upgrade: implement partial upgrade for upgrading persistent-nodemap

Upgrading repositories to use persistent nodemap should be fast and easy as it
requires only two things:

1) Updating the requirements
2) Writing a persistent-nodemap on disk

For both of the steps above, we don't need to edit existing revlogs.

This patch makes upgrade only do the above mentioned two steps if we are
only upgarding to use persistent-nodemap feature.

Since `nodemap.persist_nodemap()` assumes that there exists a nodemap file for
the given revlog if we are trying to call it, this patch adds `force` argument
to create a file if does not exist which is true in our upgrade case.

The test changes demonstrate that we no longer write nodemap files for manifest
after upgrade which I think is desirable.

Differential Revision: https://phab.mercurial-scm.org/D9936

mercurial/revlogutils/nodemap.py

@@ -1,647 +1,652 b''
 # nodemap.py - nodemap related code and utilities
 #
 # Copyright 2019 Pierre-Yves David <[e-mail]>
 # Copyright 2019 George Racinet <[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 os
 import re
 import struct
 
 from ..node import hex
 
 from .. import (
     error,
     util,
 )
 
 
 class NodeMap(dict):
     def __missing__(self, x):
         raise error.RevlogError(b'unknown node: %s' % x)
 
 
 def persisted_data(revlog):
     """read the nodemap for a revlog from disk"""
     if revlog.nodemap_file is None:
         return None
     pdata = revlog.opener.tryread(revlog.nodemap_file)
     if not pdata:
         return None
     offset = 0
     (version,) = S_VERSION.unpack(pdata[offset : offset + S_VERSION.size])
     if version != ONDISK_VERSION:
         return None
     offset += S_VERSION.size
     headers = S_HEADER.unpack(pdata[offset : offset + S_HEADER.size])
     uid_size, tip_rev, data_length, data_unused, tip_node_size = headers
     offset += S_HEADER.size
     docket = NodeMapDocket(pdata[offset : offset + uid_size])
     offset += uid_size
     docket.tip_rev = tip_rev
     docket.tip_node = pdata[offset : offset + tip_node_size]
     docket.data_length = data_length
     docket.data_unused = data_unused
 
     filename = _rawdata_filepath(revlog, docket)
     use_mmap = revlog.opener.options.get(b"persistent-nodemap.mmap")
     try:
         with revlog.opener(filename) as fd:
             if use_mmap:
                 data = util.buffer(util.mmapread(fd, data_length))
             else:
                 data = fd.read(data_length)
     except (IOError, OSError) as e:
         if e.errno == errno.ENOENT:
             return None
         else:
             raise
     if len(data) < data_length:
         return None
     return docket, data
 
 
 def setup_persistent_nodemap(tr, revlog):
     """Install whatever is needed transaction side to persist a nodemap on disk
 
     (only actually persist the nodemap if this is relevant for this revlog)
     """
     if revlog._inline:
         return  # inlined revlog are too small for this to be relevant
     if revlog.nodemap_file is None:
         return  # we do not use persistent_nodemap on this revlog
 
     # we need to happen after the changelog finalization, in that use "cl-"
     callback_id = b"nm-revlog-persistent-nodemap-%s" % revlog.nodemap_file
     if tr.hasfinalize(callback_id):
         return  # no need to register again
     tr.addpending(
         callback_id, lambda tr: persist_nodemap(tr, revlog, pending=True)
     )
     tr.addfinalize(callback_id, lambda tr: persist_nodemap(tr, revlog))
 
 
 class _NoTransaction(object):
     """transaction like object to update the nodemap outside a transaction"""
 
     def __init__(self):
         self._postclose = {}
 
     def addpostclose(self, callback_id, callback_func):
         self._postclose[callback_id] = callback_func
 
     def registertmp(self, *args, **kwargs):
         pass
 
     def addbackup(self, *args, **kwargs):
         pass
 
     def add(self, *args, **kwargs):
         pass
 
     def addabort(self, *args, **kwargs):
         pass
 
     def _report(self, *args):
         pass
 
 
 def update_persistent_nodemap(revlog):
     """update the persistent nodemap right now
 
     To be used for updating the nodemap on disk outside of a normal transaction
     setup (eg, `debugupdatecache`).
     """
     if revlog._inline:
         return  # inlined revlog are too small for this to be relevant
     if revlog.nodemap_file is None:
         return  # we do not use persistent_nodemap on this revlog
 
     notr = _NoTransaction()
     persist_nodemap(notr, revlog)
     for k in sorted(notr._postclose):
         notr._postclose[k](None)
 
 
-def persist_nodemap(tr, revlog, pending=False):
+def persist_nodemap(tr, revlog, pending=False, force=False):
     """Write nodemap data on disk for a given revlog"""
     if getattr(revlog, 'filteredrevs', ()):
         raise error.ProgrammingError(
             "cannot persist nodemap of a filtered changelog"
         )
     if revlog.nodemap_file is None:
-        msg = "calling persist nodemap on a revlog without the feature enabled"
-        raise error.ProgrammingError(msg)
+        if force:
+            revlog.nodemap_file = get_nodemap_file(
+                revlog.opener, revlog.indexfile
+            )
+        else:
+            msg = "calling persist nodemap on a revlog without the feature enabled"
+            raise error.ProgrammingError(msg)
 
     can_incremental = util.safehasattr(revlog.index, "nodemap_data_incremental")
     ondisk_docket = revlog._nodemap_docket
     feed_data = util.safehasattr(revlog.index, "update_nodemap_data")
     use_mmap = revlog.opener.options.get(b"persistent-nodemap.mmap")
 
     data = None
     # first attemp an incremental update of the data
     if can_incremental and ondisk_docket is not None:
         target_docket = revlog._nodemap_docket.copy()
         (
             src_docket,
             data_changed_count,
             data,
         ) = revlog.index.nodemap_data_incremental()
         new_length = target_docket.data_length + len(data)
         new_unused = target_docket.data_unused + data_changed_count
         if src_docket != target_docket:
             data = None
         elif new_length <= (new_unused * 10):  # under 10% of unused data
             data = None
         else:
             datafile = _rawdata_filepath(revlog, target_docket)
             # EXP-TODO: if this is a cache, this should use a cache vfs, not a
             # store vfs
             tr.add(datafile, target_docket.data_length)
             with revlog.opener(datafile, b'r+') as fd:
                 fd.seek(target_docket.data_length)
                 fd.write(data)
                 if feed_data:
                     if use_mmap:
                         fd.seek(0)
                         new_data = fd.read(new_length)
                     else:
                         fd.flush()
                         new_data = util.buffer(util.mmapread(fd, new_length))
             target_docket.data_length = new_length
             target_docket.data_unused = new_unused
 
