cleanup: migrate from re.escape to stringutil.reescape

This has consistent behavior on Python 2.7, 3.6, and 3.7 and has the
benefit of probably being a little faster. Test output changes are
largely because / used to be pointlessly escaped.

Differential Revision: https://phab.mercurial-scm.org/D3842

hgext/keyword.py

@@ -1,815 +1,815
 # keyword.py - $Keyword$ expansion for Mercurial
 #
 # Copyright 2007-2015 Christian Ebert <[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.
 #
 # $Id$
 #
 # Keyword expansion hack against the grain of a Distributed SCM
 #
 # There are many good reasons why this is not needed in a distributed
 # SCM, still it may be useful in very small projects based on single
 # files (like LaTeX packages), that are mostly addressed to an
 # audience not running a version control system.
 #
 # For in-depth discussion refer to
 # <https://mercurial-scm.org/wiki/KeywordPlan>.
 #
 # Keyword expansion is based on Mercurial's changeset template mappings.
 #
 # Binary files are not touched.
 #
 # Files to act upon/ignore are specified in the [keyword] section.
 # Customized keyword template mappings in the [keywordmaps] section.
 #
 # Run 'hg help keyword' and 'hg kwdemo' to get info on configuration.
 
 '''expand keywords in tracked files
 
 This extension expands RCS/CVS-like or self-customized $Keywords$ in
 tracked text files selected by your configuration.
 
 Keywords are only expanded in local repositories and not stored in the
 change history. The mechanism can be regarded as a convenience for the
 current user or for archive distribution.
 
 Keywords expand to the changeset data pertaining to the latest change
 relative to the working directory parent of each file.
 
 Configuration is done in the [keyword], [keywordset] and [keywordmaps]
 sections of hgrc files.
 
 Example::
 
     [keyword]
     # expand keywords in every python file except those matching "x*"
     **.py =
     x*    = ignore
 
     [keywordset]
     # prefer svn- over cvs-like default keywordmaps
     svn = True
 
 .. note::
 
    The more specific you are in your filename patterns the less you
    lose speed in huge repositories.
 
 For [keywordmaps] template mapping and expansion demonstration and
 control run :hg:`kwdemo`. See :hg:`help templates` for a list of
 available templates and filters.
 
 Three additional date template filters are provided:
 
 :``utcdate``:    "2006/09/18 15:13:13"
 :``svnutcdate``: "2006-09-18 15:13:13Z"
 :``svnisodate``: "2006-09-18 08:13:13 -700 (Mon, 18 Sep 2006)"
 
 The default template mappings (view with :hg:`kwdemo -d`) can be
 replaced with customized keywords and templates. Again, run
 :hg:`kwdemo` to control the results of your configuration changes.
 
 Before changing/disabling active keywords, you must run :hg:`kwshrink`
 to avoid storing expanded keywords in the change history.
 
 To force expansion after enabling it, or a configuration change, run
 :hg:`kwexpand`.
 
 Expansions spanning more than one line and incremental expansions,
 like CVS' $Log$, are not supported. A keyword template map "Log =
 {desc}" expands to the first line of the changeset description.
 '''
 
 
 from __future__ import absolute_import
 
 import os
 import re
 import weakref
 
 from mercurial.i18n import _
 from mercurial.hgweb import webcommands
 
 from mercurial import (
     cmdutil,
     context,
     dispatch,
     error,
     extensions,
     filelog,
     localrepo,
     logcmdutil,
     match,
     patch,
     pathutil,
     pycompat,
     registrar,
     scmutil,
     templatefilters,
     templateutil,
     util,
 )
 from mercurial.utils import (
     dateutil,
     stringutil,
 )
 
 cmdtable = {}
 command = registrar.command(cmdtable)
 # 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'
 
 # hg commands that do not act on keywords
 nokwcommands = ('add addremove annotate bundle export grep incoming init log'
                 ' outgoing push tip verify convert email glog')
 
 # webcommands that do not act on keywords
 nokwwebcommands = ('annotate changeset rev filediff diff comparison')
 
 # hg commands that trigger expansion only when writing to working dir,
 # not when reading filelog, and unexpand when reading from working dir
 restricted = ('merge kwexpand kwshrink record qrecord resolve transplant'
               ' unshelve rebase graft backout histedit fetch')
 
 # names of extensions using dorecord
 recordextensions = 'record'
 
 colortable = {
     'kwfiles.enabled': 'green bold',
     'kwfiles.deleted': 'cyan bold underline',
     'kwfiles.enabledunknown': 'green',
     'kwfiles.ignored': 'bold',
     'kwfiles.ignoredunknown': 'none'
 }
 
 templatefilter = registrar.templatefilter()
 
 configtable = {}
 configitem = registrar.configitem(configtable)
 
 configitem('keywordset', 'svn',
     default=False,
 )
 # date like in cvs' $Date
 @templatefilter('utcdate', intype=templateutil.date)
 def utcdate(date):
     '''Date. Returns a UTC-date in this format: "2009/08/18 11:00:13".
     '''
     dateformat = '%Y/%m/%d %H:%M:%S'
     return dateutil.datestr((date[0], 0), dateformat)
 # date like in svn's $Date
 @templatefilter('svnisodate', intype=templateutil.date)
 def svnisodate(date):
     '''Date. Returns a date in this format: "2009-08-18 13:00:13
     +0200 (Tue, 18 Aug 2009)".
     '''
     return dateutil.datestr(date, '%Y-%m-%d %H:%M:%S %1%2 (%a, %d %b %Y)')
 # date like in svn's $Id
 @templatefilter('svnutcdate', intype=templateutil.date)
 def svnutcdate(date):
     '''Date. Returns a UTC-date in this format: "2009-08-18
     11:00:13Z".
     '''
     dateformat = '%Y-%m-%d %H:%M:%SZ'
     return dateutil.datestr((date[0], 0), dateformat)
 
 # make keyword tools accessible
 kwtools = {'hgcmd': ''}
 
 def _defaultkwmaps(ui):
     '''Returns default keywordmaps according to keywordset configuration.'''
     templates = {
         'Revision': '{node|short}',
         'Author': '{author|user}',
     }
     kwsets = ({
         'Date': '{date|utcdate}',
         'RCSfile': '{file|basename},v',
         'RCSFile': '{file|basename},v', # kept for backwards compatibility
                                         # with hg-keyword
         'Source': '{root}/{file},v',
         'Id': '{file|basename},v {node|short} {date|utcdate} {author|user}',
         'Header': '{root}/{file},v {node|short} {date|utcdate} {author|user}',
     }, {
         'Date': '{date|svnisodate}',
         'Id': '{file|basename},v {node|short} {date|svnutcdate} {author|user}',
         'LastChangedRevision': '{node|short}',
         'LastChangedBy': '{author|user}',
         'LastChangedDate': '{date|svnisodate}',
     })
     templates.update(kwsets[ui.configbool('keywordset', 'svn')])
     return templates
 
 def _shrinktext(text, subfunc):
     '''Helper for keyword expansion removal in text.
     Depending on subfunc also returns number of substitutions.'''
     return subfunc(r'$\1$', text)
 
 def _preselect(wstatus, changed):
     '''Retrieves modified and added files from a working directory state
     and returns the subset of each contained in given changed files
     retrieved from a change context.'''
     modified = [f for f in wstatus.modified if f in changed]
     added = [f for f in wstatus.added if f in changed]
     return modified, added
 
 
 class kwtemplater(object):
     '''
     Sets up keyword templates, corresponding keyword regex, and
     provides keyword substitution functions.
     '''
 
     def __init__(self, ui, repo, inc, exc):
         self.ui = ui
         self._repo = weakref.ref(repo)
         self.match = match.match(repo.root, '', [], inc, exc)
         self.restrict = kwtools['hgcmd'] in restricted.split()
         self.postcommit = False
 
         kwmaps = self.ui.configitems('keywordmaps')
         if kwmaps: # override default templates
             self.templates = dict(kwmaps)
         else:
             self.templates = _defaultkwmaps(self.ui)
 
     @property
     def repo(self):
         return self._repo()
 
     @util.propertycache
     def escape(self):
         '''Returns bar-separated and escaped keywords.'''
-        return '|'.join(map(re.escape, self.templates.keys()))
+        return '|'.join(map(stringutil.reescape, self.templates.keys()))
 
     @util.propertycache
     def rekw(self):
         '''Returns regex for unexpanded keywords.'''
         return re.compile(r'\$(%s)\$' % self.escape)
 
     @util.propertycache
     def rekwexp(self):
         '''Returns regex for expanded keywords.'''
         return re.compile(r'\$(%s): [^$\n\r]*? \$' % self.escape)
 
     def substitute(self, data, path, ctx, subfunc):
         '''Replaces keywords in data with expanded template.'''
         def kwsub(mobj):
             kw = mobj.group(1)
             ct = logcmdutil.maketemplater(self.ui, self.repo,
                                           self.templates[kw])
             self.ui.pushbuffer()
             ct.show(ctx, root=self.repo.root, file=path)
             ekw = templatefilters.firstline(self.ui.popbuffer())
             return '$%s: %s $' % (kw, ekw)
         return subfunc(kwsub, data)
 
     def linkctx(self, path, fileid):
         '''Similar to filelog.linkrev, but returns a changectx.'''
         return self.repo.filectx(path, fileid=fileid).changectx()
 
     def expand(self, path, node, data):
         '''Returns data with keywords expanded.'''
         if (not self.restrict and self.match(path)
             and not stringutil.binary(data)):
             ctx = self.linkctx(path, node)
             return self.substitute(data, path, ctx, self.rekw.sub)
         return data
 
     def iskwfile(self, cand, ctx):
         '''Returns subset of candidates which are configured for keyword
         expansion but are not symbolic links.'''
         return [f for f in cand if self.match(f) and 'l' not in ctx.flags(f)]
 
     def overwrite(self, ctx, candidates, lookup, expand, rekw=False):
         '''Overwrites selected files expanding/shrinking keywords.'''
         if self.restrict or lookup or self.postcommit: # exclude kw_copy
             candidates = self.iskwfile(candidates, ctx)
         if not candidates:
             return
         kwcmd = self.restrict and lookup # kwexpand/kwshrink
         if self.restrict or expand and lookup:
             mf = ctx.manifest()
         if self.restrict or rekw:
             re_kw = self.rekw
         else:
             re_kw = self.rekwexp
         if expand:
             msg = _('overwriting %s expanding keywords\n')
         else:
             msg = _('overwriting %s shrinking keywords\n')
         for f in candidates:
             if self.restrict:
                 data = self.repo.file(f).read(mf[f])
             else:
                 data = self.repo.wread(f)
             if stringutil.binary(data):
                 continue
             if expand:
                 parents = ctx.parents()
                 if lookup:
                     ctx = self.linkctx(f, mf[f])
                 elif self.restrict and len(parents) > 1:
                     # merge commit
                     # in case of conflict f is in modified state during
                     # merge, even if f does not differ from f in parent
                     for p in parents:
                         if f in p and not p[f].cmp(ctx[f]):
                             ctx = p[f].changectx()
                             break
                 data, found = self.substitute(data, f, ctx, re_kw.subn)
             elif self.restrict:
                 found = re_kw.search(data)
             else:
                 data, found = _shrinktext(data, re_kw.subn)
             if found:
                 self.ui.note(msg % f)
                 fp = self.repo.wvfs(f, "wb", atomictemp=True)
                 fp.write(data)
                 fp.close()
                 if kwcmd:
                     self.repo.dirstate.normal(f)
                 elif self.postcommit:
                     self.repo.dirstate.normallookup(f)
 
     def shrink(self, fname, text):
         '''Returns text with all keyword substitutions removed.'''
         if self.match(fname) and not stringutil.binary(text):
             return _shrinktext(text, self.rekwexp.sub)
         return text
 
     def shrinklines(self, fname, lines):
         '''Returns lines with keyword substitutions removed.'''
         if self.match(fname):
             text = ''.join(lines)
             if not stringutil.binary(text):
                 return _shrinktext(text, self.rekwexp.sub).splitlines(True)
         return lines
 
     def wread(self, fname, data):
         '''If in restricted mode returns data read from wdir with
         keyword substitutions removed.'''
         if self.restrict:
             return self.shrink(fname, data)
         return data
 
 class kwfilelog(filelog.filelog):
     '''
     Subclass of filelog to hook into its read, add, cmp methods.
     Keywords are "stored" unexpanded, and processed on reading.
     '''
     def __init__(self, opener, kwt, path):
         super(kwfilelog, self).__init__(opener, path)
         self.kwt = kwt
         self.path = path
 
     def read(self, node):
         '''Expands keywords when reading filelog.'''
         data = super(kwfilelog, self).read(node)
         if self.renamed(node):
             return data
         return self.kwt.expand(self.path, node, data)
 
     def add(self, text, meta, tr, link, p1=None, p2=None):
         '''Removes keyword substitutions when adding to filelog.'''
         text = self.kwt.shrink(self.path, text)
         return super(kwfilelog, self).add(text, meta, tr, link, p1, p2)
 
     def cmp(self, node, text):
         '''Removes keyword substitutions for comparison.'''
         text = self.kwt.shrink(self.path, text)
         return super(kwfilelog, self).cmp(node, text)
 
 def _status(ui, repo, wctx, kwt, *pats, **opts):
     '''Bails out if [keyword] configuration is not active.
     Returns status of working directory.'''
     if kwt:
         opts = pycompat.byteskwargs(opts)
         return repo.status(match=scmutil.match(wctx, pats, opts), clean=True,
                            unknown=opts.get('unknown') or opts.get('all'))
     if ui.configitems('keyword'):
         raise error.Abort(_('[keyword] patterns cannot match'))
     raise error.Abort(_('no [keyword] patterns configured'))
 
 def _kwfwrite(ui, repo, expand, *pats, **opts):
     '''Selects files and passes them to kwtemplater.overwrite.'''
     wctx = repo[None]
     if len(wctx.parents()) > 1:
         raise error.Abort(_('outstanding uncommitted merge'))
     kwt = getattr(repo, '_keywordkwt', None)
     with repo.wlock():
         status = _status(ui, repo, wctx, kwt, *pats, **opts)
         if status.modified or status.added or status.removed or status.deleted:
             raise error.Abort(_('outstanding uncommitted changes'))
         kwt.overwrite(wctx, status.clean, True, expand)
 
 @command('kwdemo',
          [('d', 'default', None, _('show default keyword template maps')),
           ('f', 'rcfile', '',
            _('read maps from rcfile'), _('FILE'))],
          _('hg kwdemo [-d] [-f RCFILE] [TEMPLATEMAP]...'),
          optionalrepo=True)
 def demo(ui, repo, *args, **opts):
     '''print [keywordmaps] configuration and an expansion example
 
     Show current, custom, or default keyword template maps and their
     expansions.
 
