mercurial/verify.py
author Arseniy Alekseyev <aalekseyev@janestreet.com>
Fri, 26 Apr 2024 19:10:35 +0100
changeset 51626 865efc020c33
parent 51015 812cd3dfa4cb
permissions -rw-r--r--
dirstate: remove the python-side whitelist of allowed matchers This whitelist is too permissive because it allows matchers that contain disallowed ones deep inside, for example through `intersectionmatcher`. It is also too restrictive because it doesn't pass through some of the matchers we support, such as `patternmatcher`. It's also unnecessary because unsupported matchers raise `FallbackError` and we fall back anyway. Making this change makes more of the tests use rust code path, and therefore subtly change behavior. For example, rust status in largefiles repos seems to have strange behavior.
Ignore whitespace changes - Everywhere: Within whitespace: At end of lines:
2778
fdc232d8a193 Move repo.verify
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
     1
# verify.py - repository integrity checking for Mercurial
fdc232d8a193 Move repo.verify
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
     2
#
46819
d4ba4d51f85f contributor: change mentions of mpm to olivia
Raphaël Gomès <rgomes@octobus.net>
parents: 46790
diff changeset
     3
# Copyright 2006, 2007 Olivia Mackall <olivia@selenic.com>
2778
fdc232d8a193 Move repo.verify
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
     4
#
8225
46293a0c7e9f updated license to be explicit about GPL version 2
Martin Geisler <mg@lazybytes.net>
parents: 8209
diff changeset
     5
# This software may be used and distributed according to the terms of the
10263
25e572394f5c Update license to GPLv2+
Matt Mackall <mpm@selenic.com>
parents: 9690
diff changeset
     6
# GNU General Public License version 2 or any later version.
2778
fdc232d8a193 Move repo.verify
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
     7
25991
d21d1774c73b verify: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 25846
diff changeset
     8
17860
a45b33f12627 verify: fix all doubled-slash sites (issue3665)
Bryan O'Sullivan <bryano@fb.com>
parents: 17851
diff changeset
     9
import os
25991
d21d1774c73b verify: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 25846
diff changeset
    10
d21d1774c73b verify: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 25846
diff changeset
    11
from .i18n import _
47012
d55b71393907 node: replace nullid and friends with nodeconstants class
Joerg Sonnenberger <joerg@bec.de>
parents: 46895
diff changeset
    12
from .node import short
d55b71393907 node: replace nullid and friends with nodeconstants class
Joerg Sonnenberger <joerg@bec.de>
parents: 46895
diff changeset
    13
from .utils import stringutil
25991
d21d1774c73b verify: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 25846
diff changeset
    14
d21d1774c73b verify: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 25846
diff changeset
    15
from . import (
d21d1774c73b verify: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 25846
diff changeset
    16
    error,
35585
35fb3367f72d py3: use pycompat.bytestr() instead of str()
Pulkit Goyal <7895pulkit@gmail.com>
parents: 33499
diff changeset
    17
    pycompat,
49826
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
    18
    requirements,
25991
d21d1774c73b verify: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 25846
diff changeset
    19
    revlog,
50988
cf47b83d8ad0 transaction: abstract away the detection of an abandoned transaction
Raphaël Gomès <rgomes@octobus.net>
parents: 50504
diff changeset
    20
    transaction,
25991
d21d1774c73b verify: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 25846
diff changeset
    21
    util,
d21d1774c73b verify: use absolute_import
Gregory Szorc <gregory.szorc@gmail.com>
parents: 25846
diff changeset
    22
)
2778
fdc232d8a193 Move repo.verify
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
    23
42155
57539e5ea2e0 verify: introduce a notion of "level"
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41877
diff changeset
    24
VERIFY_DEFAULT = 0
42156
496ac8a02380 verify: introduce an experimental --full flag
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 42155
diff changeset
    25
VERIFY_FULL = 1
42155
57539e5ea2e0 verify: introduce a notion of "level"
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41877
diff changeset
    26
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
    27
42155
57539e5ea2e0 verify: introduce a notion of "level"
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41877
diff changeset
    28
def verify(repo, level=None):
27849
900d36a3e4dd with: use context manager in verify
Bryan O'Sullivan <bryano@fb.com>
parents: 27695
diff changeset
    29
    with repo.lock():
42155
57539e5ea2e0 verify: introduce a notion of "level"
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41877
diff changeset
    30
        v = verifier(repo, level)
57539e5ea2e0 verify: introduce a notion of "level"
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41877
diff changeset
    31
        return v.verify()
4915
97b734fb9c6f Use try/finally pattern to cleanup locks and transactions
Matt Mackall <mpm@selenic.com>
parents: 4635
diff changeset
    32
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
    33
17860
a45b33f12627 verify: fix all doubled-slash sites (issue3665)
Bryan O'Sullivan <bryano@fb.com>
parents: 17851
diff changeset
    34
def _normpath(f):
a45b33f12627 verify: fix all doubled-slash sites (issue3665)
Bryan O'Sullivan <bryano@fb.com>
parents: 17851
diff changeset
    35
    # under hg < 2.4, convert didn't sanitize paths properly, so a
a45b33f12627 verify: fix all doubled-slash sites (issue3665)
Bryan O'Sullivan <bryano@fb.com>
parents: 17851
diff changeset
    36
    # converted repo may contain repeated slashes
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
    37
    while b'//' in f:
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
    38
        f = f.replace(b'//', b'/')
17860
a45b33f12627 verify: fix all doubled-slash sites (issue3665)
Bryan O'Sullivan <bryano@fb.com>
parents: 17851
diff changeset
    39
    return f
a45b33f12627 verify: fix all doubled-slash sites (issue3665)
Bryan O'Sullivan <bryano@fb.com>
parents: 17851
diff changeset
    40
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
    41
47358
ba8a9fbed897 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47357
diff changeset
    42
HINT_FNCACHE = _(
ba8a9fbed897 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47357
diff changeset
    43
    b'hint: run "hg debugrebuildfncache" to recover from corrupt fncache\n'
ba8a9fbed897 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47357
diff changeset
    44
)
ba8a9fbed897 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47357
diff changeset
    45
47363
9823b3489e90 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47362
diff changeset
    46
WARN_PARENT_DIR_UNKNOWN_REV = _(
9823b3489e90 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47362
diff changeset
    47
    b"parent-directory manifest refers to unknown revision %s"
9823b3489e90 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47362
diff changeset
    48
)
9823b3489e90 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47362
diff changeset
    49
47367
34a92e84267e verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47366
diff changeset
    50
WARN_UNKNOWN_COPY_SOURCE = _(
34a92e84267e verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47366
diff changeset
    51
    b"warning: copy source of '%s' not in parents of %s"
34a92e84267e verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47366
diff changeset
    52
)
34a92e84267e verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47366
diff changeset
    53
47369
041d6515bb0f verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47368
diff changeset
    54
WARN_NULLID_COPY_SOURCE = _(
041d6515bb0f verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47368
diff changeset
    55
    b"warning: %s@%s: copy source revision is nullid %s:%s\n"
041d6515bb0f verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47368
diff changeset
    56
)
041d6515bb0f verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47368
diff changeset
    57
47358
ba8a9fbed897 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47357
diff changeset
    58
48946
642e31cb55f0 py3: use class X: instead of class X(object):
Gregory Szorc <gregory.szorc@gmail.com>
parents: 48931
diff changeset
    59
class verifier:
42155
57539e5ea2e0 verify: introduce a notion of "level"
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41877
diff changeset
    60
    def __init__(self, repo, level=None):
27444
6647401858ab verify: move widely used variables into class members
Durham Goode <durham@fb.com>
parents: 27443
diff changeset
    61
        self.repo = repo.unfiltered()
6647401858ab verify: move widely used variables into class members
Durham Goode <durham@fb.com>
parents: 27443
diff changeset
    62
        self.ui = repo.ui
39938
fec944719324 narrow: move support for `hg verify` into core
Martin von Zweigbergk <martinvonz@google.com>
parents: 39872
diff changeset
    63
        self.match = repo.narrowmatch()
42155
57539e5ea2e0 verify: introduce a notion of "level"
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41877
diff changeset
    64
        if level is None:
57539e5ea2e0 verify: introduce a notion of "level"
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41877
diff changeset
    65
            level = VERIFY_DEFAULT
57539e5ea2e0 verify: introduce a notion of "level"
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41877
diff changeset
    66
        self._level = level
27444
6647401858ab verify: move widely used variables into class members
Durham Goode <durham@fb.com>
parents: 27443
diff changeset
    67
        self.badrevs = set()
27453
8462d7f2c4fe verify: clean up weird error/warning lists
Matt Mackall <mpm@selenic.com>
parents: 27450
diff changeset
    68
        self.errors = 0
8462d7f2c4fe verify: clean up weird error/warning lists
Matt Mackall <mpm@selenic.com>
parents: 27450
diff changeset
    69
        self.warnings = 0
27444
6647401858ab verify: move widely used variables into class members
Durham Goode <durham@fb.com>
parents: 27443
diff changeset
    70
        self.havecl = len(repo.changelog) > 0
39244
73cf21b2e8a6 manifest: add getstorage() to manifestlog and use it globally
Gregory Szorc <gregory.szorc@gmail.com>
parents: 38397
diff changeset
    71
        self.havemf = len(repo.manifestlog.getstorage(b'')) > 0