     if data is None:
         # otherwise fallback to a full new export
         target_docket = NodeMapDocket()
         datafile = _rawdata_filepath(revlog, target_docket)
         if util.safehasattr(revlog.index, "nodemap_data_all"):
             data = revlog.index.nodemap_data_all()
         else:
             data = persistent_data(revlog.index)
         # EXP-TODO: if this is a cache, this should use a cache vfs, not a
         # store vfs
 
         tryunlink = revlog.opener.tryunlink
 
         def abortck(tr):
             tryunlink(datafile)
 
         callback_id = b"delete-%s" % datafile
 
         # some flavor of the transaction abort does not cleanup new file, it
         # simply empty them.
         tr.addabort(callback_id, abortck)
         with revlog.opener(datafile, b'w+') as fd:
             fd.write(data)
             if feed_data:
                 if use_mmap:
                     new_data = data
                 else:
                     fd.flush()
                     new_data = util.buffer(util.mmapread(fd, len(data)))
         target_docket.data_length = len(data)
     target_docket.tip_rev = revlog.tiprev()
     target_docket.tip_node = revlog.node(target_docket.tip_rev)
     # EXP-TODO: if this is a cache, this should use a cache vfs, not a
     # store vfs
     file_path = revlog.nodemap_file
     if pending:
         file_path += b'.a'
         tr.registertmp(file_path)
     else:
         tr.addbackup(file_path)
 
     with revlog.opener(file_path, b'w', atomictemp=True) as fp:
         fp.write(target_docket.serialize())
     revlog._nodemap_docket = target_docket
     if feed_data:
         revlog.index.update_nodemap_data(target_docket, new_data)
 
     # search for old index file in all cases, some older process might have
     # left one behind.
     olds = _other_rawdata_filepath(revlog, target_docket)
     if olds:
         realvfs = getattr(revlog, '_realopener', revlog.opener)
 
         def cleanup(tr):
             for oldfile in olds:
                 realvfs.tryunlink(oldfile)
 
         callback_id = b"revlog-cleanup-nodemap-%s" % revlog.nodemap_file
         tr.addpostclose(callback_id, cleanup)
 
 
 ### Nodemap docket file
 #
 # The nodemap data are stored on disk using 2 files:
 #
 # * a raw data files containing a persistent nodemap
 #   (see `Nodemap Trie` section)
 #
 # * a small "docket" file containing medatadata
 #
 # While the nodemap data can be multiple tens of megabytes, the "docket" is
 # small, it is easy to update it automatically or to duplicated its content
 # during a transaction.
 #
 # Multiple raw data can exist at the same time (The currently valid one and a
 # new one beind used by an in progress transaction). To accomodate this, the
 # filename hosting the raw data has a variable parts. The exact filename is
 # specified inside the "docket" file.
 #
 # The docket file contains information to find, qualify and validate the raw
 # data. Its content is currently very light, but it will expand as the on disk
 # nodemap gains the necessary features to be used in production.
 
 ONDISK_VERSION = 1
 S_VERSION = struct.Struct(">B")
 S_HEADER = struct.Struct(">BQQQQ")
 
 ID_SIZE = 8
 
 
 def _make_uid():
     """return a new unique identifier.
 
     The identifier is random and composed of ascii characters."""
     return hex(os.urandom(ID_SIZE))
 
 
 class NodeMapDocket(object):
     """metadata associated with persistent nodemap data
 
     The persistent data may come from disk or be on their way to disk.
     """
 
     def __init__(self, uid=None):
         if uid is None:
             uid = _make_uid()
         # a unique identifier for the data file:
         #   - When new data are appended, it is preserved.
         #   - When a new data file is created, a new identifier is generated.
         self.uid = uid
         # the tipmost revision stored in the data file. This revision and all
         # revision before it are expected to be encoded in the data file.
         self.tip_rev = None
         # the node of that tipmost revision, if it mismatch the current index
         # data the docket is not valid for the current index and should be
         # discarded.
         #
         # note: this method is not perfect as some destructive operation could
         # preserve the same tip_rev + tip_node while altering lower revision.
         # However this multiple other caches have the same vulnerability (eg:
         # brancmap cache).
         self.tip_node = None
         # the size (in bytes) of the persisted data to encode the nodemap valid
         # for `tip_rev`.
         #   - data file shorter than this are corrupted,
         #   - any extra data should be ignored.
         self.data_length = None
         # the amount (in bytes) of "dead" data, still in the data file but no
         # longer used for the nodemap.
         self.data_unused = 0
 
     def copy(self):
         new = NodeMapDocket(uid=self.uid)
         new.tip_rev = self.tip_rev
         new.tip_node = self.tip_node
         new.data_length = self.data_length
         new.data_unused = self.data_unused
         return new
 
     def __cmp__(self, other):
         if self.uid < other.uid:
             return -1
         if self.uid > other.uid:
             return 1
         elif self.data_length < other.data_length:
             return -1
         elif self.data_length > other.data_length:
             return 1
         return 0
 
     def __eq__(self, other):
         return self.uid == other.uid and self.data_length == other.data_length
 
     def serialize(self):
         """return serialized bytes for a docket using the passed uid"""
         data = []
         data.append(S_VERSION.pack(ONDISK_VERSION))
         headers = (
             len(self.uid),
             self.tip_rev,
             self.data_length,
             self.data_unused,
             len(self.tip_node),
         )
         data.append(S_HEADER.pack(*headers))
         data.append(self.uid)
         data.append(self.tip_node)
         return b''.join(data)
 
 
 def _rawdata_filepath(revlog, docket):
     """The (vfs relative) nodemap's rawdata file for a given uid"""
     if revlog.nodemap_file.endswith(b'.n.a'):
         prefix = revlog.nodemap_file[:-4]
     else:
         prefix = revlog.nodemap_file[:-2]
     return b"%s-%s.nd" % (prefix, docket.uid)
 
 
 def _other_rawdata_filepath(revlog, docket):
     prefix = revlog.nodemap_file[:-2]
     pattern = re.compile(br"(^|/)%s-[0-9a-f]+\.nd$" % prefix)
     new_file_path = _rawdata_filepath(revlog, docket)
     new_file_name = revlog.opener.basename(new_file_path)
     dirpath = revlog.opener.dirname(new_file_path)
     others = []
     for f in revlog.opener.listdir(dirpath):
         if pattern.match(f) and f != new_file_name:
             others.append(f)
     return others
 