     Extend the current configuration by specifying maps as arguments
     and using -f/--rcfile to source an external hgrc file.
 
     Use -d/--default to disable current configuration.
 
     See :hg:`help templates` for information on templates and filters.
     '''
     def demoitems(section, items):
         ui.write('[%s]\n' % section)
         for k, v in sorted(items):
             ui.write('%s = %s\n' % (k, v))
 
     fn = 'demo.txt'
     tmpdir = pycompat.mkdtemp('', 'kwdemo.')
     ui.note(_('creating temporary repository at %s\n') % tmpdir)
     if repo is None:
         baseui = ui
     else:
         baseui = repo.baseui
     repo = localrepo.localrepository(baseui, tmpdir, True)
     ui.setconfig('keyword', fn, '', 'keyword')
     svn = ui.configbool('keywordset', 'svn')
     # explicitly set keywordset for demo output
     ui.setconfig('keywordset', 'svn', svn, 'keyword')
 
     uikwmaps = ui.configitems('keywordmaps')
     if args or opts.get(r'rcfile'):
         ui.status(_('\n\tconfiguration using custom keyword template maps\n'))
         if uikwmaps:
             ui.status(_('\textending current template maps\n'))
         if opts.get(r'default') or not uikwmaps:
             if svn:
                 ui.status(_('\toverriding default svn keywordset\n'))
             else:
                 ui.status(_('\toverriding default cvs keywordset\n'))
         if opts.get(r'rcfile'):
             ui.readconfig(opts.get('rcfile'))
         if args:
             # simulate hgrc parsing
             rcmaps = '[keywordmaps]\n%s\n' % '\n'.join(args)
             repo.vfs.write('hgrc', rcmaps)
             ui.readconfig(repo.vfs.join('hgrc'))
         kwmaps = dict(ui.configitems('keywordmaps'))
     elif opts.get(r'default'):
         if svn:
             ui.status(_('\n\tconfiguration using default svn keywordset\n'))
         else:
             ui.status(_('\n\tconfiguration using default cvs keywordset\n'))
         kwmaps = _defaultkwmaps(ui)
         if uikwmaps:
             ui.status(_('\tdisabling current template maps\n'))
             for k, v in kwmaps.iteritems():
                 ui.setconfig('keywordmaps', k, v, 'keyword')
     else:
         ui.status(_('\n\tconfiguration using current keyword template maps\n'))
         if uikwmaps:
             kwmaps = dict(uikwmaps)
         else:
             kwmaps = _defaultkwmaps(ui)
 
     uisetup(ui)
     reposetup(ui, repo)
     ui.write(('[extensions]\nkeyword =\n'))
     demoitems('keyword', ui.configitems('keyword'))
     demoitems('keywordset', ui.configitems('keywordset'))
     demoitems('keywordmaps', kwmaps.iteritems())
     keywords = '$' + '$\n$'.join(sorted(kwmaps.keys())) + '$\n'
     repo.wvfs.write(fn, keywords)
     repo[None].add([fn])
     ui.note(_('\nkeywords written to %s:\n') % fn)
     ui.note(keywords)
     with repo.wlock():
         repo.dirstate.setbranch('demobranch')
     for name, cmd in ui.configitems('hooks'):
         if name.split('.', 1)[0].find('commit') > -1:
             repo.ui.setconfig('hooks', name, '', 'keyword')
     msg = _('hg keyword configuration and expansion example')
     ui.note(("hg ci -m '%s'\n" % msg))
     repo.commit(text=msg)
     ui.status(_('\n\tkeywords expanded\n'))
     ui.write(repo.wread(fn))
     repo.wvfs.rmtree(repo.root)
 
 @command('kwexpand',
     cmdutil.walkopts,
     _('hg kwexpand [OPTION]... [FILE]...'),
     inferrepo=True)
 def expand(ui, repo, *pats, **opts):
     '''expand keywords in the working directory
 
     Run after (re)enabling keyword expansion.
 
     kwexpand refuses to run if given files contain local changes.
     '''
     # 3rd argument sets expansion to True
     _kwfwrite(ui, repo, True, *pats, **opts)
 
 @command('kwfiles',
          [('A', 'all', None, _('show keyword status flags of all files')),
           ('i', 'ignore', None, _('show files excluded from expansion')),
           ('u', 'unknown', None, _('only show unknown (not tracked) files')),
          ] + cmdutil.walkopts,
          _('hg kwfiles [OPTION]... [FILE]...'),
          inferrepo=True)
 def files(ui, repo, *pats, **opts):
     '''show files configured for keyword expansion
 
     List which files in the working directory are matched by the
     [keyword] configuration patterns.
 
     Useful to prevent inadvertent keyword expansion and to speed up
     execution by including only files that are actual candidates for
     expansion.
 
     See :hg:`help keyword` on how to construct patterns both for
     inclusion and exclusion of files.
 
     With -A/--all and -v/--verbose the codes used to show the status
     of files are::
 
       K = keyword expansion candidate
       k = keyword expansion candidate (not tracked)
       I = ignored
       i = ignored (not tracked)
     '''
     kwt = getattr(repo, '_keywordkwt', None)
     wctx = repo[None]
     status = _status(ui, repo, wctx, kwt, *pats, **opts)
     if pats:
         cwd = repo.getcwd()
     else:
         cwd = ''
     files = []
     opts = pycompat.byteskwargs(opts)
     if not opts.get('unknown') or opts.get('all'):
         files = sorted(status.modified + status.added + status.clean)
     kwfiles = kwt.iskwfile(files, wctx)
     kwdeleted = kwt.iskwfile(status.deleted, wctx)
     kwunknown = kwt.iskwfile(status.unknown, wctx)
     if not opts.get('ignore') or opts.get('all'):
         showfiles = kwfiles, kwdeleted, kwunknown
     else:
         showfiles = [], [], []
     if opts.get('all') or opts.get('ignore'):
         showfiles += ([f for f in files if f not in kwfiles],
                       [f for f in status.unknown if f not in kwunknown])
     kwlabels = 'enabled deleted enabledunknown ignored ignoredunknown'.split()
     kwstates = zip(kwlabels, 'K!kIi', showfiles)
     fm = ui.formatter('kwfiles', opts)
     fmt = '%.0s%s\n'
     if opts.get('all') or ui.verbose:
         fmt = '%s %s\n'
     for kwstate, char, filenames in kwstates:
         label = 'kwfiles.' + kwstate
         for f in filenames:
             fm.startitem()
             fm.write('kwstatus path', fmt, char,
                      repo.pathto(f, cwd), label=label)
     fm.end()
 
 @command('kwshrink',
     cmdutil.walkopts,
     _('hg kwshrink [OPTION]... [FILE]...'),
     inferrepo=True)
 def shrink(ui, repo, *pats, **opts):
     '''revert expanded keywords in the working directory
 
     Must be run before changing/disabling active keywords.
 
     kwshrink refuses to run if given files contain local changes.
     '''
     # 3rd argument sets expansion to False
     _kwfwrite(ui, repo, False, *pats, **opts)
 
 # monkeypatches
 
 def kwpatchfile_init(orig, self, ui, gp, backend, store, eolmode=None):
     '''Monkeypatch/wrap patch.patchfile.__init__ to avoid
     rejects or conflicts due to expanded keywords in working dir.'''
     orig(self, ui, gp, backend, store, eolmode)
     kwt = getattr(getattr(backend, 'repo', None), '_keywordkwt', None)
     if kwt:
         # shrink keywords read from working dir
         self.lines = kwt.shrinklines(self.fname, self.lines)
 
 def kwdiff(orig, repo, *args, **kwargs):
     '''Monkeypatch patch.diff to avoid expansion.'''
     kwt = getattr(repo, '_keywordkwt', None)
     if kwt:
         restrict = kwt.restrict
         kwt.restrict = True
     try:
         for chunk in orig(repo, *args, **kwargs):
             yield chunk
     finally:
         if kwt:
             kwt.restrict = restrict
 
 def kwweb_skip(orig, web):
     '''Wraps webcommands.x turning off keyword expansion.'''
     kwt = getattr(web.repo, '_keywordkwt', None)
     if kwt:
         origmatch = kwt.match
         kwt.match = util.never
     try:
         for chunk in orig(web):
             yield chunk
     finally:
         if kwt:
             kwt.match = origmatch
 
 def kw_amend(orig, ui, repo, old, extra, pats, opts):
     '''Wraps cmdutil.amend expanding keywords after amend.'''
     kwt = getattr(repo, '_keywordkwt', None)
     if kwt is None:
         return orig(ui, repo, old, extra, pats, opts)
     with repo.wlock():
         kwt.postcommit = True
         newid = orig(ui, repo, old, extra, pats, opts)
         if newid != old.node():
             ctx = repo[newid]
             kwt.restrict = True
             kwt.overwrite(ctx, ctx.files(), False, True)
             kwt.restrict = False
         return newid
 
 def kw_copy(orig, ui, repo, pats, opts, rename=False):
     '''Wraps cmdutil.copy so that copy/rename destinations do not
     contain expanded keywords.
     Note that the source of a regular file destination may also be a
     symlink:
     hg cp sym x                -> x is symlink
     cp sym x; hg cp -A sym x   -> x is file (maybe expanded keywords)
     For the latter we have to follow the symlink to find out whether its
     target is configured for expansion and we therefore must unexpand the
     keywords in the destination.'''
     kwt = getattr(repo, '_keywordkwt', None)
     if kwt is None:
         return orig(ui, repo, pats, opts, rename)
     with repo.wlock():
         orig(ui, repo, pats, opts, rename)
         if opts.get('dry_run'):
             return
         wctx = repo[None]
         cwd = repo.getcwd()
 
         def haskwsource(dest):
             '''Returns true if dest is a regular file and configured for
             expansion or a symlink which points to a file configured for
             expansion. '''
             source = repo.dirstate.copied(dest)
             if 'l' in wctx.flags(source):
                 source = pathutil.canonpath(repo.root, cwd,
                                            os.path.realpath(source))
             return kwt.match(source)
 
         candidates = [f for f in repo.dirstate.copies() if
                       'l' not in wctx.flags(f) and haskwsource(f)]
         kwt.overwrite(wctx, candidates, False, False)
 
 def kw_dorecord(orig, ui, repo, commitfunc, *pats, **opts):
     '''Wraps record.dorecord expanding keywords after recording.'''
     kwt = getattr(repo, '_keywordkwt', None)
     if kwt is None:
         return orig(ui, repo, commitfunc, *pats, **opts)
     with repo.wlock():
         # record returns 0 even when nothing has changed
         # therefore compare nodes before and after
         kwt.postcommit = True
         ctx = repo['.']
         wstatus = ctx.status()
         ret = orig(ui, repo, commitfunc, *pats, **opts)
         recctx = repo['.']
         if ctx != recctx:
             modified, added = _preselect(wstatus, recctx.files())
             kwt.restrict = False
             kwt.overwrite(recctx, modified, False, True)
             kwt.overwrite(recctx, added, False, True, True)
             kwt.restrict = True
         return ret
 
 def kwfilectx_cmp(orig, self, fctx):
     if fctx._customcmp:
         return fctx.cmp(self)
     kwt = getattr(self._repo, '_keywordkwt', None)
     if kwt is None:
         return orig(self, fctx)
     # keyword affects data size, comparing wdir and filelog size does
     # not make sense
     if (fctx._filenode is None and
         (self._repo._encodefilterpats or
          kwt.match(fctx.path()) and 'l' not in fctx.flags() or
          self.size() - 4 == fctx.size()) or
         self.size() == fctx.size()):
         return self._filelog.cmp(self._filenode, fctx.data())
     return True
 
 def uisetup(ui):
     ''' Monkeypatches dispatch._parse to retrieve user command.
     Overrides file method to return kwfilelog instead of filelog
     if file matches user configuration.
     Wraps commit to overwrite configured files with updated
     keyword substitutions.
     Monkeypatches patch and webcommands.'''
 
     def kwdispatch_parse(orig, ui, args):
         '''Monkeypatch dispatch._parse to obtain running hg command.'''
         cmd, func, args, options, cmdoptions = orig(ui, args)
         kwtools['hgcmd'] = cmd
         return cmd, func, args, options, cmdoptions
 
     extensions.wrapfunction(dispatch, '_parse', kwdispatch_parse)
 
     extensions.wrapfunction(context.filectx, 'cmp', kwfilectx_cmp)
     extensions.wrapfunction(patch.patchfile, '__init__', kwpatchfile_init)
     extensions.wrapfunction(patch, 'diff', kwdiff)
     extensions.wrapfunction(cmdutil, 'amend', kw_amend)
     extensions.wrapfunction(cmdutil, 'copy', kw_copy)
     extensions.wrapfunction(cmdutil, 'dorecord', kw_dorecord)
     for c in nokwwebcommands.split():
         extensions.wrapfunction(webcommands, c, kwweb_skip)
 
 def reposetup(ui, repo):
     '''Sets up repo as kwrepo for keyword substitution.'''
 
