node: make bin() be a wrapper instead of just an alias

This includes a full backout of 59affe7e and 30d0cb27. Per the review
of the former, we'd rather adapt the API to behave like it used to (at
least for now), and take a second run at it if it shows up in our
performance numbers. I ran perfrevlogindex with and without this
change and it didn't make a measurable difference, so maybe it's fine
(despite my intuition to the contrary).

Differential Revision: https://phab.mercurial-scm.org/D2279

hgext/histedit.py

@@ -1,1651 +1,1650 b''
 # histedit.py - interactive history editing for mercurial
 #
 # Copyright 2009 Augie Fackler <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 """interactive history editing
 
 With this extension installed, Mercurial gains one new command: histedit. Usage
 is as follows, assuming the following history::
 
  @  3[tip]   7c2fd3b9020c   2009-04-27 18:04 -0500   durin42
  |    Add delta
  |
  o  2   030b686bedc4   2009-04-27 18:04 -0500   durin42
  |    Add gamma
  |
  o  1   c561b4e977df   2009-04-27 18:04 -0500   durin42
  |    Add beta
  |
  o  0   d8d2fcd0e319   2009-04-27 18:04 -0500   durin42
       Add alpha
 
 If you were to run ``hg histedit c561b4e977df``, you would see the following
 file open in your editor::
 
  pick c561b4e977df Add beta
  pick 030b686bedc4 Add gamma
  pick 7c2fd3b9020c Add delta
 
  # Edit history between c561b4e977df and 7c2fd3b9020c
  #
  # Commits are listed from least to most recent
  #
  # Commands:
  #  p, pick = use commit
  #  e, edit = use commit, but stop for amending
  #  f, fold = use commit, but combine it with the one above
  #  r, roll = like fold, but discard this commit's description and date
  #  d, drop = remove commit from history
  #  m, mess = edit commit message without changing commit content
  #  b, base = checkout changeset and apply further changesets from there
  #
 
 In this file, lines beginning with ``#`` are ignored. You must specify a rule
 for each revision in your history. For example, if you had meant to add gamma
 before beta, and then wanted to add delta in the same revision as beta, you
 would reorganize the file to look like this::
 
  pick 030b686bedc4 Add gamma
  pick c561b4e977df Add beta
  fold 7c2fd3b9020c Add delta
 
  # Edit history between c561b4e977df and 7c2fd3b9020c
  #
  # Commits are listed from least to most recent
  #
  # Commands:
  #  p, pick = use commit
  #  e, edit = use commit, but stop for amending
  #  f, fold = use commit, but combine it with the one above
  #  r, roll = like fold, but discard this commit's description and date
  #  d, drop = remove commit from history
  #  m, mess = edit commit message without changing commit content
  #  b, base = checkout changeset and apply further changesets from there
  #
 
 At which point you close the editor and ``histedit`` starts working. When you
 specify a ``fold`` operation, ``histedit`` will open an editor when it folds
 those revisions together, offering you a chance to clean up the commit message::
 
  Add beta
  ***
  Add delta
 
 Edit the commit message to your liking, then close the editor. The date used
 for the commit will be the later of the two commits' dates. For this example,
 let's assume that the commit message was changed to ``Add beta and delta.``
 After histedit has run and had a chance to remove any old or temporary
 revisions it needed, the history looks like this::
 
  @  2[tip]   989b4d060121   2009-04-27 18:04 -0500   durin42
  |    Add beta and delta.
  |
  o  1   081603921c3f   2009-04-27 18:04 -0500   durin42
  |    Add gamma
  |
  o  0   d8d2fcd0e319   2009-04-27 18:04 -0500   durin42
       Add alpha
 
 Note that ``histedit`` does *not* remove any revisions (even its own temporary
 ones) until after it has completed all the editing operations, so it will
 probably perform several strip operations when it's done. For the above example,
 it had to run strip twice. Strip can be slow depending on a variety of factors,
 so you might need to be a little patient. You can choose to keep the original
 revisions by passing the ``--keep`` flag.
 
 The ``edit`` operation will drop you back to a command prompt,
 allowing you to edit files freely, or even use ``hg record`` to commit
 some changes as a separate commit. When you're done, any remaining
 uncommitted changes will be committed as well. When done, run ``hg
 histedit --continue`` to finish this step. If there are uncommitted
 changes, you'll be prompted for a new commit message, but the default
 commit message will be the original message for the ``edit`` ed
 revision, and the date of the original commit will be preserved.
 
 The ``message`` operation will give you a chance to revise a commit
 message without changing the contents. It's a shortcut for doing
 ``edit`` immediately followed by `hg histedit --continue``.
 
 If ``histedit`` encounters a conflict when moving a revision (while
 handling ``pick`` or ``fold``), it'll stop in a similar manner to
 ``edit`` with the difference that it won't prompt you for a commit
 message when done. If you decide at this point that you don't like how
 much work it will be to rearrange history, or that you made a mistake,
 you can use ``hg histedit --abort`` to abandon the new changes you
 have made and return to the state before you attempted to edit your
 history.
 
 If we clone the histedit-ed example repository above and add four more
 changes, such that we have the following history::
 
    @  6[tip]   038383181893   2009-04-27 18:04 -0500   stefan
    |    Add theta
    |
    o  5   140988835471   2009-04-27 18:04 -0500   stefan
    |    Add eta
    |
    o  4   122930637314   2009-04-27 18:04 -0500   stefan
    |    Add zeta
    |
    o  3   836302820282   2009-04-27 18:04 -0500   stefan
    |    Add epsilon
    |
    o  2   989b4d060121   2009-04-27 18:04 -0500   durin42
    |    Add beta and delta.
    |
    o  1   081603921c3f   2009-04-27 18:04 -0500   durin42
    |    Add gamma
    |
    o  0   d8d2fcd0e319   2009-04-27 18:04 -0500   durin42
         Add alpha
 
 If you run ``hg histedit --outgoing`` on the clone then it is the same
 as running ``hg histedit 836302820282``. If you need plan to push to a
 repository that Mercurial does not detect to be related to the source
 repo, you can add a ``--force`` option.
 
 Config
 ------
 
 Histedit rule lines are truncated to 80 characters by default. You
 can customize this behavior by setting a different length in your
 configuration file::
 
   [histedit]
   linelen = 120      # truncate rule lines at 120 characters
 
 ``hg histedit`` attempts to automatically choose an appropriate base
 revision to use. To change which base revision is used, define a
 revset in your configuration file::
 
   [histedit]
   defaultrev = only(.) & draft()
 
 By default each edited revision needs to be present in histedit commands.
 To remove revision you need to use ``drop`` operation. You can configure
 the drop to be implicit for missing commits by adding::
 
   [histedit]
   dropmissing = True
 
 By default, histedit will close the transaction after each action. For
 performance purposes, you can configure histedit to use a single transaction
 across the entire histedit. WARNING: This setting introduces a significant risk
 of losing the work you've done in a histedit if the histedit aborts
 unexpectedly::
 
   [histedit]
   singletransaction = True
 
 """
 
 from __future__ import absolute_import
 
-import binascii
 import errno
 import os
 
 from mercurial.i18n import _
 from mercurial import (
     bundle2,
     cmdutil,
     context,
     copies,
     destutil,
     discovery,
     error,
     exchange,
     extensions,
     hg,
     lock,
     merge as mergemod,
     mergeutil,
     node,
     obsolete,
     pycompat,
     registrar,
     repair,
     scmutil,
     util,
 )
 
 pickle = util.pickle
 release = lock.release
 cmdtable = {}
 command = registrar.command(cmdtable)
 
 configtable = {}
 configitem = registrar.configitem(configtable)
 configitem('experimental', 'histedit.autoverb',
     default=False,
 )
 configitem('histedit', 'defaultrev',
     default=configitem.dynamicdefault,
 )
 configitem('histedit', 'dropmissing',
     default=False,
 )
 configitem('histedit', 'linelen',
     default=80,
 )
 configitem('histedit', 'singletransaction',
     default=False,
 )
 
 # Note for extension authors: ONLY specify testedwith = 'ships-with-hg-core' for
 # extensions which SHIP WITH MERCURIAL. Non-mainline extensions should
 # be specifying the version(s) of Mercurial they are tested with, or
 # leave the attribute unspecified.
 testedwith = 'ships-with-hg-core'
 
 actiontable = {}
 primaryactions = set()
 secondaryactions = set()
 tertiaryactions = set()
 internalactions = set()
 
 def geteditcomment(ui, first, last):
     """ construct the editor comment
     The comment includes::
      - an intro
      - sorted primary commands
      - sorted short commands
      - sorted long commands
      - additional hints
 
     Commands are only included once.
     """
     intro = _("""Edit history between %s and %s
 
 Commits are listed from least to most recent
 
 You can reorder changesets by reordering the lines
 
 Commands:
 """)
     actions = []
     def addverb(v):
         a = actiontable[v]
         lines = a.message.split("\n")
         if len(a.verbs):
             v = ', '.join(sorted(a.verbs, key=lambda v: len(v)))
         actions.append(" %s = %s" % (v, lines[0]))
         actions.extend(['  %s' for l in lines[1:]])
 
     for v in (
          sorted(primaryactions) +
          sorted(secondaryactions) +
          sorted(tertiaryactions)
         ):
         addverb(v)
     actions.append('')
 
     hints = []
     if ui.configbool('histedit', 'dropmissing'):
         hints.append("Deleting a changeset from the list "
                      "will DISCARD it from the edited history!")
 
     lines = (intro % (first, last)).split('\n') + actions + hints
 
     return ''.join(['# %s\n' % l if l else '#\n' for l in lines])
 
 class histeditstate(object):
     def __init__(self, repo, parentctxnode=None, actions=None, keep=None,
             topmost=None, replacements=None, lock=None, wlock=None):
         self.repo = repo
         self.actions = actions
         self.keep = keep
         self.topmost = topmost
         self.parentctxnode = parentctxnode
         self.lock = lock
         self.wlock = wlock
         self.backupfile = None
         if replacements is None:
             self.replacements = []
         else:
             self.replacements = replacements
 
     def read(self):
         """Load histedit state from disk and set fields appropriately."""
         try:
             state = self.repo.vfs.read('histedit-state')
         except IOError as err:
             if err.errno != errno.ENOENT:
                 raise
             cmdutil.wrongtooltocontinue(self.repo, _('histedit'))
 
         if state.startswith('v1\n'):
             data = self._load()
             parentctxnode, rules, keep, topmost, replacements, backupfile = data
         else:
             data = pickle.loads(state)
             parentctxnode, rules, keep, topmost, replacements = data
             backupfile = None
 
         self.parentctxnode = parentctxnode
         rules = "\n".join(["%s %s" % (verb, rest) for [verb, rest] in rules])
         actions = parserules(rules, self)
         self.actions = actions
         self.keep = keep
         self.topmost = topmost
         self.replacements = replacements
         self.backupfile = backupfile
 
     def write(self, tr=None):
         if tr:
             tr.addfilegenerator('histedit-state', ('histedit-state',),
                                 self._write, location='plain')
         else:
             with self.repo.vfs("histedit-state", "w") as f:
                 self._write(f)
 
     def _write(self, fp):
         fp.write('v1\n')
         fp.write('%s\n' % node.hex(self.parentctxnode))
         fp.write('%s\n' % node.hex(self.topmost))
         fp.write('%s\n' % ('True' if self.keep else 'False'))
         fp.write('%d\n' % len(self.actions))
         for action in self.actions:
             fp.write('%s\n' % action.tostate())
         fp.write('%d\n' % len(self.replacements))
         for replacement in self.replacements:
             fp.write('%s%s\n' % (node.hex(replacement[0]), ''.join(node.hex(r)
                 for r in replacement[1])))
         backupfile = self.backupfile
         if not backupfile:
             backupfile = ''
         fp.write('%s\n' % backupfile)
 
     def _load(self):
         fp = self.repo.vfs('histedit-state', 'r')
         lines = [l[:-1] for l in fp.readlines()]
 
         index = 0
         lines[index] # version number
         index += 1
 
         parentctxnode = node.bin(lines[index])
         index += 1
 
         topmost = node.bin(lines[index])
         index += 1
 
         keep = lines[index] == 'True'
         index += 1
 
         # Rules
         rules = []
         rulelen = int(lines[index])
         index += 1
         for i in xrange(rulelen):
             ruleaction = lines[index]
             index += 1
             rule = lines[index]
             index += 1
             rules.append((ruleaction, rule))
 
         # Replacements
         replacements = []
         replacementlen = int(lines[index])
         index += 1
         for i in xrange(replacementlen):
             replacement = lines[index]
             original = node.bin(replacement[:40])
             succ = [node.bin(replacement[i:i + 40]) for i in
                     range(40, len(replacement), 40)]
             replacements.append((original, succ))
             index += 1
 
         backupfile = lines[index]
         index += 1
 
         fp.close()
 
         return parentctxnode, rules, keep, topmost, replacements, backupfile
 
     def clear(self):
         if self.inprogress():
             self.repo.vfs.unlink('histedit-state')
 
     def inprogress(self):
         return self.repo.vfs.exists('histedit-state')
 
 
 class histeditaction(object):
     def __init__(self, state, node):
         self.state = state
         self.repo = state.repo
         self.node = node
 
     @classmethod
     def fromrule(cls, state, rule):
         """Parses the given rule, returning an instance of the histeditaction.
         """
         rulehash = rule.strip().split(' ', 1)[0]
         try:
             rev = node.bin(rulehash)
-        except (TypeError, binascii.Error):
+        except TypeError:
             raise error.ParseError("invalid changeset %s" % rulehash)
         return cls(state, rev)
 
     def verify(self, prev, expected, seen):
         """ Verifies semantic correctness of the rule"""
         repo = self.repo
         ha = node.hex(self.node)
         try:
             self.node = repo[ha].node()
         except error.RepoError:
             raise error.ParseError(_('unknown changeset %s listed')
                               % ha[:12])
         if self.node is not None:
             self._verifynodeconstraints(prev, expected, seen)
 
     def _verifynodeconstraints(self, prev, expected, seen):
         # by default command need a node in the edited list
         if self.node not in expected:
             raise error.ParseError(_('%s "%s" changeset was not a candidate')
                                    % (self.verb, node.short(self.node)),
                                    hint=_('only use listed changesets'))
         # and only one command per node
         if self.node in seen:
             raise error.ParseError(_('duplicated command for changeset %s') %
                                    node.short(self.node))
 
     def torule(self):
         """build a histedit rule line for an action
 
         by default lines are in the form:
         <hash> <rev> <summary>
         """
         ctx = self.repo[self.node]
         summary = _getsummary(ctx)
         line = '%s %s %d %s' % (self.verb, ctx, ctx.rev(), summary)
         # trim to 75 columns by default so it's not stupidly wide in my editor
         # (the 5 more are left for verb)
         maxlen = self.repo.ui.configint('histedit', 'linelen')
         maxlen = max(maxlen, 22) # avoid truncating hash
         return util.ellipsis(line, maxlen)
 
     def tostate(self):
         """Print an action in format used by histedit state files
            (the first line is a verb, the remainder is the second)
         """
         return "%s\n%s" % (self.verb, node.hex(self.node))
 
     def run(self):
         """Runs the action. The default behavior is simply apply the action's
         rulectx onto the current parentctx."""
         self.applychange()
         self.continuedirty()
         return self.continueclean()
 
     def applychange(self):
         """Applies the changes from this action's rulectx onto the current
         parentctx, but does not commit them."""
         repo = self.repo
         rulectx = repo[self.node]
         repo.ui.pushbuffer(error=True, labeled=True)
         hg.update(repo, self.state.parentctxnode, quietempty=True)
         stats = applychanges(repo.ui, repo, rulectx, {})
         repo.dirstate.setbranch(rulectx.branch())
         if stats and stats[3] > 0:
             buf = repo.ui.popbuffer()
             repo.ui.write(buf)
             raise error.InterventionRequired(
                 _('Fix up the change (%s %s)') %
                 (self.verb, node.short(self.node)),
                 hint=_('hg histedit --continue to resume'))
         else:
             repo.ui.popbuffer()
 
     def continuedirty(self):
         """Continues the action when changes have been applied to the working
         copy. The default behavior is to commit the dirty changes."""
         repo = self.repo
         rulectx = repo[self.node]
 
         editor = self.commiteditor()
         commit = commitfuncfor(repo, rulectx)
 
         commit(text=rulectx.description(), user=rulectx.user(),
                date=rulectx.date(), extra=rulectx.extra(), editor=editor)
 
     def commiteditor(self):
         """The editor to be used to edit the commit message."""
         return False
 
     def continueclean(self):
         """Continues the action when the working copy is clean. The default
         behavior is to accept the current commit as the new version of the
         rulectx."""
         ctx = self.repo['.']
         if ctx.node() == self.state.parentctxnode:
             self.repo.ui.warn(_('%s: skipping changeset (no changes)\n') %
                               node.short(self.node))
             return ctx, [(self.node, tuple())]
         if ctx.node() == self.node:
             # Nothing changed
             return ctx, []
         return ctx, [(self.node, (ctx.node(),))]
 
 def commitfuncfor(repo, src):
     """Build a commit function for the replacement of <src>
 
     This function ensure we apply the same treatment to all changesets.
 