47139
f58a13c52726 revlog: split the `version` attribute into its two components
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47138
diff changeset
    72
        self.revlogv1 = repo.changelog._format_version != revlog.REVLOGV0
44113
e77b57e09bfa verify: avoid spurious integrity warnings in verbose mode (issue6172)
Matt Harbison <matt_harbison@yahoo.com>
parents: 44073
diff changeset
    73
        self.lrugetctx = util.lrucachefunc(repo.unfiltered().__getitem__)
27444
6647401858ab verify: move widely used variables into class members
Durham Goode <durham@fb.com>
parents: 27443
diff changeset
    74
        self.refersmf = False
27445
cc178057ab49 verify: move fncachewarned up to a class variable
Durham Goode <durham@fb.com>
parents: 27444
diff changeset
    75
        self.fncachewarned = False
32288
a2ab9ebcd85b verify: add a config option to skip certain flag processors
Jun Wu <quark@fb.com>
parents: 32250
diff changeset
    76
        # developer config: verify.skipflags
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
    77
        self.skipflags = repo.ui.configint(b'verify', b'skipflags')
37417
76d2115cb817 verify: allow suppressing warnings about extra files
Gregory Szorc <gregory.szorc@gmail.com>
parents: 37410
diff changeset
    78
        self.warnorphanstorefiles = True
27444
6647401858ab verify: move widely used variables into class members
Durham Goode <durham@fb.com>
parents: 27443
diff changeset
    79
41862
c66037fb1bc5 verify: make the `warn` method private
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41861
diff changeset
    80
    def _warn(self, msg):
41861
a58748300e61 verify: document the `warn` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 40616
diff changeset
    81
        """record a "warning" level issue"""
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
    82
        self.ui.warn(msg + b"\n")
27453
8462d7f2c4fe verify: clean up weird error/warning lists
Matt Mackall <mpm@selenic.com>
parents: 27450
diff changeset
    83
        self.warnings += 1
27446
6b2c1a1871a6 verify: move warn() to a class level function
Durham Goode <durham@fb.com>
parents: 27445
diff changeset
    84
41864
7eaf4b1ac2a3 verify: make `err` a private method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41863
diff changeset
    85
    def _err(self, linkrev, msg, filename=None):
41863
9534f3cb6b6c verify: document the `err` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41862
diff changeset
    86
        """record a "error" level issue"""
27447
d1b91c10ce70 verify: move err() to be a class function
Durham Goode <durham@fb.com>
parents: 27446
diff changeset
    87
        if linkrev is not None:
d1b91c10ce70 verify: move err() to be a class function
Durham Goode <durham@fb.com>
parents: 27446
diff changeset
    88
            self.badrevs.add(linkrev)
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
    89
            linkrev = b"%d" % linkrev
27447
d1b91c10ce70 verify: move err() to be a class function
Durham Goode <durham@fb.com>
parents: 27446
diff changeset
    90
        else:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
    91
            linkrev = b'?'
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
    92
        msg = b"%s: %s" % (linkrev, msg)
27447
d1b91c10ce70 verify: move err() to be a class function
Durham Goode <durham@fb.com>
parents: 27446
diff changeset
    93
        if filename:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
    94
            msg = b"%s@%s" % (filename, msg)
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
    95
        self.ui.warn(b" " + msg + b"\n")
27453
8462d7f2c4fe verify: clean up weird error/warning lists
Matt Mackall <mpm@selenic.com>
parents: 27450
diff changeset
    96
        self.errors += 1
27447
d1b91c10ce70 verify: move err() to be a class function
Durham Goode <durham@fb.com>
parents: 27446
diff changeset
    97
41866
cfe08588d711 verify: make the `exc` method private
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41865
diff changeset
    98
    def _exc(self, linkrev, msg, inst, filename=None):
41865
5df8475c5343 verify: document the `exc` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41864
diff changeset
    99
        """record exception raised during the verify process"""
46696
ed0899e01628 verify: convert an exception to bytes before logging
Matt Harbison <matt_harbison@yahoo.com>
parents: 44113
diff changeset
   100
        fmsg = stringutil.forcebytestr(inst)
36577
d85ef895d5f6 verify: fix exception formatting bug in Python 3
Augie Fackler <augie@google.com>
parents: 36339
diff changeset
   101
        if not fmsg:
d85ef895d5f6 verify: fix exception formatting bug in Python 3
Augie Fackler <augie@google.com>
parents: 36339
diff changeset
   102
            fmsg = pycompat.byterepr(inst)
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   103
        self._err(linkrev, b"%s: %s" % (msg, fmsg), filename)
27448
f4f2179077cb verify: move exc() function onto class
Durham Goode <durham@fb.com>
parents: 27447
diff changeset
   104
41874
1f412223f5bb verify: rename the `checklog` to `_checkrevlog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41873
diff changeset
   105
    def _checkrevlog(self, obj, name, linkrev):
41873
08d977451f26 verify: document the `checklog` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41871
diff changeset
   106
        """verify high level property of a revlog
08d977451f26 verify: document the `checklog` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41871
diff changeset
   107
08d977451f26 verify: document the `checklog` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41871
diff changeset
   108
        - revlog is present,
08d977451f26 verify: document the `checklog` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41871
diff changeset
   109
        - revlog is non-empty,
08d977451f26 verify: document the `checklog` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41871
diff changeset
   110
        - sizes (index and data) are correct,
08d977451f26 verify: document the `checklog` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41871
diff changeset
   111
        - revlog's format version is correct.
08d977451f26 verify: document the `checklog` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41871
diff changeset
   112
        """
27642
f6457349985b verify: move checklog() onto class
Durham Goode <durham@fb.com>
parents: 27453
diff changeset
   113
        if not len(obj) and (self.havecl or self.havemf):
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   114
            self._err(linkrev, _(b"empty or missing %s") % name)
27642
f6457349985b verify: move checklog() onto class
Durham Goode <durham@fb.com>
parents: 27453
diff changeset
   115
            return
f6457349985b verify: move checklog() onto class
Durham Goode <durham@fb.com>
parents: 27453
diff changeset
   116
f6457349985b verify: move checklog() onto class
Durham Goode <durham@fb.com>
parents: 27453
diff changeset
   117
        d = obj.checksize()
f6457349985b verify: move checklog() onto class
Durham Goode <durham@fb.com>
parents: 27453
diff changeset
   118
        if d[0]:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   119
            self._err(None, _(b"data length off by %d bytes") % d[0], name)
27642
f6457349985b verify: move checklog() onto class
Durham Goode <durham@fb.com>
parents: 27453
diff changeset
   120
        if d[1]:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   121
            self._err(None, _(b"index contains %d extra bytes") % d[1], name)
27642
f6457349985b verify: move checklog() onto class
Durham Goode <durham@fb.com>
parents: 27453
diff changeset
   122
47139
f58a13c52726 revlog: split the `version` attribute into its two components
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47138
diff changeset
   123
        if obj._format_version != revlog.REVLOGV0:
27642
f6457349985b verify: move checklog() onto class
Durham Goode <durham@fb.com>
parents: 27453
diff changeset
   124
            if not self.revlogv1:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   125
                self._warn(_(b"warning: `%s' uses revlog format 1") % name)
27642
f6457349985b verify: move checklog() onto class
Durham Goode <durham@fb.com>
parents: 27453
diff changeset
   126
        elif self.revlogv1:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   127
            self._warn(_(b"warning: `%s' uses revlog format 0") % name)
27642
f6457349985b verify: move checklog() onto class
Durham Goode <durham@fb.com>
parents: 27453
diff changeset
   128
41871
9e737ca539f6 verify: make `checkentry` a private method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41870
diff changeset
   129
    def _checkentry(self, obj, i, node, seen, linkrevs, f):
41870
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   130
        """verify a single revlog entry
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   131
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   132
        arguments are:
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   133
        - obj:      the source revlog
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   134
        - i:        the revision number
47353
5ed2aaab58b0 verify: align a comment line
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47139
diff changeset
   135
        - node:     the revision node id
41870
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   136
        - seen:     nodes previously seen for this revlog
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   137
        - linkrevs: [changelog-revisions] introducing "node"
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   138
        - f:        string label ("changelog", "manifest", or filename)
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   139
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   140
        Performs the following checks:
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   141
        - linkrev points to an existing changelog revision,
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   142
        - linkrev points to a changelog revision that introduces this revision,
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   143
        - linkrev points to the lowest of these changesets,
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   144
        - both parents exist in the revlog,
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   145
        - the revision is not duplicated.