     try:
         if (not repo.local() or kwtools['hgcmd'] in nokwcommands.split()
             or '.hg' in util.splitpath(repo.root)
             or repo._url.startswith('bundle:')):
             return
     except AttributeError:
         pass
 
     inc, exc = [], ['.hg*']
     for pat, opt in ui.configitems('keyword'):
         if opt != 'ignore':
             inc.append(pat)
         else:
             exc.append(pat)
     if not inc:
         return
 
     kwt = kwtemplater(ui, repo, inc, exc)
 
     class kwrepo(repo.__class__):
         def file(self, f):
             if f[0] == '/':
                 f = f[1:]
             return kwfilelog(self.svfs, kwt, f)
 
         def wread(self, filename):
             data = super(kwrepo, self).wread(filename)
             return kwt.wread(filename, data)
 
         def commit(self, *args, **opts):
             # use custom commitctx for user commands
             # other extensions can still wrap repo.commitctx directly
             self.commitctx = self.kwcommitctx
             try:
                 return super(kwrepo, self).commit(*args, **opts)
             finally:
                 del self.commitctx
 
         def kwcommitctx(self, ctx, error=False):
             n = super(kwrepo, self).commitctx(ctx, error)
             # no lock needed, only called from repo.commit() which already locks
             if not kwt.postcommit:
                 restrict = kwt.restrict
                 kwt.restrict = True
                 kwt.overwrite(self[n], sorted(ctx.added() + ctx.modified()),
                               False, True)
                 kwt.restrict = restrict
             return n
 
         def rollback(self, dryrun=False, force=False):
             with self.wlock():
                 origrestrict = kwt.restrict
                 try:
                     if not dryrun:
                         changed = self['.'].files()
                     ret = super(kwrepo, self).rollback(dryrun, force)
                     if not dryrun:
                         ctx = self['.']
                         modified, added = _preselect(ctx.status(), changed)
                         kwt.restrict = False
                         kwt.overwrite(ctx, modified, True, True)
                         kwt.overwrite(ctx, added, True, False)
                     return ret
                 finally:
                     kwt.restrict = origrestrict
 
     repo.__class__ = kwrepo
     repo._keywordkwt = kwt

hgext/lfs/pointer.py

@@ -1,83 +1,83
 # pointer.py - Git-LFS pointer serialization
 #
 # Copyright 2017 Facebook, Inc.
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import re
 
 from mercurial.i18n import _
 
 from mercurial import (
     error,
     pycompat,
 )
 from mercurial.utils import (
     stringutil,
 )
 
 class InvalidPointer(error.RevlogError):
     pass
 
 class gitlfspointer(dict):
     VERSION = 'https://git-lfs.github.com/spec/v1'
 
     def __init__(self, *args, **kwargs):
         self['version'] = self.VERSION
         super(gitlfspointer, self).__init__(*args)
         self.update(pycompat.byteskwargs(kwargs))
 
     @classmethod
     def deserialize(cls, text):
         try:
             return cls(l.split(' ', 1) for l in text.splitlines()).validate()
         except ValueError: # l.split returns 1 item instead of 2
             raise InvalidPointer(_('cannot parse git-lfs text: %s')
                                  % stringutil.pprint(text))
 
     def serialize(self):
         sortkeyfunc = lambda x: (x[0] != 'version', x)
         items = sorted(self.validate().iteritems(), key=sortkeyfunc)
         return ''.join('%s %s\n' % (k, v) for k, v in items)
 
     def oid(self):
         return self['oid'].split(':')[-1]
 
     def size(self):
         return int(self['size'])
 
     # regular expressions used by _validate
     # see https://github.com/git-lfs/git-lfs/blob/master/docs/spec.md
     _keyre = re.compile(br'\A[a-z0-9.-]+\Z')
     _valuere = re.compile(br'\A[^\n]*\Z')
     _requiredre = {
         'size': re.compile(br'\A[0-9]+\Z'),
         'oid': re.compile(br'\Asha256:[0-9a-f]{64}\Z'),
-        'version': re.compile(br'\A%s\Z' % re.escape(VERSION)),
+        'version': re.compile(br'\A%s\Z' % stringutil.reescape(VERSION)),
     }
 
     def validate(self):
         """raise InvalidPointer on error. return self if there is no error"""
         requiredcount = 0
         for k, v in self.iteritems():
             if k in self._requiredre:
                 if not self._requiredre[k].match(v):
                     raise InvalidPointer(
                         _('unexpected lfs pointer value: %s=%s')
                         % (k, stringutil.pprint(v)))
                 requiredcount += 1
             elif not self._keyre.match(k):
                 raise InvalidPointer(_('unexpected lfs pointer key: %s') % k)
             if not self._valuere.match(v):
                 raise InvalidPointer(_('unexpected lfs pointer value: %s=%s')
                                      % (k, stringutil.pprint(v)))
         if len(self._requiredre) != requiredcount:
             miss = sorted(set(self._requiredre.keys()).difference(self.keys()))
             raise InvalidPointer(_('missing lfs pointer keys: %s')
                                  % ', '.join(miss))
         return self
 
 deserialize = gitlfspointer.deserialize

mercurial/hgweb/webutil.py

@@ -1,804 +1,804
 # hgweb/webutil.py - utility library for the web interface.
 #
 # Copyright 21 May 2005 - (c) 2005 Jake Edge <[e-mail]>
 # Copyright 2005-2007 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import copy
 import difflib
 import os
 import re
 
 from ..i18n import _
 from ..node import hex, nullid, short
 
 from .common import (
     ErrorResponse,
     HTTP_BAD_REQUEST,
     HTTP_NOT_FOUND,
     paritygen,
 )
 
 from .. import (
     context,
     error,
     match,
     mdiff,
     obsutil,
     patch,
     pathutil,
     pycompat,
     scmutil,
     templatefilters,
     templatekw,
     templateutil,
     ui as uimod,
     util,
 )
 
 from ..utils import (
     stringutil,
 )
 
 archivespecs = util.sortdict((
     ('zip', ('application/zip', 'zip', '.zip', None)),
     ('gz', ('application/x-gzip', 'tgz', '.tar.gz', None)),
     ('bz2', ('application/x-bzip2', 'tbz2', '.tar.bz2', None)),
 ))
 
 def archivelist(ui, nodeid, url=None):
     allowed = ui.configlist('web', 'allow-archive', untrusted=True)
     archives = []
 
     for typ, spec in archivespecs.iteritems():
         if typ in allowed or ui.configbool('web', 'allow' + typ,
                                            untrusted=True):
             archives.append({
                 'type': typ,
                 'extension': spec[2],
                 'node': nodeid,
                 'url': url,
             })
 
     return templateutil.mappinglist(archives)
 
 def up(p):
     if p[0:1] != "/":
         p = "/" + p
     if p[-1:] == "/":
         p = p[:-1]
     up = os.path.dirname(p)
     if up == "/":
         return "/"
     return up + "/"
 
 def _navseq(step, firststep=None):
     if firststep:
         yield firststep
         if firststep >= 20 and firststep <= 40:
             firststep = 50
             yield firststep
         assert step > 0
         assert firststep > 0
         while step <= firststep:
             step *= 10
     while True:
         yield 1 * step
         yield 3 * step
         step *= 10
 
 class revnav(object):
 
     def __init__(self, repo):
         """Navigation generation object
 
         :repo: repo object we generate nav for
         """
         # used for hex generation
         self._revlog = repo.changelog
 
     def __nonzero__(self):
         """return True if any revision to navigate over"""
         return self._first() is not None
 
     __bool__ = __nonzero__
 
     def _first(self):
         """return the minimum non-filtered changeset or None"""
         try:
             return next(iter(self._revlog))
         except StopIteration:
             return None
 
     def hex(self, rev):
         return hex(self._revlog.node(rev))
 
     def gen(self, pos, pagelen, limit):
         """computes label and revision id for navigation link
 
         :pos: is the revision relative to which we generate navigation.
         :pagelen: the size of each navigation page
         :limit: how far shall we link
 
         The return is:
             - a single element mappinglist
             - containing a dictionary with a `before` and `after` key
             - values are dictionaries with `label` and `node` keys
         """
         if not self:
             # empty repo
             return templateutil.mappinglist([
                 {'before': templateutil.mappinglist([]),
                  'after': templateutil.mappinglist([])},
             ])
 
         targets = []
         for f in _navseq(1, pagelen):
             if f > limit:
                 break
             targets.append(pos + f)
             targets.append(pos - f)
         targets.sort()
 
         first = self._first()
         navbefore = [{'label': '(%i)' % first, 'node': self.hex(first)}]
         navafter = []
         for rev in targets:
             if rev not in self._revlog:
                 continue
             if pos < rev < limit:
                 navafter.append({'label': '+%d' % abs(rev - pos),
                                  'node': self.hex(rev)})
             if 0 < rev < pos:
                 navbefore.append({'label': '-%d' % abs(rev - pos),
                                   'node': self.hex(rev)})
 
         navafter.append({'label': 'tip', 'node': 'tip'})
 
         # TODO: maybe this can be a scalar object supporting tomap()
         return templateutil.mappinglist([
             {'before': templateutil.mappinglist(navbefore),
              'after': templateutil.mappinglist(navafter)},
         ])
 
 class filerevnav(revnav):
 
     def __init__(self, repo, path):
         """Navigation generation object
 
         :repo: repo object we generate nav for
         :path: path of the file we generate nav for
         """
         # used for iteration
         self._changelog = repo.unfiltered().changelog
         # used for hex generation
         self._revlog = repo.file(path)
 
     def hex(self, rev):
         return hex(self._changelog.node(self._revlog.linkrev(rev)))
 
 # TODO: maybe this can be a wrapper class for changectx/filectx list, which
 # yields {'ctx': ctx}
 def _ctxsgen(context, ctxs):
     for s in ctxs:
         d = {
             'node': s.hex(),
             'rev': s.rev(),
             'user': s.user(),
             'date': s.date(),
             'description': s.description(),
             'branch': s.branch(),
         }
         if util.safehasattr(s, 'path'):
             d['file'] = s.path()
         yield d
 
 def _siblings(siblings=None, hiderev=None):
     if siblings is None:
         siblings = []
     siblings = [s for s in siblings if s.node() != nullid]
     if len(siblings) == 1 and siblings[0].rev() == hiderev:
         siblings = []
     return templateutil.mappinggenerator(_ctxsgen, args=(siblings,))
 
 def difffeatureopts(req, ui, section):
     diffopts = patch.difffeatureopts(ui, untrusted=True,
                                      section=section, whitespace=True)
 
     for k in ('ignorews', 'ignorewsamount', 'ignorewseol', 'ignoreblanklines'):
         v = req.qsparams.get(k)
         if v is not None:
             v = stringutil.parsebool(v)
             setattr(diffopts, k, v if v is not None else True)
 
     return diffopts
 
 def annotate(req, fctx, ui):
     diffopts = difffeatureopts(req, ui, 'annotate')
     return fctx.annotate(follow=True, diffopts=diffopts)
 
 def parents(ctx, hide=None):
     if isinstance(ctx, context.basefilectx):
         introrev = ctx.introrev()
         if ctx.changectx().rev() != introrev:
             return _siblings([ctx.repo()[introrev]], hide)
     return _siblings(ctx.parents(), hide)
 
 def children(ctx, hide=None):
     return _siblings(ctx.children(), hide)
 
 def renamelink(fctx):
     r = fctx.renamed()
     if r:
         return templateutil.mappinglist([{'file': r[0], 'node': hex(r[1])}])
     return templateutil.mappinglist([])
 
 def nodetagsdict(repo, node):
     return templateutil.hybridlist(repo.nodetags(node), name='name')
 
 def nodebookmarksdict(repo, node):
     return templateutil.hybridlist(repo.nodebookmarks(node), name='name')
 
 def nodebranchdict(repo, ctx):
     branches = []
     branch = ctx.branch()
     # If this is an empty repo, ctx.node() == nullid,
     # ctx.branch() == 'default'.
     try:
         branchnode = repo.branchtip(branch)
     except error.RepoLookupError:
         branchnode = None
     if branchnode == ctx.node():
         branches.append(branch)
     return templateutil.hybridlist(branches, name='name')
 
 def nodeinbranch(repo, ctx):
     branches = []
     branch = ctx.branch()
     try:
         branchnode = repo.branchtip(branch)
     except error.RepoLookupError:
         branchnode = None
     if branch != 'default' and branchnode != ctx.node():
         branches.append(branch)
     return templateutil.hybridlist(branches, name='name')
 
 def nodebranchnodefault(ctx):
     branches = []
     branch = ctx.branch()
     if branch != 'default':
         branches.append(branch)
     return templateutil.hybridlist(branches, name='name')
 
 def _nodenamesgen(context, f, node, name):
     for t in f(node):
         yield {name: t}
 
 def showtag(repo, t1, node=nullid):
     args = (repo.nodetags, node, 'tag')
     return templateutil.mappinggenerator(_nodenamesgen, args=args, name=t1)
 
 def showbookmark(repo, t1, node=nullid):
     args = (repo.nodebookmarks, node, 'bookmark')
     return templateutil.mappinggenerator(_nodenamesgen, args=args, name=t1)
 
 def branchentries(repo, stripecount, limit=0):
     tips = []
     heads = repo.heads()
     parity = paritygen(stripecount)
     sortkey = lambda item: (not item[1], item[0].rev())
 
     def entries(context):
         count = 0
         if not tips:
             for tag, hs, tip, closed in repo.branchmap().iterbranches():
                 tips.append((repo[tip], closed))
         for ctx, closed in sorted(tips, key=sortkey, reverse=True):
             if limit > 0 and count >= limit:
                 return
             count += 1
             if closed:
                 status = 'closed'
             elif ctx.node() not in heads:
                 status = 'inactive'
             else:
                 status = 'open'
             yield {
                 'parity': next(parity),
                 'branch': ctx.branch(),
                 'status': status,
                 'node': ctx.hex(),
                 'date': ctx.date()
             }
 
     return templateutil.mappinggenerator(entries)
 
 def cleanpath(repo, path):
     path = path.lstrip('/')
     return pathutil.canonpath(repo.root, '', path)
 
 def changectx(repo, req):
     changeid = "tip"
     if 'node' in req.qsparams:
         changeid = req.qsparams['node']
         ipos = changeid.find(':')
         if ipos != -1:
             changeid = changeid[(ipos + 1):]
 