     - Add a 'histedit_source' entry in extra.
 
     Note that fold has its own separated logic because its handling is a bit
     different and not easily factored out of the fold method.
     """
     phasemin = src.phase()
     def commitfunc(**kwargs):
         overrides = {('phases', 'new-commit'): phasemin}
         with repo.ui.configoverride(overrides, 'histedit'):
             extra = kwargs.get(r'extra', {}).copy()
             extra['histedit_source'] = src.hex()
             kwargs[r'extra'] = extra
             return repo.commit(**kwargs)
     return commitfunc
 
 def applychanges(ui, repo, ctx, opts):
     """Merge changeset from ctx (only) in the current working directory"""
     wcpar = repo.dirstate.parents()[0]
     if ctx.p1().node() == wcpar:
         # edits are "in place" we do not need to make any merge,
         # just applies changes on parent for editing
         cmdutil.revert(ui, repo, ctx, (wcpar, node.nullid), all=True)
         stats = None
     else:
         try:
             # ui.forcemerge is an internal variable, do not document
             repo.ui.setconfig('ui', 'forcemerge', opts.get('tool', ''),
                               'histedit')
             stats = mergemod.graft(repo, ctx, ctx.p1(), ['local', 'histedit'])
         finally:
             repo.ui.setconfig('ui', 'forcemerge', '', 'histedit')
     return stats
 
 def collapse(repo, first, last, commitopts, skipprompt=False):
     """collapse the set of revisions from first to last as new one.
 
     Expected commit options are:
         - message
         - date
         - username
     Commit message is edited in all cases.
 
     This function works in memory."""
     ctxs = list(repo.set('%d::%d', first, last))
     if not ctxs:
         return None
     for c in ctxs:
         if not c.mutable():
             raise error.ParseError(
                 _("cannot fold into public change %s") % node.short(c.node()))
     base = first.parents()[0]
 
     # commit a new version of the old changeset, including the update
     # collect all files which might be affected
     files = set()
     for ctx in ctxs:
         files.update(ctx.files())
 
     # Recompute copies (avoid recording a -> b -> a)
     copied = copies.pathcopies(base, last)
 
     # prune files which were reverted by the updates
     files = [f for f in files if not cmdutil.samefile(f, last, base)]
     # commit version of these files as defined by head
     headmf = last.manifest()
     def filectxfn(repo, ctx, path):
         if path in headmf:
             fctx = last[path]
             flags = fctx.flags()
             mctx = context.memfilectx(repo, ctx,
                                       fctx.path(), fctx.data(),
                                       islink='l' in flags,
                                       isexec='x' in flags,
                                       copied=copied.get(path))
             return mctx
         return None
 
     if commitopts.get('message'):
         message = commitopts['message']
     else:
         message = first.description()
     user = commitopts.get('user')
     date = commitopts.get('date')
     extra = commitopts.get('extra')
 
     parents = (first.p1().node(), first.p2().node())
     editor = None
     if not skipprompt:
         editor = cmdutil.getcommiteditor(edit=True, editform='histedit.fold')
     new = context.memctx(repo,
                          parents=parents,
                          text=message,
                          files=files,
                          filectxfn=filectxfn,
                          user=user,
                          date=date,
                          extra=extra,
                          editor=editor)
     return repo.commitctx(new)
 
 def _isdirtywc(repo):
     return repo[None].dirty(missing=True)
 
 def abortdirty():
     raise error.Abort(_('working copy has pending changes'),
         hint=_('amend, commit, or revert them and run histedit '
             '--continue, or abort with histedit --abort'))
 
 def action(verbs, message, priority=False, internal=False):
     def wrap(cls):
         assert not priority or not internal
         verb = verbs[0]
         if priority:
             primaryactions.add(verb)
         elif internal:
             internalactions.add(verb)
         elif len(verbs) > 1:
             secondaryactions.add(verb)
         else:
             tertiaryactions.add(verb)
 
         cls.verb = verb
         cls.verbs = verbs
         cls.message = message
         for verb in verbs:
             actiontable[verb] = cls
         return cls
     return wrap
 
 @action(['pick', 'p'],
         _('use commit'),
         priority=True)
 class pick(histeditaction):
     def run(self):
         rulectx = self.repo[self.node]
         if rulectx.parents()[0].node() == self.state.parentctxnode:
             self.repo.ui.debug('node %s unchanged\n' % node.short(self.node))
             return rulectx, []
 
         return super(pick, self).run()
 
 @action(['edit', 'e'],
         _('use commit, but stop for amending'),
         priority=True)
 class edit(histeditaction):
     def run(self):
         repo = self.repo
         rulectx = repo[self.node]
         hg.update(repo, self.state.parentctxnode, quietempty=True)
         applychanges(repo.ui, repo, rulectx, {})
         raise error.InterventionRequired(
             _('Editing (%s), you may commit or record as needed now.')
             % node.short(self.node),
             hint=_('hg histedit --continue to resume'))
 
     def commiteditor(self):
         return cmdutil.getcommiteditor(edit=True, editform='histedit.edit')
 
 @action(['fold', 'f'],
         _('use commit, but combine it with the one above'))
 class fold(histeditaction):
     def verify(self, prev, expected, seen):
         """ Verifies semantic correctness of the fold rule"""
         super(fold, self).verify(prev, expected, seen)
         repo = self.repo
         if not prev:
             c = repo[self.node].parents()[0]
         elif not prev.verb in ('pick', 'base'):
             return
         else:
             c = repo[prev.node]
         if not c.mutable():
             raise error.ParseError(
                 _("cannot fold into public change %s") % node.short(c.node()))
 
 
     def continuedirty(self):
         repo = self.repo
         rulectx = repo[self.node]
 
         commit = commitfuncfor(repo, rulectx)
         commit(text='fold-temp-revision %s' % node.short(self.node),
                user=rulectx.user(), date=rulectx.date(),
                extra=rulectx.extra())
 
     def continueclean(self):
         repo = self.repo
         ctx = repo['.']
         rulectx = repo[self.node]
         parentctxnode = self.state.parentctxnode
         if ctx.node() == parentctxnode:
             repo.ui.warn(_('%s: empty changeset\n') %
                               node.short(self.node))
             return ctx, [(self.node, (parentctxnode,))]
 
         parentctx = repo[parentctxnode]
         newcommits = set(c.node() for c in repo.set('(%d::. - %d)', parentctx,
                                                  parentctx))
         if not newcommits:
             repo.ui.warn(_('%s: cannot fold - working copy is not a '
                            'descendant of previous commit %s\n') %
                            (node.short(self.node), node.short(parentctxnode)))
             return ctx, [(self.node, (ctx.node(),))]
 
         middlecommits = newcommits.copy()
         middlecommits.discard(ctx.node())
 
         return self.finishfold(repo.ui, repo, parentctx, rulectx, ctx.node(),
                                middlecommits)
 
     def skipprompt(self):
         """Returns true if the rule should skip the message editor.
 
         For example, 'fold' wants to show an editor, but 'rollup'
         doesn't want to.
         """
         return False
 
     def mergedescs(self):
         """Returns true if the rule should merge messages of multiple changes.
 
         This exists mainly so that 'rollup' rules can be a subclass of
         'fold'.
         """
         return True
 
     def firstdate(self):
         """Returns true if the rule should preserve the date of the first
         change.
 
         This exists mainly so that 'rollup' rules can be a subclass of
         'fold'.
         """
         return False
 
     def finishfold(self, ui, repo, ctx, oldctx, newnode, internalchanges):
         parent = ctx.parents()[0].node()
         repo.ui.pushbuffer()
         hg.update(repo, parent)
         repo.ui.popbuffer()
         ### prepare new commit data
         commitopts = {}
         commitopts['user'] = ctx.user()
         # commit message
         if not self.mergedescs():
             newmessage = ctx.description()
         else:
             newmessage = '\n***\n'.join(
                 [ctx.description()] +
                 [repo[r].description() for r in internalchanges] +
                 [oldctx.description()]) + '\n'
         commitopts['message'] = newmessage
         # date
         if self.firstdate():
             commitopts['date'] = ctx.date()
         else:
             commitopts['date'] = max(ctx.date(), oldctx.date())
         extra = ctx.extra().copy()
         # histedit_source
         # note: ctx is likely a temporary commit but that the best we can do
         #       here. This is sufficient to solve issue3681 anyway.
         extra['histedit_source'] = '%s,%s' % (ctx.hex(), oldctx.hex())
         commitopts['extra'] = extra
         phasemin = max(ctx.phase(), oldctx.phase())
         overrides = {('phases', 'new-commit'): phasemin}
         with repo.ui.configoverride(overrides, 'histedit'):
             n = collapse(repo, ctx, repo[newnode], commitopts,
                          skipprompt=self.skipprompt())
         if n is None:
             return ctx, []
         repo.ui.pushbuffer()
         hg.update(repo, n)
         repo.ui.popbuffer()
         replacements = [(oldctx.node(), (newnode,)),
                         (ctx.node(), (n,)),
                         (newnode, (n,)),
                        ]
         for ich in internalchanges:
             replacements.append((ich, (n,)))
         return repo[n], replacements
 
 @action(['base', 'b'],
         _('checkout changeset and apply further changesets from there'))
 class base(histeditaction):
 
     def run(self):
         if self.repo['.'].node() != self.node:
             mergemod.update(self.repo, self.node, False, True)
             #                                     branchmerge, force)
         return self.continueclean()
 
     def continuedirty(self):
         abortdirty()
 
     def continueclean(self):
         basectx = self.repo['.']
         return basectx, []
 
     def _verifynodeconstraints(self, prev, expected, seen):
         # base can only be use with a node not in the edited set
         if self.node in expected:
             msg = _('%s "%s" changeset was an edited list candidate')
             raise error.ParseError(
                 msg % (self.verb, node.short(self.node)),
                 hint=_('base must only use unlisted changesets'))
 
 @action(['_multifold'],
         _(
     """fold subclass used for when multiple folds happen in a row
 
     We only want to fire the editor for the folded message once when
     (say) four changes are folded down into a single change. This is
     similar to rollup, but we should preserve both messages so that
     when the last fold operation runs we can show the user all the
     commit messages in their editor.
     """),
         internal=True)
 class _multifold(fold):
     def skipprompt(self):
         return True
 
 @action(["roll", "r"],
         _("like fold, but discard this commit's description and date"))
 class rollup(fold):
     def mergedescs(self):
         return False
 
     def skipprompt(self):
         return True
 
     def firstdate(self):
         return True
 
 @action(["drop", "d"],
         _('remove commit from history'))
 class drop(histeditaction):
     def run(self):
         parentctx = self.repo[self.state.parentctxnode]
         return parentctx, [(self.node, tuple())]
 
 @action(["mess", "m"],
         _('edit commit message without changing commit content'),
         priority=True)
 class message(histeditaction):
     def commiteditor(self):
         return cmdutil.getcommiteditor(edit=True, editform='histedit.mess')
 
 def findoutgoing(ui, repo, remote=None, force=False, opts=None):
     """utility function to find the first outgoing changeset
 
     Used by initialization code"""
     if opts is None:
         opts = {}
     dest = ui.expandpath(remote or 'default-push', remote or 'default')
     dest, revs = hg.parseurl(dest, None)[:2]
     ui.status(_('comparing with %s\n') % util.hidepassword(dest))
 
     revs, checkout = hg.addbranchrevs(repo, repo, revs, None)
     other = hg.peer(repo, opts, dest)
 
     if revs:
         revs = [repo.lookup(rev) for rev in revs]
 
     outgoing = discovery.findcommonoutgoing(repo, other, revs, force=force)
     if not outgoing.missing:
         raise error.Abort(_('no outgoing ancestors'))
     roots = list(repo.revs("roots(%ln)", outgoing.missing))
     if 1 < len(roots):
         msg = _('there are ambiguous outgoing revisions')
         hint = _("see 'hg help histedit' for more detail")
         raise error.Abort(msg, hint=hint)
     return repo.lookup(roots[0])
 
 @command('histedit',
     [('', 'commands', '',
       _('read history edits from the specified file'), _('FILE')),
      ('c', 'continue', False, _('continue an edit already in progress')),
      ('', 'edit-plan', False, _('edit remaining actions list')),
      ('k', 'keep', False,
       _("don't strip old nodes after edit is complete")),
      ('', 'abort', False, _('abort an edit in progress')),
      ('o', 'outgoing', False, _('changesets not found in destination')),
      ('f', 'force', False,
       _('force outgoing even for unrelated repositories')),
      ('r', 'rev', [], _('first revision to be edited'), _('REV'))] +
     cmdutil.formatteropts,
      _("[OPTIONS] ([ANCESTOR] | --outgoing [URL])"))
 def histedit(ui, repo, *freeargs, **opts):
     """interactively edit changeset history
 
     This command lets you edit a linear series of changesets (up to
     and including the working directory, which should be clean).
     You can:
 
     - `pick` to [re]order a changeset
 
     - `drop` to omit changeset
 
     - `mess` to reword the changeset commit message
 
     - `fold` to combine it with the preceding changeset (using the later date)
 
     - `roll` like fold, but discarding this commit's description and date
 
     - `edit` to edit this changeset (preserving date)
 
     - `base` to checkout changeset and apply further changesets from there
 
     There are a number of ways to select the root changeset:
 
     - Specify ANCESTOR directly
 
     - Use --outgoing -- it will be the first linear changeset not
       included in destination. (See :hg:`help config.paths.default-push`)
 
     - Otherwise, the value from the "histedit.defaultrev" config option
       is used as a revset to select the base revision when ANCESTOR is not
       specified. The first revision returned by the revset is used. By
       default, this selects the editable history that is unique to the
       ancestry of the working directory.
 
     .. container:: verbose
 
        If you use --outgoing, this command will abort if there are ambiguous
        outgoing revisions. For example, if there are multiple branches
        containing outgoing revisions.
 
        Use "min(outgoing() and ::.)" or similar revset specification
        instead of --outgoing to specify edit target revision exactly in
        such ambiguous situation. See :hg:`help revsets` for detail about
        selecting revisions.
 
     .. container:: verbose
 
        Examples:
 
          - A number of changes have been made.
            Revision 3 is no longer needed.
 
            Start history editing from revision 3::
 
              hg histedit -r 3
 
            An editor opens, containing the list of revisions,
            with specific actions specified::
 
              pick 5339bf82f0ca 3 Zworgle the foobar
              pick 8ef592ce7cc4 4 Bedazzle the zerlog
              pick 0a9639fcda9d 5 Morgify the cromulancy
 
            Additional information about the possible actions
            to take appears below the list of revisions.
 
            To remove revision 3 from the history,
            its action (at the beginning of the relevant line)
            is changed to 'drop'::
 
              drop 5339bf82f0ca 3 Zworgle the foobar
              pick 8ef592ce7cc4 4 Bedazzle the zerlog
              pick 0a9639fcda9d 5 Morgify the cromulancy
 
          - A number of changes have been made.
            Revision 2 and 4 need to be swapped.
 
