo.i/bla.i')
     'data/foo.i.hg/bla.i'
     >>> _encodedir(b'data/foo.i.hg/bla.i')
     'data/foo.i.hg.hg/bla.i'
     >>> _encodedir(b'data/foo.i\\ndata/foo.i/bla.i\\ndata/foo.i.hg/bla.i\\n')
     'data/foo.i\\ndata/foo.i.hg/bla.i\\ndata/foo.i.hg.hg/bla.i\\n'
     """
     return (
         path.replace(b".hg/", b".hg.hg/")
         .replace(b".i/", b".i.hg/")
         .replace(b".d/", b".d.hg/")
     )
 
 
 encodedir = getattr(parsers, 'encodedir', _encodedir)
 
 
 def decodedir(path):
     """
     >>> decodedir(b'data/foo.i')
     'data/foo.i'
     >>> decodedir(b'data/foo.i.hg/bla.i')
     'data/foo.i/bla.i'
     >>> decodedir(b'data/foo.i.hg.hg/bla.i')
     'data/foo.i.hg/bla.i'
     """
     if b".hg/" not in path:
         return path
     return (
         path.replace(b".d.hg/", b".d/")
         .replace(b".i.hg/", b".i/")
         .replace(b".hg.hg/", b".hg/")
     )
 
 
 def _reserved():
     """characters that are problematic for filesystems
 
     * ascii escapes (0..31)
     * ascii hi (126..255)
     * windows specials
 
     these characters will be escaped by encodefunctions
     """
     winreserved = [ord(x) for x in u'\\:*?"<>|']
     for x in range(32):
         yield x
     for x in range(126, 256):
         yield x
     for x in winreserved:
         yield x
 
 
 def _buildencodefun():
     """
     >>> enc, dec = _buildencodefun()
 
     >>> enc(b'nothing/special.txt')
     'nothing/special.txt'
     >>> dec(b'nothing/special.txt')
     'nothing/special.txt'
 
     >>> enc(b'HELLO')
     '_h_e_l_l_o'
     >>> dec(b'_h_e_l_l_o')
     'HELLO'
 
     >>> enc(b'hello:world?')
     'hello~3aworld~3f'
     >>> dec(b'hello~3aworld~3f')
     'hello:world?'
 
     >>> enc(b'the\\x07quick\\xADshot')
     'the~07quick~adshot'
     >>> dec(b'the~07quick~adshot')
     'the\\x07quick\\xadshot'
     """
     e = b'_'
     xchr = pycompat.bytechr
     asciistr = list(map(xchr, range(127)))
     capitals = list(range(ord(b"A"), ord(b"Z") + 1))
 
     cmap = {x: x for x in asciistr}
     for x in _reserved():
         cmap[xchr(x)] = b"~%02x" % x
     for x in capitals + [ord(e)]:
         cmap[xchr(x)] = e + xchr(x).lower()
 
     dmap = {}
     for k, v in pycompat.iteritems(cmap):
         dmap[v] = k
 
     def decode(s):
         i = 0
         while i < len(s):
             for l in pycompat.xrange(1, 4):
                 try:
                     yield dmap[s[i : i + l]]
                     i += l
                     break
                 except KeyError:
                     pass
             else:
                 raise KeyError
 
     return (
         lambda s: b''.join(
             [cmap[s[c : c + 1]] for c in pycompat.xrange(len(s))]
         ),
         lambda s: b''.join(list(decode(s))),
     )
 
 
 _encodefname, _decodefname = _buildencodefun()
 
 
 def encodefilename(s):
     """
     >>> encodefilename(b'foo.i/bar.d/bla.hg/hi:world?/HELLO')
     'foo.i.hg/bar.d.hg/bla.hg.hg/hi~3aworld~3f/_h_e_l_l_o'
     """
     return _encodefname(encodedir(s))
 
 
 def decodefilename(s):
     """
     >>> decodefilename(b'foo.i.hg/bar.d.hg/bla.hg.hg/hi~3aworld~3f/_h_e_l_l_o')
     'foo.i/bar.d/bla.hg/hi:world?/HELLO'
     """
     return decodedir(_decodefname(s))
 
 
 def _buildlowerencodefun():
     """
     >>> f = _buildlowerencodefun()
     >>> f(b'nothing/special.txt')
     'nothing/special.txt'
     >>> f(b'HELLO')
     'hello'
     >>> f(b'hello:world?')
     'hello~3aworld~3f'
     >>> f(b'the\\x07quick\\xADshot')
     'the~07quick~adshot'
     """
     xchr = pycompat.bytechr
     cmap = {xchr(x): xchr(x) for x in pycompat.xrange(127)}
     for x in _reserved():
         cmap[xchr(x)] = b"~%02x" % x
     for x in range(ord(b"A"), ord(b"Z") + 1):
         cmap[xchr(x)] = xchr(x).lower()
 
     def lowerencode(s):
         return b"".join([cmap[c] for c in pycompat.iterbytestr(s)])
 
     return lowerencode
 
 
 lowerencode = getattr(parsers, 'lowerencode', None) or _buildlowerencodefun()
 
 # Windows reserved names: con, prn, aux, nul, com1..com9, lpt1..lpt9
 _winres3 = (b'aux', b'con', b'prn', b'nul')  # length 3
 _winres4 = (b'com', b'lpt')  # length 4 (with trailing 1..9)
 
 
 def _auxencode(path, dotencode):
     """
     Encodes filenames containing names reserved by Windows or which end in
     period or space. Does not touch other single reserved characters c.
     Specifically, c in '\\:*?"<>|' or ord(c) <= 31 are *not* encoded here.
     Additionally encodes space or period at the beginning, if dotencode is
     True. Parameter path is assumed to be all lowercase.
     A segment only needs encoding if a reserved name appears as a
     basename (e.g. "aux", "aux.foo"). A directory or file named "foo.aux"
     doesn't need encoding.
 
     >>> s = b'.foo/aux.txt/txt.aux/con/prn/nul/foo.'
     >>> _auxencode(s.split(b'/'), True)
     ['~2efoo', 'au~78.txt', 'txt.aux', 'co~6e', 'pr~6e', 'nu~6c', 'foo~2e']
     >>> s = b'.com1com2/lpt9.lpt4.lpt1/conprn/com0/lpt0/foo.'
     >>> _auxencode(s.split(b'/'), False)
     ['.com1com2', 'lp~749.lpt4.lpt1', 'conprn', 'com0', 'lpt0', 'foo~2e']
     >>> _auxencode([b'foo. '], True)
     ['foo.~20']
     >>> _auxencode([b' .foo'], True)
     ['~20.foo']
     """
     for i, n in enumerate(path):
         if not n:
             continue
         if dotencode and n[0] in b'. ':
             n = b"~%02x" % ord(n[0:1]) + n[1:]
             path[i] = n
         else:
             l = n.find(b'.')
             if l == -1:
                 l = len(n)
             if (l == 3 and n[:3] in _winres3) or (
                 l == 4
                 and n[3:4] <= b'9'
                 and n[3:4] >= b'1'
                 and n[:3] in _winres4
             ):
                 # encode third letter ('aux' -> 'au~78')
                 ec = b"~%02x" % ord(n[2:3])
                 n = n[0:2] + ec + n[3:]
                 path[i] = n
         if n[-1] in b'. ':
             # encode last period or space ('foo...' -> 'foo..~2e')
             path[i] = n[:-1] + b"~%02x" % ord(n[-1:])
     return path
 
 
 _maxstorepathlen = 120
 _dirprefixlen = 8
 _maxshortdirslen = 8 * (_dirprefixlen + 1) - 4
 
 
 def _hashencode(path, dotencode):
     digest = hex(hashutil.sha1(path).digest())
     le = lowerencode(path[5:]).split(b'/')  # skips prefix 'data/' or 'meta/'
     parts = _auxencode(le, dotencode)
     basename = parts[-1]
     _root, ext = os.path.splitext(basename)
     sdirs = []
     sdirslen = 0
     for p in parts[:-1]:
         d = p[:_dirprefixlen]
         if d[-1] in b'. ':
             # Windows can't access dirs ending in period or space
             d = d[:-1] + b'_'
         if sdirslen == 0:
             t = len(d)
         else:
             t = sdirslen + 1 + len(d)
             if t > _maxshortdirslen:
                 break
         sdirs.append(d)
         sdirslen = t
     dirs = b'/'.join(sdirs)
     if len(dirs) > 0:
         dirs += b'/'
     res = b'dh/' + dirs + digest + ext
     spaceleft = _maxstorepathlen - len(res)
     if spaceleft > 0:
         filler = basename[:spaceleft]
         res = b'dh/' + dirs + filler + digest + ext
     return res
 
 
 def _hybridencode(path, dotencode):
     """encodes path with a length limit
 
     Encodes all paths that begin with 'data/', according to the following.
 
     Default encoding (reversible):
 
     Encodes all uppercase letters 'X' as '_x'. All reserved or illegal
     characters are encoded as '~xx', where xx is the two digit hex code
     of the character (see encodefilename).
     Relevant path components consisting of Windows reserved filenames are
     masked by encoding the third character ('aux' -> 'au~78', see _auxencode).
 
     Hashed encoding (not reversible):
 
     If the default-encoded path is longer than _maxstorepathlen, a
     non-reversible hybrid hashing of the path is done instead.
     This encoding uses up to _dirprefixlen characters of all directory
     levels of the lowerencoded path, but not more levels than can fit into
     _maxshortdirslen.
     Then follows the filler followed by the sha digest of the full path.
     The filler is the beginning of the basename of the lowerencoded path
     (the basename is everything after the last path separator). The filler
     is as long as possible, filling in characters from the basename until
     the encoded path has _maxstorepathlen characters (or all chars of the
     basename have been taken).
     The extension (e.g. '.i' or '.d') is preserved.
 
     The string 'data/' at the beginning is replaced with 'dh/', if the hashed
     encoding was used.
     """
     path = encodedir(path)
     ef = _encodefname(path).split(b'/')
     res = b'/'.join(_auxencode(ef, dotencode))
     if len(res) > _maxstorepathlen:
         res = _hashencode(path, dotencode)
     return res
 
 
 def _pathencode(path):
     de = encodedir(path)
     if len(path) > _maxstorepathlen:
         return _hashencode(de, True)
     ef = _encodefname(de).split(b'/')
     res = b'/'.join(_auxencode(ef, True))
     if len(res) > _maxstorepathlen:
         return _hashencode(de, True)
     return res
 
 
 _pathencode = getattr(parsers, 'pathencode', _pathencode)
 
 
 def _plainhybridencode(f):
     return _hybridencode(f, False)
 
 
 def _calcmode(vfs):
     try:
         # files in .hg/ will be created using this mode
         mode = vfs.stat().st_mode
         # avoid some useless chmods
         if (0o777 & ~util.umask) == (0o777 & mode):
             mode = None
     except OSError:
         mode = None
     return mode
 
 
 _data = [
     b'bookmarks',
     b'narrowspec',
     b'data',
     b'meta',
     b'00manifest.d',
     b'00manifest.i',
     b'00changelog.d',
     b'00changelog.i',
     b'phaseroots',
     b'obsstore',
     b'requires',
 ]
 
 REVLOG_FILES_MAIN_EXT = (b'.i', b'i.tmpcensored')
 REVLOG_FILES_OTHER_EXT = (
     b'.idx',
     b'.d',
     b'.dat',
     b'.n',
     b'.nd',
     b'.sda',
     b'd.tmpcensored',
 )
 # files that are "volatile" and might change between listing and streaming
 #
 # note: the ".nd" file are nodemap data and won't "change" but they might be
 # deleted.
 REVLOG_FILES_VOLATILE_EXT = (b'.n', b'.nd')
 
 # some exception to the above matching
 #
 # XXX This is currently not in use because of issue6542
 EXCLUDED = re.compile(b'.*undo\.[^/]+\.(nd?|i)$')
 
 
 def is_revlog(f, kind, st):
     if kind != stat.S_IFREG:
         return None
     return revlog_type(f)
 
 
 def revlog_type(f):
     # XXX we need to filter `undo.` created by the transaction here, however
     # being naive about it also filter revlog for `undo.*` files, leading to
     # issue6542. So we no longer use EXCLUDED.
     if f.endswith(REVLOG_FILES_MAIN_EXT):
         return FILEFLAGS_REVLOG_MAIN
     elif f.endswith(REVLOG_FILES_OTHER_EXT):
         t = FILETYPE_FILELOG_OTHER
         if f.endswith(REVLOG_FILES_VOLATILE_EXT):
             t |= FILEFLAGS_VOLATILE
         return t
     return None
 
 
 # the file is part of changelog data
 FILEFLAGS_CHANGELOG = 1 << 13
 # the file is part of manifest data
 FILEFLAGS_MANIFESTLOG = 1 << 12
 # the file is part of filelog data
 FILEFLAGS_FILELOG = 1 << 11
 # file that are not directly part of a revlog
 FILEFLAGS_OTHER = 1 << 10
 
 # the main entry point for a revlog
 FILEFLAGS_REVLOG_MAIN = 1 << 1
 # a secondary file for a revlog
 FILEFLAGS_REVLOG_OTHER = 1 << 0
 
 # files that are "volatile" and might change between listing and streaming
 FILEFLAGS_VOLATILE = 1 << 20
 
 FILETYPE_CHANGELOG_MAIN = FILEFLAGS_CHANGELOG | FILEFLAGS_REVLOG_MAIN
 FILETYPE_CHANGELOG_OTHER = FILEFLAGS_CHANGELOG | FILEFLAGS_REVLOG_OTHER
 FILETYPE_MANIFESTLOG_MAIN = FILEFLAGS_MANIFESTLOG | FILEFLAGS_REVLOG_MAIN
 FILETYPE_MANIFESTLOG_OTHER = FILEFLAGS_MANIFESTLOG | FILEFLAGS_REVLOG_OTHER
 FILETYPE_FILELOG_MAIN = FILEFLAGS_FILELOG | FILEFLAGS_REVLOG_MAIN
 FILETYPE_FILELOG_OTHER = FILEFLAGS_FILELOG | FILEFLAGS_REVLOG_OTHER
 FILETYPE_OTHER = FILEFLAGS_OTHER
 
 
 class basicstore(object):
     '''base class for local repository stores'''
 
     def __init__(self, path, vfstype):
         vfs = vfstype(path)
         self.path = vfs.base
         self.createmode = _calcmode(vfs)
         vfs.createmode = self.createmode
         self.rawvfs = vfs
         self.vfs = vfsmod.filtervfs(vfs, encodedir)
         self.opener = self.vfs
 
     def join(self, f):
         return self.path + b'/' + encodedir(f)
 
     def _walk(self, relpath, recurse):
-        '''yields (unencoded, encoded, size)'''
+        '''yields (revlog_type, unencoded, size)'''
         path = self.path
         if relpath:
             path += b'/' + relpath
         striplen = len(self.path) + 1
         l = []
         if self.rawvfs.isdir(path):
             visit = [path]
             readdir = self.rawvfs.readdir
             while visit:
                 p = visit.pop()
                 for f, kind, st in readdir(p, stat=True):
                     fp = p + b'/' + f
                     rl_type = is_revlog(f, kind, st)
                     if rl_type is not None:
                         n = util.pconvert(fp[striplen:])
-                        l.append((rl_type, decodedir(n), n, st.st_size))
+                        l.append((rl_type, decodedir(n), st.st_size))
                     elif kind == stat.S_IFDIR and recurse:
                         visit.append(fp)
         l.sort()
         return l
 
     def changelog(self, trypending, concurrencychecker=None):
         return changelog.changelog(
             self.vfs,
             trypending=trypending,
             concurrencychecker=concurrencychecker,
         )
 
     def manifestlog(self, repo, storenarrowmatch):
         rootstore = manifest.manifestrevlog(repo.nodeconstants, self.vfs)
         return manifest.manifestlog(self.vfs, repo, rootstore, storenarrowmatch)
 
-    def datafiles(self, matcher=None):
+    def datafiles(self, matcher=None, undecodable=None):
+        """Like walk, but excluding the changelog and root manifest.
+
+        When [undecodable] is None, revlogs names that can't be
+        decoded cause an exception. When it is provided, it should
+        be a list and the filenames that can't be decoded are added
+        to it instead. This is very rarely needed."""
         files = self._walk(b'data', True) + self._walk(b'meta', True)
-        for (t, u, e, s) in files:
-            yield (FILEFLAGS_FILELOG | t, u, e, s)
+        for (t, u, s) in files:
+            yield (FILEFLAGS_FILELOG | t, u, s)
 
     def topfiles(self):
         # yield manifest before changelog
         files = reversed(self._walk(b'', False))
-        for (t, u, e, s) in files:
+        for (t, u, s) in files:
             if u.startswith(b'00changelog'):
-                yield (FILEFLAGS_CHANGELOG | t, u, e, s)
+                yield (FILEFLAGS_CHANGELOG | t, u, s)
             elif u.startswith(b'00manifest'):
-                yield (FILEFLAGS_MANIFESTLOG | t, u, e, s)
+                yield (FILEFLAGS_MANIFESTLOG | t, u, s)
             else:
-                yield (FILETYPE_OTHER | t, u, e, s)
+                yield (FILETYPE_OTHER | t, u, s)
 
     def walk(self, matcher=None):
         """return file related to data storage (ie: revlogs)
 
-        yields (file_type, unencoded, encoded, size)
+        yields (file_type, unencoded, size)
 
         if a matcher is passed, storage files of only those tracked paths
         are passed with matches the matcher
         """
         # yield data files first
         for x in self.datafiles(matcher):
             yield x
         for x in self.topfiles():
             yield x
 
     def copylist(self):
         return _data
 
     def write(self, tr):
         pass
 
     def invalidatecaches(self):
         pass
 
     def markremoved(self, fn):
         pass
 
     def __contains__(self, path):
         '''Checks if the store contains path'''
         path = b"/".join((b"data", path))
         # file?
         if self.vfs.exists(path + b".i"):
             return True
         # dir?
         if not path.endswith(b"/"):
             path = path + b"/"
         return self.vfs.exists(path)
 
 
 class encodedstore(basicstore):
     def __init__(self, path, vfstype):
         vfs = vfstype(path + b'/store')
         self.path = vfs.base
         self.createmode = _calcmode(vfs)
         vfs.createmode = self.createmode
         self.rawvfs = vfs
         self.vfs = vfsmod.filtervfs(vfs, encodefilename)
         self.opener = self.vfs
 
     # note: topfiles would also need a decode phase. It is just that in
     # practice we do not have any file outside of `data/` that needs encoding.
     # However that might change so we should probably add a test and encoding
     # decoding for it too. see issue6548
 
-    def datafiles(self, matcher=None):
-        for t, a, b, size in super(encodedstore, self).datafiles():
+    def datafiles(self, matcher=None, undecodable=None):
+        for t, f1, size in super(encodedstore, self).datafiles():
             try:
-                a = decodefilename(a)
+                f2 = decodefilename(f1)
             except KeyError:
-                a = None
-            if a is not None and not _matchtrackedpath(a, matcher):
+                if undecodable is None:
+                    msg = _(b'undecodable revlog name %s') % f1
+                    raise error.StorageError(msg)
+                else:
+                    undecodable.append(f1)
+                    continue
+            if not _matchtrackedpath(f2, matcher):
                 continue
-            yield t, a, b, size
+            yield t, f2, size
 
     def join(self, f):
         return self.path + b'/' + encodefilename(f)
 
     def copylist(self):
         return [b'requires', b'00changelog.i'] + [b'store/' + f for f in _data]
 
 
 class fncache(object):
     # the filename used to be partially encoded
     # hence the encodedir/decodedir dance
     def __init__(self, vfs):
         self.vfs = vfs
         self.entries = None
         self._dirty = False
         # set of new additions to fncache
         self.addls = set()
 
     def ensureloaded(self, warn=None):
         """read the fncache file if not already read.
 