 
 ### Nodemap Trie
 #
 # This is a simple reference implementation to compute and persist a nodemap
 # trie. This reference implementation is write only. The python version of this
 # is not expected to be actually used, since it wont provide performance
 # improvement over existing non-persistent C implementation.
 #
 # The nodemap is persisted as Trie using 4bits-address/16-entries block. each
 # revision can be adressed using its node shortest prefix.
 #
 # The trie is stored as a sequence of block. Each block contains 16 entries
 # (signed 64bit integer, big endian). Each entry can be one of the following:
 #
 #  * value >=  0 -> index of sub-block
 #  * value == -1 -> no value
 #  * value <  -1 -> encoded revision: rev = -(value+2)
 #
 # See REV_OFFSET and _transform_rev below.
 #
 # The implementation focus on simplicity, not on performance. A Rust
 # implementation should provide a efficient version of the same binary
 # persistence. This reference python implementation is never meant to be
 # extensively use in production.
 
 
 def persistent_data(index):
     """return the persistent binary form for a nodemap for a given index"""
     trie = _build_trie(index)
     return _persist_trie(trie)
 
 
 def update_persistent_data(index, root, max_idx, last_rev):
     """return the incremental update for persistent nodemap from a given index"""
     changed_block, trie = _update_trie(index, root, last_rev)
     return (
         changed_block * S_BLOCK.size,
         _persist_trie(trie, existing_idx=max_idx),
     )
 
 
 S_BLOCK = struct.Struct(">" + ("l" * 16))
 
 NO_ENTRY = -1
 # rev 0 need to be -2 because 0 is used by block, -1 is a special value.
 REV_OFFSET = 2
 
 
 def _transform_rev(rev):
     """Return the number used to represent the rev in the tree.
 
     (or retrieve a rev number from such representation)
 
     Note that this is an involution, a function equal to its inverse (i.e.
     which gives the identity when applied to itself).
     """
     return -(rev + REV_OFFSET)
 
 
 def _to_int(hex_digit):
     """turn an hexadecimal digit into a proper integer"""
     return int(hex_digit, 16)
 
 
 class Block(dict):
     """represent a block of the Trie
 
     contains up to 16 entry indexed from 0 to 15"""
 
     def __init__(self):
         super(Block, self).__init__()
         # If this block exist on disk, here is its ID
         self.ondisk_id = None
 
     def __iter__(self):
         return iter(self.get(i) for i in range(16))
 
 
 def _build_trie(index):
     """build a nodemap trie
 
     The nodemap stores revision number for each unique prefix.
 
     Each block is a dictionary with keys in `[0, 15]`. Values are either
     another block or a revision number.
     """
     root = Block()
     for rev in range(len(index)):
         current_hex = hex(index[rev][7])
         _insert_into_block(index, 0, root, rev, current_hex)
     return root
 
 
 def _update_trie(index, root, last_rev):
     """consume"""
     changed = 0
     for rev in range(last_rev + 1, len(index)):
         current_hex = hex(index[rev][7])
         changed += _insert_into_block(index, 0, root, rev, current_hex)
     return changed, root
 
 
 def _insert_into_block(index, level, block, current_rev, current_hex):
     """insert a new revision in a block
 
     index: the index we are adding revision for
     level: the depth of the current block in the trie
     block: the block currently being considered
     current_rev: the revision number we are adding
     current_hex: the hexadecimal representation of the of that revision
     """
     changed = 1
     if block.ondisk_id is not None:
         block.ondisk_id = None
     hex_digit = _to_int(current_hex[level : level + 1])
     entry = block.get(hex_digit)
     if entry is None:
         # no entry, simply store the revision number
         block[hex_digit] = current_rev
     elif isinstance(entry, dict):
         # need to recurse to an underlying block
         changed += _insert_into_block(
             index, level + 1, entry, current_rev, current_hex
         )
     else:
         # collision with a previously unique prefix, inserting new
         # vertices to fit both entry.
         other_hex = hex(index[entry][7])
         other_rev = entry
         new = Block()
         block[hex_digit] = new
         _insert_into_block(index, level + 1, new, other_rev, other_hex)
         _insert_into_block(index, level + 1, new, current_rev, current_hex)
     return changed
 
 
 def _persist_trie(root, existing_idx=None):
     """turn a nodemap trie into persistent binary data
 
     See `_build_trie` for nodemap trie structure"""
     block_map = {}
     if existing_idx is not None:
         base_idx = existing_idx + 1
     else:
         base_idx = 0
     chunks = []
     for tn in _walk_trie(root):
         if tn.ondisk_id is not None:
             block_map[id(tn)] = tn.ondisk_id
         else:
             block_map[id(tn)] = len(chunks) + base_idx
             chunks.append(_persist_block(tn, block_map))
     return b''.join(chunks)
 
 
 def _walk_trie(block):
     """yield all the block in a trie
 
     Children blocks are always yield before their parent block.
     """
     for (__, item) in sorted(block.items()):
         if isinstance(item, dict):
             for sub_block in _walk_trie(item):
                 yield sub_block
     yield block
 
 
 def _persist_block(block_node, block_map):
     """produce persistent binary data for a single block
 
     Children block are assumed to be already persisted and present in
     block_map.
     """
     data = tuple(_to_value(v, block_map) for v in block_node)
     return S_BLOCK.pack(*data)
 
 
 def _to_value(item, block_map):
     """persist any value as an integer"""
     if item is None:
         return NO_ENTRY
     elif isinstance(item, dict):
         return block_map[id(item)]
     else:
         return _transform_rev(item)
 
 
 def parse_data(data):
     """parse parse nodemap data into a nodemap Trie"""
     if (len(data) % S_BLOCK.size) != 0:
         msg = "nodemap data size is not a multiple of block size (%d): %d"
         raise error.Abort(msg % (S_BLOCK.size, len(data)))
     if not data:
         return Block(), None
     block_map = {}
     new_blocks = []
     for i in range(0, len(data), S_BLOCK.size):
         block = Block()
         block.ondisk_id = len(block_map)
         block_map[block.ondisk_id] = block
         block_data = data[i : i + S_BLOCK.size]
         values = S_BLOCK.unpack(block_data)
         new_blocks.append((block, values))
     for b, values in new_blocks:
         for idx, v in enumerate(values):
             if v == NO_ENTRY:
                 continue
             elif v >= 0:
                 b[idx] = block_map[v]
             else:
                 b[idx] = _transform_rev(v)
     return block, i // S_BLOCK.size
 