            Start history editing from revision 2::
 
              hg histedit -r 2
 
            An editor opens, containing the list of revisions,
            with specific actions specified::
 
              pick 252a1af424ad 2 Blorb a morgwazzle
              pick 5339bf82f0ca 3 Zworgle the foobar
              pick 8ef592ce7cc4 4 Bedazzle the zerlog
 
            To swap revision 2 and 4, its lines are swapped
            in the editor::
 
              pick 8ef592ce7cc4 4 Bedazzle the zerlog
              pick 5339bf82f0ca 3 Zworgle the foobar
              pick 252a1af424ad 2 Blorb a morgwazzle
 
     Returns 0 on success, 1 if user intervention is required (not only
     for intentional "edit" command, but also for resolving unexpected
     conflicts).
     """
     state = histeditstate(repo)
     try:
         state.wlock = repo.wlock()
         state.lock = repo.lock()
         _histedit(ui, repo, state, *freeargs, **opts)
     finally:
         release(state.lock, state.wlock)
 
 goalcontinue = 'continue'
 goalabort = 'abort'
 goaleditplan = 'edit-plan'
 goalnew = 'new'
 
 def _getgoal(opts):
     if opts.get('continue'):
         return goalcontinue
     if opts.get('abort'):
         return goalabort
     if opts.get('edit_plan'):
         return goaleditplan
     return goalnew
 
 def _readfile(ui, path):
     if path == '-':
         with ui.timeblockedsection('histedit'):
             return ui.fin.read()
     else:
         with open(path, 'rb') as f:
             return f.read()
 
 def _validateargs(ui, repo, state, freeargs, opts, goal, rules, revs):
     # TODO only abort if we try to histedit mq patches, not just
     # blanket if mq patches are applied somewhere
     mq = getattr(repo, 'mq', None)
     if mq and mq.applied:
         raise error.Abort(_('source has mq patches applied'))
 
     # basic argument incompatibility processing
     outg = opts.get('outgoing')
     editplan = opts.get('edit_plan')
     abort = opts.get('abort')
     force = opts.get('force')
     if force and not outg:
         raise error.Abort(_('--force only allowed with --outgoing'))
     if goal == 'continue':
         if any((outg, abort, revs, freeargs, rules, editplan)):
             raise error.Abort(_('no arguments allowed with --continue'))
     elif goal == 'abort':
         if any((outg, revs, freeargs, rules, editplan)):
             raise error.Abort(_('no arguments allowed with --abort'))
     elif goal == 'edit-plan':
         if any((outg, revs, freeargs)):
             raise error.Abort(_('only --commands argument allowed with '
                                '--edit-plan'))
     else:
         if os.path.exists(os.path.join(repo.path, 'histedit-state')):
             raise error.Abort(_('history edit already in progress, try '
                                '--continue or --abort'))
         if outg:
             if revs:
                 raise error.Abort(_('no revisions allowed with --outgoing'))
             if len(freeargs) > 1:
                 raise error.Abort(
                     _('only one repo argument allowed with --outgoing'))
         else:
             revs.extend(freeargs)
             if len(revs) == 0:
                 defaultrev = destutil.desthistedit(ui, repo)
                 if defaultrev is not None:
                     revs.append(defaultrev)
 
             if len(revs) != 1:
                 raise error.Abort(
                     _('histedit requires exactly one ancestor revision'))
 
 def _histedit(ui, repo, state, *freeargs, **opts):
     opts = pycompat.byteskwargs(opts)
     fm = ui.formatter('histedit', opts)
     fm.startitem()
     goal = _getgoal(opts)
     revs = opts.get('rev', [])
     rules = opts.get('commands', '')
     state.keep = opts.get('keep', False)
 
     _validateargs(ui, repo, state, freeargs, opts, goal, rules, revs)
 
     # rebuild state
     if goal == goalcontinue:
         state.read()
         state = bootstrapcontinue(ui, state, opts)
     elif goal == goaleditplan:
         _edithisteditplan(ui, repo, state, rules)
         return
     elif goal == goalabort:
         _aborthistedit(ui, repo, state)
         return
     else:
         # goal == goalnew
         _newhistedit(ui, repo, state, revs, freeargs, opts)
 
     _continuehistedit(ui, repo, state)
     _finishhistedit(ui, repo, state, fm)
     fm.end()
 
 def _continuehistedit(ui, repo, state):
     """This function runs after either:
     - bootstrapcontinue (if the goal is 'continue')
     - _newhistedit (if the goal is 'new')
     """
     # preprocess rules so that we can hide inner folds from the user
     # and only show one editor
     actions = state.actions[:]
     for idx, (action, nextact) in enumerate(
             zip(actions, actions[1:] + [None])):
         if action.verb == 'fold' and nextact and nextact.verb == 'fold':
             state.actions[idx].__class__ = _multifold
 
     # Force an initial state file write, so the user can run --abort/continue
     # even if there's an exception before the first transaction serialize.
     state.write()
 
     total = len(state.actions)
     pos = 0
     tr = None
     # Don't use singletransaction by default since it rolls the entire
     # transaction back if an unexpected exception happens (like a
     # pretxncommit hook throws, or the user aborts the commit msg editor).
     if ui.configbool("histedit", "singletransaction"):
         # Don't use a 'with' for the transaction, since actions may close
         # and reopen a transaction. For example, if the action executes an
         # external process it may choose to commit the transaction first.
         tr = repo.transaction('histedit')
     with util.acceptintervention(tr):
         while state.actions:
             state.write(tr=tr)
             actobj = state.actions[0]
             pos += 1
             ui.progress(_("editing"), pos, actobj.torule(),
                         _('changes'), total)
             ui.debug('histedit: processing %s %s\n' % (actobj.verb,\
                                                        actobj.torule()))
             parentctx, replacement_ = actobj.run()
             state.parentctxnode = parentctx.node()
             state.replacements.extend(replacement_)
             state.actions.pop(0)
 
     state.write()
     ui.progress(_("editing"), None)
 
 def _finishhistedit(ui, repo, state, fm):
     """This action runs when histedit is finishing its session"""
     repo.ui.pushbuffer()
     hg.update(repo, state.parentctxnode, quietempty=True)
     repo.ui.popbuffer()
 
     mapping, tmpnodes, created, ntm = processreplacement(state)
     if mapping:
         for prec, succs in mapping.iteritems():
             if not succs:
                 ui.debug('histedit: %s is dropped\n' % node.short(prec))
             else:
                 ui.debug('histedit: %s is replaced by %s\n' % (
                     node.short(prec), node.short(succs[0])))
                 if len(succs) > 1:
                     m = 'histedit:                            %s'
                     for n in succs[1:]:
                         ui.debug(m % node.short(n))
 
     if not state.keep:
         if mapping:
             movetopmostbookmarks(repo, state.topmost, ntm)
             # TODO update mq state
     else:
         mapping = {}
 
     for n in tmpnodes:
         mapping[n] = ()
 
     # remove entries about unknown nodes
     nodemap = repo.unfiltered().changelog.nodemap
     mapping = {k: v for k, v in mapping.items()
                if k in nodemap and all(n in nodemap for n in v)}
     scmutil.cleanupnodes(repo, mapping, 'histedit')
     hf = fm.hexfunc
     fl = fm.formatlist
     fd = fm.formatdict
     nodechanges = fd({hf(oldn): fl([hf(n) for n in newn], name='node')
                       for oldn, newn in mapping.iteritems()},
                      key="oldnode", value="newnodes")
     fm.data(nodechanges=nodechanges)
 
     state.clear()
     if os.path.exists(repo.sjoin('undo')):
         os.unlink(repo.sjoin('undo'))
     if repo.vfs.exists('histedit-last-edit.txt'):
         repo.vfs.unlink('histedit-last-edit.txt')
 
 def _aborthistedit(ui, repo, state):
     try:
         state.read()
         __, leafs, tmpnodes, __ = processreplacement(state)
         ui.debug('restore wc to old parent %s\n'
                 % node.short(state.topmost))
 
         # Recover our old commits if necessary
         if not state.topmost in repo and state.backupfile:
             backupfile = repo.vfs.join(state.backupfile)
             f = hg.openpath(ui, backupfile)
             gen = exchange.readbundle(ui, f, backupfile)
             with repo.transaction('histedit.abort') as tr:
                 bundle2.applybundle(repo, gen, tr, source='histedit',
                                     url='bundle:' + backupfile)
 
             os.remove(backupfile)
 
         # check whether we should update away
         if repo.unfiltered().revs('parents() and (%n  or %ln::)',
                                 state.parentctxnode, leafs | tmpnodes):
             hg.clean(repo, state.topmost, show_stats=True, quietempty=True)
         cleanupnode(ui, repo, tmpnodes)
         cleanupnode(ui, repo, leafs)
     except Exception:
         if state.inprogress():
             ui.warn(_('warning: encountered an exception during histedit '
                 '--abort; the repository may not have been completely '
                 'cleaned up\n'))
         raise
     finally:
             state.clear()
 
 def _edithisteditplan(ui, repo, state, rules):
     state.read()
     if not rules:
         comment = geteditcomment(ui,
                                  node.short(state.parentctxnode),
                                  node.short(state.topmost))
         rules = ruleeditor(repo, ui, state.actions, comment)
     else:
         rules = _readfile(ui, rules)
     actions = parserules(rules, state)
     ctxs = [repo[act.node] \
             for act in state.actions if act.node]
     warnverifyactions(ui, repo, actions, state, ctxs)
     state.actions = actions
     state.write()
 
 def _newhistedit(ui, repo, state, revs, freeargs, opts):
     outg = opts.get('outgoing')
     rules = opts.get('commands', '')
     force = opts.get('force')
 
     cmdutil.checkunfinished(repo)
     cmdutil.bailifchanged(repo)
 
     topmost, empty = repo.dirstate.parents()
     if outg:
         if freeargs:
             remote = freeargs[0]
         else:
             remote = None
         root = findoutgoing(ui, repo, remote, force, opts)
     else:
         rr = list(repo.set('roots(%ld)', scmutil.revrange(repo, revs)))
         if len(rr) != 1:
             raise error.Abort(_('The specified revisions must have '
                 'exactly one common root'))
         root = rr[0].node()
 
     revs = between(repo, root, topmost, state.keep)
     if not revs:
         raise error.Abort(_('%s is not an ancestor of working directory') %
                          node.short(root))
 
     ctxs = [repo[r] for r in revs]
     if not rules:
         comment = geteditcomment(ui, node.short(root), node.short(topmost))
         actions = [pick(state, r) for r in revs]
         rules = ruleeditor(repo, ui, actions, comment)
     else:
         rules = _readfile(ui, rules)
     actions = parserules(rules, state)
     warnverifyactions(ui, repo, actions, state, ctxs)
 
     parentctxnode = repo[root].parents()[0].node()
 
     state.parentctxnode = parentctxnode
     state.actions = actions
     state.topmost = topmost
     state.replacements = []
 
     ui.log("histedit", "%d actions to histedit", len(actions),
            histedit_num_actions=len(actions))
 
     # Create a backup so we can always abort completely.
     backupfile = None
     if not obsolete.isenabled(repo, obsolete.createmarkersopt):
         backupfile = repair._bundle(repo, [parentctxnode], [topmost], root,
                                     'histedit')
     state.backupfile = backupfile
 
 def _getsummary(ctx):
     # a common pattern is to extract the summary but default to the empty
     # string
     summary = ctx.description() or ''
     if summary:
         summary = summary.splitlines()[0]
     return summary
 
 def bootstrapcontinue(ui, state, opts):
     repo = state.repo
 
     ms = mergemod.mergestate.read(repo)
     mergeutil.checkunresolved(ms)
 
     if state.actions:
         actobj = state.actions.pop(0)
 
         if _isdirtywc(repo):
             actobj.continuedirty()
             if _isdirtywc(repo):
                 abortdirty()
 
         parentctx, replacements = actobj.continueclean()
 
         state.parentctxnode = parentctx.node()
         state.replacements.extend(replacements)
 
     return state
 
 def between(repo, old, new, keep):
     """select and validate the set of revision to edit
 
     When keep is false, the specified set can't have children."""
     ctxs = list(repo.set('%n::%n', old, new))
     if ctxs and not keep:
         if (not obsolete.isenabled(repo, obsolete.allowunstableopt) and
             repo.revs('(%ld::) - (%ld)', ctxs, ctxs)):
             raise error.Abort(_('can only histedit a changeset together '
                                 'with all its descendants'))
         if repo.revs('(%ld) and merge()', ctxs):
             raise error.Abort(_('cannot edit history that contains merges'))
         root = ctxs[0] # list is already sorted by repo.set
         if not root.mutable():
             raise error.Abort(_('cannot edit public changeset: %s') % root,
                              hint=_("see 'hg help phases' for details"))
     return [c.node() for c in ctxs]
 
 def ruleeditor(repo, ui, actions, editcomment=""):
     """open an editor to edit rules
 
     rules are in the format [ [act, ctx], ...] like in state.rules
     """
     if repo.ui.configbool("experimental", "histedit.autoverb"):
         newact = util.sortdict()
         for act in actions:
             ctx = repo[act.node]
             summary = _getsummary(ctx)
             fword = summary.split(' ', 1)[0].lower()
             added = False
 
             # if it doesn't end with the special character '!' just skip this
             if fword.endswith('!'):
                 fword = fword[:-1]
                 if fword in primaryactions | secondaryactions | tertiaryactions:
                     act.verb = fword
                     # get the target summary
                     tsum = summary[len(fword) + 1:].lstrip()
                     # safe but slow: reverse iterate over the actions so we
                     # don't clash on two commits having the same summary
                     for na, l in reversed(list(newact.iteritems())):
                         actx = repo[na.node]
                         asum = _getsummary(actx)
                         if asum == tsum:
                             added = True
                             l.append(act)
                             break
 
             if not added:
                 newact[act] = []
 
         # copy over and flatten the new list
         actions = []
         for na, l in newact.iteritems():
             actions.append(na)
             actions += l
 
     rules = '\n'.join([act.torule() for act in actions])
     rules += '\n\n'
     rules += editcomment
     rules = ui.edit(rules, ui.username(), {'prefix': 'histedit'},
                     repopath=repo.path, action='histedit')
 
     # Save edit rules in .hg/histedit-last-edit.txt in case
     # the user needs to ask for help after something
     # surprising happens.
     with repo.vfs('histedit-last-edit.txt', 'wb') as f:
         f.write(rules)
 
     return rules
 
 def parserules(rules, state):
     """Read the histedit rules string and return list of action objects """
     rules = [l for l in (r.strip() for r in rules.splitlines())
                 if l and not l.startswith('#')]
     actions = []
     for r in rules:
         if ' ' not in r:
             raise error.ParseError(_('malformed line "%s"') % r)
         verb, rest = r.split(' ', 1)
 
         if verb not in actiontable:
             raise error.ParseError(_('unknown action "%s"') % verb)
 
         action = actiontable[verb].fromrule(state, rest)
         actions.append(action)
     return actions
 
 def warnverifyactions(ui, repo, actions, state, ctxs):
     try:
         verifyactions(actions, state, ctxs)
     except error.ParseError:
         if repo.vfs.exists('histedit-last-edit.txt'):
             ui.warn(_('warning: histedit rules saved '
                       'to: .hg/histedit-last-edit.txt\n'))
         raise
 
 def verifyactions(actions, state, ctxs):
     """Verify that there exists exactly one action per given changeset and
     other constraints.
 
     Will abort if there are to many or too few rules, a malformed rule,
     or a rule on a changeset outside of the user-given range.
     """
     expected = set(c.node() for c in ctxs)
     seen = set()
     prev = None
 
     if actions and actions[0].verb in ['roll', 'fold']:
         raise error.ParseError(_('first changeset cannot use verb "%s"') %
                                actions[0].verb)
 
     for action in actions:
         action.verify(prev, expected, seen)
         prev = action
         if action.node is not None:
             seen.add(action.node)
     missing = sorted(expected - seen)  # sort to stabilize output
 
     if state.repo.ui.configbool('histedit', 'dropmissing'):
         if len(actions) == 0:
             raise error.ParseError(_('no rules provided'),
                     hint=_('use strip extension to remove commits'))
 
         drops = [drop(state, n) for n in missing]
         # put the in the beginning so they execute immediately and
         # don't show in the edit-plan in the future
         actions[:0] = drops
     elif missing:
         raise error.ParseError(_('missing rules for changeset %s') %
                 node.short(missing[0]),
                 hint=_('use "drop %s" to discard, see also: '
                        "'hg help -e histedit.config'")
                        % node.short(missing[0]))
 
 def adjustreplacementsfrommarkers(repo, oldreplacements):
     """Adjust replacements from obsolescence markers
 
     Replacements structure is originally generated based on
     histedit's state and does not account for changes that are
     not recorded there. This function fixes that by adding
     data read from obsolescence markers"""
     if not obsolete.isenabled(repo, obsolete.createmarkersopt):
         return oldreplacements
 
     unfi = repo.unfiltered()
     nm = unfi.changelog.nodemap
     obsstore = repo.obsstore
     newreplacements = list(oldreplacements)
     oldsuccs = [r[1] for r in oldreplacements]
     # successors that have already been added to succstocheck once
     seensuccs = set().union(*oldsuccs) # create a set from an iterable of tuples
     succstocheck = list(seensuccs)
     while succstocheck:
         n = succstocheck.pop()
         missing = nm.get(n) is None
         markers = obsstore.successors.get(n, ())
         if missing and not markers:
             # dead end, mark it as such
             newreplacements.append((n, ()))
         for marker in markers:
             nsuccs = marker[1]
             newreplacements.append((n, nsuccs))
             for nsucc in nsuccs:
                 if nsucc not in seensuccs:
                     seensuccs.add(nsucc)
                     succstocheck.append(nsucc)
 
     return newreplacements
 
 def processreplacement(state):
     """process the list of replacements to return
 
     1) the final mapping between original and created nodes
     2) the list of temporary node created by histedit
     3) the list of new commit created by histedit"""
     replacements = adjustreplacementsfrommarkers(state.repo, state.replacements)
     allsuccs = set()
     replaced = set()
     fullmapping = {}
     # initialize basic set
     # fullmapping records all operations recorded in replacement
     for rep in replacements:
         allsuccs.update(rep[1])
         replaced.add(rep[0])
         fullmapping.setdefault(rep[0], set()).update(rep[1])
     new = allsuccs - replaced
     tmpnodes = allsuccs & replaced
     # Reduce content fullmapping into direct relation between original nodes
     # and final node created during history edition
     # Dropped changeset are replaced by an empty list
     toproceed = set(fullmapping)
     final = {}
     while toproceed:
         for x in list(toproceed):
             succs = fullmapping[x]
             for s in list(succs):
                 if s in toproceed:
                     # non final node with unknown closure
                     # We can't process this now
                     break
                 elif s in final:
                     # non final node, replace with closure
                     succs.remove(s)
                     succs.update(final[s])
             else:
                 final[x] = succs
                 toproceed.remove(x)
     # remove tmpnodes from final mapping
     for n in tmpnodes:
         del final[n]
     # we expect all changes involved in final to exist in the repo
     # turn `final` into list (topologically sorted)
     nm = state.repo.changelog.nodemap
     for prec, succs in final.items():
         final[prec] = sorted(succs, key=nm.get)
 