         If the file on disk is corrupted, raise. If warn is provided,
         warn and keep going instead."""
         if self.entries is None:
             self._load(warn)
 
     def _load(self, warn=None):
         '''fill the entries from the fncache file'''
         self._dirty = False
         try:
             fp = self.vfs(b'fncache', mode=b'rb')
         except IOError:
             # skip nonexistent file
             self.entries = set()
             return
 
         self.entries = set()
         chunk = b''
         for c in iter(functools.partial(fp.read, fncache_chunksize), b''):
             chunk += c
             try:
                 p = chunk.rindex(b'\n')
                 self.entries.update(decodedir(chunk[: p + 1]).splitlines())
                 chunk = chunk[p + 1 :]
             except ValueError:
                 # substring '\n' not found, maybe the entry is bigger than the
                 # chunksize, so let's keep iterating
                 pass
 
         if chunk:
             msg = _(b"fncache does not ends with a newline")
             if warn:
                 warn(msg + b'\n')
             else:
                 raise error.Abort(
                     msg,
                     hint=_(
                         b"use 'hg debugrebuildfncache' to "
                         b"rebuild the fncache"
                     ),
                 )
         self._checkentries(fp, warn)
         fp.close()
 
     def _checkentries(self, fp, warn):
         """make sure there is no empty string in entries"""
         if b'' in self.entries:
             fp.seek(0)
             for n, line in enumerate(util.iterfile(fp)):
                 if not line.rstrip(b'\n'):
                     t = _(b'invalid entry in fncache, line %d') % (n + 1)
                     if warn:
                         warn(t + b'\n')
                     else:
                         raise error.Abort(t)
 
     def write(self, tr):
         if self._dirty:
             assert self.entries is not None
             self.entries = self.entries | self.addls
             self.addls = set()
             tr.addbackup(b'fncache')
             fp = self.vfs(b'fncache', mode=b'wb', atomictemp=True)
             if self.entries:
                 fp.write(encodedir(b'\n'.join(self.entries) + b'\n'))
             fp.close()
             self._dirty = False
         if self.addls:
             # if we have just new entries, let's append them to the fncache
             tr.addbackup(b'fncache')
             fp = self.vfs(b'fncache', mode=b'ab', atomictemp=True)
             if self.addls:
                 fp.write(encodedir(b'\n'.join(self.addls) + b'\n'))
             fp.close()
             self.entries = None
             self.addls = set()
 
     def add(self, fn):
         if self.entries is None:
             self._load()
         if fn not in self.entries:
             self.addls.add(fn)
 
     def remove(self, fn):
         if self.entries is None:
             self._load()
         if fn in self.addls:
             self.addls.remove(fn)
             return
         try:
             self.entries.remove(fn)
             self._dirty = True
         except KeyError:
             pass
 
     def __contains__(self, fn):
         if fn in self.addls:
             return True
         if self.entries is None:
             self._load()
         return fn in self.entries
 
     def __iter__(self):
         if self.entries is None:
             self._load()
         return iter(self.entries | self.addls)
 
 
 class _fncachevfs(vfsmod.proxyvfs):
     def __init__(self, vfs, fnc, encode):
         vfsmod.proxyvfs.__init__(self, vfs)
         self.fncache = fnc
         self.encode = encode
 
     def __call__(self, path, mode=b'r', *args, **kw):
         encoded = self.encode(path)
         if mode not in (b'r', b'rb') and (
             path.startswith(b'data/') or path.startswith(b'meta/')
         ):
             # do not trigger a fncache load when adding a file that already is
             # known to exist.
             notload = self.fncache.entries is None and self.vfs.exists(encoded)
             if notload and b'r+' in mode and not self.vfs.stat(encoded).st_size:
                 # when appending to an existing file, if the file has size zero,
                 # it should be considered as missing. Such zero-size files are
                 # the result of truncation when a transaction is aborted.
                 notload = False
             if not notload:
                 self.fncache.add(path)
         return self.vfs(encoded, mode, *args, **kw)
 
     def join(self, path):
         if path:
             return self.vfs.join(self.encode(path))
         else:
             return self.vfs.join(path)
 
     def register_file(self, path):
         """generic hook point to lets fncache steer its stew"""
         if path.startswith(b'data/') or path.startswith(b'meta/'):
             self.fncache.add(path)
 
 
 class fncachestore(basicstore):
     def __init__(self, path, vfstype, dotencode):
         if dotencode:
             encode = _pathencode
         else:
             encode = _plainhybridencode
         self.encode = encode
         vfs = vfstype(path + b'/store')
         self.path = vfs.base
         self.pathsep = self.path + b'/'
         self.createmode = _calcmode(vfs)
         vfs.createmode = self.createmode
         self.rawvfs = vfs
         fnc = fncache(vfs)
         self.fncache = fnc
         self.vfs = _fncachevfs(vfs, fnc, encode)
         self.opener = self.vfs
 
     def join(self, f):
         return self.pathsep + self.encode(f)
 
     def getsize(self, path):
         return self.rawvfs.stat(path).st_size
 
-    def datafiles(self, matcher=None):
+    def datafiles(self, matcher=None, undecodable=None):
         for f in sorted(self.fncache):
             if not _matchtrackedpath(f, matcher):
                 continue
             ef = self.encode(f)
             try:
                 t = revlog_type(f)
                 assert t is not None, f
                 t |= FILEFLAGS_FILELOG
-                yield t, f, ef, self.getsize(ef)
+                yield t, f, self.getsize(ef)
             except OSError as err:
                 if err.errno != errno.ENOENT:
                     raise
 
     def copylist(self):
         d = (
             b'bookmarks',
             b'narrowspec',
             b'data',
             b'meta',
             b'dh',
             b'fncache',
             b'phaseroots',
             b'obsstore',
             b'00manifest.d',
             b'00manifest.i',
             b'00changelog.d',
             b'00changelog.i',
             b'requires',
         )
         return [b'requires', b'00changelog.i'] + [b'store/' + f for f in d]
 
     def write(self, tr):
         self.fncache.write(tr)
 
     def invalidatecaches(self):
         self.fncache.entries = None
         self.fncache.addls = set()
 
     def markremoved(self, fn):
         self.fncache.remove(fn)
 
     def _exists(self, f):
         ef = self.encode(f)
         try:
             self.getsize(ef)
             return True
         except OSError as err:
             if err.errno != errno.ENOENT:
                 raise
             # nonexistent entry
             return False
 
     def __contains__(self, path):
         '''Checks if the store contains path'''
         path = b"/".join((b"data", path))
         # check for files (exact match)
         e = path + b'.i'
         if e in self.fncache and self._exists(e):
             return True
         # now check for directories (prefix match)
         if not path.endswith(b'/'):
             path += b'/'
         for e in self.fncache:
             if e.startswith(path) and self._exists(e):
                 return True
         return False

mercurial/streamclone.py

@@ -1,918 +1,918 b''
 # streamclone.py - producing and consuming streaming repository data
 #
 # Copyright 2015 Gregory Szorc <[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 contextlib
 import errno
 import os
 import struct
 
 from .i18n import _
 from .pycompat import open
 from .interfaces import repository
 from . import (
     bookmarks,
     cacheutil,
     error,
     narrowspec,
     phases,
     pycompat,
     requirements as requirementsmod,
     scmutil,
     store,
     util,
 )
 from .utils import (
     stringutil,
 )
 
 
 def canperformstreamclone(pullop, bundle2=False):
     """Whether it is possible to perform a streaming clone as part of pull.
 
     ``bundle2`` will cause the function to consider stream clone through
     bundle2 and only through bundle2.
 
     Returns a tuple of (supported, requirements). ``supported`` is True if
     streaming clone is supported and False otherwise. ``requirements`` is
     a set of repo requirements from the remote, or ``None`` if stream clone
     isn't supported.
     """
     repo = pullop.repo
     remote = pullop.remote
 
     bundle2supported = False
     if pullop.canusebundle2:
         if b'v2' in pullop.remotebundle2caps.get(b'stream', []):
             bundle2supported = True
         # else
         # Server doesn't support bundle2 stream clone or doesn't support
         # the versions we support. Fall back and possibly allow legacy.
 
     # Ensures legacy code path uses available bundle2.
     if bundle2supported and not bundle2:
         return False, None
     # Ensures bundle2 doesn't try to do a stream clone if it isn't supported.
     elif bundle2 and not bundle2supported:
         return False, None
 
     # Streaming clone only works on empty repositories.
     if len(repo):
         return False, None
 
     # Streaming clone only works if all data is being requested.
     if pullop.heads:
         return False, None
 
     streamrequested = pullop.streamclonerequested
 
     # If we don't have a preference, let the server decide for us. This
     # likely only comes into play in LANs.
     if streamrequested is None:
         # The server can advertise whether to prefer streaming clone.
         streamrequested = remote.capable(b'stream-preferred')
 
     if not streamrequested:
         return False, None
 
     # In order for stream clone to work, the client has to support all the
     # requirements advertised by the server.
     #
     # The server advertises its requirements via the "stream" and "streamreqs"
     # capability. "stream" (a value-less capability) is advertised if and only
     # if the only requirement is "revlogv1." Else, the "streamreqs" capability
     # is advertised and contains a comma-delimited list of requirements.
     requirements = set()
     if remote.capable(b'stream'):
         requirements.add(requirementsmod.REVLOGV1_REQUIREMENT)
     else:
         streamreqs = remote.capable(b'streamreqs')
         # This is weird and shouldn't happen with modern servers.
         if not streamreqs:
             pullop.repo.ui.warn(
                 _(
                     b'warning: stream clone requested but server has them '
                     b'disabled\n'
                 )
             )
             return False, None
 
         streamreqs = set(streamreqs.split(b','))
         # Server requires something we don't support. Bail.
         missingreqs = streamreqs - repo.supportedformats
         if missingreqs:
             pullop.repo.ui.warn(
                 _(
                     b'warning: stream clone requested but client is missing '
                     b'requirements: %s\n'
                 )
                 % b', '.join(sorted(missingreqs))
             )
             pullop.repo.ui.warn(
                 _(
                     b'(see https://www.mercurial-scm.org/wiki/MissingRequirement '
                     b'for more information)\n'
                 )
             )
             return False, None
         requirements = streamreqs
 
     return True, requirements
 
 
 def maybeperformlegacystreamclone(pullop):
     """Possibly perform a legacy stream clone operation.
 
     Legacy stream clones are performed as part of pull but before all other
     operations.
 
     A legacy stream clone will not be performed if a bundle2 stream clone is
     supported.
     """
     from . import localrepo
 
     supported, requirements = canperformstreamclone(pullop)
 
     if not supported:
         return
 
     repo = pullop.repo
     remote = pullop.remote
 
     # Save remote branchmap. We will use it later to speed up branchcache
     # creation.
     rbranchmap = None
     if remote.capable(b'branchmap'):
         with remote.commandexecutor() as e:
             rbranchmap = e.callcommand(b'branchmap', {}).result()
 
     repo.ui.status(_(b'streaming all changes\n'))
 
     with remote.commandexecutor() as e:
         fp = e.callcommand(b'stream_out', {}).result()
 
     # TODO strictly speaking, this code should all be inside the context
     # manager because the context manager is supposed to ensure all wire state
     # is flushed when exiting. But the legacy peers don't do this, so it
     # doesn't matter.
     l = fp.readline()
     try:
         resp = int(l)
     except ValueError:
         raise error.ResponseError(
             _(b'unexpected response from remote server:'), l
         )
     if resp == 1:
         raise error.Abort(_(b'operation forbidden by server'))
     elif resp == 2:
         raise error.Abort(_(b'locking the remote repository failed'))
     elif resp != 0:
         raise error.Abort(_(b'the server sent an unknown error code'))
 
     l = fp.readline()
     try:
         filecount, bytecount = map(int, l.split(b' ', 1))
     except (ValueError, TypeError):
         raise error.ResponseError(
             _(b'unexpected response from remote server:'), l
         )
 
     with repo.lock():
         consumev1(repo, fp, filecount, bytecount)
 
         # new requirements = old non-format requirements +
         #                    new format-related remote requirements
         # requirements from the streamed-in repository
         repo.requirements = requirements | (
             repo.requirements - repo.supportedformats
         )
         repo.svfs.options = localrepo.resolvestorevfsoptions(
             repo.ui, repo.requirements, repo.features
         )
         scmutil.writereporequirements(repo)
 
         if rbranchmap:
             repo._branchcaches.replace(repo, rbranchmap)
 
         repo.invalidate()
 
 
 def allowservergeneration(repo):
     """Whether streaming clones are allowed from the server."""
     if repository.REPO_FEATURE_STREAM_CLONE not in repo.features:
         return False
 
     if not repo.ui.configbool(b'server', b'uncompressed', untrusted=True):
         return False
 
     # The way stream clone works makes it impossible to hide secret changesets.
     # So don't allow this by default.
     secret = phases.hassecret(repo)
     if secret:
         return repo.ui.configbool(b'server', b'uncompressedallowsecret')
 
     return True
 
 
 # This is it's own function so extensions can override it.
 def _walkstreamfiles(repo, matcher=None):
     return repo.store.walk(matcher)
 
 
 def generatev1(repo):
     """Emit content for version 1 of a streaming clone.
 
     This returns a 3-tuple of (file count, byte size, data iterator).
 
     The data iterator consists of N entries for each file being transferred.
     Each file entry starts as a line with the file name and integer size
     delimited by a null byte.
 
     The raw file data follows. Following the raw file data is the next file
     entry, or EOF.
 
     When used on the wire protocol, an additional line indicating protocol
     success will be prepended to the stream. This function is not responsible
     for adding it.
 
     This function will obtain a repository lock to ensure a consistent view of
     the store is captured. It therefore may raise LockError.
     """
     entries = []
     total_bytes = 0
     # Get consistent snapshot of repo, lock during scan.
     with repo.lock():
         repo.ui.debug(b'scanning\n')
-        for file_type, name, ename, size in _walkstreamfiles(repo):
+        for file_type, name, size in _walkstreamfiles(repo):
             if size:
                 entries.append((name, size))
                 total_bytes += size
         _test_sync_point_walk_1(repo)
     _test_sync_point_walk_2(repo)
 
     repo.ui.debug(
         b'%d files, %d bytes to transfer\n' % (len(entries), total_bytes)
     )
 
     svfs = repo.svfs
     debugflag = repo.ui.debugflag
 
     def emitrevlogdata():
         for name, size in entries:
             if debugflag:
                 repo.ui.debug(b'sending %s (%d bytes)\n' % (name, size))
             # partially encode name over the wire for backwards compat
             yield b'%s\0%d\n' % (store.encodedir(name), size)
             # auditing at this stage is both pointless (paths are already
             # trusted by the local repo) and expensive
             with svfs(name, b'rb', auditpath=False) as fp:
                 if size <= 65536:
                     yield fp.read(size)
                 else:
                     for chunk in util.filechunkiter(fp, limit=size):
                         yield chunk
 
     return len(entries), total_bytes, emitrevlogdata()
 
 
 def generatev1wireproto(repo):
     """Emit content for version 1 of streaming clone suitable for the wire.
 
     This is the data output from ``generatev1()`` with 2 header lines. The
     first line indicates overall success. The 2nd contains the file count and
     byte size of payload.
 
     The success line contains "0" for success, "1" for stream generation not
     allowed, and "2" for error locking the repository (possibly indicating
     a permissions error for the server process).
     """
     if not allowservergeneration(repo):
         yield b'1\n'
         return
 
     try:
         filecount, bytecount, it = generatev1(repo)
     except error.LockError:
         yield b'2\n'
         return
 
     # Indicates successful response.
     yield b'0\n'
     yield b'%d %d\n' % (filecount, bytecount)
     for chunk in it:
         yield chunk
 
 
 def generatebundlev1(repo, compression=b'UN'):
     """Emit content for version 1 of a stream clone bundle.
 
     The first 4 bytes of the output ("HGS1") denote this as stream clone
     bundle version 1.
 
     The next 2 bytes indicate the compression type. Only "UN" is currently
     supported.
 
     The next 16 bytes are two 64-bit big endian unsigned integers indicating
     file count and byte count, respectively.
 
     The next 2 bytes is a 16-bit big endian unsigned short declaring the length
     of the requirements string, including a trailing \0. The following N bytes
     are the requirements string, which is ASCII containing a comma-delimited
     list of repo requirements that are needed to support the data.
 
     The remaining content is the output of ``generatev1()`` (which may be
     compressed in the future).
 
     Returns a tuple of (requirements, data generator).
     """
     if compression != b'UN':
         raise ValueError(b'we do not support the compression argument yet')
 
     requirements = repo.requirements & repo.supportedformats
     requires = b','.join(sorted(requirements))
 
     def gen():
         yield b'HGS1'
         yield compression
 
         filecount, bytecount, it = generatev1(repo)
         repo.ui.status(
             _(b'writing %d bytes for %d files\n') % (bytecount, filecount)
         )
 
         yield struct.pack(b'>QQ', filecount, bytecount)
         yield struct.pack(b'>H', len(requires) + 1)
         yield requires + b'\0'
 
         # This is where we'll add compression in the future.
         assert compression == b'UN'
 
         progress = repo.ui.makeprogress(
             _(b'bundle'), total=bytecount, unit=_(b'bytes')
         )
         progress.update(0)
 
         for chunk in it:
             progress.increment(step=len(chunk))
             yield chunk
 
         progress.complete()
 
     return requirements, gen()
 
 
 def consumev1(repo, fp, filecount, bytecount):
     """Apply the contents from version 1 of a streaming clone file handle.
 
     This takes the output from "stream_out" and applies it to the specified
     repository.
 
     Like "stream_out," the status line added by the wire protocol is not
     handled by this function.
     """
     with repo.lock():
         repo.ui.status(
             _(b'%d files to transfer, %s of data\n')
             % (filecount, util.bytecount(bytecount))
         )
         progress = repo.ui.makeprogress(
             _(b'clone'), total=bytecount, unit=_(b'bytes')
         )
         progress.update(0)
         start = util.timer()
 
         # TODO: get rid of (potential) inconsistency
         #
         # If transaction is started and any @filecache property is
         # changed at this point, it causes inconsistency between
         # in-memory cached property and streamclone-ed file on the
         # disk. Nested transaction prevents transaction scope "clone"
         # below from writing in-memory changes out at the end of it,
         # even though in-memory changes are discarded at the end of it
         # regardless of transaction nesting.
         #
         # But transaction nesting can't be simply prohibited, because
         # nesting occurs also in ordinary case (e.g. enabling
         # clonebundles).
 
         with repo.transaction(b'clone'):
             with repo.svfs.backgroundclosing(repo.ui, expectedcount=filecount):
                 for i in pycompat.xrange(filecount):
                     # XXX doesn't support '\n' or '\r' in filenames
                     l = fp.readline()
                     try:
                         name, size = l.split(b'\0', 1)
                         size = int(size)
                     except (ValueError, TypeError):
                         raise error.ResponseError(
                             _(b'unexpected response from remote server:'), l
                         )
                     if repo.ui.debugflag:
                         repo.ui.debug(
                             b'adding %s (%s)\n' % (name, util.bytecount(size))
                         )
                     # for backwards compat, name was partially encoded
                     path = store.decodedir(name)
                     with repo.svfs(path, b'w', backgroundclose=True) as ofp:
                         for chunk in util.filechunkiter(fp, limit=size):
                             progress.increment(step=len(chunk))
                             ofp.write(chunk)
 
             # force @filecache properties to be reloaded from
             # streamclone-ed file at next access
             repo.invalidate(clearfilecache=True)
 
         elapsed = util.timer() - start
         if elapsed <= 0:
             elapsed = 0.001
         progress.complete()
         repo.ui.status(
             _(b'transferred %s in %.1f seconds (%s/sec)\n')
             % (
                 util.bytecount(bytecount),
                 elapsed,
                 util.bytecount(bytecount / elapsed),
             )
         )
 
 
 def readbundle1header(fp):
     compression = fp.read(2)
     if compression != b'UN':
         raise error.Abort(
             _(
                 b'only uncompressed stream clone bundles are '
                 b'supported; got %s'
             )
             % compression
         )
 
     filecount, bytecount = struct.unpack(b'>QQ', fp.read(16))
     requireslen = struct.unpack(b'>H', fp.read(2))[0]
     requires = fp.read(requireslen)
 
     if not requires.endswith(b'\0'):
         raise error.Abort(
             _(
                 b'malformed stream clone bundle: '
                 b'requirements not properly encoded'
             )
         )
 
     requirements = set(requires.rstrip(b'\0').split(b','))
 
     return filecount, bytecount, requirements
 
 
 def applybundlev1(repo, fp):
     """Apply the content from a stream clone bundle version 1.
 