 
 # debug utility
 
 
 def check_data(ui, index, data):
     """verify that the provided nodemap data are valid for the given idex"""
     ret = 0
     ui.status((b"revision in index:   %d\n") % len(index))
     root, __ = parse_data(data)
     all_revs = set(_all_revisions(root))
     ui.status((b"revision in nodemap: %d\n") % len(all_revs))
     for r in range(len(index)):
         if r not in all_revs:
             msg = b"  revision missing from nodemap: %d\n" % r
             ui.write_err(msg)
             ret = 1
         else:
             all_revs.remove(r)
         nm_rev = _find_node(root, hex(index[r][7]))
         if nm_rev is None:
             msg = b"  revision node does not match any entries: %d\n" % r
             ui.write_err(msg)
             ret = 1
         elif nm_rev != r:
             msg = (
                 b"  revision node does not match the expected revision: "
                 b"%d != %d\n" % (r, nm_rev)
             )
             ui.write_err(msg)
             ret = 1
 
     if all_revs:
         for r in sorted(all_revs):
             msg = b"  extra revision in  nodemap: %d\n" % r
             ui.write_err(msg)
         ret = 1
     return ret
 
 
 def _all_revisions(root):
     """return all revisions stored in a Trie"""
     for block in _walk_trie(root):
         for v in block:
             if v is None or isinstance(v, Block):
                 continue
             yield v
 
 
 def _find_node(block, node):
     """find the revision associated with a given node"""
     entry = block.get(_to_int(node[0:1]))
     if isinstance(entry, dict):
         return _find_node(entry, node[1:])
     return entry
 
 
 def get_nodemap_file(opener, indexfile):
     if indexfile.endswith(b'.a'):
         pending_path = indexfile[:-4] + b".n.a"
         if opener.exists(pending_path):
             return pending_path
         else:
             return indexfile[:-4] + b".n"
     else:
         return indexfile[:-2] + b".n"

mercurial/upgrade_utils/engine.py

@@ -1,539 +1,556 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,
     revlog,
     scmutil,
     util,
     vfs as vfsmod,
 )
+from ..revlogutils import nodemap
 
 
 def _revlogfrompath(repo, path):
     """Obtain a revlog from a repo path.
 
     An instance of the appropriate class is returned.
     """
     if path == b'00changelog.i':
         return changelog.changelog(repo.svfs)
     elif path.endswith(b'00manifest.i'):
         mandir = path[: -len(b'00manifest.i')]
         return manifest.manifestrevlog(repo.svfs, tree=mandir)
     else:
         # reverse of "/".join(("data", path + ".i"))
         return filelog.filelog(repo.svfs, path[5:-2])
 
 
 def _copyrevlog(tr, destrepo, oldrl, 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, 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 not (
         unencodedname.endswith(b'00changelog.i')
         or unencodedname.endswith(b'00manifest.i')
     ):
         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 getsidedatacompanion(srcrepo, dstrepo):
     sidedatacompanion = None
     removedreqs = srcrepo.requirements - dstrepo.requirements
     addedreqs = dstrepo.requirements - srcrepo.requirements
     if requirements.SIDEDATA_REQUIREMENT in removedreqs:
 
         def sidedatacompanion(rl, rev):
             rl = getattr(rl, '_revlog', rl)
             if rl.flags(rev) & revlog.REVIDX_SIDEDATA:
                 return True, (), {}, 0, 0
             return False, (), {}, 0, 0
 
     elif requirements.COPIESSDC_REQUIREMENT in addedreqs:
         sidedatacompanion = metadata.getsidedataadder(srcrepo, dstrepo)
     elif requirements.COPIESSDC_REQUIREMENT in removedreqs:
         sidedatacompanion = metadata.getsidedataremover(srcrepo, dstrepo)
     return sidedatacompanion
 
 
 def matchrevlog(revlogfilter, entry):
     """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 entry.endswith(b'00changelog.i'):
         return UPGRADE_CHANGELOG in revlogfilter
     elif entry.endswith(b'00manifest.i'):
         return UPGRADE_MANIFEST in revlogfilter
     return UPGRADE_FILELOGS in revlogfilter
 
 
 def _perform_clone(
     ui,
     dstrepo,
     tr,
     old_revlog,
     unencoded,
     upgrade_op,
     sidedatacompanion,
     oncopiedrevision,
 ):
     """ returns the new revlog object created"""
     newrl = None
     if matchrevlog(upgrade_op.revlogs_to_process, unencoded):
         ui.note(
             _(b'cloning %d revisions from %s\n') % (len(old_revlog), unencoded)
         )
         newrl = _revlogfrompath(dstrepo, unencoded)
         old_revlog.clone(
             tr,
             newrl,
             addrevisioncb=oncopiedrevision,
             deltareuse=upgrade_op.delta_reuse_mode,
             forcedeltabothparents=upgrade_op.force_re_delta_both_parents,
             sidedatacompanion=sidedatacompanion,
         )
     else:
         msg = _(b'blindly copying %s containing %i revisions\n')
         ui.note(msg % (unencoded, len(old_revlog)))
         _copyrevlog(tr, dstrepo, old_revlog, unencoded)
 
         newrl = _revlogfrompath(dstrepo, 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 unencoded, encoded, size in alldatafiles:
         if not unencoded.endswith(b'.i'):
             continue
 
         rl = _revlogfrompath(srcrepo, 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 isinstance(rl, changelog.changelog):
             changelogs[unencoded] = rl
             crevcount += len(rl)
             csrcsize += datasize
             crawsize += rawsize
         elif isinstance(rl, manifest.manifestrevlog):
             manifests[unencoded] = rl
             mcount += 1
             mrevcount += len(rl)
             msrcsize += datasize
             mrawsize += rawsize
         elif isinstance(rl, filelog.filelog):
             filelogs[unencoded] = 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()
 
     sidedatacompanion = getsidedatacompanion(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, oldrl in sorted(filelogs.items()):
         newrl = _perform_clone(
             ui,
             dstrepo,
             tr,
             oldrl,
             unencoded,
             upgrade_op,
             sidedatacompanion,
             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, oldrl in sorted(manifests.items()):
         newrl = _perform_clone(
             ui,
             dstrepo,
             tr,
             oldrl,
             unencoded,
             upgrade_op,
             sidedatacompanion,
             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, oldrl in sorted(changelogs.items()):
         newrl = _perform_clone(
             ui,
             dstrepo,
             tr,
             oldrl,
             unencoded,
             upgrade_op,
             sidedatacompanion,
             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 path.endswith((b'.i', b'.d', b'.n', b'.nd')):
             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_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)
+        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.
             backupvfs.unlink(b'store/lock')
 
     return backuppath

tests/test-persistent-nodemap.t

@@ -1,768 +1,751 b''
 ===================================
 Test the persistent on-disk nodemap
 ===================================
 
   $ cat << EOF >> $HGRCPATH
   > [format]
   > use-persistent-nodemap=yes
   > [devel]
   > persistent-nodemap=yes
   > EOF
 
   $ hg init test-repo --config storage.revlog.persistent-nodemap.slow-path=allow
   $ cd test-repo
 
 Check handling of the default slow-path value
 
 #if no-pure no-rust
 
   $ hg id
   abort: accessing `persistent-nodemap` repository without associated fast implementation.
   (check `hg help config.format.use-persistent-nodemap` for details)
   [255]
 
 Unlock further check (we are here to test the feature)
 
   $ cat << EOF >> $HGRCPATH
   > [storage]
   > # to avoid spamming the test
   > revlog.persistent-nodemap.slow-path=allow
   > EOF
 
 #endif
 
 #if rust
 
 Regression test for a previous bug in Rust/C FFI for the `Revlog_CAPI` capsule:
 in places where `mercurial/cext/revlog.c` function signatures use `Py_ssize_t`
 (64 bits on Linux x86_64), corresponding declarations in `rust/hg-cpython/src/cindex.rs`
 incorrectly used `libc::c_int` (32 bits).
 As a result, -1 passed from Rust for the null revision became 4294967295 in C.
 