     # computed topmost element (necessary for bookmark)
     if new:
         newtopmost = sorted(new, key=state.repo.changelog.rev)[-1]
     elif not final:
         # Nothing rewritten at all. we won't need `newtopmost`
         # It is the same as `oldtopmost` and `processreplacement` know it
         newtopmost = None
     else:
         # every body died. The newtopmost is the parent of the root.
         r = state.repo.changelog.rev
         newtopmost = state.repo[sorted(final, key=r)[0]].p1().node()
 
     return final, tmpnodes, new, newtopmost
 
 def movetopmostbookmarks(repo, oldtopmost, newtopmost):
     """Move bookmark from oldtopmost to newly created topmost
 
     This is arguably a feature and we may only want that for the active
     bookmark. But the behavior is kept compatible with the old version for now.
     """
     if not oldtopmost or not newtopmost:
         return
     oldbmarks = repo.nodebookmarks(oldtopmost)
     if oldbmarks:
         with repo.lock(), repo.transaction('histedit') as tr:
             marks = repo._bookmarks
             changes = []
             for name in oldbmarks:
                 changes.append((name, newtopmost))
             marks.applychanges(repo, tr, changes)
 
 def cleanupnode(ui, repo, nodes):
     """strip a group of nodes from the repository
 
     The set of node to strip may contains unknown nodes."""
     with repo.lock():
         # do not let filtering get in the way of the cleanse
         # we should probably get rid of obsolescence marker created during the
         # histedit, but we currently do not have such information.
         repo = repo.unfiltered()
         # Find all nodes that need to be stripped
         # (we use %lr instead of %ln to silently ignore unknown items)
         nm = repo.changelog.nodemap
         nodes = sorted(n for n in nodes if n in nm)
         roots = [c.node() for c in repo.set("roots(%ln)", nodes)]
         if roots:
             repair.strip(ui, repo, roots)
 
 def stripwrapper(orig, ui, repo, nodelist, *args, **kwargs):
     if isinstance(nodelist, str):
         nodelist = [nodelist]
     if os.path.exists(os.path.join(repo.path, 'histedit-state')):
         state = histeditstate(repo)
         state.read()
         histedit_nodes = {action.node for action
                           in state.actions if action.node}
         common_nodes = histedit_nodes & set(nodelist)
         if common_nodes:
             raise error.Abort(_("histedit in progress, can't strip %s")
                              % ', '.join(node.short(x) for x in common_nodes))
     return orig(ui, repo, nodelist, *args, **kwargs)
 
 extensions.wrapfunction(repair, 'strip', stripwrapper)
 
 def summaryhook(ui, repo):
     if not os.path.exists(repo.vfs.join('histedit-state')):
         return
     state = histeditstate(repo)
     state.read()
     if state.actions:
         # i18n: column positioning for "hg summary"
         ui.write(_('hist:   %s (histedit --continue)\n') %
                  (ui.label(_('%d remaining'), 'histedit.remaining') %
                   len(state.actions)))
 
 def extsetup(ui):
     cmdutil.summaryhooks.add('histedit', summaryhook)
     cmdutil.unfinishedstates.append(
         ['histedit-state', False, True, _('histedit in progress'),
          _("use 'hg histedit --continue' or 'hg histedit --abort'")])
     cmdutil.afterresolvedstates.append(
         ['histedit-state', _('hg histedit --continue')])

mercurial/node.py

@@ -1,35 +1,42 b''
 # node.py - basic nodeid manipulation for mercurial
 #
 # Copyright 2005, 2006 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import binascii
 
 # This ugly style has a noticeable effect in manifest parsing
 hex = binascii.hexlify
-bin = binascii.unhexlify
+# Adapt to Python 3 API changes. If this ends up showing up in
+# profiles, we can use this version only on Python 3, and forward
+# binascii.unhexlify like we used to on Python 2.
+def bin(s):
+    try:
+        return binascii.unhexlify(s)
+    except binascii.Error as e:
+        raise TypeError(e)
 
 nullrev = -1
 nullid = b"\0" * 20
 nullhex = hex(nullid)
 
 # Phony node value to stand-in for new files in some uses of
 # manifests.
 newnodeid = '!' * 20
 addednodeid = ('0' * 15) + 'added'
 modifiednodeid = ('0' * 12) + 'modified'
 
 wdirnodes = {newnodeid, addednodeid, modifiednodeid}
 
 # pseudo identifiers for working directory
 # (they are experimental, so don't add too many dependencies on them)
 wdirrev = 0x7fffffff
 wdirid = b"\xff" * 20
 wdirhex = hex(wdirid)
 
 def short(node):
     return hex(node[:6])

mercurial/revlog.py

@@ -1,2501 +1,2500 b''
 # revlog.py - storage back-end for mercurial
 #
 # Copyright 2005-2007 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 """Storage back-end for Mercurial.
 
 This provides efficient delta storage with O(1) retrieve and append
 and O(changes) merge between branches.
 """
 
 from __future__ import absolute_import
 
-import binascii
 import collections
 import contextlib
 import errno
 import hashlib
 import heapq
 import os
 import struct
 import zlib
 
 # import stuff from node for others to import from revlog
 from .node import (
     bin,
     hex,
     nullid,
     nullrev,
     wdirhex,
     wdirid,
     wdirrev,
 )
 from .i18n import _
 from .thirdparty import (
     attr,
 )
 from . import (
     ancestor,
     error,
     mdiff,
     policy,
     pycompat,
     templatefilters,
     util,
 )
 
 parsers = policy.importmod(r'parsers')
 
 # Aliased for performance.
 _zlibdecompress = zlib.decompress
 
 # revlog header flags
 REVLOGV0 = 0
 REVLOGV1 = 1
 # Dummy value until file format is finalized.
 # Reminder: change the bounds check in revlog.__init__ when this is changed.
 REVLOGV2 = 0xDEAD
 FLAG_INLINE_DATA = (1 << 16)
 FLAG_GENERALDELTA = (1 << 17)
 REVLOG_DEFAULT_FLAGS = FLAG_INLINE_DATA
 REVLOG_DEFAULT_FORMAT = REVLOGV1
 REVLOG_DEFAULT_VERSION = REVLOG_DEFAULT_FORMAT | REVLOG_DEFAULT_FLAGS
 REVLOGV1_FLAGS = FLAG_INLINE_DATA | FLAG_GENERALDELTA
 REVLOGV2_FLAGS = REVLOGV1_FLAGS
 
 # revlog index flags
 REVIDX_ISCENSORED = (1 << 15) # revision has censor metadata, must be verified
 REVIDX_ELLIPSIS = (1 << 14) # revision hash does not match data (narrowhg)
 REVIDX_EXTSTORED = (1 << 13) # revision data is stored externally
 REVIDX_DEFAULT_FLAGS = 0
 # stable order in which flags need to be processed and their processors applied
 REVIDX_FLAGS_ORDER = [
     REVIDX_ISCENSORED,
     REVIDX_ELLIPSIS,
     REVIDX_EXTSTORED,
 ]
 REVIDX_KNOWN_FLAGS = util.bitsfrom(REVIDX_FLAGS_ORDER)
 
 # max size of revlog with inline data
 _maxinline = 131072
 _chunksize = 1048576
 
 RevlogError = error.RevlogError
 LookupError = error.LookupError
 CensoredNodeError = error.CensoredNodeError
 ProgrammingError = error.ProgrammingError
 
 # Store flag processors (cf. 'addflagprocessor()' to register)
 _flagprocessors = {
     REVIDX_ISCENSORED: None,
 }
 
 def addflagprocessor(flag, processor):
     """Register a flag processor on a revision data flag.
 
     Invariant:
     - Flags need to be defined in REVIDX_KNOWN_FLAGS and REVIDX_FLAGS_ORDER.
     - Only one flag processor can be registered on a specific flag.
     - flagprocessors must be 3-tuples of functions (read, write, raw) with the
       following signatures:
           - (read)  f(self, rawtext) -> text, bool
           - (write) f(self, text) -> rawtext, bool
           - (raw)   f(self, rawtext) -> bool
       "text" is presented to the user. "rawtext" is stored in revlog data, not
       directly visible to the user.
       The boolean returned by these transforms is used to determine whether
       the returned text can be used for hash integrity checking. For example,
       if "write" returns False, then "text" is used to generate hash. If
       "write" returns True, that basically means "rawtext" returned by "write"
       should be used to generate hash. Usually, "write" and "read" return
       different booleans. And "raw" returns a same boolean as "write".
 
       Note: The 'raw' transform is used for changegroup generation and in some
       debug commands. In this case the transform only indicates whether the
       contents can be used for hash integrity checks.
     """
     if not flag & REVIDX_KNOWN_FLAGS:
         msg = _("cannot register processor on unknown flag '%#x'.") % (flag)
         raise ProgrammingError(msg)
     if flag not in REVIDX_FLAGS_ORDER:
         msg = _("flag '%#x' undefined in REVIDX_FLAGS_ORDER.") % (flag)
         raise ProgrammingError(msg)
     if flag in _flagprocessors:
         msg = _("cannot register multiple processors on flag '%#x'.") % (flag)
         raise error.Abort(msg)
     _flagprocessors[flag] = processor
 
 def getoffset(q):
     return int(q >> 16)
 
 def gettype(q):
     return int(q & 0xFFFF)
 
 def offset_type(offset, type):
     if (type & ~REVIDX_KNOWN_FLAGS) != 0:
         raise ValueError('unknown revlog index flags')
     return int(int(offset) << 16 | type)
 
 _nullhash = hashlib.sha1(nullid)
 
 def hash(text, p1, p2):
     """generate a hash from the given text and its parent hashes
 
     This hash combines both the current file contents and its history
     in a manner that makes it easy to distinguish nodes with the same
     content in the revision graph.
     """
     # As of now, if one of the parent node is null, p2 is null
     if p2 == nullid:
         # deep copy of a hash is faster than creating one
         s = _nullhash.copy()
         s.update(p1)
     else:
         # none of the parent nodes are nullid
         if p1 < p2:
             a = p1
             b = p2
         else:
             a = p2
             b = p1
         s = hashlib.sha1(a)
         s.update(b)
     s.update(text)
     return s.digest()
 
 def _trimchunk(revlog, revs, startidx, endidx=None):
     """returns revs[startidx:endidx] without empty trailing revs
     """
     length = revlog.length
 
     if endidx is None:
         endidx = len(revs)
 
     # Trim empty revs at the end, but never the very first revision of a chain
     while endidx > 1 and endidx > startidx and length(revs[endidx - 1]) == 0:
         endidx -= 1
 
     return revs[startidx:endidx]
 
 def _slicechunk(revlog, revs):
     """slice revs to reduce the amount of unrelated data to be read from disk.
 
     ``revs`` is sliced into groups that should be read in one time.
     Assume that revs are sorted.
     """
     start = revlog.start
     length = revlog.length
 
     if len(revs) <= 1:
         yield revs
         return
 
     startbyte = start(revs[0])
     endbyte = start(revs[-1]) + length(revs[-1])
     readdata = deltachainspan = endbyte - startbyte
 
     chainpayload = sum(length(r) for r in revs)
 
     if deltachainspan:
         density = chainpayload / float(deltachainspan)
     else:
         density = 1.0
 
     # Store the gaps in a heap to have them sorted by decreasing size
     gapsheap = []
     heapq.heapify(gapsheap)
     prevend = None
     for i, rev in enumerate(revs):
         revstart = start(rev)
         revlen = length(rev)
 
         # Skip empty revisions to form larger holes
         if revlen == 0:
             continue
 
         if prevend is not None:
             gapsize = revstart - prevend
             # only consider holes that are large enough
             if gapsize > revlog._srmingapsize:
                 heapq.heappush(gapsheap, (-gapsize, i))
 
         prevend = revstart + revlen
 
     # Collect the indices of the largest holes until the density is acceptable
     indicesheap = []
     heapq.heapify(indicesheap)
     while gapsheap and density < revlog._srdensitythreshold:
         oppgapsize, gapidx = heapq.heappop(gapsheap)
 
         heapq.heappush(indicesheap, gapidx)
 
         # the gap sizes are stored as negatives to be sorted decreasingly
         # by the heap
         readdata -= (-oppgapsize)
         if readdata > 0:
             density = chainpayload / float(readdata)
         else:
             density = 1.0
 
     # Cut the revs at collected indices
     previdx = 0
     while indicesheap:
         idx = heapq.heappop(indicesheap)
 
         chunk = _trimchunk(revlog, revs, previdx, idx)
         if chunk:
             yield chunk
 
         previdx = idx
 
     chunk = _trimchunk(revlog, revs, previdx)
     if chunk:
         yield chunk
 
 @attr.s(slots=True, frozen=True)
 class _deltainfo(object):
     distance = attr.ib()
     deltalen = attr.ib()
     data = attr.ib()
     base = attr.ib()
     chainbase = attr.ib()
     chainlen = attr.ib()
     compresseddeltalen = attr.ib()
 
 class _deltacomputer(object):
     def __init__(self, revlog):
         self.revlog = revlog
 
     def _getcandidaterevs(self, p1, p2, cachedelta):
         """
         Provides revisions that present an interest to be diffed against,
         grouped by level of easiness.
         """
         revlog = self.revlog
         curr = len(revlog)
         prev = curr - 1
         p1r, p2r = revlog.rev(p1), revlog.rev(p2)
 
         # should we try to build a delta?
         if prev != nullrev and revlog.storedeltachains:
             tested = set()
             # This condition is true most of the time when processing
             # changegroup data into a generaldelta repo. The only time it
             # isn't true is if this is the first revision in a delta chain
             # or if ``format.generaldelta=true`` disabled ``lazydeltabase``.
             if cachedelta and revlog._generaldelta and revlog._lazydeltabase:
                 # Assume what we received from the server is a good choice
                 # build delta will reuse the cache
                 yield (cachedelta[0],)
                 tested.add(cachedelta[0])
 
             if revlog._generaldelta:
                 # exclude already lazy tested base if any
                 parents = [p for p in (p1r, p2r)
                            if p != nullrev and p not in tested]
                 if parents and not revlog._aggressivemergedeltas:
                     # Pick whichever parent is closer to us (to minimize the
                     # chance of having to build a fulltext).
                     parents = [max(parents)]
                 tested.update(parents)
                 yield parents
 
             if prev not in tested:
                 # other approach failed try against prev to hopefully save us a
                 # fulltext.
                 yield (prev,)
 
     def buildtext(self, revinfo, fh):
         """Builds a fulltext version of a revision
 
         revinfo: _revisioninfo instance that contains all needed info
         fh:      file handle to either the .i or the .d revlog file,
                  depending on whether it is inlined or not
         """
         btext = revinfo.btext
         if btext[0] is not None:
             return btext[0]
 
         revlog = self.revlog
         cachedelta = revinfo.cachedelta
         flags = revinfo.flags
         node = revinfo.node
 
         baserev = cachedelta[0]
         delta = cachedelta[1]
         # special case deltas which replace entire base; no need to decode
         # base revision. this neatly avoids censored bases, which throw when
         # they're decoded.
         hlen = struct.calcsize(">lll")
         if delta[:hlen] == mdiff.replacediffheader(revlog.rawsize(baserev),
                                                    len(delta) - hlen):
             btext[0] = delta[hlen:]
         else:
             basetext = revlog.revision(baserev, _df=fh, raw=True)
             btext[0] = mdiff.patch(basetext, delta)
 
         try:
             res = revlog._processflags(btext[0], flags, 'read', raw=True)
             btext[0], validatehash = res
             if validatehash:
                 revlog.checkhash(btext[0], node, p1=revinfo.p1, p2=revinfo.p2)
             if flags & REVIDX_ISCENSORED:
                 raise RevlogError(_('node %s is not censored') % node)
         except CensoredNodeError:
             # must pass the censored index flag to add censored revisions
             if not flags & REVIDX_ISCENSORED:
                 raise
         return btext[0]
 
     def _builddeltadiff(self, base, revinfo, fh):
         revlog = self.revlog
         t = self.buildtext(revinfo, fh)
         if revlog.iscensored(base):
             # deltas based on a censored revision must replace the
             # full content in one patch, so delta works everywhere
             header = mdiff.replacediffheader(revlog.rawsize(base), len(t))
             delta = header + t
         else:
             ptext = revlog.revision(base, _df=fh, raw=True)
             delta = mdiff.textdiff(ptext, t)
 
         return delta
 
     def _builddeltainfo(self, revinfo, base, fh):
         # can we use the cached delta?
         if revinfo.cachedelta and revinfo.cachedelta[0] == base:
             delta = revinfo.cachedelta[1]
         else:
             delta = self._builddeltadiff(base, revinfo, fh)
         revlog = self.revlog
         header, data = revlog.compress(delta)
         deltalen = len(header) + len(data)
         chainbase = revlog.chainbase(base)
         offset = revlog.end(len(revlog) - 1)
         dist = deltalen + offset - revlog.start(chainbase)
         if revlog._generaldelta:
             deltabase = base
         else:
             deltabase = chainbase
         chainlen, compresseddeltalen = revlog._chaininfo(base)
         chainlen += 1
         compresseddeltalen += deltalen
         return _deltainfo(dist, deltalen, (header, data), deltabase,
                          chainbase, chainlen, compresseddeltalen)
 
     def finddeltainfo(self, revinfo, fh):
         """Find an acceptable delta against a candidate revision
 
         revinfo: information about the revision (instance of _revisioninfo)
         fh:      file handle to either the .i or the .d revlog file,
                  depending on whether it is inlined or not
 