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   146
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   147
        Return the linkrev of the revision (or None for changelog's revisions).
00c9fde75c1a verify: document the `checkentry` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41869
diff changeset
   148
        """
27643
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   149
        lr = obj.linkrev(obj.rev(node))
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   150
        if lr < 0 or (self.havecl and lr not in linkrevs):
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   151
            if lr < 0 or lr >= len(self.repo.changelog):
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   152
                msg = _(b"rev %d points to nonexistent changeset %d")
27643
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   153
            else:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   154
                msg = _(b"rev %d points to unexpected changeset %d")
41864
7eaf4b1ac2a3 verify: make `err` a private method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41863
diff changeset
   155
            self._err(None, msg % (i, lr), f)
27643
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   156
            if linkrevs:
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   157
                if f and len(linkrevs) > 1:
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   158
                    try:
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   159
                        # attempt to filter down to real linkrevs
47354
03dea8553114 verify: expand a one liner into explicit commands
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47353
diff changeset
   160
                        linkrevs = []
03dea8553114 verify: expand a one liner into explicit commands
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47353
diff changeset
   161
                        for lr in linkrevs:
03dea8553114 verify: expand a one liner into explicit commands
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47353
diff changeset
   162
                            if self.lrugetctx(lr)[f].filenode() == node:
03dea8553114 verify: expand a one liner into explicit commands
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47353
diff changeset
   163
                                linkrevs.append(lr)
27643
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   164
                    except Exception:
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   165
                        pass
47355
0f4beb88ec18 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47354
diff changeset
   166
                msg = _(b" (expected %s)")
0f4beb88ec18 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47354
diff changeset
   167
                msg %= b" ".join(map(pycompat.bytestr, linkrevs))
0f4beb88ec18 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47354
diff changeset
   168
                self._warn(msg)
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   169
            lr = None  # can't be trusted
27643
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   170
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   171
        try:
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   172
            p1, p2 = obj.parents(node)
47012
d55b71393907 node: replace nullid and friends with nodeconstants class
Joerg Sonnenberger <joerg@bec.de>
parents: 46895
diff changeset
   173
            if p1 not in seen and p1 != self.repo.nullid:
47356
a3c3924ccc7f verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47355
diff changeset
   174
                msg = _(b"unknown parent 1 %s of %s") % (short(p1), short(node))
a3c3924ccc7f verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47355
diff changeset
   175
                self._err(lr, msg, f)
47012
d55b71393907 node: replace nullid and friends with nodeconstants class
Joerg Sonnenberger <joerg@bec.de>
parents: 46895
diff changeset
   176
            if p2 not in seen and p2 != self.repo.nullid:
47357
fde1df74d73d verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47356
diff changeset
   177
                msg = _(b"unknown parent 2 %s of %s") % (short(p2), short(node))
fde1df74d73d verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47356
diff changeset
   178
                self._err(lr, msg, f)
27643
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   179
        except Exception as inst:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   180
            self._exc(lr, _(b"checking parents of %s") % short(node), inst, f)
27643
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   181
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   182
        if node in seen:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   183
            self._err(lr, _(b"duplicate revision %d (%d)") % (i, seen[node]), f)
27643
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   184
        seen[node] = i
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   185
        return lr
62ce86fcfd06 verify: move checkentry() to be a class function
Durham Goode <durham@fb.com>
parents: 27642
diff changeset
   186
27444
6647401858ab verify: move widely used variables into class members
Durham Goode <durham@fb.com>
parents: 27443
diff changeset
   187
    def verify(self):
41867
e8c4a9f5b986 verify: minimal documentation for `verifier.verify`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41866
diff changeset
   188
        """verify the content of the Mercurial repository
e8c4a9f5b986 verify: minimal documentation for `verifier.verify`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41866
diff changeset
   189
e8c4a9f5b986 verify: minimal documentation for `verifier.verify`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41866
diff changeset
   190
        This method run all verifications, displaying issues as they are found.
e8c4a9f5b986 verify: minimal documentation for `verifier.verify`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41866
diff changeset
   191
41868
567892b4306c verify: explicitly return 0 if no error are encountered
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41867
diff changeset
   192
        return 1 if any error have been encountered, 0 otherwise."""
41869
4da2261e949b verify: add some inline documentation to the top level `verify` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41868
diff changeset
   193
        # initial validation and generic report
27444
6647401858ab verify: move widely used variables into class members
Durham Goode <durham@fb.com>
parents: 27443
diff changeset
   194
        repo = self.repo
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   195
        ui = repo.ui
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   196
        if not repo.url().startswith(b'file:'):
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   197
            raise error.Abort(_(b"cannot verify bundle or remote repos"))
6752
e79a8f36c2a5 verify: lots of refactoring
Matt Mackall <mpm@selenic.com>
parents: 6751
diff changeset
   198
50988
cf47b83d8ad0 transaction: abstract away the detection of an abandoned transaction
Raphaël Gomès <rgomes@octobus.net>
parents: 50504
diff changeset
   199
        if transaction.has_abandoned_transaction(repo):
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   200
            ui.warn(_(b"abandoned transaction found - run hg recover\n"))
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   201
27648
e72e669dd51f verify: get rid of some unnecessary local variables
Durham Goode <durham@fb.com>
parents: 27647
diff changeset
   202
        if ui.verbose or not self.revlogv1:
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   203
            ui.status(
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   204
                _(b"repository uses revlog format %d\n")
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   205
                % (self.revlogv1 and 1 or 0)
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   206
            )
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   207
41869
4da2261e949b verify: add some inline documentation to the top level `verify` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41868
diff changeset
   208
        # data verification
27695
fb0cc863d172 verify: replace "output parameters" by return values
Martin von Zweigbergk <martinvonz@google.com>
parents: 27648
diff changeset
   209
        mflinkrevs, filelinkrevs = self._verifychangelog()
fb0cc863d172 verify: replace "output parameters" by return values
Martin von Zweigbergk <martinvonz@google.com>
parents: 27648
diff changeset
   210
        filenodes = self._verifymanifest(mflinkrevs)
28111
06205989264b verify: move cross-checking of changeset/manifest out of _crosscheckfiles()
Martin von Zweigbergk <martinvonz@google.com>
parents: 28007
diff changeset
   211
        del mflinkrevs
06205989264b verify: move cross-checking of changeset/manifest out of _crosscheckfiles()
Martin von Zweigbergk <martinvonz@google.com>
parents: 28007
diff changeset
   212
        self._crosscheckfiles(filelinkrevs, filenodes)
27647
2c2858f3c1bb verify: move changelog verificaiton to its own function
Durham Goode <durham@fb.com>
parents: 27646
diff changeset
   213
        totalfiles, filerevisions = self._verifyfiles(filenodes, filelinkrevs)
2c2858f3c1bb verify: move changelog verificaiton to its own function
Durham Goode <durham@fb.com>
parents: 27646
diff changeset
   214
49826
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   215
        if self.errors:
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   216
            ui.warn(_(b"not checking dirstate because of previous errors\n"))
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   217
            dirstate_errors = 0
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   218
        else:
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   219
            dirstate_errors = self._verify_dirstate()
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   220
41869
4da2261e949b verify: add some inline documentation to the top level `verify` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41868
diff changeset
   221
        # final report
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   222
        ui.status(
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   223
            _(b"checked %d changesets with %d changes to %d files\n")
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   224
            % (len(repo.changelog), filerevisions, totalfiles)
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   225
        )
27647
2c2858f3c1bb verify: move changelog verificaiton to its own function
Durham Goode <durham@fb.com>
parents: 27646
diff changeset
   226
        if self.warnings:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   227
            ui.warn(_(b"%d warnings encountered!\n") % self.warnings)
27647
2c2858f3c1bb verify: move changelog verificaiton to its own function
Durham Goode <durham@fb.com>
parents: 27646
diff changeset
   228
        if self.fncachewarned:
47358
ba8a9fbed897 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47357
diff changeset
   229
            ui.warn(HINT_FNCACHE)
27647
2c2858f3c1bb verify: move changelog verificaiton to its own function
Durham Goode <durham@fb.com>
parents: 27646
diff changeset
   230
        if self.errors:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   231
            ui.warn(_(b"%d integrity errors encountered!\n") % self.errors)
27648
e72e669dd51f verify: get rid of some unnecessary local variables
Durham Goode <durham@fb.com>
parents: 27647
diff changeset
   232
            if self.badrevs:
47359
9305824d3a97 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47358
diff changeset
   233
                msg = _(b"(first damaged changeset appears to be %d)\n")
9305824d3a97 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47358
diff changeset
   234
                msg %= min(self.badrevs)
9305824d3a97 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47358
diff changeset
   235
                ui.warn(msg)
49826
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   236
            if dirstate_errors:
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   237
                ui.warn(
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   238
                    _(b"dirstate inconsistent with current parent's manifest\n")
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   239
                )
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   240
                ui.warn(_(b"%d dirstate errors\n") % dirstate_errors)
27647
2c2858f3c1bb verify: move changelog verificaiton to its own function
Durham Goode <durham@fb.com>
parents: 27646
diff changeset
   241
            return 1
41868
567892b4306c verify: explicitly return 0 if no error are encountered
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41867
diff changeset
   242
        return 0
27647
2c2858f3c1bb verify: move changelog verificaiton to its own function
Durham Goode <durham@fb.com>
parents: 27646
diff changeset
   243
27695
fb0cc863d172 verify: replace "output parameters" by return values
Martin von Zweigbergk <martinvonz@google.com>
parents: 27648
diff changeset
   244
    def _verifychangelog(self):
41875
2f1f475e9646 verify: document `_verifychangelog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41874
diff changeset
   245
        """verify the changelog of a repository
2f1f475e9646 verify: document `_verifychangelog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41874
diff changeset
   246
2f1f475e9646 verify: document `_verifychangelog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41874
diff changeset
   247
        The following checks are performed:
2f1f475e9646 verify: document `_verifychangelog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41874
diff changeset
   248
        - all of `_checkrevlog` checks,
2f1f475e9646 verify: document `_verifychangelog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41874
diff changeset
   249
        - all of `_checkentry` checks (for each revisions),
2f1f475e9646 verify: document `_verifychangelog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41874
diff changeset
   250
        - each revision can be read.