     return scmutil.revsymbol(repo, changeid)
 
 def basechangectx(repo, req):
     if 'node' in req.qsparams:
         changeid = req.qsparams['node']
         ipos = changeid.find(':')
         if ipos != -1:
             changeid = changeid[:ipos]
             return scmutil.revsymbol(repo, changeid)
 
     return None
 
 def filectx(repo, req):
     if 'file' not in req.qsparams:
         raise ErrorResponse(HTTP_NOT_FOUND, 'file not given')
     path = cleanpath(repo, req.qsparams['file'])
     if 'node' in req.qsparams:
         changeid = req.qsparams['node']
     elif 'filenode' in req.qsparams:
         changeid = req.qsparams['filenode']
     else:
         raise ErrorResponse(HTTP_NOT_FOUND, 'node or filenode not given')
     try:
         fctx = scmutil.revsymbol(repo, changeid)[path]
     except error.RepoError:
         fctx = repo.filectx(path, fileid=changeid)
 
     return fctx
 
 def linerange(req):
     linerange = req.qsparams.getall('linerange')
     if not linerange:
         return None
     if len(linerange) > 1:
         raise ErrorResponse(HTTP_BAD_REQUEST,
                             'redundant linerange parameter')
     try:
         fromline, toline = map(int, linerange[0].split(':', 1))
     except ValueError:
         raise ErrorResponse(HTTP_BAD_REQUEST,
                             'invalid linerange parameter')
     try:
         return util.processlinerange(fromline, toline)
     except error.ParseError as exc:
         raise ErrorResponse(HTTP_BAD_REQUEST, pycompat.bytestr(exc))
 
 def formatlinerange(fromline, toline):
     return '%d:%d' % (fromline + 1, toline)
 
 def _succsandmarkersgen(context, mapping):
     repo = context.resource(mapping, 'repo')
     itemmappings = templatekw.showsuccsandmarkers(context, mapping)
     for item in itemmappings.tovalue(context, mapping):
         item['successors'] = _siblings(repo[successor]
                                        for successor in item['successors'])
         yield item
 
 def succsandmarkers(context, mapping):
     return templateutil.mappinggenerator(_succsandmarkersgen, args=(mapping,))
 
 # teach templater succsandmarkers is switched to (context, mapping) API
 succsandmarkers._requires = {'repo', 'ctx'}
 
 def _whyunstablegen(context, mapping):
     repo = context.resource(mapping, 'repo')
     ctx = context.resource(mapping, 'ctx')
 
     entries = obsutil.whyunstable(repo, ctx)
     for entry in entries:
         if entry.get('divergentnodes'):
             entry['divergentnodes'] = _siblings(entry['divergentnodes'])
         yield entry
 
 def whyunstable(context, mapping):
     return templateutil.mappinggenerator(_whyunstablegen, args=(mapping,))
 
 whyunstable._requires = {'repo', 'ctx'}
 
 def commonentry(repo, ctx):
     node = ctx.node()
     return {
         # TODO: perhaps ctx.changectx() should be assigned if ctx is a
         # filectx, but I'm not pretty sure if that would always work because
         # fctx.parents() != fctx.changectx.parents() for example.
         'ctx': ctx,
         'rev': ctx.rev(),
         'node': hex(node),
         'author': ctx.user(),
         'desc': ctx.description(),
         'date': ctx.date(),
         'extra': ctx.extra(),
         'phase': ctx.phasestr(),
         'obsolete': ctx.obsolete(),
         'succsandmarkers': succsandmarkers,
         'instabilities': templateutil.hybridlist(ctx.instabilities(),
                                                  name='instability'),
         'whyunstable': whyunstable,
         'branch': nodebranchnodefault(ctx),
         'inbranch': nodeinbranch(repo, ctx),
         'branches': nodebranchdict(repo, ctx),
         'tags': nodetagsdict(repo, node),
         'bookmarks': nodebookmarksdict(repo, node),
         'parent': lambda **x: parents(ctx),
         'child': lambda **x: children(ctx),
     }
 
 def changelistentry(web, ctx):
     '''Obtain a dictionary to be used for entries in a changelist.
 
     This function is called when producing items for the "entries" list passed
     to the "shortlog" and "changelog" templates.
     '''
     repo = web.repo
     rev = ctx.rev()
     n = ctx.node()
     showtags = showtag(repo, 'changelogtag', n)
     files = listfilediffs(ctx.files(), n, web.maxfiles)
 
     entry = commonentry(repo, ctx)
     entry.update(
         allparents=lambda **x: parents(ctx),
         parent=lambda **x: parents(ctx, rev - 1),
         child=lambda **x: children(ctx, rev + 1),
         changelogtag=showtags,
         files=files,
     )
     return entry
 
 def changelistentries(web, revs, maxcount, parityfn):
     """Emit up to N records for an iterable of revisions."""
     repo = web.repo
 
     count = 0
     for rev in revs:
         if count >= maxcount:
             break
 
         count += 1
 
         entry = changelistentry(web, repo[rev])
         entry['parity'] = next(parityfn)
 
         yield entry
 
 def symrevorshortnode(req, ctx):
     if 'node' in req.qsparams:
         return templatefilters.revescape(req.qsparams['node'])
     else:
         return short(ctx.node())
 
 def _listfilesgen(context, ctx, stripecount):
     parity = paritygen(stripecount)
     for blockno, f in enumerate(ctx.files()):
         template = 'filenodelink' if f in ctx else 'filenolink'
         yield context.process(template, {
             'node': ctx.hex(),
             'file': f,
             'blockno': blockno + 1,
             'parity': next(parity),
         })
 
 def changesetentry(web, ctx):
     '''Obtain a dictionary to be used to render the "changeset" template.'''
 
     showtags = showtag(web.repo, 'changesettag', ctx.node())
     showbookmarks = showbookmark(web.repo, 'changesetbookmark', ctx.node())
     showbranch = nodebranchnodefault(ctx)
 
     basectx = basechangectx(web.repo, web.req)
     if basectx is None:
         basectx = ctx.p1()
 
     style = web.config('web', 'style')
     if 'style' in web.req.qsparams:
         style = web.req.qsparams['style']
 
     diff = diffs(web, ctx, basectx, None, style)
 
     parity = paritygen(web.stripecount)
     diffstatsgen = diffstatgen(ctx, basectx)
     diffstats = diffstat(ctx, diffstatsgen, parity)
 
     return dict(
         diff=diff,
         symrev=symrevorshortnode(web.req, ctx),
         basenode=basectx.hex(),
         changesettag=showtags,
         changesetbookmark=showbookmarks,
         changesetbranch=showbranch,
         files=templateutil.mappedgenerator(_listfilesgen,
                                            args=(ctx, web.stripecount)),
         diffsummary=lambda **x: diffsummary(diffstatsgen),
         diffstat=diffstats,
         archives=web.archivelist(ctx.hex()),
         **pycompat.strkwargs(commonentry(web.repo, ctx)))
 
 def _listfilediffsgen(context, files, node, max):
     for f in files[:max]:
         yield context.process('filedifflink', {'node': hex(node), 'file': f})
     if len(files) > max:
         yield context.process('fileellipses', {})
 
 def listfilediffs(files, node, max):
     return templateutil.mappedgenerator(_listfilediffsgen,
                                         args=(files, node, max))
 
 def _prettyprintdifflines(context, lines, blockno, lineidprefix):
     for lineno, l in enumerate(lines, 1):
         difflineno = "%d.%d" % (blockno, lineno)
         if l.startswith('+'):
             ltype = "difflineplus"
         elif l.startswith('-'):
             ltype = "difflineminus"
         elif l.startswith('@'):
             ltype = "difflineat"
         else:
             ltype = "diffline"
         yield context.process(ltype, {
             'line': l,
             'lineno': lineno,
             'lineid': lineidprefix + "l%s" % difflineno,
             'linenumber': "% 8s" % difflineno,
         })
 
 def _diffsgen(context, repo, ctx, basectx, files, style, stripecount,
               linerange, lineidprefix):
     if files:
         m = match.exact(repo.root, repo.getcwd(), files)
     else:
         m = match.always(repo.root, repo.getcwd())
 
     diffopts = patch.diffopts(repo.ui, untrusted=True)
     node1 = basectx.node()
     node2 = ctx.node()
     parity = paritygen(stripecount)
 
     diffhunks = patch.diffhunks(repo, node1, node2, m, opts=diffopts)
     for blockno, (fctx1, fctx2, header, hunks) in enumerate(diffhunks, 1):
         if style != 'raw':
             header = header[1:]
         lines = [h + '\n' for h in header]
         for hunkrange, hunklines in hunks:
             if linerange is not None and hunkrange is not None:
                 s1, l1, s2, l2 = hunkrange
                 if not mdiff.hunkinrange((s2, l2), linerange):
                     continue
             lines.extend(hunklines)
         if lines:
             l = templateutil.mappedgenerator(_prettyprintdifflines,
                                              args=(lines, blockno,
                                                    lineidprefix))
             yield {
                 'parity': next(parity),
                 'blockno': blockno,
                 'lines': l,
             }
 
 def diffs(web, ctx, basectx, files, style, linerange=None, lineidprefix=''):
     args = (web.repo, ctx, basectx, files, style, web.stripecount,
             linerange, lineidprefix)
     return templateutil.mappinggenerator(_diffsgen, args=args, name='diffblock')
 
 def _compline(type, leftlineno, leftline, rightlineno, rightline):
     lineid = leftlineno and ("l%d" % leftlineno) or ''
     lineid += rightlineno and ("r%d" % rightlineno) or ''
     llno = '%d' % leftlineno if leftlineno else ''
     rlno = '%d' % rightlineno if rightlineno else ''
     return {
         'type': type,
         'lineid': lineid,
         'leftlineno': leftlineno,
         'leftlinenumber': "% 6s" % llno,
         'leftline': leftline or '',
         'rightlineno': rightlineno,
         'rightlinenumber': "% 6s" % rlno,
         'rightline': rightline or '',
     }
 
 def _getcompblockgen(context, leftlines, rightlines, opcodes):
     for type, llo, lhi, rlo, rhi in opcodes:
         len1 = lhi - llo
         len2 = rhi - rlo
         count = min(len1, len2)
         for i in xrange(count):
             yield _compline(type=type,
                             leftlineno=llo + i + 1,
                             leftline=leftlines[llo + i],
                             rightlineno=rlo + i + 1,
                             rightline=rightlines[rlo + i])
         if len1 > len2:
             for i in xrange(llo + count, lhi):
                 yield _compline(type=type,
                                 leftlineno=i + 1,
                                 leftline=leftlines[i],
                                 rightlineno=None,
                                 rightline=None)
         elif len2 > len1:
             for i in xrange(rlo + count, rhi):
                 yield _compline(type=type,
                                 leftlineno=None,
                                 leftline=None,
                                 rightlineno=i + 1,
                                 rightline=rightlines[i])
 
 def _getcompblock(leftlines, rightlines, opcodes):
     args = (leftlines, rightlines, opcodes)
     return templateutil.mappinggenerator(_getcompblockgen, args=args,
                                          name='comparisonline')
 
 def _comparegen(context, contextnum, leftlines, rightlines):
     '''Generator function that provides side-by-side comparison data.'''
     s = difflib.SequenceMatcher(None, leftlines, rightlines)
     if contextnum < 0:
         l = _getcompblock(leftlines, rightlines, s.get_opcodes())
         yield {'lines': l}
     else:
         for oc in s.get_grouped_opcodes(n=contextnum):
             l = _getcompblock(leftlines, rightlines, oc)
             yield {'lines': l}
 
 def compare(contextnum, leftlines, rightlines):
     args = (contextnum, leftlines, rightlines)
     return templateutil.mappinggenerator(_comparegen, args=args,
                                          name='comparisonblock')
 
 def diffstatgen(ctx, basectx):
     '''Generator function that provides the diffstat data.'''
 
     stats = patch.diffstatdata(
         util.iterlines(ctx.diff(basectx, noprefix=False)))
     maxname, maxtotal, addtotal, removetotal, binary = patch.diffstatsum(stats)
     while True:
         yield stats, maxname, maxtotal, addtotal, removetotal, binary
 
 def diffsummary(statgen):
     '''Return a short summary of the diff.'''
 
     stats, maxname, maxtotal, addtotal, removetotal, binary = next(statgen)
     return _(' %d files changed, %d insertions(+), %d deletions(-)\n') % (
              len(stats), addtotal, removetotal)
 
 def _diffstattmplgen(context, ctx, statgen, parity):
     stats, maxname, maxtotal, addtotal, removetotal, binary = next(statgen)
     files = ctx.files()
 
     def pct(i):
         if maxtotal == 0:
             return 0
         return (float(i) / maxtotal) * 100
 
     fileno = 0
     for filename, adds, removes, isbinary in stats:
         template = 'diffstatlink' if filename in files else 'diffstatnolink'
         total = adds + removes
         fileno += 1
         yield context.process(template, {
             'node': ctx.hex(),
             'file': filename,
             'fileno': fileno,
             'total': total,
             'addpct': pct(adds),
             'removepct': pct(removes),
             'parity': next(parity),
         })
 
 def diffstat(ctx, statgen, parity):
     '''Return a diffstat template for each file in the diff.'''
     args = (ctx, statgen, parity)
     return templateutil.mappedgenerator(_diffstattmplgen, args=args)
 
 class sessionvars(templateutil.wrapped):
     def __init__(self, vars, start='?'):
         self._start = start
         self._vars = vars
 
     def __getitem__(self, key):
         return self._vars[key]
 
     def __setitem__(self, key, value):
         self._vars[key] = value
 
     def __copy__(self):
         return sessionvars(copy.copy(self._vars), self._start)
 
     def contains(self, context, mapping, item):
         item = templateutil.unwrapvalue(context, mapping, item)
         return item in self._vars
 
     def getmember(self, context, mapping, key):
         key = templateutil.unwrapvalue(context, mapping, key)
         return self._vars.get(key)
 
     def getmin(self, context, mapping):
         raise error.ParseError(_('not comparable'))
 
     def getmax(self, context, mapping):
         raise error.ParseError(_('not comparable'))
 
     def filter(self, context, mapping, select):
         # implement if necessary
         raise error.ParseError(_('not filterable'))
 
     def itermaps(self, context):
         separator = self._start
         for key, value in sorted(self._vars.iteritems()):
             yield {'name': key,
                    'value': pycompat.bytestr(value),
                    'separator': separator,
             }
             separator = '&'
 
     def join(self, context, mapping, sep):
         # could be '{separator}{name}={value|urlescape}'
         raise error.ParseError(_('not displayable without template'))
 
     def show(self, context, mapping):
         return self.join(context, '')
 
     def tobool(self, context, mapping):
         return bool(self._vars)
 
     def tovalue(self, context, mapping):
         return self._vars
 
 class wsgiui(uimod.ui):
     # default termwidth breaks under mod_wsgi
     def termwidth(self):
         return 80
 
 def getwebsubs(repo):
     websubtable = []
     websubdefs = repo.ui.configitems('websub')
     # we must maintain interhg backwards compatibility
     websubdefs += repo.ui.configitems('interhg')
     for key, pattern in websubdefs:
         # grab the delimiter from the character after the "s"
         unesc = pattern[1:2]
-        delim = re.escape(unesc)
+        delim = stringutil.reescape(unesc)
 