         Returns the first acceptable candidate revision, as ordered by
         _getcandidaterevs
         """
         cachedelta = revinfo.cachedelta
         p1 = revinfo.p1
         p2 = revinfo.p2
         revlog = self.revlog
 
         deltainfo = None
         for candidaterevs in self._getcandidaterevs(p1, p2, cachedelta):
             nominateddeltas = []
             for candidaterev in candidaterevs:
                 candidatedelta = self._builddeltainfo(revinfo, candidaterev, fh)
                 if revlog._isgooddeltainfo(candidatedelta, revinfo.textlen):
                     nominateddeltas.append(candidatedelta)
             if nominateddeltas:
                 deltainfo = min(nominateddeltas, key=lambda x: x.deltalen)
                 break
 
         return deltainfo
 
 @attr.s(slots=True, frozen=True)
 class _revisioninfo(object):
     """Information about a revision that allows building its fulltext
     node:       expected hash of the revision
     p1, p2:     parent revs of the revision
     btext:      built text cache consisting of a one-element list
     cachedelta: (baserev, uncompressed_delta) or None
     flags:      flags associated to the revision storage
 
     One of btext[0] or cachedelta must be set.
     """
     node = attr.ib()
     p1 = attr.ib()
     p2 = attr.ib()
     btext = attr.ib()
     textlen = attr.ib()
     cachedelta = attr.ib()
     flags = attr.ib()
 
 # index v0:
 #  4 bytes: offset
 #  4 bytes: compressed length
 #  4 bytes: base rev
 #  4 bytes: link rev
 # 20 bytes: parent 1 nodeid
 # 20 bytes: parent 2 nodeid
 # 20 bytes: nodeid
 indexformatv0 = struct.Struct(">4l20s20s20s")
 indexformatv0_pack = indexformatv0.pack
 indexformatv0_unpack = indexformatv0.unpack
 
 class revlogoldio(object):
     def __init__(self):
         self.size = indexformatv0.size
 
     def parseindex(self, data, inline):
         s = self.size
         index = []
         nodemap = {nullid: nullrev}
         n = off = 0
         l = len(data)
         while off + s <= l:
             cur = data[off:off + s]
             off += s
             e = indexformatv0_unpack(cur)
             # transform to revlogv1 format
             e2 = (offset_type(e[0], 0), e[1], -1, e[2], e[3],
                   nodemap.get(e[4], nullrev), nodemap.get(e[5], nullrev), e[6])
             index.append(e2)
             nodemap[e[6]] = n
             n += 1
 
         # add the magic null revision at -1
         index.append((0, 0, 0, -1, -1, -1, -1, nullid))
 
         return index, nodemap, None
 
     def packentry(self, entry, node, version, rev):
         if gettype(entry[0]):
             raise RevlogError(_('index entry flags need revlog version 1'))
         e2 = (getoffset(entry[0]), entry[1], entry[3], entry[4],
               node(entry[5]), node(entry[6]), entry[7])
         return indexformatv0_pack(*e2)
 
 # index ng:
 #  6 bytes: offset
 #  2 bytes: flags
 #  4 bytes: compressed length
 #  4 bytes: uncompressed length
 #  4 bytes: base rev
 #  4 bytes: link rev
 #  4 bytes: parent 1 rev
 #  4 bytes: parent 2 rev
 # 32 bytes: nodeid
 indexformatng = struct.Struct(">Qiiiiii20s12x")
 indexformatng_pack = indexformatng.pack
 versionformat = struct.Struct(">I")
 versionformat_pack = versionformat.pack
 versionformat_unpack = versionformat.unpack
 
 # corresponds to uncompressed length of indexformatng (2 gigs, 4-byte
 # signed integer)
 _maxentrysize = 0x7fffffff
 
 class revlogio(object):
     def __init__(self):
         self.size = indexformatng.size
 
     def parseindex(self, data, inline):
         # call the C implementation to parse the index data
         index, cache = parsers.parse_index2(data, inline)
         return index, getattr(index, 'nodemap', None), cache
 
     def packentry(self, entry, node, version, rev):
         p = indexformatng_pack(*entry)
         if rev == 0:
             p = versionformat_pack(version) + p[4:]
         return p
 
 class revlog(object):
     """
     the underlying revision storage object
 
     A revlog consists of two parts, an index and the revision data.
 
     The index is a file with a fixed record size containing
     information on each revision, including its nodeid (hash), the
     nodeids of its parents, the position and offset of its data within
     the data file, and the revision it's based on. Finally, each entry
     contains a linkrev entry that can serve as a pointer to external
     data.
 
     The revision data itself is a linear collection of data chunks.
     Each chunk represents a revision and is usually represented as a
     delta against the previous chunk. To bound lookup time, runs of
     deltas are limited to about 2 times the length of the original
     version data. This makes retrieval of a version proportional to
     its size, or O(1) relative to the number of revisions.
 
     Both pieces of the revlog are written to in an append-only
     fashion, which means we never need to rewrite a file to insert or
     remove data, and can use some simple techniques to avoid the need
     for locking while reading.
 
     If checkambig, indexfile is opened with checkambig=True at
     writing, to avoid file stat ambiguity.
 
     If mmaplargeindex is True, and an mmapindexthreshold is set, the
     index will be mmapped rather than read if it is larger than the
     configured threshold.
     """
     def __init__(self, opener, indexfile, datafile=None, checkambig=False,
                  mmaplargeindex=False):
         """
         create a revlog object
 
         opener is a function that abstracts the file opening operation
         and can be used to implement COW semantics or the like.
         """
         self.indexfile = indexfile
         self.datafile = datafile or (indexfile[:-2] + ".d")
         self.opener = opener
         #  When True, indexfile is opened with checkambig=True at writing, to
         #  avoid file stat ambiguity.
         self._checkambig = checkambig
         # 3-tuple of (node, rev, text) for a raw revision.
         self._cache = None
         # Maps rev to chain base rev.
         self._chainbasecache = util.lrucachedict(100)
         # 2-tuple of (offset, data) of raw data from the revlog at an offset.
         self._chunkcache = (0, '')
         # How much data to read and cache into the raw revlog data cache.
         self._chunkcachesize = 65536
         self._maxchainlen = None
         self._aggressivemergedeltas = False
         self.index = []
         # Mapping of partial identifiers to full nodes.
         self._pcache = {}
         # Mapping of revision integer to full node.
         self._nodecache = {nullid: nullrev}
         self._nodepos = None
         self._compengine = 'zlib'
         self._maxdeltachainspan = -1
         self._withsparseread = False
         self._srdensitythreshold = 0.25
         self._srmingapsize = 262144
 
         mmapindexthreshold = None
         v = REVLOG_DEFAULT_VERSION
         opts = getattr(opener, 'options', None)
         if opts is not None:
             if 'revlogv2' in opts:
                 # version 2 revlogs always use generaldelta.
                 v = REVLOGV2 | FLAG_GENERALDELTA | FLAG_INLINE_DATA
             elif 'revlogv1' in opts:
                 if 'generaldelta' in opts:
                     v |= FLAG_GENERALDELTA
             else:
                 v = 0
             if 'chunkcachesize' in opts:
                 self._chunkcachesize = opts['chunkcachesize']
             if 'maxchainlen' in opts:
                 self._maxchainlen = opts['maxchainlen']
             if 'aggressivemergedeltas' in opts:
                 self._aggressivemergedeltas = opts['aggressivemergedeltas']
             self._lazydeltabase = bool(opts.get('lazydeltabase', False))
             if 'compengine' in opts:
                 self._compengine = opts['compengine']
             if 'maxdeltachainspan' in opts:
                 self._maxdeltachainspan = opts['maxdeltachainspan']
             if mmaplargeindex and 'mmapindexthreshold' in opts:
                 mmapindexthreshold = opts['mmapindexthreshold']
             self._withsparseread = bool(opts.get('with-sparse-read', False))
             if 'sparse-read-density-threshold' in opts:
                 self._srdensitythreshold = opts['sparse-read-density-threshold']
             if 'sparse-read-min-gap-size' in opts:
                 self._srmingapsize = opts['sparse-read-min-gap-size']
 
         if self._chunkcachesize <= 0:
             raise RevlogError(_('revlog chunk cache size %r is not greater '
                                 'than 0') % self._chunkcachesize)
         elif self._chunkcachesize & (self._chunkcachesize - 1):
             raise RevlogError(_('revlog chunk cache size %r is not a power '
                                 'of 2') % self._chunkcachesize)
 
         indexdata = ''
         self._initempty = True
         try:
             with self._indexfp() as f:
                 if (mmapindexthreshold is not None and
                         self.opener.fstat(f).st_size >= mmapindexthreshold):
                     indexdata = util.buffer(util.mmapread(f))
                 else:
                     indexdata = f.read()
             if len(indexdata) > 0:
                 v = versionformat_unpack(indexdata[:4])[0]
                 self._initempty = False
         except IOError as inst:
             if inst.errno != errno.ENOENT:
                 raise
 
         self.version = v
         self._inline = v & FLAG_INLINE_DATA
         self._generaldelta = v & FLAG_GENERALDELTA
         flags = v & ~0xFFFF
         fmt = v & 0xFFFF
         if fmt == REVLOGV0:
             if flags:
                 raise RevlogError(_('unknown flags (%#04x) in version %d '
                                     'revlog %s') %
                                   (flags >> 16, fmt, self.indexfile))
         elif fmt == REVLOGV1:
             if flags & ~REVLOGV1_FLAGS:
                 raise RevlogError(_('unknown flags (%#04x) in version %d '
                                     'revlog %s') %
                                   (flags >> 16, fmt, self.indexfile))
         elif fmt == REVLOGV2:
             if flags & ~REVLOGV2_FLAGS:
                 raise RevlogError(_('unknown flags (%#04x) in version %d '
                                     'revlog %s') %
                                   (flags >> 16, fmt, self.indexfile))
         else:
             raise RevlogError(_('unknown version (%d) in revlog %s') %
                               (fmt, self.indexfile))
 
         self.storedeltachains = True
 
         self._io = revlogio()
         if self.version == REVLOGV0:
             self._io = revlogoldio()
         try:
             d = self._io.parseindex(indexdata, self._inline)
         except (ValueError, IndexError):
             raise RevlogError(_("index %s is corrupted") % (self.indexfile))
         self.index, nodemap, self._chunkcache = d
         if nodemap is not None:
             self.nodemap = self._nodecache = nodemap
         if not self._chunkcache:
             self._chunkclear()
         # revnum -> (chain-length, sum-delta-length)
         self._chaininfocache = {}
         # revlog header -> revlog compressor
         self._decompressors = {}
 
     @util.propertycache
     def _compressor(self):
         return util.compengines[self._compengine].revlogcompressor()
 
     def _indexfp(self, mode='r'):
         """file object for the revlog's index file"""
         args = {r'mode': mode}
         if mode != 'r':
             args[r'checkambig'] = self._checkambig
         if mode == 'w':
             args[r'atomictemp'] = True
         return self.opener(self.indexfile, **args)
 
     def _datafp(self, mode='r'):
         """file object for the revlog's data file"""
         return self.opener(self.datafile, mode=mode)
 
     @contextlib.contextmanager
     def _datareadfp(self, existingfp=None):
         """file object suitable to read data"""
         if existingfp is not None:
             yield existingfp
         else:
             if self._inline:
                 func = self._indexfp
             else:
                 func = self._datafp
             with func() as fp:
                 yield fp
 
     def tip(self):
         return self.node(len(self.index) - 2)
     def __contains__(self, rev):
         return 0 <= rev < len(self)
     def __len__(self):
         return len(self.index) - 1
     def __iter__(self):
         return iter(xrange(len(self)))
     def revs(self, start=0, stop=None):
         """iterate over all rev in this revlog (from start to stop)"""
         step = 1
         if stop is not None:
             if start > stop:
                 step = -1
             stop += step
         else:
             stop = len(self)
         return xrange(start, stop, step)
 
     @util.propertycache
     def nodemap(self):
         self.rev(self.node(0))
         return self._nodecache
 
     def hasnode(self, node):
         try:
             self.rev(node)
             return True
         except KeyError:
             return False
 
     def clearcaches(self):
         self._cache = None
         self._chainbasecache.clear()
         self._chunkcache = (0, '')
         self._pcache = {}
 
         try:
             self._nodecache.clearcaches()
         except AttributeError:
             self._nodecache = {nullid: nullrev}
             self._nodepos = None
 
     def rev(self, node):
         try:
             return self._nodecache[node]
         except TypeError:
             raise
         except RevlogError:
             # parsers.c radix tree lookup failed
             if node == wdirid:
                 raise error.WdirUnsupported
             raise LookupError(node, self.indexfile, _('no node'))
         except KeyError:
             # pure python cache lookup failed
             n = self._nodecache
             i = self.index
             p = self._nodepos
             if p is None:
                 p = len(i) - 2
             for r in xrange(p, -1, -1):
                 v = i[r][7]
                 n[v] = r
                 if v == node:
                     self._nodepos = r - 1
                     return r
             if node == wdirid:
                 raise error.WdirUnsupported
             raise LookupError(node, self.indexfile, _('no node'))
 
     # Accessors for index entries.
 
     # First tuple entry is 8 bytes. First 6 bytes are offset. Last 2 bytes
     # are flags.
     def start(self, rev):
         return int(self.index[rev][0] >> 16)
 
     def flags(self, rev):
         return self.index[rev][0] & 0xFFFF
 
     def length(self, rev):
         return self.index[rev][1]
 
     def rawsize(self, rev):
         """return the length of the uncompressed text for a given revision"""
         l = self.index[rev][2]
         if l >= 0:
             return l
 
         t = self.revision(rev, raw=True)
         return len(t)
 
     def size(self, rev):
         """length of non-raw text (processed by a "read" flag processor)"""
         # fast path: if no "read" flag processor could change the content,
         # size is rawsize. note: ELLIPSIS is known to not change the content.
         flags = self.flags(rev)
         if flags & (REVIDX_KNOWN_FLAGS ^ REVIDX_ELLIPSIS) == 0:
             return self.rawsize(rev)
 
         return len(self.revision(rev, raw=False))
 
     def chainbase(self, rev):
         base = self._chainbasecache.get(rev)
         if base is not None:
             return base
 
         index = self.index
         base = index[rev][3]
         while base != rev:
             rev = base
             base = index[rev][3]
 
         self._chainbasecache[rev] = base
         return base
 
     def linkrev(self, rev):
         return self.index[rev][4]
 
     def parentrevs(self, rev):
         try:
             entry = self.index[rev]
         except IndexError:
             if rev == wdirrev:
                 raise error.WdirUnsupported
             raise
 
         return entry[5], entry[6]
 
     def node(self, rev):
         try:
             return self.index[rev][7]
         except IndexError:
             if rev == wdirrev:
                 raise error.WdirUnsupported
             raise
 
     # Derived from index values.
 
     def end(self, rev):
         return self.start(rev) + self.length(rev)
 
     def parents(self, node):
         i = self.index
         d = i[self.rev(node)]
         return i[d[5]][7], i[d[6]][7] # map revisions to nodes inline
 
     def chainlen(self, rev):
         return self._chaininfo(rev)[0]
 
     def _chaininfo(self, rev):
         chaininfocache = self._chaininfocache
         if rev in chaininfocache:
             return chaininfocache[rev]
         index = self.index
         generaldelta = self._generaldelta
         iterrev = rev
         e = index[iterrev]
         clen = 0
         compresseddeltalen = 0
         while iterrev != e[3]:
             clen += 1
             compresseddeltalen += e[1]
             if generaldelta:
                 iterrev = e[3]
             else:
                 iterrev -= 1
             if iterrev in chaininfocache:
                 t = chaininfocache[iterrev]
                 clen += t[0]
                 compresseddeltalen += t[1]
                 break
             e = index[iterrev]
         else:
             # Add text length of base since decompressing that also takes
             # work. For cache hits the length is already included.
             compresseddeltalen += e[1]
         r = (clen, compresseddeltalen)
         chaininfocache[rev] = r
         return r
 
     def _deltachain(self, rev, stoprev=None):
         """Obtain the delta chain for a revision.
 