   $ hg log -r 00000000
   changeset:   -1:000000000000
   tag:         tip
   user:        
   date:        Thu Jan 01 00:00:00 1970 +0000
   
 
 #endif
 
 
   $ hg debugformat
   format-variant     repo
   fncache:            yes
   dotencode:          yes
   generaldelta:       yes
   share-safe:          no
   sparserevlog:       yes
   sidedata:            no
   persistent-nodemap: yes
   copies-sdc:          no
   plain-cl-delta:     yes
   compression:        zlib
   compression-level:  default
   $ hg debugbuilddag .+5000 --new-file
 
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5000
   tip-node: 6b02b8c7b96654c25e86ba69eda198d7e6ad8b3c
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
   $ f --size .hg/store/00changelog.n
   .hg/store/00changelog.n: size=70
 
 Simple lookup works
 
   $ ANYNODE=`hg log --template '{node|short}\n' --rev tip`
   $ hg log -r "$ANYNODE" --template '{rev}\n'
   5000
 
 
 #if rust
 
   $ f --sha256 .hg/store/00changelog-*.nd
   .hg/store/00changelog-????????????????.nd: sha256=2e029d3200bd1a986b32784fc2ef1a3bd60dc331f025718bcf5ff44d93f026fd (glob)
 
   $ f --sha256 .hg/store/00manifest-*.nd
   .hg/store/00manifest-????????????????.nd: sha256=97117b1c064ea2f86664a124589e47db0e254e8d34739b5c5cc5bf31c9da2b51 (glob)
   $ hg debugnodemap --dump-new | f --sha256 --size
   size=121088, sha256=2e029d3200bd1a986b32784fc2ef1a3bd60dc331f025718bcf5ff44d93f026fd
   $ hg debugnodemap --dump-disk | f --sha256 --bytes=256 --hexdump --size
   size=121088, sha256=2e029d3200bd1a986b32784fc2ef1a3bd60dc331f025718bcf5ff44d93f026fd
   0000: 00 00 00 91 00 00 00 20 00 00 00 bb 00 00 00 e7 |....... ........|
   0010: 00 00 00 66 00 00 00 a1 00 00 01 13 00 00 01 22 |...f..........."|
   0020: 00 00 00 23 00 00 00 fc 00 00 00 ba 00 00 00 5e |...#...........^|
   0030: 00 00 00 df 00 00 01 4e 00 00 01 65 00 00 00 ab |.......N...e....|
   0040: 00 00 00 a9 00 00 00 95 00 00 00 73 00 00 00 38 |...........s...8|
   0050: 00 00 00 cc 00 00 00 92 00 00 00 90 00 00 00 69 |...............i|
   0060: 00 00 00 ec 00 00 00 8d 00 00 01 4f 00 00 00 12 |...........O....|
   0070: 00 00 02 0c 00 00 00 77 00 00 00 9c 00 00 00 8f |.......w........|
   0080: 00 00 00 d5 00 00 00 6b 00 00 00 48 00 00 00 b3 |.......k...H....|
   0090: 00 00 00 e5 00 00 00 b5 00 00 00 8e 00 00 00 ad |................|
   00a0: 00 00 00 7b 00 00 00 7c 00 00 00 0b 00 00 00 2b |...{...|.......+|
   00b0: 00 00 00 c6 00 00 00 1e 00 00 01 08 00 00 00 11 |................|
   00c0: 00 00 01 30 00 00 00 26 00 00 01 9c 00 00 00 35 |...0...&.......5|
   00d0: 00 00 00 b8 00 00 01 31 00 00 00 2c 00 00 00 55 |.......1...,...U|
   00e0: 00 00 00 8a 00 00 00 9a 00 00 00 0c 00 00 01 1e |................|
   00f0: 00 00 00 a4 00 00 00 83 00 00 00 c9 00 00 00 8c |................|
 
 
 #else
 
   $ f --sha256 .hg/store/00changelog-*.nd
   .hg/store/00changelog-????????????????.nd: sha256=f544f5462ff46097432caf6d764091f6d8c46d6121be315ead8576d548c9dd79 (glob)
   $ hg debugnodemap --dump-new | f --sha256 --size
   size=121088, sha256=f544f5462ff46097432caf6d764091f6d8c46d6121be315ead8576d548c9dd79
   $ hg debugnodemap --dump-disk | f --sha256 --bytes=256 --hexdump --size
   size=121088, sha256=f544f5462ff46097432caf6d764091f6d8c46d6121be315ead8576d548c9dd79
   0000: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   0010: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   0020: ff ff ff ff ff ff f5 06 ff ff ff ff ff ff f3 e7 |................|
   0030: ff ff ef ca ff ff ff ff ff ff ff ff ff ff ff ff |................|
   0040: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   0050: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ed 08 |................|
   0060: ff ff ed 66 ff ff ff ff ff ff ff ff ff ff ff ff |...f............|
   0070: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   0080: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   0090: ff ff ff ff ff ff ff ff ff ff ff ff ff ff f6 ed |................|
   00a0: ff ff ff ff ff ff fe 61 ff ff ff ff ff ff ff ff |.......a........|
   00b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   00c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   00d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff |................|
   00e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff f1 02 |................|
   00f0: ff ff ff ff ff ff ed 1b ff ff ff ff ff ff ff ff |................|
 
 #endif
 
   $ hg debugnodemap --check
   revision in index:   5001
   revision in nodemap: 5001
 
 add a new commit
 
   $ hg up
   5001 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ echo foo > foo
   $ hg add foo
 
 
 Check slow-path config value handling
 -------------------------------------
 
 #if no-pure no-rust
 
   $ hg id --config "storage.revlog.persistent-nodemap.slow-path=invalid-value"
   unknown value for config "storage.revlog.persistent-nodemap.slow-path": "invalid-value"
   falling back to default value: abort
   abort: accessing `persistent-nodemap` repository without associated fast implementation.
   (check `hg help config.format.use-persistent-nodemap` for details)
   [255]
 