         # identify portions of the pattern, taking care to avoid escaped
         # delimiters. the replace format and flags are optional, but
         # delimiters are required.
         match = re.match(
             br'^s%s(.+)(?:(?<=\\\\)|(?<!\\))%s(.*)%s([ilmsux])*$'
             % (delim, delim, delim), pattern)
         if not match:
             repo.ui.warn(_("websub: invalid pattern for %s: %s\n")
                               % (key, pattern))
             continue
 
         # we need to unescape the delimiter for regexp and format
         delim_re = re.compile(br'(?<!\\)\\%s' % delim)
         regexp = delim_re.sub(unesc, match.group(1))
         format = delim_re.sub(unesc, match.group(2))
 
         # the pattern allows for 6 regexp flags, so set them if necessary
         flagin = match.group(3)
         flags = 0
         if flagin:
             for flag in flagin.upper():
                 flags |= re.__dict__[flag]
 
         try:
             regexp = re.compile(regexp, flags)
             websubtable.append((regexp, format))
         except re.error:
             repo.ui.warn(_("websub: invalid regexp for %s: %s\n")
                          % (key, regexp))
     return websubtable
 
 def getgraphnode(repo, ctx):
     return (templatekw.getgraphnodecurrent(repo, ctx) +
             templatekw.getgraphnodesymbol(ctx))

mercurial/match.py

@@ -1,1032 +1,1032
 # match.py - filename matching
 #
 #  Copyright 2008, 2009 Matt Mackall <[e-mail]> and others
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import, print_function
 
 import copy
 import os
 import re
 
 from .i18n import _
 from . import (
     encoding,
     error,
     pathutil,
     pycompat,
     util,
 )
 from .utils import (
     stringutil,
 )
 
 allpatternkinds = ('re', 'glob', 'path', 'relglob', 'relpath', 'relre',
                    'listfile', 'listfile0', 'set', 'include', 'subinclude',
                    'rootfilesin')
 cwdrelativepatternkinds = ('relpath', 'glob')
 
 propertycache = util.propertycache
 
 def _rematcher(regex):
     '''compile the regexp with the best available regexp engine and return a
     matcher function'''
     m = util.re.compile(regex)
     try:
         # slightly faster, provided by facebook's re2 bindings
         return m.test_match
     except AttributeError:
         return m.match
 
 def _expandsets(kindpats, ctx, listsubrepos):
     '''Returns the kindpats list with the 'set' patterns expanded.'''
     fset = set()
     other = []
 
     for kind, pat, source in kindpats:
         if kind == 'set':
             if not ctx:
                 raise error.ProgrammingError("fileset expression with no "
                                              "context")
             s = ctx.getfileset(pat)
             fset.update(s)
 
             if listsubrepos:
                 for subpath in ctx.substate:
                     s = ctx.sub(subpath).getfileset(pat)
                     fset.update(subpath + '/' + f for f in s)
 
             continue
         other.append((kind, pat, source))
     return fset, other
 
 def _expandsubinclude(kindpats, root):
     '''Returns the list of subinclude matcher args and the kindpats without the
     subincludes in it.'''
     relmatchers = []
     other = []
 
     for kind, pat, source in kindpats:
         if kind == 'subinclude':
             sourceroot = pathutil.dirname(util.normpath(source))
             pat = util.pconvert(pat)
             path = pathutil.join(sourceroot, pat)
 
             newroot = pathutil.dirname(path)
             matcherargs = (newroot, '', [], ['include:%s' % path])
 
             prefix = pathutil.canonpath(root, root, newroot)
             if prefix:
                 prefix += '/'
             relmatchers.append((prefix, matcherargs))
         else:
             other.append((kind, pat, source))
 
     return relmatchers, other
 
 def _kindpatsalwaysmatch(kindpats):
     """"Checks whether the kindspats match everything, as e.g.
     'relpath:.' does.
     """
     for kind, pat, source in kindpats:
         if pat != '' or kind not in ['relpath', 'glob']:
             return False
     return True
 
 def match(root, cwd, patterns=None, include=None, exclude=None, default='glob',
           exact=False, auditor=None, ctx=None, listsubrepos=False, warn=None,
           badfn=None, icasefs=False):
     """build an object to match a set of file patterns
 
     arguments:
     root - the canonical root of the tree you're matching against
     cwd - the current working directory, if relevant
     patterns - patterns to find
     include - patterns to include (unless they are excluded)
     exclude - patterns to exclude (even if they are included)
     default - if a pattern in patterns has no explicit type, assume this one
     exact - patterns are actually filenames (include/exclude still apply)
     warn - optional function used for printing warnings
     badfn - optional bad() callback for this matcher instead of the default
     icasefs - make a matcher for wdir on case insensitive filesystems, which
         normalizes the given patterns to the case in the filesystem
 
     a pattern is one of:
     'glob:<glob>' - a glob relative to cwd
     're:<regexp>' - a regular expression
     'path:<path>' - a path relative to repository root, which is matched
                     recursively
     'rootfilesin:<path>' - a path relative to repository root, which is
                     matched non-recursively (will not match subdirectories)
     'relglob:<glob>' - an unrooted glob (*.c matches C files in all dirs)
     'relpath:<path>' - a path relative to cwd
     'relre:<regexp>' - a regexp that needn't match the start of a name
     'set:<fileset>' - a fileset expression
     'include:<path>' - a file of patterns to read and include
     'subinclude:<path>' - a file of patterns to match against files under
                           the same directory
     '<something>' - a pattern of the specified default type
     """
     normalize = _donormalize
     if icasefs:
         if exact:
             raise error.ProgrammingError("a case-insensitive exact matcher "
                                          "doesn't make sense")
         dirstate = ctx.repo().dirstate
         dsnormalize = dirstate.normalize
 
         def normalize(patterns, default, root, cwd, auditor, warn):
             kp = _donormalize(patterns, default, root, cwd, auditor, warn)
             kindpats = []
             for kind, pats, source in kp:
                 if kind not in ('re', 'relre'):  # regex can't be normalized
                     p = pats
                     pats = dsnormalize(pats)
 
                     # Preserve the original to handle a case only rename.
                     if p != pats and p in dirstate:
                         kindpats.append((kind, p, source))
 
                 kindpats.append((kind, pats, source))
             return kindpats
 
     if exact:
         m = exactmatcher(root, cwd, patterns, badfn)
     elif patterns:
         kindpats = normalize(patterns, default, root, cwd, auditor, warn)
         if _kindpatsalwaysmatch(kindpats):
             m = alwaysmatcher(root, cwd, badfn, relativeuipath=True)
         else:
             m = patternmatcher(root, cwd, kindpats, ctx=ctx,
                                listsubrepos=listsubrepos, badfn=badfn)
     else:
         # It's a little strange that no patterns means to match everything.
         # Consider changing this to match nothing (probably using nevermatcher).
         m = alwaysmatcher(root, cwd, badfn)
 
     if include:
         kindpats = normalize(include, 'glob', root, cwd, auditor, warn)
         im = includematcher(root, cwd, kindpats, ctx=ctx,
                             listsubrepos=listsubrepos, badfn=None)
         m = intersectmatchers(m, im)
     if exclude:
         kindpats = normalize(exclude, 'glob', root, cwd, auditor, warn)
         em = includematcher(root, cwd, kindpats, ctx=ctx,
                             listsubrepos=listsubrepos, badfn=None)
         m = differencematcher(m, em)
     return m
 
 def exact(root, cwd, files, badfn=None):
     return exactmatcher(root, cwd, files, badfn=badfn)
 
 def always(root, cwd):
     return alwaysmatcher(root, cwd)
 
 def never(root, cwd):
     return nevermatcher(root, cwd)
 
 def badmatch(match, badfn):
     """Make a copy of the given matcher, replacing its bad method with the given
     one.
     """
     m = copy.copy(match)
     m.bad = badfn
     return m
 
 def _donormalize(patterns, default, root, cwd, auditor, warn):
     '''Convert 'kind:pat' from the patterns list to tuples with kind and
     normalized and rooted patterns and with listfiles expanded.'''
     kindpats = []
     for kind, pat in [_patsplit(p, default) for p in patterns]:
         if kind in cwdrelativepatternkinds:
             pat = pathutil.canonpath(root, cwd, pat, auditor)
         elif kind in ('relglob', 'path', 'rootfilesin'):
             pat = util.normpath(pat)
         elif kind in ('listfile', 'listfile0'):
             try:
                 files = util.readfile(pat)
                 if kind == 'listfile0':
                     files = files.split('\0')
                 else:
                     files = files.splitlines()
                 files = [f for f in files if f]
             except EnvironmentError:
                 raise error.Abort(_("unable to read file list (%s)") % pat)
             for k, p, source in _donormalize(files, default, root, cwd,
                                              auditor, warn):
                 kindpats.append((k, p, pat))
             continue
         elif kind == 'include':
             try:
                 fullpath = os.path.join(root, util.localpath(pat))
                 includepats = readpatternfile(fullpath, warn)
                 for k, p, source in _donormalize(includepats, default,
                                                  root, cwd, auditor, warn):
                     kindpats.append((k, p, source or pat))
             except error.Abort as inst:
                 raise error.Abort('%s: %s' % (pat, inst[0]))
             except IOError as inst:
                 if warn:
                     warn(_("skipping unreadable pattern file '%s': %s\n") %
                          (pat, stringutil.forcebytestr(inst.strerror)))
             continue
         # else: re or relre - which cannot be normalized
         kindpats.append((kind, pat, ''))
     return kindpats
 
 class basematcher(object):
 
     def __init__(self, root, cwd, badfn=None, relativeuipath=True):
         self._root = root
         self._cwd = cwd
         if badfn is not None:
             self.bad = badfn
         self._relativeuipath = relativeuipath
 
     def __call__(self, fn):
         return self.matchfn(fn)
     def __iter__(self):
         for f in self._files:
             yield f
     # Callbacks related to how the matcher is used by dirstate.walk.
     # Subscribers to these events must monkeypatch the matcher object.
     def bad(self, f, msg):
         '''Callback from dirstate.walk for each explicit file that can't be
         found/accessed, with an error message.'''
 
     # If an explicitdir is set, it will be called when an explicitly listed
     # directory is visited.
     explicitdir = None
 
     # If an traversedir is set, it will be called when a directory discovered
     # by recursive traversal is visited.
     traversedir = None
 
     def abs(self, f):
         '''Convert a repo path back to path that is relative to the root of the
         matcher.'''
         return f
 
     def rel(self, f):
         '''Convert repo path back to path that is relative to cwd of matcher.'''
         return util.pathto(self._root, self._cwd, f)
 
     def uipath(self, f):
         '''Convert repo path to a display path.  If patterns or -I/-X were used
         to create this matcher, the display path will be relative to cwd.
         Otherwise it is relative to the root of the repo.'''
         return (self._relativeuipath and self.rel(f)) or self.abs(f)
 
     @propertycache
     def _files(self):
         return []
 
     def files(self):
         '''Explicitly listed files or patterns or roots:
         if no patterns or .always(): empty list,
         if exact: list exact files,
         if not .anypats(): list all files and dirs,
         else: optimal roots'''
         return self._files
 
     @propertycache
     def _fileset(self):
         return set(self._files)
 
     def exact(self, f):
         '''Returns True if f is in .files().'''
         return f in self._fileset
 
     def matchfn(self, f):
         return False
 
     def visitdir(self, dir):
         '''Decides whether a directory should be visited based on whether it
         has potential matches in it or one of its subdirectories. This is
         based on the match's primary, included, and excluded patterns.
 