     We assume the 4 byte header has been read and validated and the file handle
     is at the 2 byte compression identifier.
     """
     if len(repo):
         raise error.Abort(
             _(b'cannot apply stream clone bundle on non-empty repo')
         )
 
     filecount, bytecount, requirements = readbundle1header(fp)
     missingreqs = requirements - repo.supportedformats
     if missingreqs:
         raise error.Abort(
             _(b'unable to apply stream clone: unsupported format: %s')
             % b', '.join(sorted(missingreqs))
         )
 
     consumev1(repo, fp, filecount, bytecount)
 
 
 class streamcloneapplier(object):
     """Class to manage applying streaming clone bundles.
 
     We need to wrap ``applybundlev1()`` in a dedicated type to enable bundle
     readers to perform bundle type-specific functionality.
     """
 
     def __init__(self, fh):
         self._fh = fh
 
     def apply(self, repo):
         return applybundlev1(repo, self._fh)
 
 
 # type of file to stream
 _fileappend = 0  # append only file
 _filefull = 1  # full snapshot file
 
 # Source of the file
 _srcstore = b's'  # store (svfs)
 _srccache = b'c'  # cache (cache)
 
 # This is it's own function so extensions can override it.
 def _walkstreamfullstorefiles(repo):
     """list snapshot file from the store"""
     fnames = []
     if not repo.publishing():
         fnames.append(b'phaseroots')
     return fnames
 
 
 def _filterfull(entry, copy, vfsmap):
     """actually copy the snapshot files"""
     src, name, ftype, data = entry
     if ftype != _filefull:
         return entry
     return (src, name, ftype, copy(vfsmap[src].join(name)))
 
 
 @contextlib.contextmanager
 def maketempcopies():
     """return a function to temporary copy file"""
     files = []
     try:
 
         def copy(src):
             fd, dst = pycompat.mkstemp()
             os.close(fd)
             files.append(dst)
             util.copyfiles(src, dst, hardlink=True)
             return dst
 
         yield copy
     finally:
         for tmp in files:
             util.tryunlink(tmp)
 
 
 def _makemap(repo):
     """make a (src -> vfs) map for the repo"""
     vfsmap = {
         _srcstore: repo.svfs,
         _srccache: repo.cachevfs,
     }
     # we keep repo.vfs out of the on purpose, ther are too many danger there
     # (eg: .hg/hgrc)
     assert repo.vfs not in vfsmap.values()
 
     return vfsmap
 
 
 def _emit2(repo, entries, totalfilesize):
     """actually emit the stream bundle"""
     vfsmap = _makemap(repo)
     # we keep repo.vfs out of the on purpose, ther are too many danger there
     # (eg: .hg/hgrc),
     #
     # this assert is duplicated (from _makemap) as author might think this is
     # fine, while this is really not fine.
     if repo.vfs in vfsmap.values():
         raise error.ProgrammingError(
             b'repo.vfs must not be added to vfsmap for security reasons'
         )
 
     progress = repo.ui.makeprogress(
         _(b'bundle'), total=totalfilesize, unit=_(b'bytes')
     )
     progress.update(0)
     with maketempcopies() as copy, progress:
         # copy is delayed until we are in the try
         entries = [_filterfull(e, copy, vfsmap) for e in entries]
         yield None  # this release the lock on the repository
         totalbytecount = 0
 
         for src, name, ftype, data in entries:
             vfs = vfsmap[src]
             yield src
             yield util.uvarintencode(len(name))
             if ftype == _fileappend:
                 fp = vfs(name)
                 size = data
             elif ftype == _filefull:
                 fp = open(data, b'rb')
                 size = util.fstat(fp).st_size
             bytecount = 0
             try:
                 yield util.uvarintencode(size)
                 yield name
                 if size <= 65536:
                     chunks = (fp.read(size),)
                 else:
                     chunks = util.filechunkiter(fp, limit=size)
                 for chunk in chunks:
                     bytecount += len(chunk)
                     totalbytecount += len(chunk)
                     progress.update(totalbytecount)
                     yield chunk
                 if bytecount != size:
                     # Would most likely be caused by a race due to `hg strip` or
                     # a revlog split
                     raise error.Abort(
                         _(
                             b'clone could only read %d bytes from %s, but '
                             b'expected %d bytes'
                         )
                         % (bytecount, name, size)
                     )
             finally:
                 fp.close()
 
 
 def _test_sync_point_walk_1(repo):
     """a function for synchronisation during tests"""
 
 
 def _test_sync_point_walk_2(repo):
     """a function for synchronisation during tests"""
 
 
 def _v2_walk(repo, includes, excludes, includeobsmarkers):
     """emit a seris of files information useful to clone a repo
 
     return (entries, totalfilesize)
 
     entries is a list of tuple (vfs-key, file-path, file-type, size)
 
     - `vfs-key`: is a key to the right vfs to write the file (see _makemap)
     - `name`: file path of the file to copy (to be feed to the vfss)
     - `file-type`: do this file need to be copied with the source lock ?
     - `size`: the size of the file (or None)
     """
     assert repo._currentlock(repo._lockref) is not None
     entries = []
     totalfilesize = 0
 
     matcher = None
     if includes or excludes:
         matcher = narrowspec.match(repo.root, includes, excludes)
 
-    for rl_type, name, ename, size in _walkstreamfiles(repo, matcher):
+    for rl_type, name, size in _walkstreamfiles(repo, matcher):
         if size:
             ft = _fileappend
             if rl_type & store.FILEFLAGS_VOLATILE:
                 ft = _filefull
             entries.append((_srcstore, name, ft, size))
             totalfilesize += size
     for name in _walkstreamfullstorefiles(repo):
         if repo.svfs.exists(name):
             totalfilesize += repo.svfs.lstat(name).st_size
             entries.append((_srcstore, name, _filefull, None))
     if includeobsmarkers and repo.svfs.exists(b'obsstore'):
         totalfilesize += repo.svfs.lstat(b'obsstore').st_size
         entries.append((_srcstore, b'obsstore', _filefull, None))
     for name in cacheutil.cachetocopy(repo):
         if repo.cachevfs.exists(name):
             totalfilesize += repo.cachevfs.lstat(name).st_size
             entries.append((_srccache, name, _filefull, None))
     return entries, totalfilesize
 
 
 def generatev2(repo, includes, excludes, includeobsmarkers):
     """Emit content for version 2 of a streaming clone.
 
     the data stream consists the following entries:
     1) A char representing the file destination (eg: store or cache)
     2) A varint containing the length of the filename
     3) A varint containing the length of file data
     4) N bytes containing the filename (the internal, store-agnostic form)
     5) N bytes containing the file data
 
     Returns a 3-tuple of (file count, file size, data iterator).
     """
 
     with repo.lock():
 
         repo.ui.debug(b'scanning\n')
 
         entries, totalfilesize = _v2_walk(
             repo,
             includes=includes,
             excludes=excludes,
             includeobsmarkers=includeobsmarkers,
         )
 
         chunks = _emit2(repo, entries, totalfilesize)
         first = next(chunks)
         assert first is None
         _test_sync_point_walk_1(repo)
     _test_sync_point_walk_2(repo)
 
     return len(entries), totalfilesize, chunks
 
 
 @contextlib.contextmanager
 def nested(*ctxs):
     this = ctxs[0]
     rest = ctxs[1:]
     with this:
         if rest:
             with nested(*rest):
                 yield
         else:
             yield
 
 
 def consumev2(repo, fp, filecount, filesize):
     """Apply the contents from a version 2 streaming clone.
 
     Data is read from an object that only needs to provide a ``read(size)``
     method.
     """
     with repo.lock():
         repo.ui.status(
             _(b'%d files to transfer, %s of data\n')
             % (filecount, util.bytecount(filesize))
         )
 
         start = util.timer()
         progress = repo.ui.makeprogress(
             _(b'clone'), total=filesize, unit=_(b'bytes')
         )
         progress.update(0)
 
         vfsmap = _makemap(repo)
         # we keep repo.vfs out of the on purpose, ther are too many danger
         # there (eg: .hg/hgrc),
         #
         # this assert is duplicated (from _makemap) as author might think this
         # is fine, while this is really not fine.
         if repo.vfs in vfsmap.values():
             raise error.ProgrammingError(
                 b'repo.vfs must not be added to vfsmap for security reasons'
             )
 
         with repo.transaction(b'clone'):
             ctxs = (vfs.backgroundclosing(repo.ui) for vfs in vfsmap.values())
             with nested(*ctxs):
                 for i in range(filecount):
                     src = util.readexactly(fp, 1)
                     vfs = vfsmap[src]
                     namelen = util.uvarintdecodestream(fp)
                     datalen = util.uvarintdecodestream(fp)
 
                     name = util.readexactly(fp, namelen)
 
                     if repo.ui.debugflag:
                         repo.ui.debug(
                             b'adding [%s] %s (%s)\n'
                             % (src, name, util.bytecount(datalen))
                         )
 
                     with vfs(name, b'w') as ofp:
                         for chunk in util.filechunkiter(fp, limit=datalen):
                             progress.increment(step=len(chunk))
                             ofp.write(chunk)
 
             # force @filecache properties to be reloaded from
             # streamclone-ed file at next access
             repo.invalidate(clearfilecache=True)
 
         elapsed = util.timer() - start
         if elapsed <= 0:
             elapsed = 0.001
         repo.ui.status(
             _(b'transferred %s in %.1f seconds (%s/sec)\n')
             % (
                 util.bytecount(progress.pos),
                 elapsed,
                 util.bytecount(progress.pos / elapsed),
             )
         )
         progress.complete()
 
 
 def applybundlev2(repo, fp, filecount, filesize, requirements):
     from . import localrepo
 
     missingreqs = [r for r in requirements if r not in repo.supported]
     if missingreqs:
         raise error.Abort(
             _(b'unable to apply stream clone: unsupported format: %s')
             % b', '.join(sorted(missingreqs))
         )
 
     consumev2(repo, fp, filecount, filesize)
 
     # new requirements = old non-format requirements +
     #                    new format-related remote requirements
     # requirements from the streamed-in repository
     repo.requirements = set(requirements) | (
         repo.requirements - repo.supportedformats
     )
     repo.svfs.options = localrepo.resolvestorevfsoptions(
         repo.ui, repo.requirements, repo.features
     )
     scmutil.writereporequirements(repo)
 
 
 def _copy_files(src_vfs_map, dst_vfs_map, entries, progress):
     hardlink = [True]
 
     def copy_used():
         hardlink[0] = False
         progress.topic = _(b'copying')
 
     for k, path, size in entries:
         src_vfs = src_vfs_map[k]
         dst_vfs = dst_vfs_map[k]
         src_path = src_vfs.join(path)
         dst_path = dst_vfs.join(path)
         dirname = dst_vfs.dirname(path)
         if not dst_vfs.exists(dirname):
             dst_vfs.makedirs(dirname)
         dst_vfs.register_file(path)
         # XXX we could use the #nb_bytes argument.
         util.copyfile(
             src_path,
             dst_path,
             hardlink=hardlink[0],
             no_hardlink_cb=copy_used,
             check_fs_hardlink=False,
         )
         progress.increment()
     return hardlink[0]
 
 
 def local_copy(src_repo, dest_repo):
     """copy all content from one local repository to another
 
     This is useful for local clone"""
     src_store_requirements = {
         r
         for r in src_repo.requirements
         if r not in requirementsmod.WORKING_DIR_REQUIREMENTS
     }
     dest_store_requirements = {
         r
         for r in dest_repo.requirements
         if r not in requirementsmod.WORKING_DIR_REQUIREMENTS
     }
     assert src_store_requirements == dest_store_requirements
 
     with dest_repo.lock():
         with src_repo.lock():
 
             # bookmark is not integrated to the streaming as it might use the
             # `repo.vfs` and they are too many sentitive data accessible
             # through `repo.vfs` to expose it to streaming clone.
             src_book_vfs = bookmarks.bookmarksvfs(src_repo)
             srcbookmarks = src_book_vfs.join(b'bookmarks')
             bm_count = 0
             if os.path.exists(srcbookmarks):
                 bm_count = 1
 
             entries, totalfilesize = _v2_walk(
                 src_repo,
                 includes=None,
                 excludes=None,
                 includeobsmarkers=True,
             )
             src_vfs_map = _makemap(src_repo)
             dest_vfs_map = _makemap(dest_repo)
             progress = src_repo.ui.makeprogress(
                 topic=_(b'linking'),
                 total=len(entries) + bm_count,
                 unit=_(b'files'),
             )
             # copy  files
             #
             # We could copy the full file while the source repository is locked
             # and the other one without the lock. However, in the linking case,
             # this would also requires checks that nobody is appending any data
             # to the files while we do the clone, so this is not done yet. We
             # could do this blindly when copying files.
             files = ((k, path, size) for k, path, ftype, size in entries)
             hardlink = _copy_files(src_vfs_map, dest_vfs_map, files, progress)
 
             # copy bookmarks over
             if bm_count:
                 dst_book_vfs = bookmarks.bookmarksvfs(dest_repo)
                 dstbookmarks = dst_book_vfs.join(b'bookmarks')
                 util.copyfile(srcbookmarks, dstbookmarks)
         progress.complete()
         if hardlink:
             msg = b'linked %d files\n'
         else:
             msg = b'copied %d files\n'
         src_repo.ui.debug(msg % (len(entries) + bm_count))
 
         with dest_repo.transaction(b"localclone") as tr:
             dest_repo.store.write(tr)
 
         # clean up transaction file as they do not make sense
         undo_files = [(dest_repo.svfs, b'undo.backupfiles')]
         undo_files.extend(dest_repo.undofiles())
         for undovfs, undofile in undo_files:
             try:
                 undovfs.unlink(undofile)
             except OSError as e:
                 if e.errno != errno.ENOENT:
                     msg = _(b'error removing %s: %s\n')
                     path = undovfs.join(undofile)
                     e_msg = stringutil.forcebytestr(e)
                     msg %= (path, e_msg)
                     dest_repo.ui.warn(msg)

mercurial/upgrade_utils/engine.py

@@ -1,649 +1,649 b''
 # upgrade.py - functions for in place upgrade of Mercurial repository
 #
 # Copyright (c) 2016-present, Gregory Szorc
 #
 # 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 stat
 
 from ..i18n import _
 from ..pycompat import getattr
 from .. import (
     changelog,
     error,
     filelog,
     manifest,
     metadata,
     pycompat,
     requirements,
     scmutil,
     store,
     util,
     vfs as vfsmod,
 )
 from ..revlogutils import (
     constants as revlogconst,
     flagutil,
     nodemap,
     sidedata as sidedatamod,
 )
 from . import actions as upgrade_actions
 
 
 def get_sidedata_helpers(srcrepo, dstrepo):
     use_w = srcrepo.ui.configbool(b'experimental', b'worker.repository-upgrade')
     sequential = pycompat.iswindows or not use_w
     if not sequential:
         srcrepo.register_sidedata_computer(
             revlogconst.KIND_CHANGELOG,
             sidedatamod.SD_FILES,
             (sidedatamod.SD_FILES,),
             metadata._get_worker_sidedata_adder(srcrepo, dstrepo),
             flagutil.REVIDX_HASCOPIESINFO,
             replace=True,
         )
     return sidedatamod.get_sidedata_helpers(srcrepo, dstrepo._wanted_sidedata)
 
 
 def _revlogfrompath(repo, rl_type, path):
     """Obtain a revlog from a repo path.
 
     An instance of the appropriate class is returned.
     """
     if rl_type & store.FILEFLAGS_CHANGELOG:
         return changelog.changelog(repo.svfs)
     elif rl_type & store.FILEFLAGS_MANIFESTLOG:
         mandir = b''
         if b'/' in path:
             mandir = path.rsplit(b'/', 1)[0]
         return manifest.manifestrevlog(
             repo.nodeconstants, repo.svfs, tree=mandir
         )
     else:
         # drop the extension and the `data/` prefix
         path_part = path.rsplit(b'.', 1)[0].split(b'/', 1)
         if len(path_part) < 2:
             msg = _(b'cannot recognize revlog from filename: %s')
             msg %= path
             raise error.Abort(msg)
         path = path_part[1]
         return filelog.filelog(repo.svfs, path)
 
 
 def _copyrevlog(tr, destrepo, oldrl, rl_type, unencodedname):
     """copy all relevant files for `oldrl` into `destrepo` store
 
     Files are copied "as is" without any transformation. The copy is performed
     without extra checks. Callers are responsible for making sure the copied
     content is compatible with format of the destination repository.
     """
     oldrl = getattr(oldrl, '_revlog', oldrl)
     newrl = _revlogfrompath(destrepo, rl_type, unencodedname)
     newrl = getattr(newrl, '_revlog', newrl)
 
     oldvfs = oldrl.opener
     newvfs = newrl.opener
     oldindex = oldvfs.join(oldrl._indexfile)
     newindex = newvfs.join(newrl._indexfile)
     olddata = oldvfs.join(oldrl._datafile)
     newdata = newvfs.join(newrl._datafile)
 
     with newvfs(newrl._indexfile, b'w'):
         pass  # create all the directories
 
     util.copyfile(oldindex, newindex)
     copydata = oldrl.opener.exists(oldrl._datafile)
     if copydata:
         util.copyfile(olddata, newdata)
 
     if rl_type & store.FILEFLAGS_FILELOG:
         destrepo.svfs.fncache.add(unencodedname)
         if copydata:
             destrepo.svfs.fncache.add(unencodedname[:-2] + b'.d')
 
 
 UPGRADE_CHANGELOG = b"changelog"
 UPGRADE_MANIFEST = b"manifest"
 UPGRADE_FILELOGS = b"all-filelogs"
 
 UPGRADE_ALL_REVLOGS = frozenset(
     [UPGRADE_CHANGELOG, UPGRADE_MANIFEST, UPGRADE_FILELOGS]
 )
 
 
 def matchrevlog(revlogfilter, rl_type):
     """check if a revlog is selected for cloning.
 
     In other words, are there any updates which need to be done on revlog
     or it can be blindly copied.
 