2f1f475e9646 verify: document `_verifychangelog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41874
diff changeset
   251
2f1f475e9646 verify: document `_verifychangelog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41874
diff changeset
   252
        The function returns some of the data observed in the changesets as a
2f1f475e9646 verify: document `_verifychangelog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41874
diff changeset
   253
        (mflinkrevs, filelinkrevs) tuples:
2f1f475e9646 verify: document `_verifychangelog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41874
diff changeset
   254
        - mflinkrevs:   is a { manifest-node -> [changelog-rev] } mapping
2f1f475e9646 verify: document `_verifychangelog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41874
diff changeset
   255
        - filelinkrevs: is a { file-path -> [changelog-rev] } mapping
2f1f475e9646 verify: document `_verifychangelog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41874
diff changeset
   256
2f1f475e9646 verify: document `_verifychangelog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41874
diff changeset
   257
        If a matcher was specified, filelinkrevs will only contains matched
2f1f475e9646 verify: document `_verifychangelog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41874
diff changeset
   258
        files.
2f1f475e9646 verify: document `_verifychangelog`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41874
diff changeset
   259
        """
27647
2c2858f3c1bb verify: move changelog verificaiton to its own function
Durham Goode <durham@fb.com>
parents: 27646
diff changeset
   260
        ui = self.ui
2c2858f3c1bb verify: move changelog verificaiton to its own function
Durham Goode <durham@fb.com>
parents: 27646
diff changeset
   261
        repo = self.repo
30866
5249b6470de9 verify: replace _validpath() by matcher
Martin von Zweigbergk <martinvonz@google.com>
parents: 30393
diff changeset
   262
        match = self.match
27647
2c2858f3c1bb verify: move changelog verificaiton to its own function
Durham Goode <durham@fb.com>
parents: 27646
diff changeset
   263
        cl = repo.changelog
2c2858f3c1bb verify: move changelog verificaiton to its own function
Durham Goode <durham@fb.com>
parents: 27646
diff changeset
   264
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   265
        ui.status(_(b"checking changesets\n"))
27695
fb0cc863d172 verify: replace "output parameters" by return values
Martin von Zweigbergk <martinvonz@google.com>
parents: 27648
diff changeset
   266
        mflinkrevs = {}
fb0cc863d172 verify: replace "output parameters" by return values
Martin von Zweigbergk <martinvonz@google.com>
parents: 27648
diff changeset
   267
        filelinkrevs = {}
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   268
        seen = {}
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   269
        self._checkrevlog(cl, b"changelog", 0)
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   270
        progress = ui.makeprogress(
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   271
            _(b'checking'), unit=_(b'changesets'), total=len(repo)
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   272
        )
51015
812cd3dfa4cb verify: keep the revlog open for reading while verifying it
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50988
diff changeset
   273
        with cl.reading():
812cd3dfa4cb verify: keep the revlog open for reading while verifying it
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50988
diff changeset
   274
            for i in repo:
812cd3dfa4cb verify: keep the revlog open for reading while verifying it
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50988
diff changeset
   275
                progress.update(i)
812cd3dfa4cb verify: keep the revlog open for reading while verifying it
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50988
diff changeset
   276
                n = cl.node(i)
812cd3dfa4cb verify: keep the revlog open for reading while verifying it
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50988
diff changeset
   277
                self._checkentry(cl, i, n, seen, [i], b"changelog")
2778
fdc232d8a193 Move repo.verify
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   278
51015
812cd3dfa4cb verify: keep the revlog open for reading while verifying it
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50988
diff changeset
   279
                try:
812cd3dfa4cb verify: keep the revlog open for reading while verifying it
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50988
diff changeset
   280
                    changes = cl.read(n)
812cd3dfa4cb verify: keep the revlog open for reading while verifying it
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50988
diff changeset
   281
                    if changes[0] != self.repo.nullid:
812cd3dfa4cb verify: keep the revlog open for reading while verifying it
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50988
diff changeset
   282
                        mflinkrevs.setdefault(changes[0], []).append(i)
812cd3dfa4cb verify: keep the revlog open for reading while verifying it
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50988
diff changeset
   283
                        self.refersmf = True
812cd3dfa4cb verify: keep the revlog open for reading while verifying it
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50988
diff changeset
   284
                    for f in changes[3]:
812cd3dfa4cb verify: keep the revlog open for reading while verifying it
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50988
diff changeset
   285
                        if match(f):
812cd3dfa4cb verify: keep the revlog open for reading while verifying it
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50988
diff changeset
   286
                            filelinkrevs.setdefault(_normpath(f), []).append(i)
812cd3dfa4cb verify: keep the revlog open for reading while verifying it
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50988
diff changeset
   287
                except Exception as inst:
27444
6647401858ab verify: move widely used variables into class members
Durham Goode <durham@fb.com>
parents: 27443
diff changeset
   288
                    self.refersmf = True
51015
812cd3dfa4cb verify: keep the revlog open for reading while verifying it
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50988
diff changeset
   289
                    self._exc(i, _(b"unpacking changeset %s") % short(n), inst)
38397
1249475f0bd6 verify: use progress helper
Martin von Zweigbergk <martinvonz@google.com>
parents: 38396
diff changeset
   290
        progress.complete()
27695
fb0cc863d172 verify: replace "output parameters" by return values
Martin von Zweigbergk <martinvonz@google.com>
parents: 27648
diff changeset
   291
        return mflinkrevs, filelinkrevs
2778
fdc232d8a193 Move repo.verify
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   292
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   293
    def _verifymanifest(
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   294
        self, mflinkrevs, dir=b"", storefiles=None, subdirprogress=None
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   295
    ):
41876
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   296
        """verify the manifestlog content
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   297
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   298
        Inputs:
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   299
        - mflinkrevs:     a {manifest-node -> [changelog-revisions]} mapping
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   300
        - dir:            a subdirectory to check (for tree manifest repo)
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   301
        - storefiles:     set of currently "orphan" files.