   $ hg log -r . --config "storage.revlog.persistent-nodemap.slow-path=warn"
   warning: accessing `persistent-nodemap` repository without associated fast implementation.
   (check `hg help config.format.use-persistent-nodemap` for details)
   changeset:   5000:6b02b8c7b966
   tag:         tip
   user:        debugbuilddag
   date:        Thu Jan 01 01:23:20 1970 +0000
   summary:     r5000
   
   $ hg ci -m 'foo' --config "storage.revlog.persistent-nodemap.slow-path=abort"
   abort: accessing `persistent-nodemap` repository without associated fast implementation.
   (check `hg help config.format.use-persistent-nodemap` for details)
   [255]
 
 #else
 
   $ hg id --config "storage.revlog.persistent-nodemap.slow-path=invalid-value"
   unknown value for config "storage.revlog.persistent-nodemap.slow-path": "invalid-value"
   falling back to default value: abort
   6b02b8c7b966+ tip
 
 #endif
 
   $ hg ci -m 'foo'
 
 #if no-pure no-rust
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5001
   tip-node: 16395c3cf7e231394735e6b1717823ada303fb0c
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 #else
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5001
   tip-node: 16395c3cf7e231394735e6b1717823ada303fb0c
   data-length: 121344
   data-unused: 256
   data-unused: 0.211%
 #endif
 
   $ f --size .hg/store/00changelog.n
   .hg/store/00changelog.n: size=70
 
 (The pure code use the debug code that perform incremental update, the C code reencode from scratch)
 
 #if pure
   $ f --sha256 .hg/store/00changelog-*.nd --size
   .hg/store/00changelog-????????????????.nd: size=121344, sha256=cce54c5da5bde3ad72a4938673ed4064c86231b9c64376b082b163fdb20f8f66 (glob)
 #endif
 
 #if rust
   $ f --sha256 .hg/store/00changelog-*.nd --size
   .hg/store/00changelog-????????????????.nd: size=121344, sha256=952b042fcf614ceb37b542b1b723e04f18f83efe99bee4e0f5ccd232ef470e58 (glob)
 #endif
 
 #if no-pure no-rust
   $ f --sha256 .hg/store/00changelog-*.nd --size
   .hg/store/00changelog-????????????????.nd: size=121088, sha256=df7c06a035b96cb28c7287d349d603baef43240be7736fe34eea419a49702e17 (glob)
 #endif
 
   $ hg debugnodemap --check
   revision in index:   5002
   revision in nodemap: 5002
 
 Test code path without mmap
 ---------------------------
 
   $ echo bar > bar
   $ hg add bar
   $ hg ci -m 'bar' --config storage.revlog.persistent-nodemap.mmap=no
 
   $ hg debugnodemap --check --config storage.revlog.persistent-nodemap.mmap=yes
   revision in index:   5003
   revision in nodemap: 5003
   $ hg debugnodemap --check --config storage.revlog.persistent-nodemap.mmap=no
   revision in index:   5003
   revision in nodemap: 5003
 
 
 #if pure
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5002
   tip-node: 880b18d239dfa9f632413a2071bfdbcc4806a4fd
   data-length: 121600
   data-unused: 512
   data-unused: 0.421%
   $ f --sha256 .hg/store/00changelog-*.nd --size
   .hg/store/00changelog-????????????????.nd: size=121600, sha256=def52503d049ccb823974af313a98a935319ba61f40f3aa06a8be4d35c215054 (glob)
 #endif
 #if rust
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5002
   tip-node: 880b18d239dfa9f632413a2071bfdbcc4806a4fd
   data-length: 121600
   data-unused: 512
   data-unused: 0.421%
   $ f --sha256 .hg/store/00changelog-*.nd --size
   .hg/store/00changelog-????????????????.nd: size=121600, sha256=dacf5b5f1d4585fee7527d0e67cad5b1ba0930e6a0928f650f779aefb04ce3fb (glob)
 #endif
 #if no-pure no-rust
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5002
   tip-node: 880b18d239dfa9f632413a2071bfdbcc4806a4fd
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
   $ f --sha256 .hg/store/00changelog-*.nd --size
   .hg/store/00changelog-????????????????.nd: size=121088, sha256=59fcede3e3cc587755916ceed29e3c33748cd1aa7d2f91828ac83e7979d935e8 (glob)
 #endif
 
 Test force warming the cache
 
   $ rm .hg/store/00changelog.n
   $ hg debugnodemap --metadata
   $ hg debugupdatecache
 #if pure
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5002
   tip-node: 880b18d239dfa9f632413a2071bfdbcc4806a4fd
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 #else
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5002
   tip-node: 880b18d239dfa9f632413a2071bfdbcc4806a4fd
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 #endif
 
 Check out of sync nodemap
 =========================
 
 First copy old data on the side.
 
   $ mkdir ../tmp-copies
   $ cp .hg/store/00changelog-????????????????.nd .hg/store/00changelog.n ../tmp-copies
 
 Nodemap lagging behind
 ----------------------
 
 make a new commit
 
   $ echo bar2 > bar
   $ hg ci -m 'bar2'
   $ NODE=`hg log -r tip -T '{node}\n'`
   $ hg log -r "$NODE" -T '{rev}\n'
   5003
 
 If the nodemap is lagging behind, it can catch up fine
 
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5003
   tip-node: c9329770f979ade2d16912267c38ba5f82fd37b3
   data-length: 121344 (pure !)
   data-length: 121344 (rust !)
   data-length: 121152 (no-rust no-pure !)
   data-unused: 192 (pure !)
   data-unused: 192 (rust !)
   data-unused: 0 (no-rust no-pure !)
   data-unused: 0.158% (pure !)
   data-unused: 0.158% (rust !)
   data-unused: 0.000% (no-rust no-pure !)
   $ cp -f ../tmp-copies/* .hg/store/
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5002
   tip-node: 880b18d239dfa9f632413a2071bfdbcc4806a4fd
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
   $ hg log -r "$NODE" -T '{rev}\n'
   5003
 
 changelog altered
 -----------------
 
 If the nodemap is not gated behind a requirements, an unaware client can alter
 the repository so the revlog used to generate the nodemap is not longer
 compatible with the persistent nodemap. We need to detect that.
 
   $ hg up "$NODE~5"
   0 files updated, 0 files merged, 4 files removed, 0 files unresolved
   $ echo bar > babar
   $ hg add babar
   $ hg ci -m 'babar'
   created new head
   $ OTHERNODE=`hg log -r tip -T '{node}\n'`
   $ hg log -r "$OTHERNODE" -T '{rev}\n'
   5004
 
   $ hg --config extensions.strip= strip --rev "$NODE~1" --no-backup
 
 the nodemap should detect the changelog have been tampered with and recover.
 