         ``stoprev`` specifies a revision to stop at. If not specified, we
         stop at the base of the chain.
 
         Returns a 2-tuple of (chain, stopped) where ``chain`` is a list of
         revs in ascending order and ``stopped`` is a bool indicating whether
         ``stoprev`` was hit.
         """
         # Try C implementation.
         try:
             return self.index.deltachain(rev, stoprev, self._generaldelta)
         except AttributeError:
             pass
 
         chain = []
 
         # Alias to prevent attribute lookup in tight loop.
         index = self.index
         generaldelta = self._generaldelta
 
         iterrev = rev
         e = index[iterrev]
         while iterrev != e[3] and iterrev != stoprev:
             chain.append(iterrev)
             if generaldelta:
                 iterrev = e[3]
             else:
                 iterrev -= 1
             e = index[iterrev]
 
         if iterrev == stoprev:
             stopped = True
         else:
             chain.append(iterrev)
             stopped = False
 
         chain.reverse()
         return chain, stopped
 
     def ancestors(self, revs, stoprev=0, inclusive=False):
         """Generate the ancestors of 'revs' in reverse topological order.
         Does not generate revs lower than stoprev.
 
         See the documentation for ancestor.lazyancestors for more details."""
 
         return ancestor.lazyancestors(self.parentrevs, revs, stoprev=stoprev,
                                       inclusive=inclusive)
 
     def descendants(self, revs):
         """Generate the descendants of 'revs' in revision order.
 
         Yield a sequence of revision numbers starting with a child of
         some rev in revs, i.e., each revision is *not* considered a
         descendant of itself.  Results are ordered by revision number (a
         topological sort)."""
         first = min(revs)
         if first == nullrev:
             for i in self:
                 yield i
             return
 
         seen = set(revs)
         for i in self.revs(start=first + 1):
             for x in self.parentrevs(i):
                 if x != nullrev and x in seen:
                     seen.add(i)
                     yield i
                     break
 
     def findcommonmissing(self, common=None, heads=None):
         """Return a tuple of the ancestors of common and the ancestors of heads
         that are not ancestors of common. In revset terminology, we return the
         tuple:
 
           ::common, (::heads) - (::common)
 
         The list is sorted by revision number, meaning it is
         topologically sorted.
 
         'heads' and 'common' are both lists of node IDs.  If heads is
         not supplied, uses all of the revlog's heads.  If common is not
         supplied, uses nullid."""
         if common is None:
             common = [nullid]
         if heads is None:
             heads = self.heads()
 
         common = [self.rev(n) for n in common]
         heads = [self.rev(n) for n in heads]
 
         # we want the ancestors, but inclusive
         class lazyset(object):
             def __init__(self, lazyvalues):
                 self.addedvalues = set()
                 self.lazyvalues = lazyvalues
 
             def __contains__(self, value):
                 return value in self.addedvalues or value in self.lazyvalues
 
             def __iter__(self):
                 added = self.addedvalues
                 for r in added:
                     yield r
                 for r in self.lazyvalues:
                     if not r in added:
                         yield r
 
             def add(self, value):
                 self.addedvalues.add(value)
 
             def update(self, values):
                 self.addedvalues.update(values)
 
         has = lazyset(self.ancestors(common))
         has.add(nullrev)
         has.update(common)
 
         # take all ancestors from heads that aren't in has
         missing = set()
         visit = collections.deque(r for r in heads if r not in has)
         while visit:
             r = visit.popleft()
             if r in missing:
                 continue
             else:
                 missing.add(r)
                 for p in self.parentrevs(r):
                     if p not in has:
                         visit.append(p)
         missing = list(missing)
         missing.sort()
         return has, [self.node(miss) for miss in missing]
 
     def incrementalmissingrevs(self, common=None):
         """Return an object that can be used to incrementally compute the
         revision numbers of the ancestors of arbitrary sets that are not
         ancestors of common. This is an ancestor.incrementalmissingancestors
         object.
 
         'common' is a list of revision numbers. If common is not supplied, uses
         nullrev.
         """
         if common is None:
             common = [nullrev]
 
         return ancestor.incrementalmissingancestors(self.parentrevs, common)
 
     def findmissingrevs(self, common=None, heads=None):
         """Return the revision numbers of the ancestors of heads that
         are not ancestors of common.
 
         More specifically, return a list of revision numbers corresponding to
         nodes N such that every N satisfies the following constraints:
 
           1. N is an ancestor of some node in 'heads'
           2. N is not an ancestor of any node in 'common'
 
         The list is sorted by revision number, meaning it is
         topologically sorted.
 
         'heads' and 'common' are both lists of revision numbers.  If heads is
         not supplied, uses all of the revlog's heads.  If common is not
         supplied, uses nullid."""
         if common is None:
             common = [nullrev]
         if heads is None:
             heads = self.headrevs()
 
         inc = self.incrementalmissingrevs(common=common)
         return inc.missingancestors(heads)
 
     def findmissing(self, common=None, heads=None):
         """Return the ancestors of heads that are not ancestors of common.
 
         More specifically, return a list of nodes N such that every N
         satisfies the following constraints:
 
           1. N is an ancestor of some node in 'heads'
           2. N is not an ancestor of any node in 'common'
 
         The list is sorted by revision number, meaning it is
         topologically sorted.
 
         'heads' and 'common' are both lists of node IDs.  If heads is
         not supplied, uses all of the revlog's heads.  If common is not
         supplied, uses nullid."""
         if common is None:
             common = [nullid]
         if heads is None:
             heads = self.heads()
 
         common = [self.rev(n) for n in common]
         heads = [self.rev(n) for n in heads]
 
         inc = self.incrementalmissingrevs(common=common)
         return [self.node(r) for r in inc.missingancestors(heads)]
 
     def nodesbetween(self, roots=None, heads=None):
         """Return a topological path from 'roots' to 'heads'.
 
         Return a tuple (nodes, outroots, outheads) where 'nodes' is a
         topologically sorted list of all nodes N that satisfy both of
         these constraints:
 
           1. N is a descendant of some node in 'roots'
           2. N is an ancestor of some node in 'heads'
 
         Every node is considered to be both a descendant and an ancestor
         of itself, so every reachable node in 'roots' and 'heads' will be
         included in 'nodes'.
 
         'outroots' is the list of reachable nodes in 'roots', i.e., the
         subset of 'roots' that is returned in 'nodes'.  Likewise,
         'outheads' is the subset of 'heads' that is also in 'nodes'.
 
         'roots' and 'heads' are both lists of node IDs.  If 'roots' is
         unspecified, uses nullid as the only root.  If 'heads' is
         unspecified, uses list of all of the revlog's heads."""
         nonodes = ([], [], [])
         if roots is not None:
             roots = list(roots)
             if not roots:
                 return nonodes
             lowestrev = min([self.rev(n) for n in roots])
         else:
             roots = [nullid] # Everybody's a descendant of nullid
             lowestrev = nullrev
         if (lowestrev == nullrev) and (heads is None):
             # We want _all_ the nodes!
             return ([self.node(r) for r in self], [nullid], list(self.heads()))
         if heads is None:
             # All nodes are ancestors, so the latest ancestor is the last
             # node.
             highestrev = len(self) - 1
             # Set ancestors to None to signal that every node is an ancestor.
             ancestors = None
             # Set heads to an empty dictionary for later discovery of heads
             heads = {}
         else:
             heads = list(heads)
             if not heads:
                 return nonodes
             ancestors = set()
             # Turn heads into a dictionary so we can remove 'fake' heads.
             # Also, later we will be using it to filter out the heads we can't
             # find from roots.
             heads = dict.fromkeys(heads, False)
             # Start at the top and keep marking parents until we're done.
             nodestotag = set(heads)
             # Remember where the top was so we can use it as a limit later.
             highestrev = max([self.rev(n) for n in nodestotag])
             while nodestotag:
                 # grab a node to tag
                 n = nodestotag.pop()
                 # Never tag nullid
                 if n == nullid:
                     continue
                 # A node's revision number represents its place in a
                 # topologically sorted list of nodes.
                 r = self.rev(n)
                 if r >= lowestrev:
                     if n not in ancestors:
                         # If we are possibly a descendant of one of the roots
                         # and we haven't already been marked as an ancestor
                         ancestors.add(n) # Mark as ancestor
                         # Add non-nullid parents to list of nodes to tag.
                         nodestotag.update([p for p in self.parents(n) if
                                            p != nullid])
                     elif n in heads: # We've seen it before, is it a fake head?
                         # So it is, real heads should not be the ancestors of
                         # any other heads.
                         heads.pop(n)
             if not ancestors:
                 return nonodes
             # Now that we have our set of ancestors, we want to remove any
             # roots that are not ancestors.
 
             # If one of the roots was nullid, everything is included anyway.
             if lowestrev > nullrev:
                 # But, since we weren't, let's recompute the lowest rev to not
                 # include roots that aren't ancestors.
 
                 # Filter out roots that aren't ancestors of heads
                 roots = [root for root in roots if root in ancestors]
                 # Recompute the lowest revision
                 if roots:
                     lowestrev = min([self.rev(root) for root in roots])
                 else:
                     # No more roots?  Return empty list
                     return nonodes
             else:
                 # We are descending from nullid, and don't need to care about
                 # any other roots.
                 lowestrev = nullrev
                 roots = [nullid]
         # Transform our roots list into a set.
         descendants = set(roots)
         # Also, keep the original roots so we can filter out roots that aren't
         # 'real' roots (i.e. are descended from other roots).
         roots = descendants.copy()
         # Our topologically sorted list of output nodes.
         orderedout = []
         # Don't start at nullid since we don't want nullid in our output list,
         # and if nullid shows up in descendants, empty parents will look like
         # they're descendants.
         for r in self.revs(start=max(lowestrev, 0), stop=highestrev + 1):
             n = self.node(r)
             isdescendant = False
             if lowestrev == nullrev:  # Everybody is a descendant of nullid
                 isdescendant = True
             elif n in descendants:
                 # n is already a descendant
                 isdescendant = True
                 # This check only needs to be done here because all the roots
                 # will start being marked is descendants before the loop.
                 if n in roots:
                     # If n was a root, check if it's a 'real' root.
                     p = tuple(self.parents(n))
                     # If any of its parents are descendants, it's not a root.
                     if (p[0] in descendants) or (p[1] in descendants):
                         roots.remove(n)
             else:
                 p = tuple(self.parents(n))
                 # A node is a descendant if either of its parents are
                 # descendants.  (We seeded the dependents list with the roots
                 # up there, remember?)
                 if (p[0] in descendants) or (p[1] in descendants):
                     descendants.add(n)
                     isdescendant = True
             if isdescendant and ((ancestors is None) or (n in ancestors)):
                 # Only include nodes that are both descendants and ancestors.
                 orderedout.append(n)
                 if (ancestors is not None) and (n in heads):
                     # We're trying to figure out which heads are reachable
                     # from roots.
                     # Mark this head as having been reached
                     heads[n] = True
                 elif ancestors is None:
                     # Otherwise, we're trying to discover the heads.
                     # Assume this is a head because if it isn't, the next step
                     # will eventually remove it.
                     heads[n] = True
                     # But, obviously its parents aren't.
                     for p in self.parents(n):
                         heads.pop(p, None)
         heads = [head for head, flag in heads.iteritems() if flag]
         roots = list(roots)
         assert orderedout
         assert roots
         assert heads
         return (orderedout, roots, heads)
 
     def headrevs(self):
         try:
             return self.index.headrevs()
         except AttributeError:
             return self._headrevs()
 
     def computephases(self, roots):
         return self.index.computephasesmapsets(roots)
 
     def _headrevs(self):
         count = len(self)
         if not count:
             return [nullrev]
         # we won't iter over filtered rev so nobody is a head at start
         ishead = [0] * (count + 1)
         index = self.index
         for r in self:
             ishead[r] = 1  # I may be an head
             e = index[r]
             ishead[e[5]] = ishead[e[6]] = 0  # my parent are not
         return [r for r, val in enumerate(ishead) if val]
 
     def heads(self, start=None, stop=None):
         """return the list of all nodes that have no children
 
         if start is specified, only heads that are descendants of
         start will be returned
         if stop is specified, it will consider all the revs from stop
         as if they had no children
         """
         if start is None and stop is None:
             if not len(self):
                 return [nullid]
             return [self.node(r) for r in self.headrevs()]
 
         if start is None:
             start = nullid
         if stop is None:
             stop = []
         stoprevs = set([self.rev(n) for n in stop])
         startrev = self.rev(start)
         reachable = {startrev}
         heads = {startrev}
 
         parentrevs = self.parentrevs
         for r in self.revs(start=startrev + 1):
             for p in parentrevs(r):
                 if p in reachable:
                     if r not in stoprevs:
                         reachable.add(r)
                     heads.add(r)
                 if p in heads and p not in stoprevs:
                     heads.remove(p)
 
         return [self.node(r) for r in heads]
 
     def children(self, node):
         """find the children of a given node"""
         c = []
         p = self.rev(node)
         for r in self.revs(start=p + 1):
             prevs = [pr for pr in self.parentrevs(r) if pr != nullrev]
             if prevs:
                 for pr in prevs:
                     if pr == p:
                         c.append(self.node(r))
             elif p == nullrev:
                 c.append(self.node(r))
         return c
 
     def descendant(self, start, end):
         if start == nullrev:
             return True
         for i in self.descendants([start]):
             if i == end:
                 return True
             elif i > end:
                 break
         return False
 
     def commonancestorsheads(self, a, b):
         """calculate all the heads of the common ancestors of nodes a and b"""
         a, b = self.rev(a), self.rev(b)
         try:
             ancs = self.index.commonancestorsheads(a, b)
         except (AttributeError, OverflowError): # C implementation failed
             ancs = ancestor.commonancestorsheads(self.parentrevs, a, b)
         return pycompat.maplist(self.node, ancs)
 
     def isancestor(self, a, b):
         """return True if node a is an ancestor of node b
 
         The implementation of this is trivial but the use of
         commonancestorsheads is not."""
         return a in self.commonancestorsheads(a, b)
 
     def ancestor(self, a, b):
         """calculate the "best" common ancestor of nodes a and b"""
 
         a, b = self.rev(a), self.rev(b)
         try:
             ancs = self.index.ancestors(a, b)
         except (AttributeError, OverflowError):
             ancs = ancestor.ancestors(self.parentrevs, a, b)
         if ancs:
             # choose a consistent winner when there's a tie
             return min(map(self.node, ancs))
         return nullid
 
     def _match(self, id):
         if isinstance(id, int):
             # rev
             return self.node(id)
         if len(id) == 20:
             # possibly a binary node
             # odds of a binary node being all hex in ASCII are 1 in 10**25
             try:
                 node = id
                 self.rev(node) # quick search the index
                 return node
             except LookupError:
                 pass # may be partial hex id
         try:
             # str(rev)
             rev = int(id)
             if str(rev) != id:
                 raise ValueError
             if rev < 0:
                 rev = len(self) + rev
             if rev < 0 or rev >= len(self):
                 raise ValueError
             return self.node(rev)
         except (ValueError, OverflowError):
             pass
         if len(id) == 40:
             try:
                 # a full hex nodeid?
                 node = bin(id)
                 self.rev(node)
                 return node
             except (TypeError, LookupError):
                 pass
 
     def _partialmatch(self, id):
         maybewdir = wdirhex.startswith(id)
         try:
             partial = self.index.partialmatch(id)
             if partial and self.hasnode(partial):
                 if maybewdir:
                     # single 'ff...' match in radix tree, ambiguous with wdir
                     raise RevlogError
                 return partial
             if maybewdir:
                 # no 'ff...' match in radix tree, wdir identified
                 raise error.WdirUnsupported
             return None
         except RevlogError:
             # parsers.c radix tree lookup gave multiple matches
             # fast path: for unfiltered changelog, radix tree is accurate
             if not getattr(self, 'filteredrevs', None):
                 raise LookupError(id, self.indexfile,
                                   _('ambiguous identifier'))
             # fall through to slow path that filters hidden revisions
         except (AttributeError, ValueError):
             # we are pure python, or key was too short to search radix tree
             pass
 
         if id in self._pcache:
             return self._pcache[id]
 
         if len(id) < 40:
             try:
                 # hex(node)[:...]
                 l = len(id) // 2  # grab an even number of digits
                 prefix = bin(id[:l * 2])
                 nl = [e[7] for e in self.index if e[7].startswith(prefix)]
                 nl = [n for n in nl if hex(n).startswith(id) and
                       self.hasnode(n)]
                 if len(nl) > 0:
                     if len(nl) == 1 and not maybewdir:
                         self._pcache[id] = nl[0]
                         return nl[0]
                     raise LookupError(id, self.indexfile,
                                       _('ambiguous identifier'))
                 if maybewdir:
                     raise error.WdirUnsupported
                 return None
-            except (TypeError, binascii.Error):
+            except TypeError:
                 pass
 
     def lookup(self, id):
         """locate a node based on:
             - revision number or str(revision number)
             - nodeid or subset of hex nodeid
         """
         n = self._match(id)
         if n is not None:
             return n
         n = self._partialmatch(id)
         if n:
             return n
 
         raise LookupError(id, self.indexfile, _('no match found'))
 
     def shortest(self, hexnode, minlength=1):
         """Find the shortest unambiguous prefix that matches hexnode."""
         def isvalid(test):
             try:
                 if self._partialmatch(test) is None:
                     return False
 
                 try:
                     i = int(test)
                     # if we are a pure int, then starting with zero will not be
                     # confused as a rev; or, obviously, if the int is larger
                     # than the value of the tip rev
                     if test[0] == '0' or i > len(self):
                         return True
                     return False
                 except ValueError:
                     return True
             except error.RevlogError:
                 return False
             except error.WdirUnsupported:
                 # single 'ff...' match
                 return True
 
         shortest = hexnode
         startlength = max(6, minlength)
         length = startlength
         while True:
             test = hexnode[:length]
             if isvalid(test):
                 shortest = test
                 if length == minlength or length > startlength:
                     return shortest
                 length -= 1
             else:
                 length += 1
                 if len(shortest) <= length:
                     return shortest
 
     def cmp(self, node, text):
         """compare text with a given file revision
 
         returns True if text is different than what is stored.
         """
         p1, p2 = self.parents(node)
         return hash(text, p1, p2) != node
 
     def _cachesegment(self, offset, data):
         """Add a segment to the revlog cache.
 