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   302
        - subdirprogress: a progress object
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   303
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   304
        This function checks:
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   305
        * all of `_checkrevlog` checks (for all manifest related revlogs)
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   306
        * all of `_checkentry` checks (for all manifest related revisions)
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   307
        * nodes for subdirectory exists in the sub-directory manifest
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   308
        * each manifest entries have a file path
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   309
        * each manifest node refered in mflinkrevs exist in the manifest log
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   310
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   311
        If tree manifest is in use and a matchers is specified, only the
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   312
        sub-directories matching it will be verified.
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   313
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   314
        return a two level mapping:
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   315
            {"path" -> { filenode -> changelog-revision}}
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   316
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   317
        This mapping primarily contains entries for every files in the
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   318
        repository. In addition, when tree-manifest is used, it also contains
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   319
        sub-directory entries.
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   320
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   321
        If a matcher is provided, only matching paths will be included.
5ad5a70df2f7 verify: document the `_verifymanifest` method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41875
diff changeset
   322
        """
27646
8f43793382c6 verify: move manifest verification to its own function
Durham Goode <durham@fb.com>
parents: 27645
diff changeset
   323
        repo = self.repo
8f43793382c6 verify: move manifest verification to its own function
Durham Goode <durham@fb.com>
parents: 27645
diff changeset
   324
        ui = self.ui
30866
5249b6470de9 verify: replace _validpath() by matcher
Martin von Zweigbergk <martinvonz@google.com>
parents: 30393
diff changeset
   325
        match = self.match
30295
f65faa4422c8 manifest: remove manifest.readshallowdelta
Durham Goode <durham@fb.com>
parents: 28467
diff changeset
   326
        mfl = self.repo.manifestlog
39244
73cf21b2e8a6 manifest: add getstorage() to manifestlog and use it globally
Gregory Szorc <gregory.szorc@gmail.com>
parents: 38397
diff changeset
   327
        mf = mfl.getstorage(dir)
27646
8f43793382c6 verify: move manifest verification to its own function
Durham Goode <durham@fb.com>
parents: 27645
diff changeset
   328
28203
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   329
        if not dir:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   330
            self.ui.status(_(b"checking manifests\n"))
28203
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   331
27695
fb0cc863d172 verify: replace "output parameters" by return values
Martin von Zweigbergk <martinvonz@google.com>
parents: 27648
diff changeset
   332
        filenodes = {}
28203
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   333
        subdirnodes = {}
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   334
        seen = {}
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   335
        label = b"manifest"
28203
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   336
        if dir:
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   337
            label = dir
28204
962921c330b0 verify: check for orphaned dirlogs
Martin von Zweigbergk <martinvonz@google.com>
parents: 28203
diff changeset
   338
            revlogfiles = mf.files()
962921c330b0 verify: check for orphaned dirlogs
Martin von Zweigbergk <martinvonz@google.com>
parents: 28203
diff changeset
   339
            storefiles.difference_update(revlogfiles)
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   340
            if subdirprogress:  # should be true since we're in a subdirectory
38396
0ddbe03c5aaa verify: use progress helper for subdirectory progress
Martin von Zweigbergk <martinvonz@google.com>
parents: 37417
diff changeset
   341
                subdirprogress.increment()
27444
6647401858ab verify: move widely used variables into class members
Durham Goode <durham@fb.com>
parents: 27443
diff changeset
   342
        if self.refersmf:
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   343
            # Do not check manifest if there are only changelog entries with
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   344
            # null manifests.
47138
bc138f2a2e47 verify: pass a revlog to `_checkrevlog` in `_verifymanifest`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47012
diff changeset
   345
            self._checkrevlog(mf._revlog, label, 0)
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   346
        progress = ui.makeprogress(
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   347
            _(b'checking'), unit=_(b'manifests'), total=len(mf)
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   348
        )
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   349
        for i in mf:
28203
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   350
            if not dir:
38397
1249475f0bd6 verify: use progress helper
Martin von Zweigbergk <martinvonz@google.com>
parents: 38396
diff changeset
   351
                progress.update(i)
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   352
            n = mf.node(i)
41871
9e737ca539f6 verify: make `checkentry` a private method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41870
diff changeset
   353
            lr = self._checkentry(mf, i, n, seen, mflinkrevs.get(n, []), label)
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   354
            if n in mflinkrevs:
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   355
                del mflinkrevs[n]
28203
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   356
            elif dir:
47360
0693dc0b44fb verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47359
diff changeset
   357
                msg = _(b"%s not in parent-directory manifest") % short(n)
0693dc0b44fb verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47359
diff changeset
   358
                self._err(lr, msg, label)
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   359
            else:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   360
                self._err(lr, _(b"%s not in changesets") % short(n), label)
2778
fdc232d8a193 Move repo.verify
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   361
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   362
            try:
30295
f65faa4422c8 manifest: remove manifest.readshallowdelta
Durham Goode <durham@fb.com>
parents: 28467
diff changeset
   363
                mfdelta = mfl.get(dir, n).readdelta(shallow=True)
f65faa4422c8 manifest: remove manifest.readshallowdelta
Durham Goode <durham@fb.com>
parents: 28467
diff changeset
   364
                for f, fn, fl in mfdelta.iterentries():
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   365
                    if not f:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   366
                        self._err(lr, _(b"entry without name in manifest"))
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   367
                    elif f == b"/dev/null":  # ignore this in very old repos
28203
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   368
                        continue
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   369
                    fullpath = dir + _normpath(f)
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   370
                    if fl == b't':
30866
5249b6470de9 verify: replace _validpath() by matcher
Martin von Zweigbergk <martinvonz@google.com>
parents: 30393
diff changeset
   371
                        if not match.visitdir(fullpath):
5249b6470de9 verify: replace _validpath() by matcher
Martin von Zweigbergk <martinvonz@google.com>
parents: 30393
diff changeset
   372
                            continue
47362
fb43853975b4 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47361
diff changeset
   373
                        sdn = subdirnodes.setdefault(fullpath + b'/', {})
fb43853975b4 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47361
diff changeset
   374
                        sdn.setdefault(fn, []).append(lr)
28203
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   375
                    else:
30866
5249b6470de9 verify: replace _validpath() by matcher
Martin von Zweigbergk <martinvonz@google.com>
parents: 30393
diff changeset
   376
                        if not match(fullpath):
5249b6470de9 verify: replace _validpath() by matcher
Martin von Zweigbergk <martinvonz@google.com>
parents: 30393
diff changeset
   377
                            continue
28203
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   378
                        filenodes.setdefault(fullpath, {}).setdefault(fn, lr)
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   379
            except Exception as inst:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   380
                self._exc(lr, _(b"reading delta %s") % short(n), inst, label)
42157
7755b89cadaf verify: also check full manifest validity during verify runs
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 42156
diff changeset
   381
            if self._level >= VERIFY_FULL:
7755b89cadaf verify: also check full manifest validity during verify runs
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 42156
diff changeset
   382
                try:
7755b89cadaf verify: also check full manifest validity during verify runs
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 42156
diff changeset
   383
                    # Various issues can affect manifest. So we read each full
7755b89cadaf verify: also check full manifest validity during verify runs
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 42156
diff changeset
   384
                    # text from storage. This triggers the checks from the core
7755b89cadaf verify: also check full manifest validity during verify runs
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 42156
diff changeset
   385
                    # code (eg: hash verification, filename are ordered, etc.)
7755b89cadaf verify: also check full manifest validity during verify runs
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 42156
diff changeset
   386
                    mfdelta = mfl.get(dir, n).read()
7755b89cadaf verify: also check full manifest validity during verify runs
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 42156
diff changeset
   387
                except Exception as inst:
47361
bfb0ed91bb5e verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47360
diff changeset
   388
                    msg = _(b"reading full manifest %s") % short(n)
bfb0ed91bb5e verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47360
diff changeset
   389
                    self._exc(lr, msg, inst, label)
42157
7755b89cadaf verify: also check full manifest validity during verify runs
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 42156
diff changeset
   390
28203
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   391
        if not dir:
38397
1249475f0bd6 verify: use progress helper
Martin von Zweigbergk <martinvonz@google.com>
parents: 38396
diff changeset
   392
            progress.complete()
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   393
28111
06205989264b verify: move cross-checking of changeset/manifest out of _crosscheckfiles()
Martin von Zweigbergk <martinvonz@google.com>
parents: 28007
diff changeset
   394
        if self.havemf:
41877
9c5a6af74afa verify: small refactoring and documentation in `_verifymanifest`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41876
diff changeset
   395
            # since we delete entry in `mflinkrevs` during iteration, any
9c5a6af74afa verify: small refactoring and documentation in `_verifymanifest`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41876