         Returns the string 'all' if the given directory and all subdirectories
         should be visited. Otherwise returns True or False indicating whether
         the given directory should be visited.
         '''
         return True
 
     def always(self):
         '''Matcher will match everything and .files() will be empty --
         optimization might be possible.'''
         return False
 
     def isexact(self):
         '''Matcher will match exactly the list of files in .files() --
         optimization might be possible.'''
         return False
 
     def prefix(self):
         '''Matcher will match the paths in .files() recursively --
         optimization might be possible.'''
         return False
 
     def anypats(self):
         '''None of .always(), .isexact(), and .prefix() is true --
         optimizations will be difficult.'''
         return not self.always() and not self.isexact() and not self.prefix()
 
 class alwaysmatcher(basematcher):
     '''Matches everything.'''
 
     def __init__(self, root, cwd, badfn=None, relativeuipath=False):
         super(alwaysmatcher, self).__init__(root, cwd, badfn,
                                             relativeuipath=relativeuipath)
 
     def always(self):
         return True
 
     def matchfn(self, f):
         return True
 
     def visitdir(self, dir):
         return 'all'
 
     def __repr__(self):
         return r'<alwaysmatcher>'
 
 class nevermatcher(basematcher):
     '''Matches nothing.'''
 
     def __init__(self, root, cwd, badfn=None):
         super(nevermatcher, self).__init__(root, cwd, badfn)
 
     # It's a little weird to say that the nevermatcher is an exact matcher
     # or a prefix matcher, but it seems to make sense to let callers take
     # fast paths based on either. There will be no exact matches, nor any
     # prefixes (files() returns []), so fast paths iterating over them should
     # be efficient (and correct).
     def isexact(self):
         return True
 
     def prefix(self):
         return True
 
     def visitdir(self, dir):
         return False
 
     def __repr__(self):
         return r'<nevermatcher>'
 
 class patternmatcher(basematcher):
 
     def __init__(self, root, cwd, kindpats, ctx=None, listsubrepos=False,
                  badfn=None):
         super(patternmatcher, self).__init__(root, cwd, badfn)
 
         self._files = _explicitfiles(kindpats)
         self._prefix = _prefix(kindpats)
         self._pats, self.matchfn = _buildmatch(ctx, kindpats, '$', listsubrepos,
                                                root)
 
     @propertycache
     def _dirs(self):
         return set(util.dirs(self._fileset)) | {'.'}
 
     def visitdir(self, dir):
         if self._prefix and dir in self._fileset:
             return 'all'
         return ('.' in self._fileset or
                 dir in self._fileset or
                 dir in self._dirs or
                 any(parentdir in self._fileset
                     for parentdir in util.finddirs(dir)))
 
     def prefix(self):
         return self._prefix
 
     @encoding.strmethod
     def __repr__(self):
         return ('<patternmatcher patterns=%r>' % pycompat.bytestr(self._pats))
 
 class includematcher(basematcher):
 
     def __init__(self, root, cwd, kindpats, ctx=None, listsubrepos=False,
                  badfn=None):
         super(includematcher, self).__init__(root, cwd, badfn)
 
         self._pats, self.matchfn = _buildmatch(ctx, kindpats, '(?:/|$)',
                                                listsubrepos, root)
         self._prefix = _prefix(kindpats)
         roots, dirs = _rootsanddirs(kindpats)
         # roots are directories which are recursively included.
         self._roots = set(roots)
         # dirs are directories which are non-recursively included.
         self._dirs = set(dirs)
 
     def visitdir(self, dir):
         if self._prefix and dir in self._roots:
             return 'all'
         return ('.' in self._roots or
                 dir in self._roots or
                 dir in self._dirs or
                 any(parentdir in self._roots
                     for parentdir in util.finddirs(dir)))
 
     @encoding.strmethod
     def __repr__(self):
         return ('<includematcher includes=%r>' % pycompat.bytestr(self._pats))
 
 class exactmatcher(basematcher):
     '''Matches the input files exactly. They are interpreted as paths, not
     patterns (so no kind-prefixes).
     '''
 
     def __init__(self, root, cwd, files, badfn=None):
         super(exactmatcher, self).__init__(root, cwd, badfn)
 
         if isinstance(files, list):
             self._files = files
         else:
             self._files = list(files)
 
     matchfn = basematcher.exact
 
     @propertycache
     def _dirs(self):
         return set(util.dirs(self._fileset)) | {'.'}
 
     def visitdir(self, dir):
         return dir in self._dirs
 
     def isexact(self):
         return True
 
     @encoding.strmethod
     def __repr__(self):
         return ('<exactmatcher files=%r>' % self._files)
 
 class differencematcher(basematcher):
     '''Composes two matchers by matching if the first matches and the second
     does not.
 
     The second matcher's non-matching-attributes (root, cwd, bad, explicitdir,
     traversedir) are ignored.
     '''
     def __init__(self, m1, m2):
         super(differencematcher, self).__init__(m1._root, m1._cwd)
         self._m1 = m1
         self._m2 = m2
         self.bad = m1.bad
         self.explicitdir = m1.explicitdir
         self.traversedir = m1.traversedir
 
     def matchfn(self, f):
         return self._m1(f) and not self._m2(f)
 
     @propertycache
     def _files(self):
         if self.isexact():
             return [f for f in self._m1.files() if self(f)]
         # If m1 is not an exact matcher, we can't easily figure out the set of
         # files, because its files() are not always files. For example, if
         # m1 is "path:dir" and m2 is "rootfileins:.", we don't
         # want to remove "dir" from the set even though it would match m2,
         # because the "dir" in m1 may not be a file.
         return self._m1.files()
 
     def visitdir(self, dir):
         if self._m2.visitdir(dir) == 'all':
             return False
         return bool(self._m1.visitdir(dir))
 
     def isexact(self):
         return self._m1.isexact()
 
     @encoding.strmethod
     def __repr__(self):
         return ('<differencematcher m1=%r, m2=%r>' % (self._m1, self._m2))
 
 def intersectmatchers(m1, m2):
     '''Composes two matchers by matching if both of them match.
 
     The second matcher's non-matching-attributes (root, cwd, bad, explicitdir,
     traversedir) are ignored.
     '''
     if m1 is None or m2 is None:
         return m1 or m2
     if m1.always():
         m = copy.copy(m2)
         # TODO: Consider encapsulating these things in a class so there's only
         # one thing to copy from m1.
         m.bad = m1.bad
         m.explicitdir = m1.explicitdir
         m.traversedir = m1.traversedir
         m.abs = m1.abs
         m.rel = m1.rel
         m._relativeuipath |= m1._relativeuipath
         return m
     if m2.always():
         m = copy.copy(m1)
         m._relativeuipath |= m2._relativeuipath
         return m
     return intersectionmatcher(m1, m2)
 
 class intersectionmatcher(basematcher):
     def __init__(self, m1, m2):
         super(intersectionmatcher, self).__init__(m1._root, m1._cwd)
         self._m1 = m1
         self._m2 = m2
         self.bad = m1.bad
         self.explicitdir = m1.explicitdir
         self.traversedir = m1.traversedir
 
     @propertycache
     def _files(self):
         if self.isexact():
             m1, m2 = self._m1, self._m2
             if not m1.isexact():
                 m1, m2 = m2, m1
             return [f for f in m1.files() if m2(f)]
         # It neither m1 nor m2 is an exact matcher, we can't easily intersect
         # the set of files, because their files() are not always files. For
         # example, if intersecting a matcher "-I glob:foo.txt" with matcher of
         # "path:dir2", we don't want to remove "dir2" from the set.
         return self._m1.files() + self._m2.files()
 
     def matchfn(self, f):
         return self._m1(f) and self._m2(f)
 
     def visitdir(self, dir):
         visit1 = self._m1.visitdir(dir)
         if visit1 == 'all':
             return self._m2.visitdir(dir)
         # bool() because visit1=True + visit2='all' should not be 'all'
         return bool(visit1 and self._m2.visitdir(dir))
 
     def always(self):
         return self._m1.always() and self._m2.always()
 
     def isexact(self):
         return self._m1.isexact() or self._m2.isexact()
 
     @encoding.strmethod
     def __repr__(self):
         return ('<intersectionmatcher m1=%r, m2=%r>' % (self._m1, self._m2))
 
 class subdirmatcher(basematcher):
     """Adapt a matcher to work on a subdirectory only.
 
     The paths are remapped to remove/insert the path as needed:
 
     >>> from . import pycompat
     >>> m1 = match(b'root', b'', [b'a.txt', b'sub/b.txt'])
     >>> m2 = subdirmatcher(b'sub', m1)
     >>> bool(m2(b'a.txt'))
     False
     >>> bool(m2(b'b.txt'))
     True
     >>> bool(m2.matchfn(b'a.txt'))
     False
     >>> bool(m2.matchfn(b'b.txt'))
     True
     >>> m2.files()
     ['b.txt']
     >>> m2.exact(b'b.txt')
     True
     >>> util.pconvert(m2.rel(b'b.txt'))
     'sub/b.txt'
     >>> def bad(f, msg):
     ...     print(pycompat.sysstr(b"%s: %s" % (f, msg)))
     >>> m1.bad = bad
     >>> m2.bad(b'x.txt', b'No such file')
     sub/x.txt: No such file
     >>> m2.abs(b'c.txt')
     'sub/c.txt'
     """
 
     def __init__(self, path, matcher):
         super(subdirmatcher, self).__init__(matcher._root, matcher._cwd)
         self._path = path
         self._matcher = matcher
         self._always = matcher.always()
 
         self._files = [f[len(path) + 1:] for f in matcher._files
                        if f.startswith(path + "/")]
 
         # If the parent repo had a path to this subrepo and the matcher is
         # a prefix matcher, this submatcher always matches.
         if matcher.prefix():
             self._always = any(f == path for f in matcher._files)
 
     def bad(self, f, msg):
         self._matcher.bad(self._path + "/" + f, msg)
 
     def abs(self, f):
         return self._matcher.abs(self._path + "/" + f)
 
     def rel(self, f):
         return self._matcher.rel(self._path + "/" + f)
 
     def uipath(self, f):
         return self._matcher.uipath(self._path + "/" + f)
 
     def matchfn(self, f):
         # Some information is lost in the superclass's constructor, so we
         # can not accurately create the matching function for the subdirectory
         # from the inputs. Instead, we override matchfn() and visitdir() to
         # call the original matcher with the subdirectory path prepended.
         return self._matcher.matchfn(self._path + "/" + f)
 
     def visitdir(self, dir):
         if dir == '.':
             dir = self._path
         else:
             dir = self._path + "/" + dir
         return self._matcher.visitdir(dir)
 
     def always(self):
         return self._always
 
     def prefix(self):
         return self._matcher.prefix() and not self._always
 
     @encoding.strmethod
     def __repr__(self):
         return ('<subdirmatcher path=%r, matcher=%r>' %
                 (self._path, self._matcher))
 
 class unionmatcher(basematcher):
     """A matcher that is the union of several matchers.
 
     The non-matching-attributes (root, cwd, bad, explicitdir, traversedir) are
     taken from the first matcher.
     """
 
     def __init__(self, matchers):
         m1 = matchers[0]
         super(unionmatcher, self).__init__(m1._root, m1._cwd)
         self.explicitdir = m1.explicitdir
         self.traversedir = m1.traversedir
         self._matchers = matchers
 
     def matchfn(self, f):
         for match in self._matchers:
             if match(f):
                 return True
         return False
 
     def visitdir(self, dir):
         r = False
         for m in self._matchers:
             v = m.visitdir(dir)
             if v == 'all':
                 return v
             r |= v
         return r
 
     @encoding.strmethod
     def __repr__(self):
         return ('<unionmatcher matchers=%r>' % self._matchers)
 
 def patkind(pattern, default=None):
     '''If pattern is 'kind:pat' with a known kind, return kind.'''
     return _patsplit(pattern, default)[0]
 
 def _patsplit(pattern, default):
     """Split a string into the optional pattern kind prefix and the actual
     pattern."""
     if ':' in pattern:
         kind, pat = pattern.split(':', 1)
         if kind in allpatternkinds:
             return kind, pat
     return default, pattern
 
 def _globre(pat):
     r'''Convert an extended glob string to a regexp string.
 
     >>> from . import pycompat
     >>> def bprint(s):
     ...     print(pycompat.sysstr(s))
     >>> bprint(_globre(br'?'))
     .
     >>> bprint(_globre(br'*'))
     [^/]*
     >>> bprint(_globre(br'**'))
     .*
     >>> bprint(_globre(br'**/a'))
     (?:.*/)?a
     >>> bprint(_globre(br'a/**/b'))
-    a\/(?:.*/)?b
+    a/(?:.*/)?b
     >>> bprint(_globre(br'[a*?!^][^b][!c]'))
     [a*?!^][\^b][^c]
     >>> bprint(_globre(br'{a,b}'))
     (?:a|b)
     >>> bprint(_globre(br'.\*\?'))
     \.\*\?
     '''
     i, n = 0, len(pat)
     res = ''
     group = 0
-    escape = util.re.escape
+    escape = util.stringutil.reescape
     def peek():
         return i < n and pat[i:i + 1]
     while i < n:
         c = pat[i:i + 1]
         i += 1
         if c not in '*?[{},\\':
             res += escape(c)
         elif c == '*':
             if peek() == '*':
                 i += 1
                 if peek() == '/':
                     i += 1
                     res += '(?:.*/)?'
                 else:
                     res += '.*'
             else:
                 res += '[^/]*'
         elif c == '?':
             res += '.'
         elif c == '[':
             j = i
             if j < n and pat[j:j + 1] in '!]':
                 j += 1
             while j < n and pat[j:j + 1] != ']':
                 j += 1
             if j >= n:
                 res += '\\['
             else:
                 stuff = pat[i:j].replace('\\','\\\\')
                 i = j + 1
                 if stuff[0:1] == '!':
                     stuff = '^' + stuff[1:]
                 elif stuff[0:1] == '^':
                     stuff = '\\' + stuff
                 res = '%s[%s]' % (res, stuff)
         elif c == '{':
             group += 1
             res += '(?:'
         elif c == '}' and group:
             res += ')'
             group -= 1
         elif c == ',' and group:
             res += '|'
         elif c == '\\':
             p = peek()
             if p:
                 i += 1
                 res += escape(p)
             else:
                 res += escape(c)
         else:
             res += escape(c)
     return res
 
 def _regex(kind, pat, globsuffix):
     '''Convert a (normalized) pattern of any kind into a regular expression.
     globsuffix is appended to the regexp of globs.'''
     if not pat:
         return ''
     if kind == 're':
         return pat
     if kind in ('path', 'relpath'):
         if pat == '.':
             return ''
-        return util.re.escape(pat) + '(?:/|$)'
+        return util.stringutil.reescape(pat) + '(?:/|$)'
     if kind == 'rootfilesin':
         if pat == '.':
             escaped = ''
         else:
             # Pattern is a directory name.
-            escaped = util.re.escape(pat) + '/'
+            escaped = util.stringutil.reescape(pat) + '/'
         # Anything after the pattern must be a non-directory.
         return escaped + '[^/]+$'
     if kind == 'relglob':
         return '(?:|.*/)' + _globre(pat) + globsuffix
     if kind == 'relre':
         if pat.startswith('^'):
             return pat
         return '.*' + pat
     return _globre(pat) + globsuffix
 