     The store entry is checked against the passed filter"""
     if rl_type & store.FILEFLAGS_CHANGELOG:
         return UPGRADE_CHANGELOG in revlogfilter
     elif rl_type & store.FILEFLAGS_MANIFESTLOG:
         return UPGRADE_MANIFEST in revlogfilter
     assert rl_type & store.FILEFLAGS_FILELOG
     return UPGRADE_FILELOGS in revlogfilter
 
 
 def _perform_clone(
     ui,
     dstrepo,
     tr,
     old_revlog,
     rl_type,
     unencoded,
     upgrade_op,
     sidedata_helpers,
     oncopiedrevision,
 ):
     """returns the new revlog object created"""
     newrl = None
     if matchrevlog(upgrade_op.revlogs_to_process, rl_type):
         ui.note(
             _(b'cloning %d revisions from %s\n') % (len(old_revlog), unencoded)
         )
         newrl = _revlogfrompath(dstrepo, rl_type, unencoded)
         old_revlog.clone(
             tr,
             newrl,
             addrevisioncb=oncopiedrevision,
             deltareuse=upgrade_op.delta_reuse_mode,
             forcedeltabothparents=upgrade_op.force_re_delta_both_parents,
             sidedata_helpers=sidedata_helpers,
         )
     else:
         msg = _(b'blindly copying %s containing %i revisions\n')
         ui.note(msg % (unencoded, len(old_revlog)))
         _copyrevlog(tr, dstrepo, old_revlog, rl_type, unencoded)
 
         newrl = _revlogfrompath(dstrepo, rl_type, unencoded)
     return newrl
 
 
 def _clonerevlogs(
     ui,
     srcrepo,
     dstrepo,
     tr,
     upgrade_op,
 ):
     """Copy revlogs between 2 repos."""
     revcount = 0
     srcsize = 0
     srcrawsize = 0
     dstsize = 0
     fcount = 0
     frevcount = 0
     fsrcsize = 0
     frawsize = 0
     fdstsize = 0
     mcount = 0
     mrevcount = 0
     msrcsize = 0
     mrawsize = 0
     mdstsize = 0
     crevcount = 0
     csrcsize = 0
     crawsize = 0
     cdstsize = 0
 
     alldatafiles = list(srcrepo.store.walk())
     # mapping of data files which needs to be cloned
     # key is unencoded filename
     # value is revlog_object_from_srcrepo
     manifests = {}
     changelogs = {}
     filelogs = {}
 
     # Perform a pass to collect metadata. This validates we can open all
     # source files and allows a unified progress bar to be displayed.
-    for rl_type, unencoded, encoded, size in alldatafiles:
+    for rl_type, unencoded, size in alldatafiles:
         if not rl_type & store.FILEFLAGS_REVLOG_MAIN:
             continue
 
         # the store.walk function will wrongly pickup transaction backup and
         # get confused. As a quick fix for 5.9 release, we ignore those.
         # (this is not a module constants because it seems better to keep the
         # hack together)
         skip_undo = (
             b'undo.backup.00changelog.i',
             b'undo.backup.00manifest.i',
         )
         if unencoded in skip_undo:
             continue
 
         rl = _revlogfrompath(srcrepo, rl_type, unencoded)
 
         info = rl.storageinfo(
             exclusivefiles=True,
             revisionscount=True,
             trackedsize=True,
             storedsize=True,
         )
 
         revcount += info[b'revisionscount'] or 0
         datasize = info[b'storedsize'] or 0
         rawsize = info[b'trackedsize'] or 0
 
         srcsize += datasize
         srcrawsize += rawsize
 
         # This is for the separate progress bars.
         if rl_type & store.FILEFLAGS_CHANGELOG:
             changelogs[unencoded] = (rl_type, rl)
             crevcount += len(rl)
             csrcsize += datasize
             crawsize += rawsize
         elif rl_type & store.FILEFLAGS_MANIFESTLOG:
             manifests[unencoded] = (rl_type, rl)
             mcount += 1
             mrevcount += len(rl)
             msrcsize += datasize
             mrawsize += rawsize
         elif rl_type & store.FILEFLAGS_FILELOG:
             filelogs[unencoded] = (rl_type, rl)
             fcount += 1
             frevcount += len(rl)
             fsrcsize += datasize
             frawsize += rawsize
         else:
             error.ProgrammingError(b'unknown revlog type')
 
     if not revcount:
         return
 
     ui.status(
         _(
             b'migrating %d total revisions (%d in filelogs, %d in manifests, '
             b'%d in changelog)\n'
         )
         % (revcount, frevcount, mrevcount, crevcount)
     )
     ui.status(
         _(b'migrating %s in store; %s tracked data\n')
         % ((util.bytecount(srcsize), util.bytecount(srcrawsize)))
     )
 
     # Used to keep track of progress.
     progress = None
 
     def oncopiedrevision(rl, rev, node):
         progress.increment()
 
     sidedata_helpers = get_sidedata_helpers(srcrepo, dstrepo)
 
     # Migrating filelogs
     ui.status(
         _(
             b'migrating %d filelogs containing %d revisions '
             b'(%s in store; %s tracked data)\n'
         )
         % (
             fcount,
             frevcount,
             util.bytecount(fsrcsize),
             util.bytecount(frawsize),
         )
     )
     progress = srcrepo.ui.makeprogress(_(b'file revisions'), total=frevcount)
     for unencoded, (rl_type, oldrl) in sorted(filelogs.items()):
         newrl = _perform_clone(
             ui,
             dstrepo,
             tr,
             oldrl,
             rl_type,
             unencoded,
             upgrade_op,
             sidedata_helpers,
             oncopiedrevision,
         )
         info = newrl.storageinfo(storedsize=True)
         fdstsize += info[b'storedsize'] or 0
     ui.status(
         _(
             b'finished migrating %d filelog revisions across %d '
             b'filelogs; change in size: %s\n'
         )
         % (frevcount, fcount, util.bytecount(fdstsize - fsrcsize))
     )
 
     # Migrating manifests
     ui.status(
         _(
             b'migrating %d manifests containing %d revisions '
             b'(%s in store; %s tracked data)\n'
         )
         % (
             mcount,
             mrevcount,
             util.bytecount(msrcsize),
             util.bytecount(mrawsize),
         )
     )
     if progress:
         progress.complete()
     progress = srcrepo.ui.makeprogress(
         _(b'manifest revisions'), total=mrevcount
     )
     for unencoded, (rl_type, oldrl) in sorted(manifests.items()):
         newrl = _perform_clone(
             ui,
             dstrepo,
             tr,
             oldrl,
             rl_type,
             unencoded,
             upgrade_op,
             sidedata_helpers,
             oncopiedrevision,
         )
         info = newrl.storageinfo(storedsize=True)
         mdstsize += info[b'storedsize'] or 0
     ui.status(
         _(
             b'finished migrating %d manifest revisions across %d '
             b'manifests; change in size: %s\n'
         )
         % (mrevcount, mcount, util.bytecount(mdstsize - msrcsize))
     )
 
     # Migrating changelog
     ui.status(
         _(
             b'migrating changelog containing %d revisions '
             b'(%s in store; %s tracked data)\n'
         )
         % (
             crevcount,
             util.bytecount(csrcsize),
             util.bytecount(crawsize),
         )
     )
     if progress:
         progress.complete()
     progress = srcrepo.ui.makeprogress(
         _(b'changelog revisions'), total=crevcount
     )
     for unencoded, (rl_type, oldrl) in sorted(changelogs.items()):
         newrl = _perform_clone(
             ui,
             dstrepo,
             tr,
             oldrl,
             rl_type,
             unencoded,
             upgrade_op,
             sidedata_helpers,
             oncopiedrevision,
         )
         info = newrl.storageinfo(storedsize=True)
         cdstsize += info[b'storedsize'] or 0
     progress.complete()
     ui.status(
         _(
             b'finished migrating %d changelog revisions; change in size: '
             b'%s\n'
         )
         % (crevcount, util.bytecount(cdstsize - csrcsize))
     )
 
     dstsize = fdstsize + mdstsize + cdstsize
     ui.status(
         _(
             b'finished migrating %d total revisions; total change in store '
             b'size: %s\n'
         )
         % (revcount, util.bytecount(dstsize - srcsize))
     )
 
 
 def _files_to_copy_post_revlog_clone(srcrepo):
     """yields files which should be copied to destination after revlogs
     are cloned"""
     for path, kind, st in sorted(srcrepo.store.vfs.readdir(b'', stat=True)):
         # don't copy revlogs as they are already cloned
         if store.revlog_type(path) is not None:
             continue
         # Skip transaction related files.
         if path.startswith(b'undo'):
             continue
         # Only copy regular files.
         if kind != stat.S_IFREG:
             continue
         # Skip other skipped files.
         if path in (b'lock', b'fncache'):
             continue
         # TODO: should we skip cache too?
 
         yield path
 
 
 def _replacestores(currentrepo, upgradedrepo, backupvfs, upgrade_op):
     """Replace the stores after current repository is upgraded
 
     Creates a backup of current repository store at backup path
     Replaces upgraded store files in current repo from upgraded one
 
     Arguments:
       currentrepo: repo object of current repository
       upgradedrepo: repo object of the upgraded data
       backupvfs: vfs object for the backup path
       upgrade_op: upgrade operation object
                   to be used to decide what all is upgraded
     """
     # TODO: don't blindly rename everything in store
     # There can be upgrades where store is not touched at all
     if upgrade_op.backup_store:
         util.rename(currentrepo.spath, backupvfs.join(b'store'))
     else:
         currentrepo.vfs.rmtree(b'store', forcibly=True)
     util.rename(upgradedrepo.spath, currentrepo.spath)
 
 
 def finishdatamigration(ui, srcrepo, dstrepo, requirements):
     """Hook point for extensions to perform additional actions during upgrade.
 
     This function is called after revlogs and store files have been copied but
     before the new store is swapped into the original location.
     """
 
 
 def upgrade(ui, srcrepo, dstrepo, upgrade_op):
     """Do the low-level work of upgrading a repository.
 
     The upgrade is effectively performed as a copy between a source
     repository and a temporary destination repository.
 
     The source repository is unmodified for as long as possible so the
     upgrade can abort at any time without causing loss of service for
     readers and without corrupting the source repository.
     """
     assert srcrepo.currentwlock()
     assert dstrepo.currentwlock()
     backuppath = None
     backupvfs = None
 
     ui.status(
         _(
             b'(it is safe to interrupt this process any time before '
             b'data migration completes)\n'
         )
     )
 
     if upgrade_actions.dirstatev2 in upgrade_op.upgrade_actions:
         ui.status(_(b'upgrading to dirstate-v2 from v1\n'))
         upgrade_dirstate(ui, srcrepo, upgrade_op, b'v1', b'v2')
         upgrade_op.upgrade_actions.remove(upgrade_actions.dirstatev2)
 
     if upgrade_actions.dirstatev2 in upgrade_op.removed_actions:
         ui.status(_(b'downgrading from dirstate-v2 to v1\n'))
         upgrade_dirstate(ui, srcrepo, upgrade_op, b'v2', b'v1')
         upgrade_op.removed_actions.remove(upgrade_actions.dirstatev2)
 
     if not (upgrade_op.upgrade_actions or upgrade_op.removed_actions):
         return
 
     if upgrade_op.requirements_only:
         ui.status(_(b'upgrading repository requirements\n'))
         scmutil.writereporequirements(srcrepo, upgrade_op.new_requirements)
     # if there is only one action and that is persistent nodemap upgrade
     # directly write the nodemap file and update requirements instead of going
     # through the whole cloning process
     elif (
         len(upgrade_op.upgrade_actions) == 1
         and b'persistent-nodemap' in upgrade_op.upgrade_actions_names
         and not upgrade_op.removed_actions
     ):
         ui.status(
             _(b'upgrading repository to use persistent nodemap feature\n')
         )
         with srcrepo.transaction(b'upgrade') as tr:
             unfi = srcrepo.unfiltered()
             cl = unfi.changelog
             nodemap.persist_nodemap(tr, cl, force=True)
             # we want to directly operate on the underlying revlog to force
             # create a nodemap file. This is fine since this is upgrade code
             # and it heavily relies on repository being revlog based
             # hence accessing private attributes can be justified
             nodemap.persist_nodemap(
                 tr, unfi.manifestlog._rootstore._revlog, force=True
             )
         scmutil.writereporequirements(srcrepo, upgrade_op.new_requirements)
     elif (
         len(upgrade_op.removed_actions) == 1
         and [
             x
             for x in upgrade_op.removed_actions
             if x.name == b'persistent-nodemap'
         ]
         and not upgrade_op.upgrade_actions
     ):
         ui.status(
             _(b'downgrading repository to not use persistent nodemap feature\n')
         )
         with srcrepo.transaction(b'upgrade') as tr:
             unfi = srcrepo.unfiltered()
             cl = unfi.changelog
             nodemap.delete_nodemap(tr, srcrepo, cl)
             # check comment 20 lines above for accessing private attributes
             nodemap.delete_nodemap(
                 tr, srcrepo, unfi.manifestlog._rootstore._revlog
             )
         scmutil.writereporequirements(srcrepo, upgrade_op.new_requirements)
     else:
         with dstrepo.transaction(b'upgrade') as tr:
             _clonerevlogs(
                 ui,
                 srcrepo,
                 dstrepo,
                 tr,
                 upgrade_op,
             )
 
         # Now copy other files in the store directory.
         for p in _files_to_copy_post_revlog_clone(srcrepo):
             srcrepo.ui.status(_(b'copying %s\n') % p)
             src = srcrepo.store.rawvfs.join(p)
             dst = dstrepo.store.rawvfs.join(p)
             util.copyfile(src, dst, copystat=True)
 
         finishdatamigration(ui, srcrepo, dstrepo, requirements)
 
         ui.status(_(b'data fully upgraded in a temporary repository\n'))
 
         if upgrade_op.backup_store:
             backuppath = pycompat.mkdtemp(
                 prefix=b'upgradebackup.', dir=srcrepo.path
             )
             backupvfs = vfsmod.vfs(backuppath)
 
             # Make a backup of requires file first, as it is the first to be modified.
             util.copyfile(
                 srcrepo.vfs.join(b'requires'), backupvfs.join(b'requires')
             )
 
         # We install an arbitrary requirement that clients must not support
         # as a mechanism to lock out new clients during the data swap. This is
         # better than allowing a client to continue while the repository is in
         # an inconsistent state.
         ui.status(
             _(
                 b'marking source repository as being upgraded; clients will be '
                 b'unable to read from repository\n'
             )
         )
         scmutil.writereporequirements(
             srcrepo, srcrepo.requirements | {b'upgradeinprogress'}
         )
 
         ui.status(_(b'starting in-place swap of repository data\n'))
         if upgrade_op.backup_store:
             ui.status(
                 _(b'replaced files will be backed up at %s\n') % backuppath
             )
 
         # Now swap in the new store directory. Doing it as a rename should make
         # the operation nearly instantaneous and atomic (at least in well-behaved
         # environments).
         ui.status(_(b'replacing store...\n'))
         tstart = util.timer()
         _replacestores(srcrepo, dstrepo, backupvfs, upgrade_op)
         elapsed = util.timer() - tstart
         ui.status(
             _(
                 b'store replacement complete; repository was inconsistent for '
                 b'%0.1fs\n'
             )
             % elapsed
         )
 
         # We first write the requirements file. Any new requirements will lock
         # out legacy clients.
         ui.status(
             _(
                 b'finalizing requirements file and making repository readable '
                 b'again\n'
             )
         )
         scmutil.writereporequirements(srcrepo, upgrade_op.new_requirements)
 
         if upgrade_op.backup_store:
             # The lock file from the old store won't be removed because nothing has a
             # reference to its new location. So clean it up manually. Alternatively, we
             # could update srcrepo.svfs and other variables to point to the new
             # location. This is simpler.
             assert backupvfs is not None  # help pytype
             backupvfs.unlink(b'store/lock')
 
     return backuppath
 
 
 def upgrade_dirstate(ui, srcrepo, upgrade_op, old, new):
     if upgrade_op.backup_store:
         backuppath = pycompat.mkdtemp(
             prefix=b'upgradebackup.', dir=srcrepo.path
         )
         ui.status(_(b'replaced files will be backed up at %s\n') % backuppath)
         backupvfs = vfsmod.vfs(backuppath)
         util.copyfile(
             srcrepo.vfs.join(b'requires'), backupvfs.join(b'requires')
         )
         util.copyfile(
             srcrepo.vfs.join(b'dirstate'), backupvfs.join(b'dirstate')
         )
 
     assert srcrepo.dirstate._use_dirstate_v2 == (old == b'v2')
     srcrepo.dirstate._map._use_dirstate_tree = True
     srcrepo.dirstate._map.preload()
     srcrepo.dirstate._use_dirstate_v2 = new == b'v2'
     srcrepo.dirstate._map._use_dirstate_v2 = srcrepo.dirstate._use_dirstate_v2
     srcrepo.dirstate._dirty = True
     srcrepo.vfs.unlink(b'dirstate')
     srcrepo.dirstate.write(None)
 
     scmutil.writereporequirements(srcrepo, upgrade_op.new_requirements)

mercurial/verify.py

@@ -1,586 +1,588 b''
 # verify.py - repository integrity checking for Mercurial
 #
 # Copyright 2006, 2007 Olivia 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 os
 
 from .i18n import _
 from .node import short
 from .utils import stringutil
 
 from . import (
     error,
     pycompat,
     revlog,
     util,
 )
 
 VERIFY_DEFAULT = 0
 VERIFY_FULL = 1
 
 
 def verify(repo, level=None):
     with repo.lock():
         v = verifier(repo, level)
         return v.verify()
 
 
 def _normpath(f):
     # under hg < 2.4, convert didn't sanitize paths properly, so a
     # converted repo may contain repeated slashes
     while b'//' in f:
         f = f.replace(b'//', b'/')
     return f
 
 
 HINT_FNCACHE = _(
     b'hint: run "hg debugrebuildfncache" to recover from corrupt fncache\n'
 )
 
 WARN_PARENT_DIR_UNKNOWN_REV = _(
     b"parent-directory manifest refers to unknown revision %s"
 )
 
 WARN_UNKNOWN_COPY_SOURCE = _(
     b"warning: copy source of '%s' not in parents of %s"
 )
 
 WARN_NULLID_COPY_SOURCE = _(
     b"warning: %s@%s: copy source revision is nullid %s:%s\n"
 )
 
 
 class verifier(object):
     def __init__(self, repo, level=None):
         self.repo = repo.unfiltered()
         self.ui = repo.ui
         self.match = repo.narrowmatch()
         if level is None:
             level = VERIFY_DEFAULT
         self._level = level
         self.badrevs = set()
         self.errors = 0
         self.warnings = 0
         self.havecl = len(repo.changelog) > 0
         self.havemf = len(repo.manifestlog.getstorage(b'')) > 0
         self.revlogv1 = repo.changelog._format_version != revlog.REVLOGV0
         self.lrugetctx = util.lrucachefunc(repo.unfiltered().__getitem__)
         self.refersmf = False
         self.fncachewarned = False
         # developer config: verify.skipflags
         self.skipflags = repo.ui.configint(b'verify', b'skipflags')
         self.warnorphanstorefiles = True
 
     def _warn(self, msg):
         """record a "warning" level issue"""
         self.ui.warn(msg + b"\n")
         self.warnings += 1
 
     def _err(self, linkrev, msg, filename=None):
         """record a "error" level issue"""
         if linkrev is not None:
             self.badrevs.add(linkrev)
             linkrev = b"%d" % linkrev
         else:
             linkrev = b'?'
         msg = b"%s: %s" % (linkrev, msg)
         if filename:
             msg = b"%s@%s" % (filename, msg)
         self.ui.warn(b" " + msg + b"\n")
         self.errors += 1
 
     def _exc(self, linkrev, msg, inst, filename=None):
         """record exception raised during the verify process"""
         fmsg = stringutil.forcebytestr(inst)
         if not fmsg:
             fmsg = pycompat.byterepr(inst)
         self._err(linkrev, b"%s: %s" % (msg, fmsg), filename)
 
     def _checkrevlog(self, obj, name, linkrev):
         """verify high level property of a revlog
 
         - revlog is present,
         - revlog is non-empty,
         - sizes (index and data) are correct,
         - revlog's format version is correct.
         """
         if not len(obj) and (self.havecl or self.havemf):
             self._err(linkrev, _(b"empty or missing %s") % name)
             return
 
         d = obj.checksize()
         if d[0]:
             self._err(None, _(b"data length off by %d bytes") % d[0], name)
         if d[1]:
             self._err(None, _(b"index contains %d extra bytes") % d[1], name)
 
         if obj._format_version != revlog.REVLOGV0:
             if not self.revlogv1:
                 self._warn(_(b"warning: `%s' uses revlog format 1") % name)
         elif self.revlogv1:
             self._warn(_(b"warning: `%s' uses revlog format 0") % name)
 
     def _checkentry(self, obj, i, node, seen, linkrevs, f):
         """verify a single revlog entry
 
         arguments are:
         - obj:      the source revlog
         - i:        the revision number
         - node:     the revision node id
         - seen:     nodes previously seen for this revlog
         - linkrevs: [changelog-revisions] introducing "node"
         - f:        string label ("changelog", "manifest", or filename)
 
         Performs the following checks:
         - linkrev points to an existing changelog revision,
         - linkrev points to a changelog revision that introduces this revision,
         - linkrev points to the lowest of these changesets,
         - both parents exist in the revlog,
         - the revision is not duplicated.
 