         Accepts an absolute offset and the data that is at that location.
         """
         o, d = self._chunkcache
         # try to add to existing cache
         if o + len(d) == offset and len(d) + len(data) < _chunksize:
             self._chunkcache = o, d + data
         else:
             self._chunkcache = offset, data
 
     def _readsegment(self, offset, length, df=None):
         """Load a segment of raw data from the revlog.
 
         Accepts an absolute offset, length to read, and an optional existing
         file handle to read from.
 
         If an existing file handle is passed, it will be seeked and the
         original seek position will NOT be restored.
 
         Returns a str or buffer of raw byte data.
         """
         # Cache data both forward and backward around the requested
         # data, in a fixed size window. This helps speed up operations
         # involving reading the revlog backwards.
         cachesize = self._chunkcachesize
         realoffset = offset & ~(cachesize - 1)
         reallength = (((offset + length + cachesize) & ~(cachesize - 1))
                       - realoffset)
         with self._datareadfp(df) as df:
             df.seek(realoffset)
             d = df.read(reallength)
         self._cachesegment(realoffset, d)
         if offset != realoffset or reallength != length:
             return util.buffer(d, offset - realoffset, length)
         return d
 
     def _getsegment(self, offset, length, df=None):
         """Obtain a segment of raw data from the revlog.
 
         Accepts an absolute offset, length of bytes to obtain, and an
         optional file handle to the already-opened revlog. If the file
         handle is used, it's original seek position will not be preserved.
 
         Requests for data may be returned from a cache.
 
         Returns a str or a buffer instance of raw byte data.
         """
         o, d = self._chunkcache
         l = len(d)
 
         # is it in the cache?
         cachestart = offset - o
         cacheend = cachestart + length
         if cachestart >= 0 and cacheend <= l:
             if cachestart == 0 and cacheend == l:
                 return d # avoid a copy
             return util.buffer(d, cachestart, cacheend - cachestart)
 
         return self._readsegment(offset, length, df=df)
 
     def _getsegmentforrevs(self, startrev, endrev, df=None):
         """Obtain a segment of raw data corresponding to a range of revisions.
 
         Accepts the start and end revisions and an optional already-open
         file handle to be used for reading. If the file handle is read, its
         seek position will not be preserved.
 
         Requests for data may be satisfied by a cache.
 
         Returns a 2-tuple of (offset, data) for the requested range of
         revisions. Offset is the integer offset from the beginning of the
         revlog and data is a str or buffer of the raw byte data.
 
         Callers will need to call ``self.start(rev)`` and ``self.length(rev)``
         to determine where each revision's data begins and ends.
         """
         # Inlined self.start(startrev) & self.end(endrev) for perf reasons
         # (functions are expensive).
         index = self.index
         istart = index[startrev]
         start = int(istart[0] >> 16)
         if startrev == endrev:
             end = start + istart[1]
         else:
             iend = index[endrev]
             end = int(iend[0] >> 16) + iend[1]
 
         if self._inline:
             start += (startrev + 1) * self._io.size
             end += (endrev + 1) * self._io.size
         length = end - start
 
         return start, self._getsegment(start, length, df=df)
 
     def _chunk(self, rev, df=None):
         """Obtain a single decompressed chunk for a revision.
 
         Accepts an integer revision and an optional already-open file handle
         to be used for reading. If used, the seek position of the file will not
         be preserved.
 
         Returns a str holding uncompressed data for the requested revision.
         """
         return self.decompress(self._getsegmentforrevs(rev, rev, df=df)[1])
 
     def _chunks(self, revs, df=None):
         """Obtain decompressed chunks for the specified revisions.
 
         Accepts an iterable of numeric revisions that are assumed to be in
         ascending order. Also accepts an optional already-open file handle
         to be used for reading. If used, the seek position of the file will
         not be preserved.
 
         This function is similar to calling ``self._chunk()`` multiple times,
         but is faster.
 
         Returns a list with decompressed data for each requested revision.
         """
         if not revs:
             return []
         start = self.start
         length = self.length
         inline = self._inline
         iosize = self._io.size
         buffer = util.buffer
 
         l = []
         ladd = l.append
 
         if not self._withsparseread:
             slicedchunks = (revs,)
         else:
             slicedchunks = _slicechunk(self, revs)
 
         for revschunk in slicedchunks:
             firstrev = revschunk[0]
             # Skip trailing revisions with empty diff
             for lastrev in revschunk[::-1]:
                 if length(lastrev) != 0:
                     break
 
             try:
                 offset, data = self._getsegmentforrevs(firstrev, lastrev, df=df)
             except OverflowError:
                 # issue4215 - we can't cache a run of chunks greater than
                 # 2G on Windows
                 return [self._chunk(rev, df=df) for rev in revschunk]
 
             decomp = self.decompress
             for rev in revschunk:
                 chunkstart = start(rev)
                 if inline:
                     chunkstart += (rev + 1) * iosize
                 chunklength = length(rev)
                 ladd(decomp(buffer(data, chunkstart - offset, chunklength)))
 
         return l
 
     def _chunkclear(self):
         """Clear the raw chunk cache."""
         self._chunkcache = (0, '')
 
     def deltaparent(self, rev):
         """return deltaparent of the given revision"""
         base = self.index[rev][3]
         if base == rev:
             return nullrev
         elif self._generaldelta:
             return base
         else:
             return rev - 1
 
     def revdiff(self, rev1, rev2):
         """return or calculate a delta between two revisions
 
         The delta calculated is in binary form and is intended to be written to
         revlog data directly. So this function needs raw revision data.
         """
         if rev1 != nullrev and self.deltaparent(rev2) == rev1:
             return bytes(self._chunk(rev2))
 
         return mdiff.textdiff(self.revision(rev1, raw=True),
                               self.revision(rev2, raw=True))
 
     def revision(self, nodeorrev, _df=None, raw=False):
         """return an uncompressed revision of a given node or revision
         number.
 
         _df - an existing file handle to read from. (internal-only)
         raw - an optional argument specifying if the revision data is to be
         treated as raw data when applying flag transforms. 'raw' should be set
         to True when generating changegroups or in debug commands.
         """
         if isinstance(nodeorrev, int):
             rev = nodeorrev
             node = self.node(rev)
         else:
             node = nodeorrev
             rev = None
 
         cachedrev = None
         flags = None
         rawtext = None
         if node == nullid:
             return ""
         if self._cache:
             if self._cache[0] == node:
                 # _cache only stores rawtext
                 if raw:
                     return self._cache[2]
                 # duplicated, but good for perf
                 if rev is None:
                     rev = self.rev(node)
                 if flags is None:
                     flags = self.flags(rev)
                 # no extra flags set, no flag processor runs, text = rawtext
                 if flags == REVIDX_DEFAULT_FLAGS:
                     return self._cache[2]
                 # rawtext is reusable. need to run flag processor
                 rawtext = self._cache[2]
 
             cachedrev = self._cache[1]
 
         # look up what we need to read
         if rawtext is None:
             if rev is None:
                 rev = self.rev(node)
 
             chain, stopped = self._deltachain(rev, stoprev=cachedrev)
             if stopped:
                 rawtext = self._cache[2]
 
             # drop cache to save memory
             self._cache = None
 
             bins = self._chunks(chain, df=_df)
             if rawtext is None:
                 rawtext = bytes(bins[0])
                 bins = bins[1:]
 
             rawtext = mdiff.patches(rawtext, bins)
             self._cache = (node, rev, rawtext)
 
         if flags is None:
             if rev is None:
                 rev = self.rev(node)
             flags = self.flags(rev)
 
         text, validatehash = self._processflags(rawtext, flags, 'read', raw=raw)
         if validatehash:
             self.checkhash(text, node, rev=rev)
 
         return text
 
     def hash(self, text, p1, p2):
         """Compute a node hash.
 
         Available as a function so that subclasses can replace the hash
         as needed.
         """
         return hash(text, p1, p2)
 
     def _processflags(self, text, flags, operation, raw=False):
         """Inspect revision data flags and applies transforms defined by
         registered flag processors.
 
         ``text`` - the revision data to process
         ``flags`` - the revision flags
         ``operation`` - the operation being performed (read or write)
         ``raw`` - an optional argument describing if the raw transform should be
         applied.
 
         This method processes the flags in the order (or reverse order if
         ``operation`` is 'write') defined by REVIDX_FLAGS_ORDER, applying the
         flag processors registered for present flags. The order of flags defined
         in REVIDX_FLAGS_ORDER needs to be stable to allow non-commutativity.
 
         Returns a 2-tuple of ``(text, validatehash)`` where ``text`` is the
         processed text and ``validatehash`` is a bool indicating whether the
         returned text should be checked for hash integrity.
 
         Note: If the ``raw`` argument is set, it has precedence over the
         operation and will only update the value of ``validatehash``.
         """
         # fast path: no flag processors will run
         if flags == 0:
             return text, True
         if not operation in ('read', 'write'):
             raise ProgrammingError(_("invalid '%s' operation ") % (operation))
         # Check all flags are known.
         if flags & ~REVIDX_KNOWN_FLAGS:
             raise RevlogError(_("incompatible revision flag '%#x'") %
                               (flags & ~REVIDX_KNOWN_FLAGS))
         validatehash = True
         # Depending on the operation (read or write), the order might be
         # reversed due to non-commutative transforms.
         orderedflags = REVIDX_FLAGS_ORDER
         if operation == 'write':
             orderedflags = reversed(orderedflags)
 
         for flag in orderedflags:
             # If a flagprocessor has been registered for a known flag, apply the
             # related operation transform and update result tuple.
             if flag & flags:
                 vhash = True
 
                 if flag not in _flagprocessors:
                     message = _("missing processor for flag '%#x'") % (flag)
                     raise RevlogError(message)
 
                 processor = _flagprocessors[flag]
                 if processor is not None:
                     readtransform, writetransform, rawtransform = processor
 
                     if raw:
                         vhash = rawtransform(self, text)
                     elif operation == 'read':
                         text, vhash = readtransform(self, text)
                     else: # write operation
                         text, vhash = writetransform(self, text)
                 validatehash = validatehash and vhash
 
         return text, validatehash
 
     def checkhash(self, text, node, p1=None, p2=None, rev=None):
         """Check node hash integrity.
 
         Available as a function so that subclasses can extend hash mismatch
         behaviors as needed.
         """
         if p1 is None and p2 is None:
             p1, p2 = self.parents(node)
         if node != self.hash(text, p1, p2):
             revornode = rev
             if revornode is None:
                 revornode = templatefilters.short(hex(node))
             raise RevlogError(_("integrity check failed on %s:%s")
                 % (self.indexfile, pycompat.bytestr(revornode)))
 
     def _enforceinlinesize(self, tr, fp=None):
         """Check if the revlog is too big for inline and convert if so.
 
         This should be called after revisions are added to the revlog. If the
         revlog has grown too large to be an inline revlog, it will convert it
         to use multiple index and data files.
         """
         if not self._inline or (self.start(-2) + self.length(-2)) < _maxinline:
             return
 
         trinfo = tr.find(self.indexfile)
         if trinfo is None:
             raise RevlogError(_("%s not found in the transaction")
                               % self.indexfile)
 
         trindex = trinfo[2]
         if trindex is not None:
             dataoff = self.start(trindex)
         else:
             # revlog was stripped at start of transaction, use all leftover data
             trindex = len(self) - 1
             dataoff = self.end(-2)
 
         tr.add(self.datafile, dataoff)
 
         if fp:
             fp.flush()
             fp.close()
 
         with self._datafp('w') as df:
             for r in self:
                 df.write(self._getsegmentforrevs(r, r)[1])
 
         with self._indexfp('w') as fp:
             self.version &= ~FLAG_INLINE_DATA
             self._inline = False
             io = self._io
             for i in self:
                 e = io.packentry(self.index[i], self.node, self.version, i)
                 fp.write(e)
 
             # the temp file replace the real index when we exit the context
             # manager
 
         tr.replace(self.indexfile, trindex * self._io.size)
         self._chunkclear()
 
     def addrevision(self, text, transaction, link, p1, p2, cachedelta=None,
                     node=None, flags=REVIDX_DEFAULT_FLAGS, deltacomputer=None):
         """add a revision to the log
 
         text - the revision data to add
         transaction - the transaction object used for rollback
         link - the linkrev data to add
         p1, p2 - the parent nodeids of the revision
         cachedelta - an optional precomputed delta
         node - nodeid of revision; typically node is not specified, and it is
             computed by default as hash(text, p1, p2), however subclasses might
             use different hashing method (and override checkhash() in such case)
         flags - the known flags to set on the revision
         deltacomputer - an optional _deltacomputer instance shared between
             multiple calls
         """
         if link == nullrev:
             raise RevlogError(_("attempted to add linkrev -1 to %s")
                               % self.indexfile)
 
         if flags:
             node = node or self.hash(text, p1, p2)
 
         rawtext, validatehash = self._processflags(text, flags, 'write')
 
         # If the flag processor modifies the revision data, ignore any provided
         # cachedelta.
         if rawtext != text:
             cachedelta = None
 
         if len(rawtext) > _maxentrysize:
             raise RevlogError(
                 _("%s: size of %d bytes exceeds maximum revlog storage of 2GiB")
                 % (self.indexfile, len(rawtext)))
 
         node = node or self.hash(rawtext, p1, p2)
         if node in self.nodemap:
             return node
 
         if validatehash:
             self.checkhash(rawtext, node, p1=p1, p2=p2)
 
         return self.addrawrevision(rawtext, transaction, link, p1, p2, node,
                                    flags, cachedelta=cachedelta,
                                    deltacomputer=deltacomputer)
 
     def addrawrevision(self, rawtext, transaction, link, p1, p2, node, flags,
                        cachedelta=None, deltacomputer=None):
         """add a raw revision with known flags, node and parents
         useful when reusing a revision not stored in this revlog (ex: received
         over wire, or read from an external bundle).
         """
         dfh = None
         if not self._inline:
             dfh = self._datafp("a+")
         ifh = self._indexfp("a+")
         try:
             return self._addrevision(node, rawtext, transaction, link, p1, p2,
                                      flags, cachedelta, ifh, dfh,
                                      deltacomputer=deltacomputer)
         finally:
             if dfh:
                 dfh.close()
             ifh.close()
 
     def compress(self, data):
         """Generate a possibly-compressed representation of data."""
         if not data:
             return '', data
 
         compressed = self._compressor.compress(data)
 
         if compressed:
             # The revlog compressor added the header in the returned data.
             return '', compressed
 
         if data[0:1] == '\0':
             return '', data
         return 'u', data
 
     def decompress(self, data):
         """Decompress a revlog chunk.
 