 def _buildmatch(ctx, kindpats, globsuffix, listsubrepos, root):
     '''Return regexp string and a matcher function for kindpats.
     globsuffix is appended to the regexp of globs.'''
     matchfuncs = []
 
     subincludes, kindpats = _expandsubinclude(kindpats, root)
     if subincludes:
         submatchers = {}
         def matchsubinclude(f):
             for prefix, matcherargs in subincludes:
                 if f.startswith(prefix):
                     mf = submatchers.get(prefix)
                     if mf is None:
                         mf = match(*matcherargs)
                         submatchers[prefix] = mf
 
                     if mf(f[len(prefix):]):
                         return True
             return False
         matchfuncs.append(matchsubinclude)
 
     fset, kindpats = _expandsets(kindpats, ctx, listsubrepos)
     if fset:
         matchfuncs.append(fset.__contains__)
 
     regex = ''
     if kindpats:
         regex, mf = _buildregexmatch(kindpats, globsuffix)
         matchfuncs.append(mf)
 
     if len(matchfuncs) == 1:
         return regex, matchfuncs[0]
     else:
         return regex, lambda f: any(mf(f) for mf in matchfuncs)
 
 def _buildregexmatch(kindpats, globsuffix):
     """Build a match function from a list of kinds and kindpats,
     return regexp string and a matcher function."""
     try:
         regex = '(?:%s)' % '|'.join([_regex(k, p, globsuffix)
                                      for (k, p, s) in kindpats])
         if len(regex) > 20000:
             raise OverflowError
         return regex, _rematcher(regex)
     except OverflowError:
         # We're using a Python with a tiny regex engine and we
         # made it explode, so we'll divide the pattern list in two
         # until it works
         l = len(kindpats)
         if l < 2:
             raise
         regexa, a = _buildregexmatch(kindpats[:l//2], globsuffix)
         regexb, b = _buildregexmatch(kindpats[l//2:], globsuffix)
         return regex, lambda s: a(s) or b(s)
     except re.error:
         for k, p, s in kindpats:
             try:
                 _rematcher('(?:%s)' % _regex(k, p, globsuffix))
             except re.error:
                 if s:
                     raise error.Abort(_("%s: invalid pattern (%s): %s") %
                                      (s, k, p))
                 else:
                     raise error.Abort(_("invalid pattern (%s): %s") % (k, p))
         raise error.Abort(_("invalid pattern"))
 
 def _patternrootsanddirs(kindpats):
     '''Returns roots and directories corresponding to each pattern.
 
     This calculates the roots and directories exactly matching the patterns and
     returns a tuple of (roots, dirs) for each. It does not return other
     directories which may also need to be considered, like the parent
     directories.
     '''
     r = []
     d = []
     for kind, pat, source in kindpats:
         if kind == 'glob': # find the non-glob prefix
             root = []
             for p in pat.split('/'):
                 if '[' in p or '{' in p or '*' in p or '?' in p:
                     break
                 root.append(p)
             r.append('/'.join(root) or '.')
         elif kind in ('relpath', 'path'):
             r.append(pat or '.')
         elif kind in ('rootfilesin',):
             d.append(pat or '.')
         else: # relglob, re, relre
             r.append('.')
     return r, d
 
 def _roots(kindpats):
     '''Returns root directories to match recursively from the given patterns.'''
     roots, dirs = _patternrootsanddirs(kindpats)
     return roots
 
 def _rootsanddirs(kindpats):
     '''Returns roots and exact directories from patterns.
 
     roots are directories to match recursively, whereas exact directories should
     be matched non-recursively. The returned (roots, dirs) tuple will also
     include directories that need to be implicitly considered as either, such as
     parent directories.
 
     >>> _rootsanddirs(
     ...     [(b'glob', b'g/h/*', b''), (b'glob', b'g/h', b''),
     ...      (b'glob', b'g*', b'')])
     (['g/h', 'g/h', '.'], ['g', '.'])
     >>> _rootsanddirs(
     ...     [(b'rootfilesin', b'g/h', b''), (b'rootfilesin', b'', b'')])
     ([], ['g/h', '.', 'g', '.'])
     >>> _rootsanddirs(
     ...     [(b'relpath', b'r', b''), (b'path', b'p/p', b''),
     ...      (b'path', b'', b'')])
     (['r', 'p/p', '.'], ['p', '.'])
     >>> _rootsanddirs(
     ...     [(b'relglob', b'rg*', b''), (b're', b're/', b''),
     ...      (b'relre', b'rr', b'')])
     (['.', '.', '.'], ['.'])
     '''
     r, d = _patternrootsanddirs(kindpats)
 
     # Append the parents as non-recursive/exact directories, since they must be
     # scanned to get to either the roots or the other exact directories.
     d.extend(util.dirs(d))
     d.extend(util.dirs(r))
     # util.dirs() does not include the root directory, so add it manually
     d.append('.')
 
     return r, d
 
 def _explicitfiles(kindpats):
     '''Returns the potential explicit filenames from the patterns.
 
     >>> _explicitfiles([(b'path', b'foo/bar', b'')])
     ['foo/bar']
     >>> _explicitfiles([(b'rootfilesin', b'foo/bar', b'')])
     []
     '''
     # Keep only the pattern kinds where one can specify filenames (vs only
     # directory names).
     filable = [kp for kp in kindpats if kp[0] not in ('rootfilesin',)]
     return _roots(filable)
 
 def _prefix(kindpats):
     '''Whether all the patterns match a prefix (i.e. recursively)'''
     for kind, pat, source in kindpats:
         if kind not in ('path', 'relpath'):
             return False
     return True
 
 _commentre = None
 
 def readpatternfile(filepath, warn, sourceinfo=False):
     '''parse a pattern file, returning a list of
     patterns. These patterns should be given to compile()
     to be validated and converted into a match function.
 
     trailing white space is dropped.
     the escape character is backslash.
     comments start with #.
     empty lines are skipped.
 
     lines can be of the following formats:
 
     syntax: regexp # defaults following lines to non-rooted regexps
     syntax: glob   # defaults following lines to non-rooted globs
     re:pattern     # non-rooted regular expression
     glob:pattern   # non-rooted glob
     pattern        # pattern of the current default type
 
     if sourceinfo is set, returns a list of tuples:
     (pattern, lineno, originalline). This is useful to debug ignore patterns.
     '''
 
     syntaxes = {'re': 'relre:', 'regexp': 'relre:', 'glob': 'relglob:',
                 'include': 'include', 'subinclude': 'subinclude'}
     syntax = 'relre:'
     patterns = []
 
     fp = open(filepath, 'rb')
     for lineno, line in enumerate(util.iterfile(fp), start=1):
         if "#" in line:
             global _commentre
             if not _commentre:
                 _commentre = util.re.compile(br'((?:^|[^\\])(?:\\\\)*)#.*')
             # remove comments prefixed by an even number of escapes
             m = _commentre.search(line)
             if m:
                 line = line[:m.end(1)]
             # fixup properly escaped comments that survived the above
             line = line.replace("\\#", "#")
         line = line.rstrip()
         if not line:
             continue
 
         if line.startswith('syntax:'):
             s = line[7:].strip()
             try:
                 syntax = syntaxes[s]
             except KeyError:
                 if warn:
                     warn(_("%s: ignoring invalid syntax '%s'\n") %
                          (filepath, s))
             continue
 
         linesyntax = syntax
         for s, rels in syntaxes.iteritems():
             if line.startswith(rels):
                 linesyntax = rels
                 line = line[len(rels):]
                 break
             elif line.startswith(s+':'):
                 linesyntax = rels
                 line = line[len(s) + 1:]
                 break
         if sourceinfo:
             patterns.append((linesyntax + line, lineno, line))
         else:
             patterns.append(linesyntax + line)
     fp.close()
     return patterns

mercurial/sshpeer.py

@@ -1,634 +1,635
 # sshpeer.py - ssh repository proxy class 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 re
 import uuid
 
 from .i18n import _
 from . import (
     error,
     pycompat,
     util,
     wireprotoserver,
     wireprototypes,
     wireprotov1peer,
     wireprotov1server,
 )
 from .utils import (
     procutil,
+    stringutil,
 )
 
 def _serverquote(s):
     """quote a string for the remote shell ... which we assume is sh"""
     if not s:
         return s
     if re.match('[a-zA-Z0-9@%_+=:,./-]*$', s):
         return s
     return "'%s'" % s.replace("'", "'\\''")
 
 def _forwardoutput(ui, pipe):
     """display all data currently available on pipe as remote output.
 
     This is non blocking."""
     if pipe:
         s = procutil.readpipe(pipe)
         if s:
             for l in s.splitlines():
                 ui.status(_("remote: "), l, '\n')
 
 class doublepipe(object):
     """Operate a side-channel pipe in addition of a main one
 
     The side-channel pipe contains server output to be forwarded to the user
     input. The double pipe will behave as the "main" pipe, but will ensure the
     content of the "side" pipe is properly processed while we wait for blocking
     call on the "main" pipe.
 
     If large amounts of data are read from "main", the forward will cease after
     the first bytes start to appear. This simplifies the implementation
     without affecting actual output of sshpeer too much as we rarely issue
     large read for data not yet emitted by the server.
 
     The main pipe is expected to be a 'bufferedinputpipe' from the util module
     that handle all the os specific bits. This class lives in this module
     because it focus on behavior specific to the ssh protocol."""
 
     def __init__(self, ui, main, side):
         self._ui = ui
         self._main = main
         self._side = side
 
     def _wait(self):
         """wait until some data are available on main or side
 
         return a pair of boolean (ismainready, issideready)
 
         (This will only wait for data if the setup is supported by `util.poll`)
         """
         if (isinstance(self._main, util.bufferedinputpipe) and
             self._main.hasbuffer):
             # Main has data. Assume side is worth poking at.
             return True, True
 
         fds = [self._main.fileno(), self._side.fileno()]
         try:
             act = util.poll(fds)
         except NotImplementedError:
             # non supported yet case, assume all have data.
             act = fds
         return (self._main.fileno() in act, self._side.fileno() in act)
 
     def write(self, data):
         return self._call('write', data)
 
     def read(self, size):
         r = self._call('read', size)
         if size != 0 and not r:
             # We've observed a condition that indicates the
             # stdout closed unexpectedly. Check stderr one
             # more time and snag anything that's there before
             # letting anyone know the main part of the pipe
             # closed prematurely.
             _forwardoutput(self._ui, self._side)
         return r
 
     def readline(self):
         return self._call('readline')
 
     def _call(self, methname, data=None):
         """call <methname> on "main", forward output of "side" while blocking
         """
         # data can be '' or 0
         if (data is not None and not data) or self._main.closed:
             _forwardoutput(self._ui, self._side)
             return ''
         while True:
             mainready, sideready = self._wait()
             if sideready:
                 _forwardoutput(self._ui, self._side)
             if mainready:
                 meth = getattr(self._main, methname)
                 if data is None:
                     return meth()
                 else:
                     return meth(data)
 
     def close(self):
         return self._main.close()
 
     def flush(self):
         return self._main.flush()
 
 def _cleanuppipes(ui, pipei, pipeo, pipee):
     """Clean up pipes used by an SSH connection."""
     if pipeo:
         pipeo.close()
     if pipei:
         pipei.close()
 
     if pipee:
         # Try to read from the err descriptor until EOF.
         try:
             for l in pipee:
                 ui.status(_('remote: '), l)
         except (IOError, ValueError):
             pass
 
         pipee.close()
 
 def _makeconnection(ui, sshcmd, args, remotecmd, path, sshenv=None):
     """Create an SSH connection to a server.
 
     Returns a tuple of (process, stdin, stdout, stderr) for the
     spawned process.
     """
     cmd = '%s %s %s' % (
         sshcmd,
         args,
         procutil.shellquote('%s -R %s serve --stdio' % (
             _serverquote(remotecmd), _serverquote(path))))
 
     ui.debug('running %s\n' % cmd)
     cmd = procutil.quotecommand(cmd)
 
     # no buffer allow the use of 'select'
     # feel free to remove buffering and select usage when we ultimately
     # move to threading.
     stdin, stdout, stderr, proc = procutil.popen4(cmd, bufsize=0, env=sshenv)
 
     return proc, stdin, stdout, stderr
 
 def _clientcapabilities():
     """Return list of capabilities of this client.
 