         Return the linkrev of the revision (or None for changelog's revisions).
         """
         lr = obj.linkrev(obj.rev(node))
         if lr < 0 or (self.havecl and lr not in linkrevs):
             if lr < 0 or lr >= len(self.repo.changelog):
                 msg = _(b"rev %d points to nonexistent changeset %d")
             else:
                 msg = _(b"rev %d points to unexpected changeset %d")
             self._err(None, msg % (i, lr), f)
             if linkrevs:
                 if f and len(linkrevs) > 1:
                     try:
                         # attempt to filter down to real linkrevs
                         linkrevs = []
                         for lr in linkrevs:
                             if self.lrugetctx(lr)[f].filenode() == node:
                                 linkrevs.append(lr)
                     except Exception:
                         pass
                 msg = _(b" (expected %s)")
                 msg %= b" ".join(map(pycompat.bytestr, linkrevs))
                 self._warn(msg)
             lr = None  # can't be trusted
 
         try:
             p1, p2 = obj.parents(node)
             if p1 not in seen and p1 != self.repo.nullid:
                 msg = _(b"unknown parent 1 %s of %s") % (short(p1), short(node))
                 self._err(lr, msg, f)
             if p2 not in seen and p2 != self.repo.nullid:
                 msg = _(b"unknown parent 2 %s of %s") % (short(p2), short(node))
                 self._err(lr, msg, f)
         except Exception as inst:
             self._exc(lr, _(b"checking parents of %s") % short(node), inst, f)
 
         if node in seen:
             self._err(lr, _(b"duplicate revision %d (%d)") % (i, seen[node]), f)
         seen[node] = i
         return lr
 
     def verify(self):
         """verify the content of the Mercurial repository
 
         This method run all verifications, displaying issues as they are found.
 
         return 1 if any error have been encountered, 0 otherwise."""
         # initial validation and generic report
         repo = self.repo
         ui = repo.ui
         if not repo.url().startswith(b'file:'):
             raise error.Abort(_(b"cannot verify bundle or remote repos"))
 
         if os.path.exists(repo.sjoin(b"journal")):
             ui.warn(_(b"abandoned transaction found - run hg recover\n"))
 
         if ui.verbose or not self.revlogv1:
             ui.status(
                 _(b"repository uses revlog format %d\n")
                 % (self.revlogv1 and 1 or 0)
             )
 
         # data verification
         mflinkrevs, filelinkrevs = self._verifychangelog()
         filenodes = self._verifymanifest(mflinkrevs)
         del mflinkrevs
         self._crosscheckfiles(filelinkrevs, filenodes)
         totalfiles, filerevisions = self._verifyfiles(filenodes, filelinkrevs)
 
         # final report
         ui.status(
             _(b"checked %d changesets with %d changes to %d files\n")
             % (len(repo.changelog), filerevisions, totalfiles)
         )
         if self.warnings:
             ui.warn(_(b"%d warnings encountered!\n") % self.warnings)
         if self.fncachewarned:
             ui.warn(HINT_FNCACHE)
         if self.errors:
             ui.warn(_(b"%d integrity errors encountered!\n") % self.errors)
             if self.badrevs:
                 msg = _(b"(first damaged changeset appears to be %d)\n")
                 msg %= min(self.badrevs)
                 ui.warn(msg)
             return 1
         return 0
 
     def _verifychangelog(self):
         """verify the changelog of a repository
 
         The following checks are performed:
         - all of `_checkrevlog` checks,
         - all of `_checkentry` checks (for each revisions),
         - each revision can be read.
 
         The function returns some of the data observed in the changesets as a
         (mflinkrevs, filelinkrevs) tuples:
         - mflinkrevs:   is a { manifest-node -> [changelog-rev] } mapping
         - filelinkrevs: is a { file-path -> [changelog-rev] } mapping
 
         If a matcher was specified, filelinkrevs will only contains matched
         files.
         """
         ui = self.ui
         repo = self.repo
         match = self.match
         cl = repo.changelog
 
         ui.status(_(b"checking changesets\n"))
         mflinkrevs = {}
         filelinkrevs = {}
         seen = {}
         self._checkrevlog(cl, b"changelog", 0)
         progress = ui.makeprogress(
             _(b'checking'), unit=_(b'changesets'), total=len(repo)
         )
         for i in repo:
             progress.update(i)
             n = cl.node(i)
             self._checkentry(cl, i, n, seen, [i], b"changelog")
 
             try:
                 changes = cl.read(n)
                 if changes[0] != self.repo.nullid:
                     mflinkrevs.setdefault(changes[0], []).append(i)
                     self.refersmf = True
                 for f in changes[3]:
                     if match(f):
                         filelinkrevs.setdefault(_normpath(f), []).append(i)
             except Exception as inst:
                 self.refersmf = True
                 self._exc(i, _(b"unpacking changeset %s") % short(n), inst)
         progress.complete()
         return mflinkrevs, filelinkrevs
 
     def _verifymanifest(
         self, mflinkrevs, dir=b"", storefiles=None, subdirprogress=None
     ):
         """verify the manifestlog content
 
         Inputs:
         - mflinkrevs:     a {manifest-node -> [changelog-revisions]} mapping
         - dir:            a subdirectory to check (for tree manifest repo)
         - storefiles:     set of currently "orphan" files.
         - subdirprogress: a progress object
 
         This function checks:
         * all of `_checkrevlog` checks (for all manifest related revlogs)
         * all of `_checkentry` checks (for all manifest related revisions)
         * nodes for subdirectory exists in the sub-directory manifest
         * each manifest entries have a file path
         * each manifest node refered in mflinkrevs exist in the manifest log
 
         If tree manifest is in use and a matchers is specified, only the
         sub-directories matching it will be verified.
 
         return a two level mapping:
             {"path" -> { filenode -> changelog-revision}}
 
         This mapping primarily contains entries for every files in the
         repository. In addition, when tree-manifest is used, it also contains
         sub-directory entries.
 
         If a matcher is provided, only matching paths will be included.
         """
         repo = self.repo
         ui = self.ui
         match = self.match
         mfl = self.repo.manifestlog
         mf = mfl.getstorage(dir)
 
         if not dir:
             self.ui.status(_(b"checking manifests\n"))
 
         filenodes = {}
         subdirnodes = {}
         seen = {}
         label = b"manifest"
         if dir:
             label = dir
             revlogfiles = mf.files()
             storefiles.difference_update(revlogfiles)
             if subdirprogress:  # should be true since we're in a subdirectory
                 subdirprogress.increment()
         if self.refersmf:
             # Do not check manifest if there are only changelog entries with
             # null manifests.
             self._checkrevlog(mf._revlog, label, 0)
         progress = ui.makeprogress(
             _(b'checking'), unit=_(b'manifests'), total=len(mf)
         )
         for i in mf:
             if not dir:
                 progress.update(i)
             n = mf.node(i)
             lr = self._checkentry(mf, i, n, seen, mflinkrevs.get(n, []), label)
             if n in mflinkrevs:
                 del mflinkrevs[n]
             elif dir:
                 msg = _(b"%s not in parent-directory manifest") % short(n)
                 self._err(lr, msg, label)
             else:
                 self._err(lr, _(b"%s not in changesets") % short(n), label)
 
             try:
                 mfdelta = mfl.get(dir, n).readdelta(shallow=True)
                 for f, fn, fl in mfdelta.iterentries():
                     if not f:
                         self._err(lr, _(b"entry without name in manifest"))
                     elif f == b"/dev/null":  # ignore this in very old repos
                         continue
                     fullpath = dir + _normpath(f)
                     if fl == b't':
                         if not match.visitdir(fullpath):
                             continue
                         sdn = subdirnodes.setdefault(fullpath + b'/', {})
                         sdn.setdefault(fn, []).append(lr)
                     else:
                         if not match(fullpath):
                             continue
                         filenodes.setdefault(fullpath, {}).setdefault(fn, lr)
             except Exception as inst:
                 self._exc(lr, _(b"reading delta %s") % short(n), inst, label)
             if self._level >= VERIFY_FULL:
                 try:
                     # Various issues can affect manifest. So we read each full
                     # text from storage. This triggers the checks from the core
                     # code (eg: hash verification, filename are ordered, etc.)
                     mfdelta = mfl.get(dir, n).read()
                 except Exception as inst:
                     msg = _(b"reading full manifest %s") % short(n)
                     self._exc(lr, msg, inst, label)
 
         if not dir:
             progress.complete()
 
         if self.havemf:
             # since we delete entry in `mflinkrevs` during iteration, any
             # remaining entries are "missing". We need to issue errors for them.
             changesetpairs = [(c, m) for m in mflinkrevs for c in mflinkrevs[m]]
             for c, m in sorted(changesetpairs):
                 if dir:
                     self._err(c, WARN_PARENT_DIR_UNKNOWN_REV % short(m), label)
                 else:
                     msg = _(b"changeset refers to unknown revision %s")
                     msg %= short(m)
                     self._err(c, msg, label)
 
         if not dir and subdirnodes:
             self.ui.status(_(b"checking directory manifests\n"))
             storefiles = set()
             subdirs = set()
             revlogv1 = self.revlogv1
-            for t, f, f2, size in repo.store.datafiles():
-                if not f:
-                    self._err(None, _(b"cannot decode filename '%s'") % f2)
-                elif (size > 0 or not revlogv1) and f.startswith(b'meta/'):
+            undecodable = []
+            for t, f, size in repo.store.datafiles(undecodable=undecodable):
+                if (size > 0 or not revlogv1) and f.startswith(b'meta/'):
                     storefiles.add(_normpath(f))
                     subdirs.add(os.path.dirname(f))
+            for f in undecodable:
+                self._err(None, _(b"cannot decode filename '%s'") % f)
             subdirprogress = ui.makeprogress(
                 _(b'checking'), unit=_(b'manifests'), total=len(subdirs)
             )
 
         for subdir, linkrevs in pycompat.iteritems(subdirnodes):
             subdirfilenodes = self._verifymanifest(
                 linkrevs, subdir, storefiles, subdirprogress
             )
             for f, onefilenodes in pycompat.iteritems(subdirfilenodes):
                 filenodes.setdefault(f, {}).update(onefilenodes)
 
         if not dir and subdirnodes:
             assert subdirprogress is not None  # help pytype
             subdirprogress.complete()
             if self.warnorphanstorefiles:
                 for f in sorted(storefiles):
                     self._warn(_(b"warning: orphan data file '%s'") % f)
 
         return filenodes
 
     def _crosscheckfiles(self, filelinkrevs, filenodes):
         repo = self.repo
         ui = self.ui
         ui.status(_(b"crosschecking files in changesets and manifests\n"))
 
         total = len(filelinkrevs) + len(filenodes)
         progress = ui.makeprogress(
             _(b'crosschecking'), unit=_(b'files'), total=total
         )
         if self.havemf:
             for f in sorted(filelinkrevs):
                 progress.increment()
                 if f not in filenodes:
                     lr = filelinkrevs[f][0]
                     self._err(lr, _(b"in changeset but not in manifest"), f)
 
         if self.havecl:
             for f in sorted(filenodes):
                 progress.increment()
                 if f not in filelinkrevs:
                     try:
                         fl = repo.file(f)
                         lr = min([fl.linkrev(fl.rev(n)) for n in filenodes[f]])
                     except Exception:
                         lr = None
                     self._err(lr, _(b"in manifest but not in changeset"), f)
 
         progress.complete()
 
     def _verifyfiles(self, filenodes, filelinkrevs):
         repo = self.repo
         ui = self.ui
         lrugetctx = self.lrugetctx
         revlogv1 = self.revlogv1
         havemf = self.havemf
         ui.status(_(b"checking files\n"))
 
         storefiles = set()
-        for rl_type, f, f2, size in repo.store.datafiles():
-            if not f:
-                self._err(None, _(b"cannot decode filename '%s'") % f2)
-            elif (size > 0 or not revlogv1) and f.startswith(b'data/'):
+        undecodable = []
+        for t, f, size in repo.store.datafiles(undecodable=undecodable):
+            if (size > 0 or not revlogv1) and f.startswith(b'data/'):
                 storefiles.add(_normpath(f))
+        for f in undecodable:
+            self._err(None, _(b"cannot decode filename '%s'") % f)
 
         state = {
             # TODO this assumes revlog storage for changelog.
             b'expectedversion': self.repo.changelog._format_version,
             b'skipflags': self.skipflags,
             # experimental config: censor.policy
             b'erroroncensored': ui.config(b'censor', b'policy') == b'abort',
         }
 
         files = sorted(set(filenodes) | set(filelinkrevs))
         revisions = 0
         progress = ui.makeprogress(
             _(b'checking'), unit=_(b'files'), total=len(files)
         )
         for i, f in enumerate(files):
             progress.update(i, item=f)
             try:
                 linkrevs = filelinkrevs[f]
             except KeyError:
                 # in manifest but not in changelog
                 linkrevs = []
 
             if linkrevs:
                 lr = linkrevs[0]
             else:
                 lr = None
 
             try:
                 fl = repo.file(f)
             except error.StorageError as e:
                 self._err(lr, _(b"broken revlog! (%s)") % e, f)
                 continue
 
             for ff in fl.files():
                 try:
                     storefiles.remove(ff)
                 except KeyError:
                     if self.warnorphanstorefiles:
                         msg = _(b" warning: revlog '%s' not in fncache!")
                         self._warn(msg % ff)
                         self.fncachewarned = True
 
             if not len(fl) and (self.havecl or self.havemf):
                 self._err(lr, _(b"empty or missing %s") % f)
             else:
                 # Guard against implementations not setting this.
                 state[b'skipread'] = set()
                 state[b'safe_renamed'] = set()
 
                 for problem in fl.verifyintegrity(state):
                     if problem.node is not None:
                         linkrev = fl.linkrev(fl.rev(problem.node))
                     else:
                         linkrev = None
 
                     if problem.warning:
                         self._warn(problem.warning)
                     elif problem.error:
                         linkrev_msg = linkrev if linkrev is not None else lr
                         self._err(linkrev_msg, problem.error, f)
                     else:
                         raise error.ProgrammingError(
                             b'problem instance does not set warning or error '
                             b'attribute: %s' % problem.msg
                         )
 
             seen = {}
             for i in fl:
                 revisions += 1
                 n = fl.node(i)
                 lr = self._checkentry(fl, i, n, seen, linkrevs, f)
                 if f in filenodes:
                     if havemf and n not in filenodes[f]:
                         self._err(lr, _(b"%s not in manifests") % (short(n)), f)
                     else:
                         del filenodes[f][n]
 
                 if n in state[b'skipread'] and n not in state[b'safe_renamed']:
                     continue
 
                 # check renames
                 try:
                     # This requires resolving fulltext (at least on revlogs,
                     # though not with LFS revisions). We may want
                     # ``verifyintegrity()`` to pass a set of nodes with
                     # rename metadata as an optimization.
                     rp = fl.renamed(n)
                     if rp:
                         if lr is not None and ui.verbose:
                             ctx = lrugetctx(lr)
                             if not any(rp[0] in pctx for pctx in ctx.parents()):
                                 self._warn(WARN_UNKNOWN_COPY_SOURCE % (f, ctx))
                         fl2 = repo.file(rp[0])
                         if not len(fl2):
                             m = _(b"empty or missing copy source revlog %s:%s")
                             self._err(lr, m % (rp[0], short(rp[1])), f)
                         elif rp[1] == self.repo.nullid:
                             msg = WARN_NULLID_COPY_SOURCE
                             msg %= (f, lr, rp[0], short(rp[1]))
                             ui.note(msg)
                         else:
                             fl2.rev(rp[1])
                 except Exception as inst:
                     self._exc(
                         lr, _(b"checking rename of %s") % short(n), inst, f
                     )
 
             # cross-check
             if f in filenodes:
                 fns = [(v, k) for k, v in pycompat.iteritems(filenodes[f])]
                 for lr, node in sorted(fns):
                     msg = _(b"manifest refers to unknown revision %s")
                     self._err(lr, msg % short(node), f)
         progress.complete()
 
         if self.warnorphanstorefiles:
             for f in sorted(storefiles):
                 self._warn(_(b"warning: orphan data file '%s'") % f)
 
         return len(files), revisions

mercurial/wireprotov2server.py

@@ -1,1613 +1,1613 b''
 # Copyright 21 May 2005 - (c) 2005 Jake Edge <[e-mail]>
 # Copyright 2005-2007 Olivia 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 collections
 import contextlib
 
 from .i18n import _
 from .node import hex
 from . import (
     discovery,
     encoding,
     error,
     match as matchmod,
     narrowspec,
     pycompat,
     streamclone,
     templatefilters,
     util,
     wireprotoframing,
     wireprototypes,
 )
 from .interfaces import util as interfaceutil
 from .utils import (
     cborutil,
     hashutil,
     stringutil,
 )
 
 FRAMINGTYPE = b'application/mercurial-exp-framing-0006'
 
 HTTP_WIREPROTO_V2 = wireprototypes.HTTP_WIREPROTO_V2
 
 COMMANDS = wireprototypes.commanddict()
 
 # Value inserted into cache key computation function. Change the value to
 # force new cache keys for every command request. This should be done when
 # there is a change to how caching works, etc.
 GLOBAL_CACHE_VERSION = 1
 
 
 def handlehttpv2request(rctx, req, res, checkperm, urlparts):
     from .hgweb import common as hgwebcommon
 
     # URL space looks like: <permissions>/<command>, where <permission> can
     # be ``ro`` or ``rw`` to signal read-only or read-write, respectively.
 
     # Root URL does nothing meaningful... yet.
     if not urlparts:
         res.status = b'200 OK'
         res.headers[b'Content-Type'] = b'text/plain'
         res.setbodybytes(_(b'HTTP version 2 API handler'))
         return
 
     if len(urlparts) == 1:
         res.status = b'404 Not Found'
         res.headers[b'Content-Type'] = b'text/plain'
         res.setbodybytes(
             _(b'do not know how to process %s\n') % req.dispatchpath
         )
         return
 
     permission, command = urlparts[0:2]
 
     if permission not in (b'ro', b'rw'):
         res.status = b'404 Not Found'
         res.headers[b'Content-Type'] = b'text/plain'
         res.setbodybytes(_(b'unknown permission: %s') % permission)
         return
 
     if req.method != b'POST':
         res.status = b'405 Method Not Allowed'
         res.headers[b'Allow'] = b'POST'
         res.setbodybytes(_(b'commands require POST requests'))
         return
 
     # At some point we'll want to use our own API instead of recycling the
     # behavior of version 1 of the wire protocol...
     # TODO return reasonable responses - not responses that overload the
     # HTTP status line message for error reporting.
     try:
         checkperm(rctx, req, b'pull' if permission == b'ro' else b'push')
     except hgwebcommon.ErrorResponse as e:
         res.status = hgwebcommon.statusmessage(
             e.code, stringutil.forcebytestr(e)
         )
         for k, v in e.headers:
             res.headers[k] = v
         res.setbodybytes(b'permission denied')
         return
 
     # We have a special endpoint to reflect the request back at the client.
     if command == b'debugreflect':
         _processhttpv2reflectrequest(rctx.repo.ui, rctx.repo, req, res)
         return
 
     # Extra commands that we handle that aren't really wire protocol
     # commands. Think extra hard before making this hackery available to
     # extension.
     extracommands = {b'multirequest'}
 
     if command not in COMMANDS and command not in extracommands:
         res.status = b'404 Not Found'
         res.headers[b'Content-Type'] = b'text/plain'
         res.setbodybytes(_(b'unknown wire protocol command: %s\n') % command)
         return
 
     repo = rctx.repo
     ui = repo.ui
 
     proto = httpv2protocolhandler(req, ui)
 
     if (
         not COMMANDS.commandavailable(command, proto)
         and command not in extracommands
     ):
         res.status = b'404 Not Found'
         res.headers[b'Content-Type'] = b'text/plain'
         res.setbodybytes(_(b'invalid wire protocol command: %s') % command)
         return
 
     # TODO consider cases where proxies may add additional Accept headers.
     if req.headers.get(b'Accept') != FRAMINGTYPE:
         res.status = b'406 Not Acceptable'
         res.headers[b'Content-Type'] = b'text/plain'
         res.setbodybytes(
             _(b'client MUST specify Accept header with value: %s\n')
             % FRAMINGTYPE
         )
         return
 
     if req.headers.get(b'Content-Type') != FRAMINGTYPE:
         res.status = b'415 Unsupported Media Type'
         # TODO we should send a response with appropriate media type,
         # since client does Accept it.
         res.headers[b'Content-Type'] = b'text/plain'
         res.setbodybytes(
             _(b'client MUST send Content-Type header with value: %s\n')
             % FRAMINGTYPE
         )
         return
 
     _processhttpv2request(ui, repo, req, res, permission, command, proto)
 
 
 def _processhttpv2reflectrequest(ui, repo, req, res):
     """Reads unified frame protocol request and dumps out state to client.
 