         The chunk is expected to begin with a header identifying the
         format type so it can be routed to an appropriate decompressor.
         """
         if not data:
             return data
 
         # Revlogs are read much more frequently than they are written and many
         # chunks only take microseconds to decompress, so performance is
         # important here.
         #
         # We can make a few assumptions about revlogs:
         #
         # 1) the majority of chunks will be compressed (as opposed to inline
         #    raw data).
         # 2) decompressing *any* data will likely by at least 10x slower than
         #    returning raw inline data.
         # 3) we want to prioritize common and officially supported compression
         #    engines
         #
         # It follows that we want to optimize for "decompress compressed data
         # when encoded with common and officially supported compression engines"
         # case over "raw data" and "data encoded by less common or non-official
         # compression engines." That is why we have the inline lookup first
         # followed by the compengines lookup.
         #
         # According to `hg perfrevlogchunks`, this is ~0.5% faster for zlib
         # compressed chunks. And this matters for changelog and manifest reads.
         t = data[0:1]
 
         if t == 'x':
             try:
                 return _zlibdecompress(data)
             except zlib.error as e:
                 raise RevlogError(_('revlog decompress error: %s') % str(e))
         # '\0' is more common than 'u' so it goes first.
         elif t == '\0':
             return data
         elif t == 'u':
             return util.buffer(data, 1)
 
         try:
             compressor = self._decompressors[t]
         except KeyError:
             try:
                 engine = util.compengines.forrevlogheader(t)
                 compressor = engine.revlogcompressor()
                 self._decompressors[t] = compressor
             except KeyError:
                 raise RevlogError(_('unknown compression type %r') % t)
 
         return compressor.decompress(data)
 
     def _isgooddeltainfo(self, d, textlen):
         """Returns True if the given delta is good. Good means that it is within
         the disk span, disk size, and chain length bounds that we know to be
         performant."""
         if d is None:
             return False
 
         # - 'd.distance' is the distance from the base revision -- bounding it
         #   limits the amount of I/O we need to do.
         # - 'd.compresseddeltalen' is the sum of the total size of deltas we
         #   need to apply -- bounding it limits the amount of CPU we consume.
 
         defaultmax = textlen * 4
         maxdist = self._maxdeltachainspan
         if not maxdist:
             maxdist = d.distance # ensure the conditional pass
         maxdist = max(maxdist, defaultmax)
         if (d.distance > maxdist or d.deltalen > textlen or
             d.compresseddeltalen > textlen * 2 or
             (self._maxchainlen and d.chainlen > self._maxchainlen)):
             return False
 
         return True
 
     def _addrevision(self, node, rawtext, transaction, link, p1, p2, flags,
                      cachedelta, ifh, dfh, alwayscache=False,
                      deltacomputer=None):
         """internal function to add revisions to the log
 
         see addrevision for argument descriptions.
 
         note: "addrevision" takes non-raw text, "_addrevision" takes raw text.
 
         if "deltacomputer" is not provided or None, a defaultdeltacomputer will
         be used.
 
         invariants:
         - rawtext is optional (can be None); if not set, cachedelta must be set.
           if both are set, they must correspond to each other.
         """
         if node == nullid:
             raise RevlogError(_("%s: attempt to add null revision") %
                               (self.indexfile))
         if node == wdirid:
             raise RevlogError(_("%s: attempt to add wdir revision") %
                               (self.indexfile))
 
         if self._inline:
             fh = ifh
         else:
             fh = dfh
 
         btext = [rawtext]
 
         curr = len(self)
         prev = curr - 1
         offset = self.end(prev)
         p1r, p2r = self.rev(p1), self.rev(p2)
 
         # full versions are inserted when the needed deltas
         # become comparable to the uncompressed text
         if rawtext is None:
             textlen = mdiff.patchedsize(self.rawsize(cachedelta[0]),
                                         cachedelta[1])
         else:
             textlen = len(rawtext)
 
         if deltacomputer is None:
             deltacomputer = _deltacomputer(self)
 
         revinfo = _revisioninfo(node, p1, p2, btext, textlen, cachedelta, flags)
         deltainfo = deltacomputer.finddeltainfo(revinfo, fh)
 
         if deltainfo is not None:
             base = deltainfo.base
             chainbase = deltainfo.chainbase
             data = deltainfo.data
             l = deltainfo.deltalen
         else:
             rawtext = deltacomputer.buildtext(revinfo, fh)
             data = self.compress(rawtext)
             l = len(data[1]) + len(data[0])
             base = chainbase = curr
 
         e = (offset_type(offset, flags), l, textlen,
              base, link, p1r, p2r, node)
         self.index.insert(-1, e)
         self.nodemap[node] = curr
 
         entry = self._io.packentry(e, self.node, self.version, curr)
         self._writeentry(transaction, ifh, dfh, entry, data, link, offset)
 
         if alwayscache and rawtext is None:
             rawtext = deltacomputer._buildtext(revinfo, fh)
 
         if type(rawtext) == bytes: # only accept immutable objects
             self._cache = (node, curr, rawtext)
         self._chainbasecache[curr] = chainbase
         return node
 
     def _writeentry(self, transaction, ifh, dfh, entry, data, link, offset):
         # Files opened in a+ mode have inconsistent behavior on various
         # platforms. Windows requires that a file positioning call be made
         # when the file handle transitions between reads and writes. See
         # 3686fa2b8eee and the mixedfilemodewrapper in windows.py. On other
         # platforms, Python or the platform itself can be buggy. Some versions
         # of Solaris have been observed to not append at the end of the file
         # if the file was seeked to before the end. See issue4943 for more.
         #
         # We work around this issue by inserting a seek() before writing.
         # Note: This is likely not necessary on Python 3.
         ifh.seek(0, os.SEEK_END)
         if dfh:
             dfh.seek(0, os.SEEK_END)
 
         curr = len(self) - 1
         if not self._inline:
             transaction.add(self.datafile, offset)
             transaction.add(self.indexfile, curr * len(entry))
             if data[0]:
                 dfh.write(data[0])
             dfh.write(data[1])
             ifh.write(entry)
         else:
             offset += curr * self._io.size
             transaction.add(self.indexfile, offset, curr)
             ifh.write(entry)
             ifh.write(data[0])
             ifh.write(data[1])
             self._enforceinlinesize(transaction, ifh)
 
     def addgroup(self, deltas, linkmapper, transaction, addrevisioncb=None):
         """
         add a delta group
 
         given a set of deltas, add them to the revision log. the
         first delta is against its parent, which should be in our
         log, the rest are against the previous delta.
 
         If ``addrevisioncb`` is defined, it will be called with arguments of
         this revlog and the node that was added.
         """
 
         nodes = []
 
         r = len(self)
         end = 0
         if r:
             end = self.end(r - 1)
         ifh = self._indexfp("a+")
         isize = r * self._io.size
         if self._inline:
             transaction.add(self.indexfile, end + isize, r)
             dfh = None
         else:
             transaction.add(self.indexfile, isize, r)
             transaction.add(self.datafile, end)
             dfh = self._datafp("a+")
         def flush():
             if dfh:
                 dfh.flush()
             ifh.flush()
         try:
             deltacomputer = _deltacomputer(self)
             # loop through our set of deltas
             for data in deltas:
                 node, p1, p2, linknode, deltabase, delta, flags = data
                 link = linkmapper(linknode)
                 flags = flags or REVIDX_DEFAULT_FLAGS
 
                 nodes.append(node)
 
                 if node in self.nodemap:
                     # this can happen if two branches make the same change
                     continue
 
                 for p in (p1, p2):
                     if p not in self.nodemap:
                         raise LookupError(p, self.indexfile,
                                           _('unknown parent'))
 
                 if deltabase not in self.nodemap:
                     raise LookupError(deltabase, self.indexfile,
                                       _('unknown delta base'))
 
                 baserev = self.rev(deltabase)
 
                 if baserev != nullrev and self.iscensored(baserev):
                     # if base is censored, delta must be full replacement in a
                     # single patch operation
                     hlen = struct.calcsize(">lll")
                     oldlen = self.rawsize(baserev)
                     newlen = len(delta) - hlen
                     if delta[:hlen] != mdiff.replacediffheader(oldlen, newlen):
                         raise error.CensoredBaseError(self.indexfile,
                                                       self.node(baserev))
 
                 if not flags and self._peek_iscensored(baserev, delta, flush):
                     flags |= REVIDX_ISCENSORED
 
                 # We assume consumers of addrevisioncb will want to retrieve
                 # the added revision, which will require a call to
                 # revision(). revision() will fast path if there is a cache
                 # hit. So, we tell _addrevision() to always cache in this case.
                 # We're only using addgroup() in the context of changegroup
                 # generation so the revision data can always be handled as raw
                 # by the flagprocessor.
                 self._addrevision(node, None, transaction, link,
                                   p1, p2, flags, (baserev, delta),
                                   ifh, dfh,
                                   alwayscache=bool(addrevisioncb),
                                   deltacomputer=deltacomputer)
 
                 if addrevisioncb:
                     addrevisioncb(self, node)
 
                 if not dfh and not self._inline:
                     # addrevision switched from inline to conventional
                     # reopen the index
                     ifh.close()
                     dfh = self._datafp("a+")
                     ifh = self._indexfp("a+")
         finally:
             if dfh:
                 dfh.close()
             ifh.close()
 
         return nodes
 
     def iscensored(self, rev):
         """Check if a file revision is censored."""
         return False
 
     def _peek_iscensored(self, baserev, delta, flush):
         """Quickly check if a delta produces a censored revision."""
         return False
 
     def getstrippoint(self, minlink):
         """find the minimum rev that must be stripped to strip the linkrev
 
         Returns a tuple containing the minimum rev and a set of all revs that
         have linkrevs that will be broken by this strip.
         """
         brokenrevs = set()
         strippoint = len(self)
 
         heads = {}
         futurelargelinkrevs = set()
         for head in self.headrevs():
             headlinkrev = self.linkrev(head)
             heads[head] = headlinkrev
             if headlinkrev >= minlink:
                 futurelargelinkrevs.add(headlinkrev)
 
         # This algorithm involves walking down the rev graph, starting at the
         # heads. Since the revs are topologically sorted according to linkrev,
         # once all head linkrevs are below the minlink, we know there are
         # no more revs that could have a linkrev greater than minlink.
         # So we can stop walking.
         while futurelargelinkrevs:
             strippoint -= 1
             linkrev = heads.pop(strippoint)
 
             if linkrev < minlink:
                 brokenrevs.add(strippoint)
             else:
                 futurelargelinkrevs.remove(linkrev)
 
             for p in self.parentrevs(strippoint):
                 if p != nullrev:
                     plinkrev = self.linkrev(p)
                     heads[p] = plinkrev
                     if plinkrev >= minlink:
                         futurelargelinkrevs.add(plinkrev)
 
         return strippoint, brokenrevs
 
     def strip(self, minlink, transaction):
         """truncate the revlog on the first revision with a linkrev >= minlink
 
         This function is called when we're stripping revision minlink and
         its descendants from the repository.
 
         We have to remove all revisions with linkrev >= minlink, because
         the equivalent changelog revisions will be renumbered after the
         strip.
 
         So we truncate the revlog on the first of these revisions, and
         trust that the caller has saved the revisions that shouldn't be
         removed and that it'll re-add them after this truncation.
         """
         if len(self) == 0:
             return
 
         rev, _ = self.getstrippoint(minlink)
         if rev == len(self):
             return
 
         # first truncate the files on disk
         end = self.start(rev)
         if not self._inline:
             transaction.add(self.datafile, end)
             end = rev * self._io.size
         else:
             end += rev * self._io.size
 
         transaction.add(self.indexfile, end)
 
         # then reset internal state in memory to forget those revisions
         self._cache = None
         self._chaininfocache = {}
         self._chunkclear()
         for x in xrange(rev, len(self)):
             del self.nodemap[self.node(x)]
 
         del self.index[rev:-1]
 
     def checksize(self):
         expected = 0
         if len(self):
             expected = max(0, self.end(len(self) - 1))
 
         try:
             with self._datafp() as f:
                 f.seek(0, 2)
                 actual = f.tell()
             dd = actual - expected
         except IOError as inst:
             if inst.errno != errno.ENOENT:
                 raise
             dd = 0
 
         try:
             f = self.opener(self.indexfile)
             f.seek(0, 2)
             actual = f.tell()
             f.close()
             s = self._io.size
             i = max(0, actual // s)
             di = actual - (i * s)
             if self._inline:
                 databytes = 0
                 for r in self:
                     databytes += max(0, self.length(r))
                 dd = 0
                 di = actual - len(self) * s - databytes
         except IOError as inst:
             if inst.errno != errno.ENOENT:
                 raise
             di = 0
 
         return (dd, di)
 
     def files(self):
         res = [self.indexfile]
         if not self._inline:
             res.append(self.datafile)
         return res
 
     DELTAREUSEALWAYS = 'always'
     DELTAREUSESAMEREVS = 'samerevs'
     DELTAREUSENEVER = 'never'
 
     DELTAREUSEFULLADD = 'fulladd'
 
     DELTAREUSEALL = {'always', 'samerevs', 'never', 'fulladd'}
 
     def clone(self, tr, destrevlog, addrevisioncb=None,
               deltareuse=DELTAREUSESAMEREVS, aggressivemergedeltas=None):
         """Copy this revlog to another, possibly with format changes.
 
         The destination revlog will contain the same revisions and nodes.
         However, it may not be bit-for-bit identical due to e.g. delta encoding
         differences.
 
         The ``deltareuse`` argument control how deltas from the existing revlog
         are preserved in the destination revlog. The argument can have the
         following values:
 
         DELTAREUSEALWAYS
            Deltas will always be reused (if possible), even if the destination
            revlog would not select the same revisions for the delta. This is the
            fastest mode of operation.
         DELTAREUSESAMEREVS
            Deltas will be reused if the destination revlog would pick the same
            revisions for the delta. This mode strikes a balance between speed
            and optimization.
         DELTAREUSENEVER
            Deltas will never be reused. This is the slowest mode of execution.
            This mode can be used to recompute deltas (e.g. if the diff/delta
            algorithm changes).
 
         Delta computation can be slow, so the choice of delta reuse policy can
         significantly affect run time.
 
         The default policy (``DELTAREUSESAMEREVS``) strikes a balance between
         two extremes. Deltas will be reused if they are appropriate. But if the
         delta could choose a better revision, it will do so. This means if you
         are converting a non-generaldelta revlog to a generaldelta revlog,
         deltas will be recomputed if the delta's parent isn't a parent of the
         revision.
 
         In addition to the delta policy, the ``aggressivemergedeltas`` argument
         controls whether to compute deltas against both parents for merges.
         By default, the current default is used.
         """
         if deltareuse not in self.DELTAREUSEALL:
             raise ValueError(_('value for deltareuse invalid: %s') % deltareuse)
 
         if len(destrevlog):
             raise ValueError(_('destination revlog is not empty'))
 
         if getattr(self, 'filteredrevs', None):
             raise ValueError(_('source revlog has filtered revisions'))
         if getattr(destrevlog, 'filteredrevs', None):
             raise ValueError(_('destination revlog has filtered revisions'))
 
         # lazydeltabase controls whether to reuse a cached delta, if possible.
         oldlazydeltabase = destrevlog._lazydeltabase
         oldamd = destrevlog._aggressivemergedeltas
 
         try:
             if deltareuse == self.DELTAREUSEALWAYS:
                 destrevlog._lazydeltabase = True
             elif deltareuse == self.DELTAREUSESAMEREVS:
                 destrevlog._lazydeltabase = False
 
             destrevlog._aggressivemergedeltas = aggressivemergedeltas or oldamd
 
             populatecachedelta = deltareuse in (self.DELTAREUSEALWAYS,
                                                 self.DELTAREUSESAMEREVS)
 
             deltacomputer = _deltacomputer(destrevlog)
             index = self.index
             for rev in self:
                 entry = index[rev]
 
                 # Some classes override linkrev to take filtered revs into
                 # account. Use raw entry from index.
                 flags = entry[0] & 0xffff
                 linkrev = entry[4]
                 p1 = index[entry[5]][7]
                 p2 = index[entry[6]][7]
                 node = entry[7]
 
                 # (Possibly) reuse the delta from the revlog if allowed and
                 # the revlog chunk is a delta.
                 cachedelta = None
                 rawtext = None
                 if populatecachedelta:
                     dp = self.deltaparent(rev)
                     if dp != nullrev:
                         cachedelta = (dp, str(self._chunk(rev)))
 
                 if not cachedelta:
                     rawtext = self.revision(rev, raw=True)
 
 
                 if deltareuse == self.DELTAREUSEFULLADD:
                     destrevlog.addrevision(rawtext, tr, linkrev, p1, p2,
                                            cachedelta=cachedelta,
                                            node=node, flags=flags,
                                            deltacomputer=deltacomputer)
                 else:
                     ifh = destrevlog.opener(destrevlog.indexfile, 'a+',
                                             checkambig=False)
                     dfh = None
                     if not destrevlog._inline:
                         dfh = destrevlog.opener(destrevlog.datafile, 'a+')
                     try:
                         destrevlog._addrevision(node, rawtext, tr, linkrev, p1,
                                                 p2, flags, cachedelta, ifh, dfh,
                                                 deltacomputer=deltacomputer)
                     finally:
                         if dfh:
                             dfh.close()
                         ifh.close()
 
                 if addrevisioncb:
                     addrevisioncb(self, rev, node)
         finally:
             destrevlog._lazydeltabase = oldlazydeltabase
             destrevlog._aggressivemergedeltas = oldamd