diff changeset
   396
            # remaining entries are "missing". We need to issue errors for them.
9c5a6af74afa verify: small refactoring and documentation in `_verifymanifest`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41876
diff changeset
   397
            changesetpairs = [(c, m) for m in mflinkrevs for c in mflinkrevs[m]]
9c5a6af74afa verify: small refactoring and documentation in `_verifymanifest`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41876
diff changeset
   398
            for c, m in sorted(changesetpairs):
28203
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   399
                if dir:
47363
9823b3489e90 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47362
diff changeset
   400
                    self._err(c, WARN_PARENT_DIR_UNKNOWN_REV % short(m), label)
28203
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   401
                else:
47364
80c690bf4953 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47363
diff changeset
   402
                    msg = _(b"changeset refers to unknown revision %s")
80c690bf4953 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47363
diff changeset
   403
                    msg %= short(m)
80c690bf4953 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47363
diff changeset
   404
                    self._err(c, msg, label)
28203
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   405
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   406
        if not dir and subdirnodes:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   407
            self.ui.status(_(b"checking directory manifests\n"))
28204
962921c330b0 verify: check for orphaned dirlogs
Martin von Zweigbergk <martinvonz@google.com>
parents: 28203
diff changeset
   408
            storefiles = set()
28205
53f42c8d5f71 verify: show progress while verifying dirlogs
Martin von Zweigbergk <martinvonz@google.com>
parents: 28204
diff changeset
   409
            subdirs = set()
28204
962921c330b0 verify: check for orphaned dirlogs
Martin von Zweigbergk <martinvonz@google.com>
parents: 28203
diff changeset
   410
            revlogv1 = self.revlogv1
47877
2174f54aab18 store: return just one filename in walk functions
Valentin Gatien-Baron <valentin.gatienbaron@gmail.com>
parents: 47370
diff changeset
   411
            undecodable = []
50504
862e3a13da44 store: rename `datafiles` to `data_entries`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50477
diff changeset
   412
            for entry in repo.store.data_entries(undecodable=undecodable):
50472
9fdc28e21b68 store: introduce a EntryFile object to actually access file info
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50471
diff changeset
   413
                for file_ in entry.files():
9fdc28e21b68 store: introduce a EntryFile object to actually access file info
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50471
diff changeset
   414
                    f = file_.unencoded_path
50477
4cbdfab6f812 store: lazily get file size on demand for the fncache case
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50472
diff changeset
   415
                    size = file_.file_size(repo.store.vfs)
50472
9fdc28e21b68 store: introduce a EntryFile object to actually access file info
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50471
diff changeset
   416
                    if (size > 0 or not revlogv1) and f.startswith(b'meta/'):
9fdc28e21b68 store: introduce a EntryFile object to actually access file info
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50471
diff changeset
   417
                        storefiles.add(_normpath(f))
9fdc28e21b68 store: introduce a EntryFile object to actually access file info
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50471
diff changeset
   418
                        subdirs.add(os.path.dirname(f))
47877
2174f54aab18 store: return just one filename in walk functions
Valentin Gatien-Baron <valentin.gatienbaron@gmail.com>
parents: 47370
diff changeset
   419
            for f in undecodable:
2174f54aab18 store: return just one filename in walk functions
Valentin Gatien-Baron <valentin.gatienbaron@gmail.com>
parents: 47370
diff changeset
   420
                self._err(None, _(b"cannot decode filename '%s'") % f)
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   421
            subdirprogress = ui.makeprogress(
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   422
                _(b'checking'), unit=_(b'manifests'), total=len(subdirs)
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   423
            )
28204
962921c330b0 verify: check for orphaned dirlogs
Martin von Zweigbergk <martinvonz@google.com>
parents: 28203
diff changeset
   424
48913
f254fc73d956 global: bulk replace simple pycompat.iteritems(x) with x.items()
Gregory Szorc <gregory.szorc@gmail.com>
parents: 48875
diff changeset
   425
        for subdir, linkrevs in subdirnodes.items():
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   426
            subdirfilenodes = self._verifymanifest(
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   427
                linkrevs, subdir, storefiles, subdirprogress
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   428
            )
48913
f254fc73d956 global: bulk replace simple pycompat.iteritems(x) with x.items()
Gregory Szorc <gregory.szorc@gmail.com>
parents: 48875
diff changeset
   429
            for f, onefilenodes in subdirfilenodes.items():
28203
7297e9e13a8a verify: check directory manifests
Martin von Zweigbergk <martinvonz@google.com>
parents: 28115
diff changeset
   430
                filenodes.setdefault(f, {}).update(onefilenodes)
28111
06205989264b verify: move cross-checking of changeset/manifest out of _crosscheckfiles()
Martin von Zweigbergk <martinvonz@google.com>
parents: 28007
diff changeset
   431
28204
962921c330b0 verify: check for orphaned dirlogs
Martin von Zweigbergk <martinvonz@google.com>
parents: 28203
diff changeset
   432
        if not dir and subdirnodes:
46790
5137896602d9 typing: add an assertion to verify.py to appease pytype
Matt Harbison <matt_harbison@yahoo.com>
parents: 46696
diff changeset
   433
            assert subdirprogress is not None  # help pytype
38396
0ddbe03c5aaa verify: use progress helper for subdirectory progress
Martin von Zweigbergk <martinvonz@google.com>
parents: 37417
diff changeset
   434
            subdirprogress.complete()
37417
76d2115cb817 verify: allow suppressing warnings about extra files
Gregory Szorc <gregory.szorc@gmail.com>
parents: 37410
diff changeset
   435
            if self.warnorphanstorefiles:
76d2115cb817 verify: allow suppressing warnings about extra files
Gregory Szorc <gregory.szorc@gmail.com>
parents: 37410
diff changeset
   436
                for f in sorted(storefiles):
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   437
                    self._warn(_(b"warning: orphan data file '%s'") % f)
28204
962921c330b0 verify: check for orphaned dirlogs
Martin von Zweigbergk <martinvonz@google.com>
parents: 28203
diff changeset
   438
27695
fb0cc863d172 verify: replace "output parameters" by return values
Martin von Zweigbergk <martinvonz@google.com>
parents: 27648
diff changeset
   439
        return filenodes
27645
df8973e1fb45 verify: move file cross checking to its own function
Durham Goode <durham@fb.com>
parents: 27644
diff changeset
   440
28111
06205989264b verify: move cross-checking of changeset/manifest out of _crosscheckfiles()
Martin von Zweigbergk <martinvonz@google.com>
parents: 28007
diff changeset
   441
    def _crosscheckfiles(self, filelinkrevs, filenodes):
27645
df8973e1fb45 verify: move file cross checking to its own function
Durham Goode <durham@fb.com>
parents: 27644
diff changeset
   442
        repo = self.repo
df8973e1fb45 verify: move file cross checking to its own function
Durham Goode <durham@fb.com>
parents: 27644
diff changeset
   443
        ui = self.ui
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   444
        ui.status(_(b"crosschecking files in changesets and manifests\n"))
2778
fdc232d8a193 Move repo.verify
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   445
28111
06205989264b verify: move cross-checking of changeset/manifest out of _crosscheckfiles()
Martin von Zweigbergk <martinvonz@google.com>
parents: 28007
diff changeset
   446
        total = len(filelinkrevs) + len(filenodes)
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   447
        progress = ui.makeprogress(
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   448
            _(b'crosschecking'), unit=_(b'files'), total=total
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   449
        )
27645
df8973e1fb45 verify: move file cross checking to its own function
Durham Goode <durham@fb.com>
parents: 27644
diff changeset
   450
        if self.havemf:
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   451
            for f in sorted(filelinkrevs):
38397
1249475f0bd6 verify: use progress helper
Martin von Zweigbergk <martinvonz@google.com>
parents: 38396
diff changeset
   452
                progress.increment()
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   453
                if f not in filenodes:
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   454
                    lr = filelinkrevs[f][0]
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   455
                    self._err(lr, _(b"in changeset but not in manifest"), f)
6892
dab95717058d verify: check repo.store
Adrian Buehlmann <adrian@cadifra.com>
parents: 6889
diff changeset
   456
27645
df8973e1fb45 verify: move file cross checking to its own function
Durham Goode <durham@fb.com>
parents: 27644
diff changeset
   457
        if self.havecl:
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   458
            for f in sorted(filenodes):
38397
1249475f0bd6 verify: use progress helper
Martin von Zweigbergk <martinvonz@google.com>
parents: 38396
diff changeset
   459
                progress.increment()