     This special endpoint can be used to help debug the wire protocol.
 
     Instead of routing the request through the normal dispatch mechanism,
     we instead read all frames, decode them, and feed them into our state
     tracker. We then dump the log of all that activity back out to the
     client.
     """
     # Reflection APIs have a history of being abused, accidentally disclosing
     # sensitive data, etc. So we have a config knob.
     if not ui.configbool(b'experimental', b'web.api.debugreflect'):
         res.status = b'404 Not Found'
         res.headers[b'Content-Type'] = b'text/plain'
         res.setbodybytes(_(b'debugreflect service not available'))
         return
 
     # We assume we have a unified framing protocol request body.
 
     reactor = wireprotoframing.serverreactor(ui)
     states = []
 
     while True:
         frame = wireprotoframing.readframe(req.bodyfh)
 
         if not frame:
             states.append(b'received: <no frame>')
             break
 
         states.append(
             b'received: %d %d %d %s'
             % (frame.typeid, frame.flags, frame.requestid, frame.payload)
         )
 
         action, meta = reactor.onframerecv(frame)
         states.append(templatefilters.json((action, meta)))
 
     action, meta = reactor.oninputeof()
     meta[b'action'] = action
     states.append(templatefilters.json(meta))
 
     res.status = b'200 OK'
     res.headers[b'Content-Type'] = b'text/plain'
     res.setbodybytes(b'\n'.join(states))
 
 
 def _processhttpv2request(ui, repo, req, res, authedperm, reqcommand, proto):
     """Post-validation handler for HTTPv2 requests.
 
     Called when the HTTP request contains unified frame-based protocol
     frames for evaluation.
     """
     # TODO Some HTTP clients are full duplex and can receive data before
     # the entire request is transmitted. Figure out a way to indicate support
     # for that so we can opt into full duplex mode.
     reactor = wireprotoframing.serverreactor(ui, deferoutput=True)
     seencommand = False
 
     outstream = None
 
     while True:
         frame = wireprotoframing.readframe(req.bodyfh)
         if not frame:
             break
 
         action, meta = reactor.onframerecv(frame)
 
         if action == b'wantframe':
             # Need more data before we can do anything.
             continue
         elif action == b'runcommand':
             # Defer creating output stream because we need to wait for
             # protocol settings frames so proper encoding can be applied.
             if not outstream:
                 outstream = reactor.makeoutputstream()
 
             sentoutput = _httpv2runcommand(
                 ui,
                 repo,
                 req,
                 res,
                 authedperm,
                 reqcommand,
                 reactor,
                 outstream,
                 meta,
                 issubsequent=seencommand,
             )
 
             if sentoutput:
                 return
 
             seencommand = True
 
         elif action == b'error':
             # TODO define proper error mechanism.
             res.status = b'200 OK'
             res.headers[b'Content-Type'] = b'text/plain'
             res.setbodybytes(meta[b'message'] + b'\n')
             return
         else:
             raise error.ProgrammingError(
                 b'unhandled action from frame processor: %s' % action
             )
 
     action, meta = reactor.oninputeof()
     if action == b'sendframes':
         # We assume we haven't started sending the response yet. If we're
         # wrong, the response type will raise an exception.
         res.status = b'200 OK'
         res.headers[b'Content-Type'] = FRAMINGTYPE
         res.setbodygen(meta[b'framegen'])
     elif action == b'noop':
         pass
     else:
         raise error.ProgrammingError(
             b'unhandled action from frame processor: %s' % action
         )
 
 
 def _httpv2runcommand(
     ui,
     repo,
     req,
     res,
     authedperm,
     reqcommand,
     reactor,
     outstream,
     command,
     issubsequent,
 ):
     """Dispatch a wire protocol command made from HTTPv2 requests.
 
     The authenticated permission (``authedperm``) along with the original
     command from the URL (``reqcommand``) are passed in.
     """
     # We already validated that the session has permissions to perform the
     # actions in ``authedperm``. In the unified frame protocol, the canonical
     # command to run is expressed in a frame. However, the URL also requested
     # to run a specific command. We need to be careful that the command we
     # run doesn't have permissions requirements greater than what was granted
     # by ``authedperm``.
     #
     # Our rule for this is we only allow one command per HTTP request and
     # that command must match the command in the URL. However, we make
     # an exception for the ``multirequest`` URL. This URL is allowed to
     # execute multiple commands. We double check permissions of each command
     # as it is invoked to ensure there is no privilege escalation.
     # TODO consider allowing multiple commands to regular command URLs
     # iff each command is the same.
 
     proto = httpv2protocolhandler(req, ui, args=command[b'args'])
 
     if reqcommand == b'multirequest':
         if not COMMANDS.commandavailable(command[b'command'], proto):
             # TODO proper error mechanism
             res.status = b'200 OK'
             res.headers[b'Content-Type'] = b'text/plain'
             res.setbodybytes(
                 _(b'wire protocol command not available: %s')
                 % command[b'command']
             )
             return True
 
         # TODO don't use assert here, since it may be elided by -O.
         assert authedperm in (b'ro', b'rw')
         wirecommand = COMMANDS[command[b'command']]
         assert wirecommand.permission in (b'push', b'pull')
 
         if authedperm == b'ro' and wirecommand.permission != b'pull':
             # TODO proper error mechanism
             res.status = b'403 Forbidden'
             res.headers[b'Content-Type'] = b'text/plain'
             res.setbodybytes(
                 _(b'insufficient permissions to execute command: %s')
                 % command[b'command']
             )
             return True
 
         # TODO should we also call checkperm() here? Maybe not if we're going
         # to overhaul that API. The granted scope from the URL check should
         # be good enough.
 
     else:
         # Don't allow multiple commands outside of ``multirequest`` URL.
         if issubsequent:
             # TODO proper error mechanism
             res.status = b'200 OK'
             res.headers[b'Content-Type'] = b'text/plain'
             res.setbodybytes(
                 _(b'multiple commands cannot be issued to this URL')
             )
             return True
 
         if reqcommand != command[b'command']:
             # TODO define proper error mechanism
             res.status = b'200 OK'
             res.headers[b'Content-Type'] = b'text/plain'
             res.setbodybytes(_(b'command in frame must match command in URL'))
             return True
 
     res.status = b'200 OK'
     res.headers[b'Content-Type'] = FRAMINGTYPE
 
     try:
         objs = dispatch(repo, proto, command[b'command'], command[b'redirect'])
 
         action, meta = reactor.oncommandresponsereadyobjects(
             outstream, command[b'requestid'], objs
         )
 
     except error.WireprotoCommandError as e:
         action, meta = reactor.oncommanderror(
             outstream, command[b'requestid'], e.message, e.messageargs
         )
 
     except Exception as e:
         action, meta = reactor.onservererror(
             outstream,
             command[b'requestid'],
             _(b'exception when invoking command: %s')
             % stringutil.forcebytestr(e),
         )
 
     if action == b'sendframes':
         res.setbodygen(meta[b'framegen'])
         return True
     elif action == b'noop':
         return False
     else:
         raise error.ProgrammingError(
             b'unhandled event from reactor: %s' % action
         )
 
 
 def getdispatchrepo(repo, proto, command):
     viewconfig = repo.ui.config(b'server', b'view')
     return repo.filtered(viewconfig)
 
 
 def dispatch(repo, proto, command, redirect):
     """Run a wire protocol command.
 
     Returns an iterable of objects that will be sent to the client.
     """
     repo = getdispatchrepo(repo, proto, command)
 
     entry = COMMANDS[command]
     func = entry.func
     spec = entry.args
 
     args = proto.getargs(spec)
 
     # There is some duplicate boilerplate code here for calling the command and
     # emitting objects. It is either that or a lot of indented code that looks
     # like a pyramid (since there are a lot of code paths that result in not
     # using the cacher).
     callcommand = lambda: func(repo, proto, **pycompat.strkwargs(args))
 
     # Request is not cacheable. Don't bother instantiating a cacher.
     if not entry.cachekeyfn:
         for o in callcommand():
             yield o
         return
 
     if redirect:
         redirecttargets = redirect[b'targets']
         redirecthashes = redirect[b'hashes']
     else:
         redirecttargets = []
         redirecthashes = []
 
     cacher = makeresponsecacher(
         repo,
         proto,
         command,
         args,
         cborutil.streamencode,
         redirecttargets=redirecttargets,
         redirecthashes=redirecthashes,
     )
 
     # But we have no cacher. Do default handling.
     if not cacher:
         for o in callcommand():
             yield o
         return
 
     with cacher:
         cachekey = entry.cachekeyfn(
             repo, proto, cacher, **pycompat.strkwargs(args)
         )
 
         # No cache key or the cacher doesn't like it. Do default handling.
         if cachekey is None or not cacher.setcachekey(cachekey):
             for o in callcommand():
                 yield o
             return
 
         # Serve it from the cache, if possible.
         cached = cacher.lookup()
 
         if cached:
             for o in cached[b'objs']:
                 yield o
             return
 
         # Else call the command and feed its output into the cacher, allowing
         # the cacher to buffer/mutate objects as it desires.
         for o in callcommand():
             for o in cacher.onobject(o):
                 yield o
 
         for o in cacher.onfinished():
             yield o
 
 
 @interfaceutil.implementer(wireprototypes.baseprotocolhandler)
 class httpv2protocolhandler(object):
     def __init__(self, req, ui, args=None):
         self._req = req
         self._ui = ui
         self._args = args
 
     @property
     def name(self):
         return HTTP_WIREPROTO_V2
 
     def getargs(self, args):
         # First look for args that were passed but aren't registered on this
         # command.
         extra = set(self._args) - set(args)
         if extra:
             raise error.WireprotoCommandError(
                 b'unsupported argument to command: %s'
                 % b', '.join(sorted(extra))
             )
 
         # And look for required arguments that are missing.
         missing = {a for a in args if args[a][b'required']} - set(self._args)
 
         if missing:
             raise error.WireprotoCommandError(
                 b'missing required arguments: %s' % b', '.join(sorted(missing))
             )
 
         # Now derive the arguments to pass to the command, taking into
         # account the arguments specified by the client.
         data = {}
         for k, meta in sorted(args.items()):
             # This argument wasn't passed by the client.
             if k not in self._args:
                 data[k] = meta[b'default']()
                 continue
 
             v = self._args[k]
 
             # Sets may be expressed as lists. Silently normalize.
             if meta[b'type'] == b'set' and isinstance(v, list):
                 v = set(v)
 
             # TODO consider more/stronger type validation.
 
             data[k] = v
 
         return data
 
     def getprotocaps(self):
         # Protocol capabilities are currently not implemented for HTTP V2.
         return set()
 
     def getpayload(self):
         raise NotImplementedError
 
     @contextlib.contextmanager
     def mayberedirectstdio(self):
         raise NotImplementedError
 
     def client(self):
         raise NotImplementedError
 
     def addcapabilities(self, repo, caps):
         return caps
 
     def checkperm(self, perm):
         raise NotImplementedError
 
 
 def httpv2apidescriptor(req, repo):
     proto = httpv2protocolhandler(req, repo.ui)
 
     return _capabilitiesv2(repo, proto)
 
 
 def _capabilitiesv2(repo, proto):
     """Obtain the set of capabilities for version 2 transports.
 
     These capabilities are distinct from the capabilities for version 1
     transports.
     """
     caps = {
         b'commands': {},
         b'framingmediatypes': [FRAMINGTYPE],
         b'pathfilterprefixes': set(narrowspec.VALID_PREFIXES),
     }
 
     for command, entry in COMMANDS.items():
         args = {}
 
         for arg, meta in entry.args.items():
             args[arg] = {
                 # TODO should this be a normalized type using CBOR's
                 # terminology?
                 b'type': meta[b'type'],
                 b'required': meta[b'required'],
             }
 
             if not meta[b'required']:
                 args[arg][b'default'] = meta[b'default']()
 
             if meta[b'validvalues']:
                 args[arg][b'validvalues'] = meta[b'validvalues']
 
         # TODO this type of check should be defined in a per-command callback.
         if (
             command == b'rawstorefiledata'
             and not streamclone.allowservergeneration(repo)
         ):
             continue
 
         caps[b'commands'][command] = {
             b'args': args,
             b'permissions': [entry.permission],
         }
 
         if entry.extracapabilitiesfn:
             extracaps = entry.extracapabilitiesfn(repo, proto)
             caps[b'commands'][command].update(extracaps)
 
     caps[b'rawrepoformats'] = sorted(repo.requirements & repo.supportedformats)
 
     targets = getadvertisedredirecttargets(repo, proto)
     if targets:
         caps[b'redirect'] = {
             b'targets': [],
             b'hashes': [b'sha256', b'sha1'],
         }
 
         for target in targets:
             entry = {
                 b'name': target[b'name'],
                 b'protocol': target[b'protocol'],
                 b'uris': target[b'uris'],
             }
 
             for key in (b'snirequired', b'tlsversions'):
                 if key in target:
                     entry[key] = target[key]
 
             caps[b'redirect'][b'targets'].append(entry)
 
     return proto.addcapabilities(repo, caps)
 
 
 def getadvertisedredirecttargets(repo, proto):
     """Obtain a list of content redirect targets.
 
     Returns a list containing potential redirect targets that will be
     advertised in capabilities data. Each dict MUST have the following
     keys:
 
     name
        The name of this redirect target. This is the identifier clients use
        to refer to a target. It is transferred as part of every command
        request.
 
     protocol
        Network protocol used by this target. Typically this is the string
        in front of the ``://`` in a URL. e.g. ``https``.
 
     uris
        List of representative URIs for this target. Clients can use the
        URIs to test parsing for compatibility or for ordering preference
        for which target to use.
 
     The following optional keys are recognized:
 
     snirequired
        Bool indicating if Server Name Indication (SNI) is required to
        connect to this target.
 
     tlsversions
        List of bytes indicating which TLS versions are supported by this
        target.
 
     By default, clients reflect the target order advertised by servers
     and servers will use the first client-advertised target when picking
     a redirect target. So targets should be advertised in the order the
     server prefers they be used.
     """
     return []
 
 
 def wireprotocommand(
     name,
     args=None,
     permission=b'push',
     cachekeyfn=None,
     extracapabilitiesfn=None,
 ):
     """Decorator to declare a wire protocol command.
 
     ``name`` is the name of the wire protocol command being provided.
 
     ``args`` is a dict defining arguments accepted by the command. Keys are
     the argument name. Values are dicts with the following keys:
 
        ``type``
           The argument data type. Must be one of the following string
           literals: ``bytes``, ``int``, ``list``, ``dict``, ``set``,
           or ``bool``.
 
        ``default``
           A callable returning the default value for this argument. If not
           specified, ``None`` will be the default value.
 
        ``example``
           An example value for this argument.
 
        ``validvalues``
           Set of recognized values for this argument.
 
     ``permission`` defines the permission type needed to run this command.
     Can be ``push`` or ``pull``. These roughly map to read-write and read-only,
     respectively. Default is to assume command requires ``push`` permissions
     because otherwise commands not declaring their permissions could modify
     a repository that is supposed to be read-only.
 
     ``cachekeyfn`` defines an optional callable that can derive the
     cache key for this request.
 
     ``extracapabilitiesfn`` defines an optional callable that defines extra
     command capabilities/parameters that are advertised next to the command
     in the capabilities data structure describing the server. The callable
     receives as arguments the repository and protocol objects. It returns
     a dict of extra fields to add to the command descriptor.
 
     Wire protocol commands are generators of objects to be serialized and
     sent to the client.
 
     If a command raises an uncaught exception, this will be translated into
     a command error.
 
     All commands can opt in to being cacheable by defining a function
     (``cachekeyfn``) that is called to derive a cache key. This function
     receives the same arguments as the command itself plus a ``cacher``
     argument containing the active cacher for the request and returns a bytes
     containing the key in a cache the response to this command may be cached
     under.
     """
     transports = {
         k for k, v in wireprototypes.TRANSPORTS.items() if v[b'version'] == 2
     }
 
     if permission not in (b'push', b'pull'):
         raise error.ProgrammingError(
             b'invalid wire protocol permission; '
             b'got %s; expected "push" or "pull"' % permission
         )
 
     if args is None:
         args = {}
 
     if not isinstance(args, dict):
         raise error.ProgrammingError(
             b'arguments for version 2 commands must be declared as dicts'
         )
 
     for arg, meta in args.items():
         if arg == b'*':
             raise error.ProgrammingError(
                 b'* argument name not allowed on version 2 commands'
             )
 
         if not isinstance(meta, dict):
             raise error.ProgrammingError(
                 b'arguments for version 2 commands '
                 b'must declare metadata as a dict'
             )
 
         if b'type' not in meta:
             raise error.ProgrammingError(
                 b'%s argument for command %s does not '
                 b'declare type field' % (arg, name)
             )
 
         if meta[b'type'] not in (
             b'bytes',
             b'int',
             b'list',
             b'dict',
             b'set',
             b'bool',
         ):
             raise error.ProgrammingError(
                 b'%s argument for command %s has '
                 b'illegal type: %s' % (arg, name, meta[b'type'])
             )
 
         if b'example' not in meta:
             raise error.ProgrammingError(
                 b'%s argument for command %s does not '
                 b'declare example field' % (arg, name)
             )
 
         meta[b'required'] = b'default' not in meta
 
         meta.setdefault(b'default', lambda: None)
         meta.setdefault(b'validvalues', None)
 
     def register(func):
         if name in COMMANDS:
             raise error.ProgrammingError(
                 b'%s command already registered for version 2' % name
             )
 
         COMMANDS[name] = wireprototypes.commandentry(
             func,
             args=args,
             transports=transports,
             permission=permission,
             cachekeyfn=cachekeyfn,
             extracapabilitiesfn=extracapabilitiesfn,
         )
 
         return func
 
     return register
 
 
 def makecommandcachekeyfn(command, localversion=None, allargs=False):
     """Construct a cache key derivation function with common features.
 
     By default, the cache key is a hash of:
 
     * The command name.
     * A global cache version number.
     * A local cache version number (passed via ``localversion``).
     * All the arguments passed to the command.
     * The media type used.
     * Wire protocol version string.
     * The repository path.
     """
     if not allargs:
         raise error.ProgrammingError(
             b'only allargs=True is currently supported'
         )
 
     if localversion is None:
         raise error.ProgrammingError(b'must set localversion argument value')
 
     def cachekeyfn(repo, proto, cacher, **args):
         spec = COMMANDS[command]
 
         # Commands that mutate the repo can not be cached.
         if spec.permission == b'push':
             return None
 
         # TODO config option to disable caching.
 
         # Our key derivation strategy is to construct a data structure
         # holding everything that could influence cacheability and to hash
         # the CBOR representation of that. Using CBOR seems like it might
         # be overkill. However, simpler hashing mechanisms are prone to
         # duplicate input issues. e.g. if you just concatenate two values,
         # "foo"+"bar" is identical to "fo"+"obar". Using CBOR provides
         # "padding" between values and prevents these problems.
 