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5002
   tip-node: b355ef8adce0949b8bdf6afc72ca853740d65944
   data-length: 121536 (pure !)
   data-length: 121088 (rust !)
   data-length: 121088 (no-pure no-rust !)
   data-unused: 448 (pure !)
   data-unused: 0 (rust !)
   data-unused: 0 (no-pure no-rust !)
   data-unused: 0.000% (rust !)
   data-unused: 0.369% (pure !)
   data-unused: 0.000% (no-pure no-rust !)
 
   $ cp -f ../tmp-copies/* .hg/store/
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5002
   tip-node: 880b18d239dfa9f632413a2071bfdbcc4806a4fd
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
   $ hg log -r "$OTHERNODE" -T '{rev}\n'
   5002
 
 missing data file
 -----------------
 
   $ UUID=`hg debugnodemap --metadata| grep 'uid:' | \
   > sed 's/uid: //'`
   $ FILE=.hg/store/00changelog-"${UUID}".nd
   $ mv $FILE ../tmp-data-file
   $ cp .hg/store/00changelog.n ../tmp-docket
 
 mercurial don't crash
 
   $ hg log -r .
   changeset:   5002:b355ef8adce0
   tag:         tip
   parent:      4998:d918ad6d18d3
   user:        test
   date:        Thu Jan 01 00:00:00 1970 +0000
   summary:     babar
   
   $ hg debugnodemap --metadata
 
   $ hg debugupdatecache
   $ hg debugnodemap --metadata
   uid: * (glob)
   tip-rev: 5002
   tip-node: b355ef8adce0949b8bdf6afc72ca853740d65944
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
   $ mv ../tmp-data-file $FILE
   $ mv ../tmp-docket .hg/store/00changelog.n
 
 Check transaction related property
 ==================================
 
 An up to date nodemap should be available to shell hooks,
 
   $ echo dsljfl > a
   $ hg add a
   $ hg ci -m a
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5003
   tip-node: a52c5079765b5865d97b993b303a18740113bbb2
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
   $ echo babar2 > babar
   $ hg ci -m 'babar2' --config "hooks.pretxnclose.nodemap-test=hg debugnodemap --metadata"
   uid: ???????????????? (glob)
   tip-rev: 5004
   tip-node: 2f5fb1c06a16834c5679d672e90da7c5f3b1a984
   data-length: 121280 (pure !)
   data-length: 121280 (rust !)
   data-length: 121088 (no-pure no-rust !)
   data-unused: 192 (pure !)
   data-unused: 192 (rust !)
   data-unused: 0 (no-pure no-rust !)
   data-unused: 0.158% (pure !)
   data-unused: 0.158% (rust !)
   data-unused: 0.000% (no-pure no-rust !)
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5004
   tip-node: 2f5fb1c06a16834c5679d672e90da7c5f3b1a984
   data-length: 121280 (pure !)
   data-length: 121280 (rust !)
   data-length: 121088 (no-pure no-rust !)
   data-unused: 192 (pure !)
   data-unused: 192 (rust !)
   data-unused: 0 (no-pure no-rust !)
   data-unused: 0.158% (pure !)
   data-unused: 0.158% (rust !)
   data-unused: 0.000% (no-pure no-rust !)
 
 Another process does not see the pending nodemap content during run.
 
   $ PATH=$RUNTESTDIR/testlib/:$PATH
   $ echo qpoasp > a
   $ hg ci -m a2 \
   > --config "hooks.pretxnclose=wait-on-file 20 sync-repo-read sync-txn-pending" \
   > --config "hooks.txnclose=touch sync-txn-close" > output.txt 2>&1 &
 
 (read the repository while the commit transaction is pending)
 
   $ wait-on-file 20 sync-txn-pending && \
   > hg debugnodemap --metadata && \
   > wait-on-file 20 sync-txn-close sync-repo-read
   uid: ???????????????? (glob)
   tip-rev: 5004
   tip-node: 2f5fb1c06a16834c5679d672e90da7c5f3b1a984
   data-length: 121280 (pure !)
   data-length: 121280 (rust !)
   data-length: 121088 (no-pure no-rust !)
   data-unused: 192 (pure !)
   data-unused: 192 (rust !)
   data-unused: 0 (no-pure no-rust !)
   data-unused: 0.158% (pure !)
   data-unused: 0.158% (rust !)
   data-unused: 0.000% (no-pure no-rust !)
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121536 (pure !)
   data-length: 121536 (rust !)
   data-length: 121088 (no-pure no-rust !)
   data-unused: 448 (pure !)
   data-unused: 448 (rust !)
   data-unused: 0 (no-pure no-rust !)
   data-unused: 0.369% (pure !)
   data-unused: 0.369% (rust !)
   data-unused: 0.000% (no-pure no-rust !)
 
   $ cat output.txt
 
 Check that a failing transaction will properly revert the data
 
   $ echo plakfe > a
   $ f --size --sha256 .hg/store/00changelog-*.nd
   .hg/store/00changelog-????????????????.nd: size=121536, sha256=bb414468d225cf52d69132e1237afba34d4346ee2eb81b505027e6197b107f03 (glob) (pure !)
   .hg/store/00changelog-????????????????.nd: size=121536, sha256=909ac727bc4d1c0fda5f7bff3c620c98bd4a2967c143405a1503439e33b377da (glob) (rust !)
   .hg/store/00changelog-????????????????.nd: size=121088, sha256=342d36d30d86dde67d3cb6c002606c4a75bcad665595d941493845066d9c8ee0 (glob) (no-pure no-rust !)
   $ hg ci -m a3 --config "extensions.abort=$RUNTESTDIR/testlib/crash_transaction_late.py"
   transaction abort!
   rollback completed
   abort: This is a late abort
   [255]
   $ hg debugnodemap --metadata
   uid: ???????????????? (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121536 (pure !)
   data-length: 121536 (rust !)
   data-length: 121088 (no-pure no-rust !)
   data-unused: 448 (pure !)
   data-unused: 448 (rust !)
   data-unused: 0 (no-pure no-rust !)
   data-unused: 0.369% (pure !)
   data-unused: 0.369% (rust !)
   data-unused: 0.000% (no-pure no-rust !)
   $ f --size --sha256 .hg/store/00changelog-*.nd
   .hg/store/00changelog-????????????????.nd: size=121536, sha256=bb414468d225cf52d69132e1237afba34d4346ee2eb81b505027e6197b107f03 (glob) (pure !)
   .hg/store/00changelog-????????????????.nd: size=121536, sha256=909ac727bc4d1c0fda5f7bff3c620c98bd4a2967c143405a1503439e33b377da (glob) (rust !)
   .hg/store/00changelog-????????????????.nd: size=121088, sha256=342d36d30d86dde67d3cb6c002606c4a75bcad665595d941493845066d9c8ee0 (glob) (no-pure no-rust !)
 