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   460
                if f not in filelinkrevs:
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   461
                    try:
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   462
                        fl = repo.file(f)
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   463
                        lr = min([fl.linkrev(fl.rev(n)) for n in filenodes[f]])
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   464
                    except Exception:
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   465
                        lr = None
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   466
                    self._err(lr, _(b"in manifest but not in changeset"), f)
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   467
38397
1249475f0bd6 verify: use progress helper
Martin von Zweigbergk <martinvonz@google.com>
parents: 38396
diff changeset
   468
        progress.complete()
8291
f5c1a9094e41 verify: avoid exception on missing file revlog
Henrik Stuart <hg@hstuart.dk>
parents: 8225
diff changeset
   469
27644
331e5c28f5f0 verify: move filelog verification to its own function
Durham Goode <durham@fb.com>
parents: 27643
diff changeset
   470
    def _verifyfiles(self, filenodes, filelinkrevs):
331e5c28f5f0 verify: move filelog verification to its own function
Durham Goode <durham@fb.com>
parents: 27643
diff changeset
   471
        repo = self.repo
331e5c28f5f0 verify: move filelog verification to its own function
Durham Goode <durham@fb.com>
parents: 27643
diff changeset
   472
        ui = self.ui
331e5c28f5f0 verify: move filelog verification to its own function
Durham Goode <durham@fb.com>
parents: 27643
diff changeset
   473
        lrugetctx = self.lrugetctx
331e5c28f5f0 verify: move filelog verification to its own function
Durham Goode <durham@fb.com>
parents: 27643
diff changeset
   474
        revlogv1 = self.revlogv1
331e5c28f5f0 verify: move filelog verification to its own function
Durham Goode <durham@fb.com>
parents: 27643
diff changeset
   475
        havemf = self.havemf
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   476
        ui.status(_(b"checking files\n"))
8291
f5c1a9094e41 verify: avoid exception on missing file revlog
Henrik Stuart <hg@hstuart.dk>
parents: 8225
diff changeset
   477
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   478
        storefiles = set()
47877
2174f54aab18 store: return just one filename in walk functions
Valentin Gatien-Baron <valentin.gatienbaron@gmail.com>
parents: 47370
diff changeset
   479
        undecodable = []
50504
862e3a13da44 store: rename `datafiles` to `data_entries`
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50477
diff changeset
   480
        for entry in repo.store.data_entries(undecodable=undecodable):
50472
9fdc28e21b68 store: introduce a EntryFile object to actually access file info
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50471
diff changeset
   481
            for file_ in entry.files():
50477
4cbdfab6f812 store: lazily get file size on demand for the fncache case
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50472
diff changeset
   482
                size = file_.file_size(repo.store.vfs)
50472
9fdc28e21b68 store: introduce a EntryFile object to actually access file info
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50471
diff changeset
   483
                f = file_.unencoded_path
9fdc28e21b68 store: introduce a EntryFile object to actually access file info
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50471
diff changeset
   484
                if (size > 0 or not revlogv1) and f.startswith(b'data/'):
9fdc28e21b68 store: introduce a EntryFile object to actually access file info
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 50471
diff changeset
   485
                    storefiles.add(_normpath(f))
47877
2174f54aab18 store: return just one filename in walk functions
Valentin Gatien-Baron <valentin.gatienbaron@gmail.com>
parents: 47370
diff changeset
   486
        for f in undecodable:
2174f54aab18 store: return just one filename in walk functions
Valentin Gatien-Baron <valentin.gatienbaron@gmail.com>
parents: 47370
diff changeset
   487
            self._err(None, _(b"cannot decode filename '%s'") % f)
6892
dab95717058d verify: check repo.store
Adrian Buehlmann <adrian@cadifra.com>
parents: 6889
diff changeset
   488
39842
97986c9c69d3 verify: start to abstract file verification
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39777
diff changeset
   489
        state = {
39845
e6d3d39cc1c7 revlog: use proper version comparison during verify
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39842
diff changeset
   490
            # TODO this assumes revlog storage for changelog.
47139
f58a13c52726 revlog: split the `version` attribute into its two components
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47138
diff changeset
   491
            b'expectedversion': self.repo.changelog._format_version,
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   492
            b'skipflags': self.skipflags,
39872
733db72f0f54 revlog: move revision verification out of verify
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39845
diff changeset
   493
            # experimental config: censor.policy
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   494
            b'erroroncensored': ui.config(b'censor', b'policy') == b'abort',
39842
97986c9c69d3 verify: start to abstract file verification
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39777
diff changeset
   495
        }
97986c9c69d3 verify: start to abstract file verification
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39777
diff changeset
   496
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   497
        files = sorted(set(filenodes) | set(filelinkrevs))
27644
331e5c28f5f0 verify: move filelog verification to its own function
Durham Goode <durham@fb.com>
parents: 27643
diff changeset
   498
        revisions = 0
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   499
        progress = ui.makeprogress(
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   500
            _(b'checking'), unit=_(b'files'), total=len(files)
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   501
        )
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   502
        for i, f in enumerate(files):
38397
1249475f0bd6 verify: use progress helper
Martin von Zweigbergk <martinvonz@google.com>
parents: 38396
diff changeset
   503
            progress.update(i, item=f)
6892
dab95717058d verify: check repo.store
Adrian Buehlmann <adrian@cadifra.com>
parents: 6889
diff changeset
   504
            try:
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   505
                linkrevs = filelinkrevs[f]
6892
dab95717058d verify: check repo.store
Adrian Buehlmann <adrian@cadifra.com>
parents: 6889
diff changeset
   506
            except KeyError:
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   507
                # in manifest but not in changelog
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   508
                linkrevs = []
2778
fdc232d8a193 Move repo.verify
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   509
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   510
            if linkrevs:
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   511
                lr = linkrevs[0]
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   512
            else:
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   513
                lr = None
2778
fdc232d8a193 Move repo.verify
Matt Mackall <mpm@selenic.com>
parents:
diff changeset
   514
3744
d626fc9e3985 verify: add rename link checking
Matt Mackall <mpm@selenic.com>
parents: 3473
diff changeset
   515
            try:
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   516
                fl = repo.file(f)
39777
b63dee7bd0d9 global: replace most uses of RevlogError with StorageError (API)
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39489
diff changeset
   517
            except error.StorageError as e:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   518
                self._err(lr, _(b"broken revlog! (%s)") % e, f)
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   519
                continue
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   520
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   521
            for ff in fl.files():
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   522
                try:
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   523
                    storefiles.remove(ff)
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   524
                except KeyError:
37417
76d2115cb817 verify: allow suppressing warnings about extra files
Gregory Szorc <gregory.szorc@gmail.com>
parents: 37410
diff changeset
   525
                    if self.warnorphanstorefiles:
47365
f39df5545cf2 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47364
diff changeset
   526
                        msg = _(b" warning: revlog '%s' not in fncache!")
f39df5545cf2 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47364
diff changeset
   527
                        self._warn(msg % ff)
37417
76d2115cb817 verify: allow suppressing warnings about extra files
Gregory Szorc <gregory.szorc@gmail.com>
parents: 37410
diff changeset
   528
                        self.fncachewarned = True
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   529
39842
97986c9c69d3 verify: start to abstract file verification
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39777
diff changeset
   530
            if not len(fl) and (self.havecl or self.havemf):
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   531
                self._err(lr, _(b"empty or missing %s") % f)
39842
97986c9c69d3 verify: start to abstract file verification
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39777
diff changeset
   532
            else:
39872
733db72f0f54 revlog: move revision verification out of verify
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39845
diff changeset
   533
                # Guard against implementations not setting this.
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   534
                state[b'skipread'] = set()
44073
b9e174d4ed11 verify: allow the storage to signal when renames can be tested on `skipread`
Matt Harbison <matt_harbison@yahoo.com>
parents: 43956
diff changeset
   535
                state[b'safe_renamed'] = set()
b9e174d4ed11 verify: allow the storage to signal when renames can be tested on `skipread`
Matt Harbison <matt_harbison@yahoo.com>
parents: 43956
diff changeset
   536
39842
97986c9c69d3 verify: start to abstract file verification
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39777
diff changeset
   537
                for problem in fl.verifyintegrity(state):
39872
733db72f0f54 revlog: move revision verification out of verify
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39845
diff changeset
   538
                    if problem.node is not None:
733db72f0f54 revlog: move revision verification out of verify
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39845
diff changeset
   539
                        linkrev = fl.linkrev(fl.rev(problem.node))
733db72f0f54 revlog: move revision verification out of verify
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39845
diff changeset
   540
                    else:
733db72f0f54 revlog: move revision verification out of verify
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39845
diff changeset
   541
                        linkrev = None
733db72f0f54 revlog: move revision verification out of verify
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39845
diff changeset
   542
39842
97986c9c69d3 verify: start to abstract file verification
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39777
diff changeset
   543
                    if problem.warning:
41862
c66037fb1bc5 verify: make the `warn` method private
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41861
diff changeset
   544
                        self._warn(problem.warning)
39842
97986c9c69d3 verify: start to abstract file verification
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39777
diff changeset
   545
                    elif problem.error:
47366
15f6887c9c68 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47365
diff changeset
   546
                        linkrev_msg = linkrev if linkrev is not None else lr
15f6887c9c68 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47365
diff changeset
   547
                        self._err(linkrev_msg, problem.error, f)
39842
97986c9c69d3 verify: start to abstract file verification
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39777
diff changeset
   548
                    else:
97986c9c69d3 verify: start to abstract file verification
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39777
diff changeset
   549
                        raise error.ProgrammingError(
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   550
                            b'problem instance does not set warning or error '
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   551
                            b'attribute: %s' % problem.msg
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   552
                        )
39842
97986c9c69d3 verify: start to abstract file verification
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39777
diff changeset
   553
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   554
            seen = {}
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   555
            for i in fl:
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   556
                revisions += 1
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   557
                n = fl.node(i)
41871
9e737ca539f6 verify: make `checkentry` a private method
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 41870
diff changeset
   558
                lr = self._checkentry(fl, i, n, seen, linkrevs, f)
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   559
                if f in filenodes:
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   560
                    if havemf and n not in filenodes[f]:
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   561
                        self._err(lr, _(b"%s not in manifests") % (short(n)), f)
6534
9b35a9f34675 verify: check copy source revlog and nodeid
Patrick Mezard <pmezard@gmail.com>
parents: 6211
diff changeset
   562
                    else:
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   563
                        del filenodes[f][n]
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   564
44073
b9e174d4ed11 verify: allow the storage to signal when renames can be tested on `skipread`
Matt Harbison <matt_harbison@yahoo.com>
parents: 43956
diff changeset
   565
                if n in state[b'skipread'] and n not in state[b'safe_renamed']:
39872
733db72f0f54 revlog: move revision verification out of verify
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39845
diff changeset
   566
                    continue
3744
d626fc9e3985 verify: add rename link checking
Matt Mackall <mpm@selenic.com>
parents: 3473
diff changeset
   567
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   568
                # check renames
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   569
                try:
43956
a447efd991b9 verify: update comment to say that lfs doesn't need fulltext to check renames
Matt Harbison <matt_harbison@yahoo.com>
parents: 43106
diff changeset
   570
                    # This requires resolving fulltext (at least on revlogs,
a447efd991b9 verify: update comment to say that lfs doesn't need fulltext to check renames
Matt Harbison <matt_harbison@yahoo.com>
parents: 43106
diff changeset
   571
                    # though not with LFS revisions). We may want
a447efd991b9 verify: update comment to say that lfs doesn't need fulltext to check renames
Matt Harbison <matt_harbison@yahoo.com>
parents: 43106
diff changeset
   572
                    # ``verifyintegrity()`` to pass a set of nodes with
39872
733db72f0f54 revlog: move revision verification out of verify
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39845
diff changeset
   573
                    # rename metadata as an optimization.
733db72f0f54 revlog: move revision verification out of verify
Gregory Szorc <gregory.szorc@gmail.com>
parents: 39845
diff changeset
   574
                    rp = fl.renamed(n)
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   575
                    if rp:
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   576
                        if lr is not None and ui.verbose:
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   577
                            ctx = lrugetctx(lr)
36339
a4d41ba4ad23 verify: don't reimplement any()
Martin von Zweigbergk <martinvonz@google.com>
parents: 36185
diff changeset
   578
                            if not any(rp[0] in pctx for pctx in ctx.parents()):
47367
34a92e84267e verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47366
diff changeset
   579
                                self._warn(WARN_UNKNOWN_COPY_SOURCE % (f, ctx))
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   580
                        fl2 = repo.file(rp[0])
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   581
                        if not len(fl2):
47368
1a0f177b300a verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47367
diff changeset
   582
                            m = _(b"empty or missing copy source revlog %s:%s")
1a0f177b300a verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47367
diff changeset
   583
                            self._err(lr, m % (rp[0], short(rp[1])), f)
47012
d55b71393907 node: replace nullid and friends with nodeconstants class
Joerg Sonnenberger <joerg@bec.de>
parents: 46895
diff changeset
   584
                        elif rp[1] == self.repo.nullid:
47369
041d6515bb0f verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47368
diff changeset
   585
                            msg = WARN_NULLID_COPY_SOURCE
041d6515bb0f verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47368
diff changeset
   586
                            msg %= (f, lr, rp[0], short(rp[1]))
041d6515bb0f verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47368
diff changeset
   587
                            ui.note(msg)
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   588
                        else:
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   589
                            fl2.rev(rp[1])
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   590
                except Exception as inst:
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   591
                    self._exc(
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   592
                        lr, _(b"checking rename of %s") % short(n), inst, f
43076
2372284d9457 formatting: blacken the codebase
Augie Fackler <augie@google.com>
parents: 42387
diff changeset
   593
                    )
6892
dab95717058d verify: check repo.store
Adrian Buehlmann <adrian@cadifra.com>
parents: 6889
diff changeset
   594
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   595
            # cross-check
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   596
            if f in filenodes:
48931
6f10a2d6adf9 verify: remove pycompat.iteritems()
Gregory Szorc <gregory.szorc@gmail.com>
parents: 48913
diff changeset
   597
                fns = [(v, k) for k, v in filenodes[f].items()]
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   598
                for lr, node in sorted(fns):
47370
ab5fd39cb402 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47369
diff changeset
   599
                    msg = _(b"manifest refers to unknown revision %s")
ab5fd39cb402 verify: use some intermediate variables instead of a multi-liner
Pierre-Yves David <pierre-yves.david@octobus.net>
parents: 47369
diff changeset
   600
                    self._err(lr, msg % short(node), f)
38397
1249475f0bd6 verify: use progress helper
Martin von Zweigbergk <martinvonz@google.com>
parents: 38396
diff changeset
   601
        progress.complete()
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   602
37417
76d2115cb817 verify: allow suppressing warnings about extra files
Gregory Szorc <gregory.szorc@gmail.com>
parents: 37410
diff changeset
   603
        if self.warnorphanstorefiles:
76d2115cb817 verify: allow suppressing warnings about extra files
Gregory Szorc <gregory.szorc@gmail.com>
parents: 37410
diff changeset
   604
            for f in sorted(storefiles):
43077
687b865b95ad formatting: byteify all mercurial/ and hgext/ string literals
Augie Fackler <augie@google.com>
parents: 43076
diff changeset
   605
                self._warn(_(b"warning: orphan data file '%s'") % f)
27443
937e73a6e4ff verify: move verify logic into a class
Durham Goode <durham@fb.com>
parents: 26900
diff changeset
   606
27644
331e5c28f5f0 verify: move filelog verification to its own function
Durham Goode <durham@fb.com>
parents: 27643
diff changeset
   607
        return len(files), revisions
49826
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   608
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   609
    def _verify_dirstate(self):
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   610
        """Check that the dirstate is consistent with the parent's manifest"""
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   611
        repo = self.repo
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   612
        ui = self.ui
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   613
        ui.status(_(b"checking dirstate\n"))
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   614
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   615
        parent1, parent2 = repo.dirstate.parents()
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   616
        m1 = repo[parent1].manifest()
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   617
        m2 = repo[parent2].manifest()
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   618
        dirstate_errors = 0
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   619
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   620
        is_narrow = requirements.NARROW_REQUIREMENT in repo.requirements
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   621
        narrow_matcher = repo.narrowmatch() if is_narrow else None
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   622
49827
d09a57ce6fc4 verify: print short `p1` node in relevant dirstate messages
Raphaël Gomès <rgomes@octobus.net>
parents: 49826
diff changeset
   623
        for err in repo.dirstate.verify(m1, m2, parent1, narrow_matcher):
49826
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   624
            ui.error(err)
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   625
            dirstate_errors += 1
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   626
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   627
        if dirstate_errors:
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   628
            self.errors += dirstate_errors
c84844cd523a verify: also check dirstate
Raphaël Gomès <rgomes@octobus.net>
parents: 48946
diff changeset
   629
        return dirstate_errors