         # Seed the hash with various data.
         state = {
             # To invalidate all cache keys.
             b'globalversion': GLOBAL_CACHE_VERSION,
             # More granular cache key invalidation.
             b'localversion': localversion,
             # Cache keys are segmented by command.
             b'command': command,
             # Throw in the media type and API version strings so changes
             # to exchange semantics invalid cache.
             b'mediatype': FRAMINGTYPE,
             b'version': HTTP_WIREPROTO_V2,
             # So same requests for different repos don't share cache keys.
             b'repo': repo.root,
         }
 
         # The arguments passed to us will have already been normalized.
         # Default values will be set, etc. This is important because it
         # means that it doesn't matter if clients send an explicit argument
         # or rely on the default value: it will all normalize to the same
         # set of arguments on the server and therefore the same cache key.
         #
         # Arguments by their very nature must support being encoded to CBOR.
         # And the CBOR encoder is deterministic. So we hash the arguments
         # by feeding the CBOR of their representation into the hasher.
         if allargs:
             state[b'args'] = pycompat.byteskwargs(args)
 
         cacher.adjustcachekeystate(state)
 
         hasher = hashutil.sha1()
         for chunk in cborutil.streamencode(state):
             hasher.update(chunk)
 
         return pycompat.sysbytes(hasher.hexdigest())
 
     return cachekeyfn
 
 
 def makeresponsecacher(
     repo, proto, command, args, objencoderfn, redirecttargets, redirecthashes
 ):
     """Construct a cacher for a cacheable command.
 
     Returns an ``iwireprotocolcommandcacher`` instance.
 
     Extensions can monkeypatch this function to provide custom caching
     backends.
     """
     return None
 
 
 def resolvenodes(repo, revisions):
     """Resolve nodes from a revisions specifier data structure."""
     cl = repo.changelog
     clhasnode = cl.hasnode
 
     seen = set()
     nodes = []
 
     if not isinstance(revisions, list):
         raise error.WireprotoCommandError(
             b'revisions must be defined as an array'
         )
 
     for spec in revisions:
         if b'type' not in spec:
             raise error.WireprotoCommandError(
                 b'type key not present in revision specifier'
             )
 
         typ = spec[b'type']
 
         if typ == b'changesetexplicit':
             if b'nodes' not in spec:
                 raise error.WireprotoCommandError(
                     b'nodes key not present in changesetexplicit revision '
                     b'specifier'
                 )
 
             for node in spec[b'nodes']:
                 if node not in seen:
                     nodes.append(node)
                     seen.add(node)
 
         elif typ == b'changesetexplicitdepth':
             for key in (b'nodes', b'depth'):
                 if key not in spec:
                     raise error.WireprotoCommandError(
                         b'%s key not present in changesetexplicitdepth revision '
                         b'specifier',
                         (key,),
                     )
 
             for rev in repo.revs(
                 b'ancestors(%ln, %s)', spec[b'nodes'], spec[b'depth'] - 1
             ):
                 node = cl.node(rev)
 
                 if node not in seen:
                     nodes.append(node)
                     seen.add(node)
 
         elif typ == b'changesetdagrange':
             for key in (b'roots', b'heads'):
                 if key not in spec:
                     raise error.WireprotoCommandError(
                         b'%s key not present in changesetdagrange revision '
                         b'specifier',
                         (key,),
                     )
 
             if not spec[b'heads']:
                 raise error.WireprotoCommandError(
                     b'heads key in changesetdagrange cannot be empty'
                 )
 
             if spec[b'roots']:
                 common = [n for n in spec[b'roots'] if clhasnode(n)]
             else:
                 common = [repo.nullid]
 
             for n in discovery.outgoing(repo, common, spec[b'heads']).missing:
                 if n not in seen:
                     nodes.append(n)
                     seen.add(n)
 
         else:
             raise error.WireprotoCommandError(
                 b'unknown revision specifier type: %s', (typ,)
             )
 
     return nodes
 
 
 @wireprotocommand(b'branchmap', permission=b'pull')
 def branchmapv2(repo, proto):
     yield {
         encoding.fromlocal(k): v
         for k, v in pycompat.iteritems(repo.branchmap())
     }
 
 
 @wireprotocommand(b'capabilities', permission=b'pull')
 def capabilitiesv2(repo, proto):
     yield _capabilitiesv2(repo, proto)
 
 
 @wireprotocommand(
     b'changesetdata',
     args={
         b'revisions': {
             b'type': b'list',
             b'example': [
                 {
                     b'type': b'changesetexplicit',
                     b'nodes': [b'abcdef...'],
                 }
             ],
         },
         b'fields': {
             b'type': b'set',
             b'default': set,
             b'example': {b'parents', b'revision'},
             b'validvalues': {b'bookmarks', b'parents', b'phase', b'revision'},
         },
     },
     permission=b'pull',
 )
 def changesetdata(repo, proto, revisions, fields):
     # TODO look for unknown fields and abort when they can't be serviced.
     # This could probably be validated by dispatcher using validvalues.
 
     cl = repo.changelog
     outgoing = resolvenodes(repo, revisions)
     publishing = repo.publishing()
 
     if outgoing:
         repo.hook(b'preoutgoing', throw=True, source=b'serve')
 
     yield {
         b'totalitems': len(outgoing),
     }
 
     # The phases of nodes already transferred to the client may have changed
     # since the client last requested data. We send phase-only records
     # for these revisions, if requested.
     # TODO actually do this. We'll probably want to emit phase heads
     # in the ancestry set of the outgoing revisions. This will ensure
     # that phase updates within that set are seen.
     if b'phase' in fields:
         pass
 
     nodebookmarks = {}
     for mark, node in repo._bookmarks.items():
         nodebookmarks.setdefault(node, set()).add(mark)
 
     # It is already topologically sorted by revision number.
     for node in outgoing:
         d = {
             b'node': node,
         }
 
         if b'parents' in fields:
             d[b'parents'] = cl.parents(node)
 
         if b'phase' in fields:
             if publishing:
                 d[b'phase'] = b'public'
             else:
                 ctx = repo[node]
                 d[b'phase'] = ctx.phasestr()
 
         if b'bookmarks' in fields and node in nodebookmarks:
             d[b'bookmarks'] = sorted(nodebookmarks[node])
             del nodebookmarks[node]
 
         followingmeta = []
         followingdata = []
 
         if b'revision' in fields:
             revisiondata = cl.revision(node)
             followingmeta.append((b'revision', len(revisiondata)))
             followingdata.append(revisiondata)
 
         # TODO make it possible for extensions to wrap a function or register
         # a handler to service custom fields.
 
         if followingmeta:
             d[b'fieldsfollowing'] = followingmeta
 
         yield d
 
         for extra in followingdata:
             yield extra
 
     # If requested, send bookmarks from nodes that didn't have revision
     # data sent so receiver is aware of any bookmark updates.
     if b'bookmarks' in fields:
         for node, marks in sorted(pycompat.iteritems(nodebookmarks)):
             yield {
                 b'node': node,
                 b'bookmarks': sorted(marks),
             }
 
 
 class FileAccessError(Exception):
     """Represents an error accessing a specific file."""
 
     def __init__(self, path, msg, args):
         self.path = path
         self.msg = msg
         self.args = args
 
 
 def getfilestore(repo, proto, path):
     """Obtain a file storage object for use with wire protocol.
 
     Exists as a standalone function so extensions can monkeypatch to add
     access control.
     """
     # This seems to work even if the file doesn't exist. So catch
     # "empty" files and return an error.
     fl = repo.file(path)
 
     if not len(fl):
         raise FileAccessError(path, b'unknown file: %s', (path,))
 
     return fl
 
 
 def emitfilerevisions(repo, path, revisions, linknodes, fields):
     for revision in revisions:
         d = {
             b'node': revision.node,
         }
 
         if b'parents' in fields:
             d[b'parents'] = [revision.p1node, revision.p2node]
 
         if b'linknode' in fields:
             d[b'linknode'] = linknodes[revision.node]
 
         followingmeta = []
         followingdata = []
 
         if b'revision' in fields:
             if revision.revision is not None:
                 followingmeta.append((b'revision', len(revision.revision)))
                 followingdata.append(revision.revision)
             else:
                 d[b'deltabasenode'] = revision.basenode
                 followingmeta.append((b'delta', len(revision.delta)))
                 followingdata.append(revision.delta)
 
         if followingmeta:
             d[b'fieldsfollowing'] = followingmeta
 
         yield d
 
         for extra in followingdata:
             yield extra
 
 
 def makefilematcher(repo, pathfilter):
     """Construct a matcher from a path filter dict."""
 
     # Validate values.
     if pathfilter:
         for key in (b'include', b'exclude'):
             for pattern in pathfilter.get(key, []):
                 if not pattern.startswith((b'path:', b'rootfilesin:')):
                     raise error.WireprotoCommandError(
                         b'%s pattern must begin with `path:` or `rootfilesin:`; '
                         b'got %s',
                         (key, pattern),
                     )
 
     if pathfilter:
         matcher = matchmod.match(
             repo.root,
             b'',
             include=pathfilter.get(b'include', []),
             exclude=pathfilter.get(b'exclude', []),
         )
     else:
         matcher = matchmod.match(repo.root, b'')
 
     # Requested patterns could include files not in the local store. So
     # filter those out.
     return repo.narrowmatch(matcher)
 
 
 @wireprotocommand(
     b'filedata',
     args={
         b'haveparents': {
             b'type': b'bool',
             b'default': lambda: False,
             b'example': True,
         },
         b'nodes': {
             b'type': b'list',
             b'example': [b'0123456...'],
         },
         b'fields': {
             b'type': b'set',
             b'default': set,
             b'example': {b'parents', b'revision'},
             b'validvalues': {b'parents', b'revision', b'linknode'},
         },
         b'path': {
             b'type': b'bytes',
             b'example': b'foo.txt',
         },
     },
     permission=b'pull',
     # TODO censoring a file revision won't invalidate the cache.
     # Figure out a way to take censoring into account when deriving
     # the cache key.
     cachekeyfn=makecommandcachekeyfn(b'filedata', 1, allargs=True),
 )
 def filedata(repo, proto, haveparents, nodes, fields, path):
     # TODO this API allows access to file revisions that are attached to
     # secret changesets. filesdata does not have this problem. Maybe this
     # API should be deleted?
 
     try:
         # Extensions may wish to access the protocol handler.
         store = getfilestore(repo, proto, path)
     except FileAccessError as e:
         raise error.WireprotoCommandError(e.msg, e.args)
 
     clnode = repo.changelog.node
     linknodes = {}
 
     # Validate requested nodes.
     for node in nodes:
         try:
             store.rev(node)
         except error.LookupError:
             raise error.WireprotoCommandError(
                 b'unknown file node: %s', (hex(node),)
             )
 
         # TODO by creating the filectx against a specific file revision
         # instead of changeset, linkrev() is always used. This is wrong for
         # cases where linkrev() may refer to a hidden changeset. But since this
         # API doesn't know anything about changesets, we're not sure how to
         # disambiguate the linknode. Perhaps we should delete this API?
         fctx = repo.filectx(path, fileid=node)
         linknodes[node] = clnode(fctx.introrev())
 
     revisions = store.emitrevisions(
         nodes,
         revisiondata=b'revision' in fields,
         assumehaveparentrevisions=haveparents,
     )
 
     yield {
         b'totalitems': len(nodes),
     }
 
     for o in emitfilerevisions(repo, path, revisions, linknodes, fields):
         yield o
 
 
 def filesdatacapabilities(repo, proto):
     batchsize = repo.ui.configint(
         b'experimental', b'server.filesdata.recommended-batch-size'
     )
     return {
         b'recommendedbatchsize': batchsize,
     }
 
 
 @wireprotocommand(
     b'filesdata',
     args={
         b'haveparents': {
             b'type': b'bool',
             b'default': lambda: False,
             b'example': True,
         },
         b'fields': {
             b'type': b'set',
             b'default': set,
             b'example': {b'parents', b'revision'},
             b'validvalues': {
                 b'firstchangeset',
                 b'linknode',
                 b'parents',
                 b'revision',
             },
         },
         b'pathfilter': {
             b'type': b'dict',
             b'default': lambda: None,
             b'example': {b'include': [b'path:tests']},
         },
         b'revisions': {
             b'type': b'list',
             b'example': [
                 {
                     b'type': b'changesetexplicit',
                     b'nodes': [b'abcdef...'],
                 }
             ],
         },
     },
     permission=b'pull',
     # TODO censoring a file revision won't invalidate the cache.
     # Figure out a way to take censoring into account when deriving
     # the cache key.
     cachekeyfn=makecommandcachekeyfn(b'filesdata', 1, allargs=True),
     extracapabilitiesfn=filesdatacapabilities,
 )
 def filesdata(repo, proto, haveparents, fields, pathfilter, revisions):
     # TODO This should operate on a repo that exposes obsolete changesets. There
     # is a race between a client making a push that obsoletes a changeset and
     # another client fetching files data for that changeset. If a client has a
     # changeset, it should probably be allowed to access files data for that
     # changeset.
 
     outgoing = resolvenodes(repo, revisions)
     filematcher = makefilematcher(repo, pathfilter)
 
     # path -> {fnode: linknode}
     fnodes = collections.defaultdict(dict)
 
     # We collect the set of relevant file revisions by iterating the changeset
     # revisions and either walking the set of files recorded in the changeset
     # or by walking the manifest at that revision. There is probably room for a
     # storage-level API to request this data, as it can be expensive to compute
     # and would benefit from caching or alternate storage from what revlogs
     # provide.
     for node in outgoing:
         ctx = repo[node]
         mctx = ctx.manifestctx()
         md = mctx.read()
 
         if haveparents:
             checkpaths = ctx.files()
         else:
             checkpaths = md.keys()
 
         for path in checkpaths:
             fnode = md[path]
 
             if path in fnodes and fnode in fnodes[path]:
                 continue
 
             if not filematcher(path):
                 continue
 
             fnodes[path].setdefault(fnode, node)
 
     yield {
         b'totalpaths': len(fnodes),
         b'totalitems': sum(len(v) for v in fnodes.values()),
     }
 
     for path, filenodes in sorted(fnodes.items()):
         try:
             store = getfilestore(repo, proto, path)
         except FileAccessError as e:
             raise error.WireprotoCommandError(e.msg, e.args)
 
         yield {
             b'path': path,
             b'totalitems': len(filenodes),
         }
 
         revisions = store.emitrevisions(
             filenodes.keys(),
             revisiondata=b'revision' in fields,
             assumehaveparentrevisions=haveparents,
         )
 
         for o in emitfilerevisions(repo, path, revisions, filenodes, fields):
             yield o
 
 
 @wireprotocommand(
     b'heads',
     args={
         b'publiconly': {
             b'type': b'bool',
             b'default': lambda: False,
             b'example': False,
         },
     },
     permission=b'pull',
 )
 def headsv2(repo, proto, publiconly):
     if publiconly:
         repo = repo.filtered(b'immutable')
 
     yield repo.heads()
 
 
 @wireprotocommand(
     b'known',
     args={
         b'nodes': {
             b'type': b'list',
             b'default': list,
             b'example': [b'deadbeef'],
         },
     },
     permission=b'pull',
 )
 def knownv2(repo, proto, nodes):
     result = b''.join(b'1' if n else b'0' for n in repo.known(nodes))
     yield result
 
 
 @wireprotocommand(
     b'listkeys',
     args={
         b'namespace': {
             b'type': b'bytes',
             b'example': b'ns',
         },
     },
     permission=b'pull',
 )
 def listkeysv2(repo, proto, namespace):
     keys = repo.listkeys(encoding.tolocal(namespace))
     keys = {
         encoding.fromlocal(k): encoding.fromlocal(v)
         for k, v in pycompat.iteritems(keys)
     }
 
     yield keys
 
 
 @wireprotocommand(
     b'lookup',
     args={
         b'key': {
             b'type': b'bytes',
             b'example': b'foo',
         },
     },
     permission=b'pull',
 )
 def lookupv2(repo, proto, key):
     key = encoding.tolocal(key)
 
     # TODO handle exception.
     node = repo.lookup(key)
 
     yield node
 
 
 def manifestdatacapabilities(repo, proto):
     batchsize = repo.ui.configint(
         b'experimental', b'server.manifestdata.recommended-batch-size'
     )
 
     return {
         b'recommendedbatchsize': batchsize,
     }
 
 
 @wireprotocommand(
     b'manifestdata',
     args={
         b'nodes': {
             b'type': b'list',
             b'example': [b'0123456...'],
         },
         b'haveparents': {
             b'type': b'bool',
             b'default': lambda: False,
             b'example': True,
         },
         b'fields': {
             b'type': b'set',
             b'default': set,
             b'example': {b'parents', b'revision'},
             b'validvalues': {b'parents', b'revision'},
         },
         b'tree': {
             b'type': b'bytes',
             b'example': b'',
         },
     },
     permission=b'pull',
     cachekeyfn=makecommandcachekeyfn(b'manifestdata', 1, allargs=True),
     extracapabilitiesfn=manifestdatacapabilities,
 )
 def manifestdata(repo, proto, haveparents, nodes, fields, tree):
     store = repo.manifestlog.getstorage(tree)
 
     # Validate the node is known and abort on unknown revisions.
     for node in nodes:
         try:
             store.rev(node)
         except error.LookupError:
             raise error.WireprotoCommandError(b'unknown node: %s', (node,))
 
     revisions = store.emitrevisions(
         nodes,
         revisiondata=b'revision' in fields,
         assumehaveparentrevisions=haveparents,
     )
 
     yield {
         b'totalitems': len(nodes),
     }
 
     for revision in revisions:
         d = {
             b'node': revision.node,
         }
 
         if b'parents' in fields:
             d[b'parents'] = [revision.p1node, revision.p2node]
 
         followingmeta = []
         followingdata = []
 
         if b'revision' in fields:
             if revision.revision is not None:
                 followingmeta.append((b'revision', len(revision.revision)))
                 followingdata.append(revision.revision)
             else:
                 d[b'deltabasenode'] = revision.basenode
                 followingmeta.append((b'delta', len(revision.delta)))
                 followingdata.append(revision.delta)
 
         if followingmeta:
             d[b'fieldsfollowing'] = followingmeta
 
         yield d
 
         for extra in followingdata:
             yield extra
 
 
 @wireprotocommand(
     b'pushkey',
     args={
         b'namespace': {
             b'type': b'bytes',
             b'example': b'ns',
         },
         b'key': {
             b'type': b'bytes',
             b'example': b'key',
         },
         b'old': {
             b'type': b'bytes',
             b'example': b'old',
         },
         b'new': {
             b'type': b'bytes',
             b'example': b'new',
         },
     },
     permission=b'push',
 )
 def pushkeyv2(repo, proto, namespace, key, old, new):
     # TODO handle ui output redirection
     yield repo.pushkey(
         encoding.tolocal(namespace),
         encoding.tolocal(key),
         encoding.tolocal(old),
         encoding.tolocal(new),
     )
 
 
 @wireprotocommand(
     b'rawstorefiledata',
     args={
         b'files': {
             b'type': b'list',
             b'example': [b'changelog', b'manifestlog'],
         },
         b'pathfilter': {
             b'type': b'list',
             b'default': lambda: None,
             b'example': {b'include': [b'path:tests']},
         },
     },
     permission=b'pull',
 )
 def rawstorefiledata(repo, proto, files, pathfilter):
     if not streamclone.allowservergeneration(repo):
         raise error.WireprotoCommandError(b'stream clone is disabled')
 
     # TODO support dynamically advertising what store files "sets" are
     # available. For now, we support changelog, manifestlog, and files.
     files = set(files)
     allowedfiles = {b'changelog', b'manifestlog'}
 
     unsupported = files - allowedfiles
     if unsupported:
         raise error.WireprotoCommandError(
             b'unknown file type: %s', (b', '.join(sorted(unsupported)),)
         )
 
     with repo.lock():
         topfiles = list(repo.store.topfiles())
 
     sendfiles = []
     totalsize = 0
 
     # TODO this is a bunch of storage layer interface abstractions because
     # it assumes revlogs.
-    for rl_type, name, encodedname, size in topfiles:
+    for rl_type, name, size in topfiles:
         # XXX use the `rl_type` for that
         if b'changelog' in files and name.startswith(b'00changelog'):
             pass
         elif b'manifestlog' in files and name.startswith(b'00manifest'):
             pass
         else:
             continue
 
         sendfiles.append((b'store', name, size))
         totalsize += size
 
     yield {
         b'filecount': len(sendfiles),
         b'totalsize': totalsize,
     }
 
     for location, name, size in sendfiles:
         yield {
             b'location': location,
             b'path': name,
             b'size': size,
         }
 
         # We have to use a closure for this to ensure the context manager is
         # closed only after sending the final chunk.
         def getfiledata():
             with repo.svfs(name, b'rb', auditpath=False) as fh:
                 for chunk in util.filechunkiter(fh, limit=size):
                     yield chunk
 
         yield wireprototypes.indefinitebytestringresponse(getfiledata())

tests/simplestorerepo.py

@@ -1,741 +1,745 b''
 # simplestorerepo.py - Extension that swaps in alternate repository storage.
 #
 # Copyright 2018 Gregory Szorc <[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.
 