 Check that removing content does not confuse the nodemap
 --------------------------------------------------------
 
 removing data with rollback
 
   $ echo aso > a
   $ hg ci -m a4
   $ hg rollback
   repository tip rolled back to revision 5005 (undo commit)
   working directory now based on revision 5005
   $ hg id -r .
   90d5d3ba2fc4 tip
 
 roming data with strip
 
   $ echo aso > a
   $ hg ci -m a4
   $ hg --config extensions.strip= strip -r . --no-backup
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg id -r . --traceback
   90d5d3ba2fc4 tip
 
 Test upgrade / downgrade
 ========================
 
 downgrading
 
   $ cat << EOF >> .hg/hgrc
   > [format]
   > use-persistent-nodemap=no
   > EOF
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   sidedata:            no     no      no
   persistent-nodemap: yes     no      no
   copies-sdc:          no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib
   compression-level:  default default default
   $ hg debugupgraderepo --run --no-backup --quiet
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store
      removed: persistent-nodemap
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   $ ls -1 .hg/store/ | egrep '00(changelog|manifest)(\.n|-.*\.nd)'
   [1]
   $ hg debugnodemap --metadata
 
 
 upgrading
 
   $ cat << EOF >> .hg/hgrc
   > [format]
   > use-persistent-nodemap=yes
   > EOF
   $ hg debugformat -v
   format-variant     repo config default
   fncache:            yes    yes     yes
   dotencode:          yes    yes     yes
   generaldelta:       yes    yes     yes
   share-safe:          no     no      no
   sparserevlog:       yes    yes     yes
   sidedata:            no     no      no
   persistent-nodemap:  no    yes      no
   copies-sdc:          no     no      no
   plain-cl-delta:     yes    yes     yes
   compression:        zlib   zlib    zlib
   compression-level:  default default default
   $ hg debugupgraderepo --run --no-backup
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, revlogv1, sparserevlog, store
      added: persistent-nodemap
   
   persistent-nodemap
      Speedup revision lookup by node id.
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   beginning upgrade...
   repository locked and read-only
   creating temporary repository to stage upgraded data: $TESTTMP/test-repo/.hg/upgrade.* (glob)
   (it is safe to interrupt this process any time before data migration completes)
-  migrating 15018 total revisions (5006 in filelogs, 5006 in manifests, 5006 in changelog)
-  migrating 1.74 MB in store; 569 MB tracked data
-  migrating 5004 filelogs containing 5006 revisions (346 KB in store; 28.2 KB tracked data)
-  finished migrating 5006 filelog revisions across 5004 filelogs; change in size: 0 bytes
-  migrating 1 manifests containing 5006 revisions (765 KB in store; 569 MB tracked data)
-  finished migrating 5006 manifest revisions across 1 manifests; change in size: 0 bytes
-  migrating changelog containing 5006 revisions (673 KB in store; 363 KB tracked data)
-  finished migrating 5006 changelog revisions; change in size: 0 bytes
-  finished migrating 15018 total revisions; total change in store size: 0 bytes
-  copying phaseroots
-  data fully upgraded in a temporary repository
-  marking source repository as being upgraded; clients will be unable to read from repository
-  starting in-place swap of repository data
-  replacing store...
-  store replacement complete; repository was inconsistent for *s (glob)
-  finalizing requirements file and making repository readable again
+  upgrading repository to use persistent nodemap feature
   removing temporary repository $TESTTMP/test-repo/.hg/upgrade.* (glob)
   $ ls -1 .hg/store/ | egrep '00(changelog|manifest)(\.n|-.*\.nd)'
   00changelog-*.nd (glob)
   00changelog.n
-  00manifest-*.nd (glob)
-  00manifest.n
 
   $ hg debugnodemap --metadata
   uid: * (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 
 Running unrelated upgrade
 
   $ hg debugupgraderepo --run --no-backup --quiet --optimize re-delta-all
   upgrade will perform the following actions:
   
   requirements
      preserved: dotencode, fncache, generaldelta, persistent-nodemap, revlogv1, sparserevlog, store
   
   optimisations: re-delta-all
   
   processed revlogs:
     - all-filelogs
     - changelog
     - manifest
   
   $ ls -1 .hg/store/ | egrep '00(changelog|manifest)(\.n|-.*\.nd)'
   00changelog-*.nd (glob)
   00changelog.n
   00manifest-*.nd (glob)
   00manifest.n
 
   $ hg debugnodemap --metadata
   uid: * (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 
 Persistent nodemap and local/streaming clone
 ============================================
 
   $ cd ..
 
 standard clone
 --------------
 
 The persistent nodemap should exist after a streaming clone
 
   $ hg clone --pull --quiet -U test-repo standard-clone
   $ ls -1 standard-clone/.hg/store/ | egrep '00(changelog|manifest)(\.n|-.*\.nd)'
   00changelog-*.nd (glob)
   00changelog.n
   00manifest-*.nd (glob)
   00manifest.n
   $ hg -R standard-clone debugnodemap --metadata
   uid: * (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 
 
 local clone
 ------------
 
 The persistent nodemap should exist after a streaming clone
 
   $ hg clone -U test-repo local-clone
   $ ls -1 local-clone/.hg/store/ | egrep '00(changelog|manifest)(\.n|-.*\.nd)'
   00changelog-*.nd (glob)
   00changelog.n
   00manifest-*.nd (glob)
   00manifest.n
   $ hg -R local-clone debugnodemap --metadata
   uid: * (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%
 
 stream clone
 ------------
 
 The persistent nodemap should exist after a streaming clone
 
   $ hg clone -U --stream --config ui.ssh="\"$PYTHON\" \"$TESTDIR/dummyssh\"" ssh://user@dummy/test-repo stream-clone --debug | egrep '00(changelog|manifest)'
   adding [s] 00manifest.n (70 bytes)
   adding [s] 00manifest.i (313 KB)
   adding [s] 00manifest.d (452 KB)
   adding [s] 00manifest-*.nd (118 KB) (glob)
   adding [s] 00changelog.n (70 bytes)
   adding [s] 00changelog.i (313 KB)
   adding [s] 00changelog.d (360 KB)
   adding [s] 00changelog-*.nd (118 KB) (glob)
   $ ls -1 stream-clone/.hg/store/ | egrep '00(changelog|manifest)(\.n|-.*\.nd)'
   00changelog-*.nd (glob)
   00changelog.n
   00manifest-*.nd (glob)
   00manifest.n
   $ hg -R stream-clone debugnodemap --metadata
   uid: * (glob)
   tip-rev: 5005
   tip-node: 90d5d3ba2fc47db50f712570487cb261a68c8ffe
   data-length: 121088
   data-unused: 0
   data-unused: 0.000%