 # To use this with the test suite:
 #
 #   $ HGREPOFEATURES="simplestore" ./run-tests.py \
 #       --extra-config-opt extensions.simplestore=`pwd`/simplestorerepo.py
 
 from __future__ import absolute_import
 
 import stat
 
 from mercurial.i18n import _
 from mercurial.node import (
     bin,
     hex,
     nullrev,
 )
 from mercurial.thirdparty import attr
 from mercurial import (
     ancestor,
     bundlerepo,
     error,
     extensions,
     localrepo,
     mdiff,
     pycompat,
     revlog,
     store,
     verify,
 )
 from mercurial.interfaces import (
     repository,
     util as interfaceutil,
 )
 from mercurial.utils import (
     cborutil,
     storageutil,
 )
 from mercurial.revlogutils import flagutil
 
 # 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'
 
 REQUIREMENT = b'testonly-simplestore'
 
 
 def validatenode(node):
     if isinstance(node, int):
         raise ValueError('expected node; got int')
 
     if len(node) != 20:
         raise ValueError('expected 20 byte node')
 
 
 def validaterev(rev):
     if not isinstance(rev, int):
         raise ValueError('expected int')
 
 
 class simplestoreerror(error.StorageError):
     pass
 
 
 @interfaceutil.implementer(repository.irevisiondelta)
 @attr.s(slots=True)
 class simplestorerevisiondelta(object):
     node = attr.ib()
     p1node = attr.ib()
     p2node = attr.ib()
     basenode = attr.ib()
     flags = attr.ib()
     baserevisionsize = attr.ib()
     revision = attr.ib()
     delta = attr.ib()
     linknode = attr.ib(default=None)
 
 
 @interfaceutil.implementer(repository.iverifyproblem)
 @attr.s(frozen=True)
 class simplefilestoreproblem(object):
     warning = attr.ib(default=None)
     error = attr.ib(default=None)
     node = attr.ib(default=None)
 
 
 @interfaceutil.implementer(repository.ifilestorage)
 class filestorage(object):
     """Implements storage for a tracked path.
 
     Data is stored in the VFS in a directory corresponding to the tracked
     path.
 
     Index data is stored in an ``index`` file using CBOR.
 
     Fulltext data is stored in files having names of the node.
     """
 
     _flagserrorclass = simplestoreerror
 
     def __init__(self, repo, svfs, path):
         self.nullid = repo.nullid
         self._repo = repo
         self._svfs = svfs
         self._path = path
 
         self._storepath = b'/'.join([b'data', path])
         self._indexpath = b'/'.join([self._storepath, b'index'])
 
         indexdata = self._svfs.tryread(self._indexpath)
         if indexdata:
             indexdata = cborutil.decodeall(indexdata)
 
         self._indexdata = indexdata or []
         self._indexbynode = {}
         self._indexbyrev = {}
         self._index = []
         self._refreshindex()
 
         self._flagprocessors = dict(flagutil.flagprocessors)
 
     def _refreshindex(self):
         self._indexbynode.clear()
         self._indexbyrev.clear()
         self._index = []
 
         for i, entry in enumerate(self._indexdata):
             self._indexbynode[entry[b'node']] = entry
             self._indexbyrev[i] = entry
 
         self._indexbynode[self._repo.nullid] = {
             b'node': self._repo.nullid,
             b'p1': self._repo.nullid,
             b'p2': self._repo.nullid,
             b'linkrev': nullrev,
             b'flags': 0,
         }
 
         self._indexbyrev[nullrev] = {
             b'node': self._repo.nullid,
             b'p1': self._repo.nullid,
             b'p2': self._repo.nullid,
             b'linkrev': nullrev,
             b'flags': 0,
         }
 
         for i, entry in enumerate(self._indexdata):
             p1rev, p2rev = self.parentrevs(self.rev(entry[b'node']))
 
             # start, length, rawsize, chainbase, linkrev, p1, p2, node
             self._index.append(
                 (0, 0, 0, -1, entry[b'linkrev'], p1rev, p2rev, entry[b'node'])
             )
 
         self._index.append((0, 0, 0, -1, -1, -1, -1, self._repo.nullid))
 
     def __len__(self):
         return len(self._indexdata)
 
     def __iter__(self):
         return iter(range(len(self)))
 
     def revs(self, start=0, stop=None):
         step = 1
         if stop is not None:
             if start > stop:
                 step = -1
 
             stop += step
         else:
             stop = len(self)
 
         return range(start, stop, step)
 
     def parents(self, node):
         validatenode(node)
 
         if node not in self._indexbynode:
             raise KeyError('unknown node')
 
         entry = self._indexbynode[node]
 
         return entry[b'p1'], entry[b'p2']
 
     def parentrevs(self, rev):
         p1, p2 = self.parents(self._indexbyrev[rev][b'node'])
         return self.rev(p1), self.rev(p2)
 
     def rev(self, node):
         validatenode(node)
 
         try:
             self._indexbynode[node]
         except KeyError:
             raise error.LookupError(node, self._indexpath, _('no node'))
 
         for rev, entry in self._indexbyrev.items():
             if entry[b'node'] == node:
                 return rev
 
         raise error.ProgrammingError(b'this should not occur')
 
     def node(self, rev):
         validaterev(rev)
 
         return self._indexbyrev[rev][b'node']
 
     def hasnode(self, node):
         validatenode(node)
         return node in self._indexbynode
 
     def censorrevision(self, tr, censornode, tombstone=b''):
         raise NotImplementedError('TODO')
 
     def lookup(self, node):
         if isinstance(node, int):
             return self.node(node)
 
         if len(node) == 20:
             self.rev(node)
             return node
 
         try:
             rev = int(node)
             if '%d' % rev != node:
                 raise ValueError
 
             if rev < 0:
                 rev = len(self) + rev
             if rev < 0 or rev >= len(self):
                 raise ValueError
 
             return self.node(rev)
         except (ValueError, OverflowError):
             pass
 
         if len(node) == 40:
             try:
                 rawnode = bin(node)
                 self.rev(rawnode)
                 return rawnode
             except TypeError:
                 pass
 
         raise error.LookupError(node, self._path, _('invalid lookup input'))
 
     def linkrev(self, rev):
         validaterev(rev)
 
         return self._indexbyrev[rev][b'linkrev']
 
     def _flags(self, rev):
         validaterev(rev)
 
         return self._indexbyrev[rev][b'flags']
 
     def _candelta(self, baserev, rev):
         validaterev(baserev)
         validaterev(rev)
 
         if (self._flags(baserev) & revlog.REVIDX_RAWTEXT_CHANGING_FLAGS) or (
             self._flags(rev) & revlog.REVIDX_RAWTEXT_CHANGING_FLAGS
         ):
             return False
 
         return True
 
     def checkhash(self, text, node, p1=None, p2=None, rev=None):
         if p1 is None and p2 is None:
             p1, p2 = self.parents(node)
         if node != storageutil.hashrevisionsha1(text, p1, p2):
             raise simplestoreerror(
                 _("integrity check failed on %s") % self._path
             )
 
     def revision(self, nodeorrev, raw=False):
         if isinstance(nodeorrev, int):
             node = self.node(nodeorrev)
         else:
             node = nodeorrev
         validatenode(node)
 
         if node == self._repo.nullid:
             return b''
 
         rev = self.rev(node)
         flags = self._flags(rev)
 
         path = b'/'.join([self._storepath, hex(node)])
         rawtext = self._svfs.read(path)
 
         if raw:
             validatehash = flagutil.processflagsraw(self, rawtext, flags)
             text = rawtext
         else:
             r = flagutil.processflagsread(self, rawtext, flags)
             text, validatehash = r
         if validatehash:
             self.checkhash(text, node, rev=rev)
 
         return text
 
     def rawdata(self, nodeorrev):
         return self.revision(raw=True)
 
     def read(self, node):
         validatenode(node)
 
         revision = self.revision(node)
 
         if not revision.startswith(b'\1\n'):
             return revision
 
         start = revision.index(b'\1\n', 2)
         return revision[start + 2 :]
 
     def renamed(self, node):
         validatenode(node)
 
         if self.parents(node)[0] != self._repo.nullid:
             return False
 
         fulltext = self.revision(node)
         m = storageutil.parsemeta(fulltext)[0]
 
         if m and 'copy' in m:
             return m['copy'], bin(m['copyrev'])
 
         return False
 
     def cmp(self, node, text):
         validatenode(node)
 
         t = text
 
         if text.startswith(b'\1\n'):
             t = b'\1\n\1\n' + text
 
         p1, p2 = self.parents(node)
 
         if storageutil.hashrevisionsha1(t, p1, p2) == node:
             return False
 
         if self.iscensored(self.rev(node)):
             return text != b''
 
         if self.renamed(node):
             t2 = self.read(node)
             return t2 != text
 
         return True
 
     def size(self, rev):
         validaterev(rev)
 
         node = self._indexbyrev[rev][b'node']
 
         if self.renamed(node):
             return len(self.read(node))
 
         if self.iscensored(rev):
             return 0
 
         return len(self.revision(node))
 
     def iscensored(self, rev):
         validaterev(rev)
 
         return self._flags(rev) & repository.REVISION_FLAG_CENSORED
 
     def commonancestorsheads(self, a, b):
         validatenode(a)
         validatenode(b)
 
         a = self.rev(a)
         b = self.rev(b)
 
         ancestors = ancestor.commonancestorsheads(self.parentrevs, a, b)
         return pycompat.maplist(self.node, ancestors)
 
     def descendants(self, revs):
         # This is a copy of revlog.descendants()
         first = min(revs)
         if first == nullrev:
             for i in self:
                 yield i
             return
 
         seen = set(revs)
         for i in self.revs(start=first + 1):
             for x in self.parentrevs(i):
                 if x != nullrev and x in seen:
                     seen.add(i)
                     yield i
                     break
 
     # Required by verify.
     def files(self):
         entries = self._svfs.listdir(self._storepath)
 
         # Strip out undo.backup.* files created as part of transaction
         # recording.
         entries = [f for f in entries if not f.startswith('undo.backup.')]
 
         return [b'/'.join((self._storepath, f)) for f in entries]
 
     def storageinfo(
         self,
         exclusivefiles=False,
         sharedfiles=False,
         revisionscount=False,
         trackedsize=False,
         storedsize=False,
     ):
         # TODO do a real implementation of this
         return {
             'exclusivefiles': [],
             'sharedfiles': [],
             'revisionscount': len(self),
             'trackedsize': 0,
             'storedsize': None,
         }
 
     def verifyintegrity(self, state):
         state['skipread'] = set()
         for rev in self:
             node = self.node(rev)
             try:
                 self.revision(node)
             except Exception as e:
                 yield simplefilestoreproblem(
                     error='unpacking %s: %s' % (node, e), node=node
                 )
                 state['skipread'].add(node)
 
     def emitrevisions(
         self,
         nodes,
         nodesorder=None,
         revisiondata=False,
         assumehaveparentrevisions=False,
         deltamode=repository.CG_DELTAMODE_STD,
         sidedata_helpers=None,
     ):
         # TODO this will probably break on some ordering options.
         nodes = [n for n in nodes if n != self._repo.nullid]
         if not nodes:
             return
         for delta in storageutil.emitrevisions(
             self,
             nodes,
             nodesorder,
             simplestorerevisiondelta,
             revisiondata=revisiondata,
             assumehaveparentrevisions=assumehaveparentrevisions,
             deltamode=deltamode,
             sidedata_helpers=sidedata_helpers,
         ):
             yield delta
 
     def add(self, text, meta, transaction, linkrev, p1, p2):
         if meta or text.startswith(b'\1\n'):
             text = storageutil.packmeta(meta, text)
 
         return self.addrevision(text, transaction, linkrev, p1, p2)
 
     def addrevision(
         self,
         text,
         transaction,
         linkrev,
         p1,
         p2,
         node=None,
         flags=revlog.REVIDX_DEFAULT_FLAGS,
         cachedelta=None,
     ):
         validatenode(p1)
         validatenode(p2)
 
         if flags:
             node = node or storageutil.hashrevisionsha1(text, p1, p2)
 
         rawtext, validatehash = flagutil.processflagswrite(self, text, flags)
 
         node = node or storageutil.hashrevisionsha1(text, p1, p2)
 
         if node in self._indexbynode:
             return node
 
         if validatehash:
             self.checkhash(rawtext, node, p1=p1, p2=p2)
 
         return self._addrawrevision(
             node, rawtext, transaction, linkrev, p1, p2, flags
         )
 
     def _addrawrevision(self, node, rawtext, transaction, link, p1, p2, flags):
         transaction.addbackup(self._indexpath)
 
         path = b'/'.join([self._storepath, hex(node)])
 
         self._svfs.write(path, rawtext)
 
         self._indexdata.append(
             {
                 b'node': node,
                 b'p1': p1,
                 b'p2': p2,
                 b'linkrev': link,
                 b'flags': flags,
             }
         )
 
         self._reflectindexupdate()
 
         return node
 
     def _reflectindexupdate(self):
         self._refreshindex()
         self._svfs.write(
             self._indexpath, ''.join(cborutil.streamencode(self._indexdata))
         )
 
     def addgroup(
         self,
         deltas,
         linkmapper,
         transaction,
         addrevisioncb=None,
         duplicaterevisioncb=None,
         maybemissingparents=False,
     ):
         if maybemissingparents:
             raise error.Abort(
                 _('simple store does not support missing parents ' 'write mode')
             )
 
         empty = True
 
         transaction.addbackup(self._indexpath)
 
         for node, p1, p2, linknode, deltabase, delta, flags in deltas:
             linkrev = linkmapper(linknode)
             flags = flags or revlog.REVIDX_DEFAULT_FLAGS
 
             if node in self._indexbynode:
                 if duplicaterevisioncb:
                     duplicaterevisioncb(self, self.rev(node))
                 empty = False
                 continue
 
             # Need to resolve the fulltext from the delta base.
             if deltabase == self._repo.nullid:
                 text = mdiff.patch(b'', delta)
             else:
                 text = mdiff.patch(self.revision(deltabase), delta)
 
             rev = self._addrawrevision(
                 node, text, transaction, linkrev, p1, p2, flags
             )
 
             if addrevisioncb:
                 addrevisioncb(self, rev)
             empty = False
         return not empty
 
     def _headrevs(self):
         # Assume all revisions are heads by default.
         revishead = {rev: True for rev in self._indexbyrev}
 
         for rev, entry in self._indexbyrev.items():
             # Unset head flag for all seen parents.
             revishead[self.rev(entry[b'p1'])] = False
             revishead[self.rev(entry[b'p2'])] = False
 
         return [rev for rev, ishead in sorted(revishead.items()) if ishead]
 
     def heads(self, start=None, stop=None):
         # This is copied from revlog.py.
         if start is None and stop is None:
             if not len(self):
                 return [self._repo.nullid]
             return [self.node(r) for r in self._headrevs()]
 
         if start is None:
             start = self._repo.nullid
         if stop is None:
             stop = []
         stoprevs = {self.rev(n) for n in stop}
         startrev = self.rev(start)
         reachable = {startrev}
         heads = {startrev}
 
         parentrevs = self.parentrevs
         for r in self.revs(start=startrev + 1):
             for p in parentrevs(r):
                 if p in reachable:
                     if r not in stoprevs:
                         reachable.add(r)
                     heads.add(r)
                 if p in heads and p not in stoprevs:
                     heads.remove(p)
 
         return [self.node(r) for r in heads]
 
     def children(self, node):
         validatenode(node)
 
         # This is a copy of revlog.children().
         c = []
         p = self.rev(node)
         for r in self.revs(start=p + 1):
             prevs = [pr for pr in self.parentrevs(r) if pr != nullrev]
             if prevs:
                 for pr in prevs:
                     if pr == p:
                         c.append(self.node(r))
             elif p == nullrev:
                 c.append(self.node(r))
         return c
 
     def getstrippoint(self, minlink):
         return storageutil.resolvestripinfo(
             minlink,
             len(self) - 1,
             self._headrevs(),
             self.linkrev,
             self.parentrevs,
         )
 
     def strip(self, minlink, transaction):
         if not len(self):
             return
 
         rev, _ignored = self.getstrippoint(minlink)
         if rev == len(self):
             return
 
         # Purge index data starting at the requested revision.
         self._indexdata[rev:] = []
         self._reflectindexupdate()
 
 
 def issimplestorefile(f, kind, st):
     if kind != stat.S_IFREG:
         return False
 
     if store.isrevlog(f, kind, st):
         return False
 
     # Ignore transaction undo files.
     if f.startswith('undo.'):
         return False
 
     # Otherwise assume it belongs to the simple store.
     return True
 
 
 class simplestore(store.encodedstore):
-    def datafiles(self):
+    def datafiles(self, undecodable=None):
         for x in super(simplestore, self).datafiles():
             yield x
 
         # Supplement with non-revlog files.
         extrafiles = self._walk('data', True, filefilter=issimplestorefile)
 
-        for unencoded, encoded, size in extrafiles:
+        for f1, size in extrafiles:
             try:
-                unencoded = store.decodefilename(unencoded)
+                f2 = store.decodefilename(f1)
             except KeyError:
-                unencoded = None
+                if undecodable is None:
+                    raise error.StorageError(b'undecodable revlog name %s' % f1)
+                else:
+                    undecodable.append(f1)
+                    continue
 
-            yield unencoded, encoded, size
+            yield f2, size
 
 
 def reposetup(ui, repo):
     if not repo.local():
         return
 
     if isinstance(repo, bundlerepo.bundlerepository):
         raise error.Abort(_('cannot use simple store with bundlerepo'))
 
     class simplestorerepo(repo.__class__):
         def file(self, f):
             return filestorage(repo, self.svfs, f)
 
     repo.__class__ = simplestorerepo
 
 
 def featuresetup(ui, supported):
     supported.add(REQUIREMENT)
 
 
 def newreporequirements(orig, ui, createopts):
     """Modifies default requirements for new repos to use the simple store."""
     requirements = orig(ui, createopts)
 
     # These requirements are only used to affect creation of the store
     # object. We have our own store. So we can remove them.
     # TODO do this once we feel like taking the test hit.
     # if 'fncache' in requirements:
     #    requirements.remove('fncache')
     # if 'dotencode' in requirements:
     #    requirements.remove('dotencode')
 
     requirements.add(REQUIREMENT)
 
     return requirements
 
 
 def makestore(orig, requirements, path, vfstype):
     if REQUIREMENT not in requirements:
         return orig(requirements, path, vfstype)
 
     return simplestore(path, vfstype)
 
 
 def verifierinit(orig, self, *args, **kwargs):
     orig(self, *args, **kwargs)
 
     # We don't care that files in the store don't align with what is
     # advertised. So suppress these warnings.
     self.warnorphanstorefiles = False
 
 
 def extsetup(ui):
     localrepo.featuresetupfuncs.add(featuresetup)
 
     extensions.wrapfunction(
         localrepo, 'newreporequirements', newreporequirements
     )
     extensions.wrapfunction(localrepo, 'makestore', makestore)
     extensions.wrapfunction(verify.verifier, '__init__', verifierinit)