     Returns a list of capabilities that are supported by this client.
     """
     protoparams = {'partial-pull'}
     comps = [e.wireprotosupport().name for e in
              util.compengines.supportedwireengines(util.CLIENTROLE)]
     protoparams.add('comp=%s' % ','.join(comps))
     return protoparams
 
 def _performhandshake(ui, stdin, stdout, stderr):
     def badresponse():
         # Flush any output on stderr.
         _forwardoutput(ui, stderr)
 
         msg = _('no suitable response from remote hg')
         hint = ui.config('ui', 'ssherrorhint')
         raise error.RepoError(msg, hint=hint)
 
     # The handshake consists of sending wire protocol commands in reverse
     # order of protocol implementation and then sniffing for a response
     # to one of them.
     #
     # Those commands (from oldest to newest) are:
     #
     # ``between``
     #   Asks for the set of revisions between a pair of revisions. Command
     #   present in all Mercurial server implementations.
     #
     # ``hello``
     #   Instructs the server to advertise its capabilities. Introduced in
     #   Mercurial 0.9.1.
     #
     # ``upgrade``
     #   Requests upgrade from default transport protocol version 1 to
     #   a newer version. Introduced in Mercurial 4.6 as an experimental
     #   feature.
     #
     # The ``between`` command is issued with a request for the null
     # range. If the remote is a Mercurial server, this request will
     # generate a specific response: ``1\n\n``. This represents the
     # wire protocol encoded value for ``\n``. We look for ``1\n\n``
     # in the output stream and know this is the response to ``between``
     # and we're at the end of our handshake reply.
     #
     # The response to the ``hello`` command will be a line with the
     # length of the value returned by that command followed by that
     # value. If the server doesn't support ``hello`` (which should be
     # rare), that line will be ``0\n``. Otherwise, the value will contain
     # RFC 822 like lines. Of these, the ``capabilities:`` line contains
     # the capabilities of the server.
     #
     # The ``upgrade`` command isn't really a command in the traditional
     # sense of version 1 of the transport because it isn't using the
     # proper mechanism for formatting insteads: instead, it just encodes
     # arguments on the line, delimited by spaces.
     #
     # The ``upgrade`` line looks like ``upgrade <token> <capabilities>``.
     # If the server doesn't support protocol upgrades, it will reply to
     # this line with ``0\n``. Otherwise, it emits an
     # ``upgraded <token> <protocol>`` line to both stdout and stderr.
     # Content immediately following this line describes additional
     # protocol and server state.
     #
     # In addition to the responses to our command requests, the server
     # may emit "banner" output on stdout. SSH servers are allowed to
     # print messages to stdout on login. Issuing commands on connection
     # allows us to flush this banner output from the server by scanning
     # for output to our well-known ``between`` command. Of course, if
     # the banner contains ``1\n\n``, this will throw off our detection.
 
     requestlog = ui.configbool('devel', 'debug.peer-request')
 
     # Generate a random token to help identify responses to version 2
     # upgrade request.
     token = pycompat.sysbytes(str(uuid.uuid4()))
     upgradecaps = [
         ('proto', wireprotoserver.SSHV2),
     ]
     upgradecaps = util.urlreq.urlencode(upgradecaps)
 
     try:
         pairsarg = '%s-%s' % ('0' * 40, '0' * 40)
         handshake = [
             'hello\n',
             'between\n',
             'pairs %d\n' % len(pairsarg),
             pairsarg,
         ]
 
         # Request upgrade to version 2 if configured.
         if ui.configbool('experimental', 'sshpeer.advertise-v2'):
             ui.debug('sending upgrade request: %s %s\n' % (token, upgradecaps))
             handshake.insert(0, 'upgrade %s %s\n' % (token, upgradecaps))
 
         if requestlog:
             ui.debug('devel-peer-request: hello+between\n')
             ui.debug('devel-peer-request:   pairs: %d bytes\n' % len(pairsarg))
         ui.debug('sending hello command\n')
         ui.debug('sending between command\n')
 
         stdin.write(''.join(handshake))
         stdin.flush()
     except IOError:
         badresponse()
 
     # Assume version 1 of wire protocol by default.
     protoname = wireprototypes.SSHV1
-    reupgraded = re.compile(b'^upgraded %s (.*)$' % re.escape(token))
+    reupgraded = re.compile(b'^upgraded %s (.*)$' % stringutil.reescape(token))
 
     lines = ['', 'dummy']
     max_noise = 500
     while lines[-1] and max_noise:
         try:
             l = stdout.readline()
             _forwardoutput(ui, stderr)
 
             # Look for reply to protocol upgrade request. It has a token
             # in it, so there should be no false positives.
             m = reupgraded.match(l)
             if m:
                 protoname = m.group(1)
                 ui.debug('protocol upgraded to %s\n' % protoname)
                 # If an upgrade was handled, the ``hello`` and ``between``
                 # requests are ignored. The next output belongs to the
                 # protocol, so stop scanning lines.
                 break
 
             # Otherwise it could be a banner, ``0\n`` response if server
             # doesn't support upgrade.
 
             if lines[-1] == '1\n' and l == '\n':
                 break
             if l:
                 ui.debug('remote: ', l)
             lines.append(l)
             max_noise -= 1
         except IOError:
             badresponse()
     else:
         badresponse()
 
     caps = set()
 
     # For version 1, we should see a ``capabilities`` line in response to the
     # ``hello`` command.
     if protoname == wireprototypes.SSHV1:
         for l in reversed(lines):
             # Look for response to ``hello`` command. Scan from the back so
             # we don't misinterpret banner output as the command reply.
             if l.startswith('capabilities:'):
                 caps.update(l[:-1].split(':')[1].split())
                 break
     elif protoname == wireprotoserver.SSHV2:
         # We see a line with number of bytes to follow and then a value
         # looking like ``capabilities: *``.
         line = stdout.readline()
         try:
             valuelen = int(line)
         except ValueError:
             badresponse()
 
         capsline = stdout.read(valuelen)
         if not capsline.startswith('capabilities: '):
             badresponse()
 
         ui.debug('remote: %s\n' % capsline)
 
         caps.update(capsline.split(':')[1].split())
         # Trailing newline.
         stdout.read(1)
 
     # Error if we couldn't find capabilities, this means:
     #
     # 1. Remote isn't a Mercurial server
     # 2. Remote is a <0.9.1 Mercurial server
     # 3. Remote is a future Mercurial server that dropped ``hello``
     #    and other attempted handshake mechanisms.
     if not caps:
         badresponse()
 
     # Flush any output on stderr before proceeding.
     _forwardoutput(ui, stderr)
 
     return protoname, caps
 
 class sshv1peer(wireprotov1peer.wirepeer):
     def __init__(self, ui, url, proc, stdin, stdout, stderr, caps,
                  autoreadstderr=True):
         """Create a peer from an existing SSH connection.
 
         ``proc`` is a handle on the underlying SSH process.
         ``stdin``, ``stdout``, and ``stderr`` are handles on the stdio
         pipes for that process.
         ``caps`` is a set of capabilities supported by the remote.
         ``autoreadstderr`` denotes whether to automatically read from
         stderr and to forward its output.
         """
         self._url = url
         self.ui = ui
         # self._subprocess is unused. Keeping a handle on the process
         # holds a reference and prevents it from being garbage collected.
         self._subprocess = proc
 
         # And we hook up our "doublepipe" wrapper to allow querying
         # stderr any time we perform I/O.
         if autoreadstderr:
             stdout = doublepipe(ui, util.bufferedinputpipe(stdout), stderr)
             stdin = doublepipe(ui, stdin, stderr)
 
         self._pipeo = stdin
         self._pipei = stdout
         self._pipee = stderr
         self._caps = caps
         self._autoreadstderr = autoreadstderr
 
     # Commands that have a "framed" response where the first line of the
     # response contains the length of that response.
     _FRAMED_COMMANDS = {
         'batch',
     }
 
     # Begin of ipeerconnection interface.
 
     def url(self):
         return self._url
 
     def local(self):
         return None
 
     def peer(self):
         return self
 
     def canpush(self):
         return True
 
     def close(self):
         pass
 
     # End of ipeerconnection interface.
 
     # Begin of ipeercommands interface.
 
     def capabilities(self):
         return self._caps
 
     # End of ipeercommands interface.
 
     def _readerr(self):
         _forwardoutput(self.ui, self._pipee)
 
     def _abort(self, exception):
         self._cleanup()
         raise exception
 
     def _cleanup(self):
         _cleanuppipes(self.ui, self._pipei, self._pipeo, self._pipee)
 
     __del__ = _cleanup
 
     def _sendrequest(self, cmd, args, framed=False):
         if (self.ui.debugflag
             and self.ui.configbool('devel', 'debug.peer-request')):
             dbg = self.ui.debug
             line = 'devel-peer-request: %s\n'
             dbg(line % cmd)
             for key, value in sorted(args.items()):
                 if not isinstance(value, dict):
                     dbg(line % '  %s: %d bytes' % (key, len(value)))
                 else:
                     for dk, dv in sorted(value.items()):
                         dbg(line % '  %s-%s: %d' % (key, dk, len(dv)))
         self.ui.debug("sending %s command\n" % cmd)
         self._pipeo.write("%s\n" % cmd)
         _func, names = wireprotov1server.commands[cmd]
         keys = names.split()
         wireargs = {}
         for k in keys:
             if k == '*':
                 wireargs['*'] = args
                 break
             else:
                 wireargs[k] = args[k]
                 del args[k]
         for k, v in sorted(wireargs.iteritems()):
             self._pipeo.write("%s %d\n" % (k, len(v)))
             if isinstance(v, dict):
                 for dk, dv in v.iteritems():
                     self._pipeo.write("%s %d\n" % (dk, len(dv)))
                     self._pipeo.write(dv)
             else:
                 self._pipeo.write(v)
         self._pipeo.flush()
 
         # We know exactly how many bytes are in the response. So return a proxy
         # around the raw output stream that allows reading exactly this many
         # bytes. Callers then can read() without fear of overrunning the
         # response.
         if framed:
             amount = self._getamount()
             return util.cappedreader(self._pipei, amount)
 
         return self._pipei
 
     def _callstream(self, cmd, **args):
         args = pycompat.byteskwargs(args)
         return self._sendrequest(cmd, args, framed=cmd in self._FRAMED_COMMANDS)
 
     def _callcompressable(self, cmd, **args):
         args = pycompat.byteskwargs(args)
         return self._sendrequest(cmd, args, framed=cmd in self._FRAMED_COMMANDS)
 
     def _call(self, cmd, **args):
         args = pycompat.byteskwargs(args)
         return self._sendrequest(cmd, args, framed=True).read()
 
     def _callpush(self, cmd, fp, **args):
         # The server responds with an empty frame if the client should
         # continue submitting the payload.
         r = self._call(cmd, **args)
         if r:
             return '', r
 
         # The payload consists of frames with content followed by an empty
         # frame.
         for d in iter(lambda: fp.read(4096), ''):
             self._writeframed(d)
         self._writeframed("", flush=True)
 
         # In case of success, there is an empty frame and a frame containing
         # the integer result (as a string).
         # In case of error, there is a non-empty frame containing the error.
         r = self._readframed()
         if r:
             return '', r
         return self._readframed(), ''
 
     def _calltwowaystream(self, cmd, fp, **args):
         # The server responds with an empty frame if the client should
         # continue submitting the payload.
         r = self._call(cmd, **args)
         if r:
             # XXX needs to be made better
             raise error.Abort(_('unexpected remote reply: %s') % r)
 
         # The payload consists of frames with content followed by an empty
         # frame.
         for d in iter(lambda: fp.read(4096), ''):
             self._writeframed(d)
         self._writeframed("", flush=True)
 
         return self._pipei
 
     def _getamount(self):
         l = self._pipei.readline()
         if l == '\n':
             if self._autoreadstderr:
                 self._readerr()
             msg = _('check previous remote output')
             self._abort(error.OutOfBandError(hint=msg))
         if self._autoreadstderr:
             self._readerr()
         try:
             return int(l)
         except ValueError:
             self._abort(error.ResponseError(_("unexpected response:"), l))
 
     def _readframed(self):
         size = self._getamount()
         if not size:
             return b''
 
         return self._pipei.read(size)
 
     def _writeframed(self, data, flush=False):
         self._pipeo.write("%d\n" % len(data))
         if data:
             self._pipeo.write(data)
         if flush:
             self._pipeo.flush()
         if self._autoreadstderr:
             self._readerr()
 
 class sshv2peer(sshv1peer):
     """A peer that speakers version 2 of the transport protocol."""
     # Currently version 2 is identical to version 1 post handshake.
     # And handshake is performed before the peer is instantiated. So
     # we need no custom code.
 
 def makepeer(ui, path, proc, stdin, stdout, stderr, autoreadstderr=True):
     """Make a peer instance from existing pipes.
 
     ``path`` and ``proc`` are stored on the eventual peer instance and may
     not be used for anything meaningful.
 
     ``stdin``, ``stdout``, and ``stderr`` are the pipes connected to the
     SSH server's stdio handles.
 
     This function is factored out to allow creating peers that don't
     actually spawn a new process. It is useful for starting SSH protocol
     servers and clients via non-standard means, which can be useful for
     testing.
     """
     try:
         protoname, caps = _performhandshake(ui, stdin, stdout, stderr)
     except Exception:
         _cleanuppipes(ui, stdout, stdin, stderr)
         raise
 
     if protoname == wireprototypes.SSHV1:
         return sshv1peer(ui, path, proc, stdin, stdout, stderr, caps,
                          autoreadstderr=autoreadstderr)
     elif protoname == wireprototypes.SSHV2:
         return sshv2peer(ui, path, proc, stdin, stdout, stderr, caps,
                          autoreadstderr=autoreadstderr)
     else:
         _cleanuppipes(ui, stdout, stdin, stderr)
         raise error.RepoError(_('unknown version of SSH protocol: %s') %
                               protoname)
 
 def instance(ui, path, create, intents=None):
     """Create an SSH peer.
 
     The returned object conforms to the ``wireprotov1peer.wirepeer`` interface.
     """
     u = util.url(path, parsequery=False, parsefragment=False)
     if u.scheme != 'ssh' or not u.host or u.path is None:
         raise error.RepoError(_("couldn't parse location %s") % path)
 
     util.checksafessh(path)
 
     if u.passwd is not None:
         raise error.RepoError(_('password in URL not supported'))
 
     sshcmd = ui.config('ui', 'ssh')
     remotecmd = ui.config('ui', 'remotecmd')
     sshaddenv = dict(ui.configitems('sshenv'))
     sshenv = procutil.shellenviron(sshaddenv)
     remotepath = u.path or '.'
 
     args = procutil.sshargs(sshcmd, u.host, u.user, u.port)
 
     if create:
         cmd = '%s %s %s' % (sshcmd, args,
             procutil.shellquote('%s init %s' %
                 (_serverquote(remotecmd), _serverquote(remotepath))))
         ui.debug('running %s\n' % cmd)
         res = ui.system(cmd, blockedtag='sshpeer', environ=sshenv)
         if res != 0:
             raise error.RepoError(_('could not create remote repo'))
 
     proc, stdin, stdout, stderr = _makeconnection(ui, sshcmd, args, remotecmd,
                                                   remotepath, sshenv)
 
     peer = makepeer(ui, path, proc, stdin, stdout, stderr)
 
     # Finally, if supported by the server, notify it about our own
     # capabilities.
     if 'protocaps' in peer.capabilities():
         try:
             peer._call("protocaps",
                        caps=' '.join(sorted(_clientcapabilities())))
         except IOError:
             peer._cleanup()
             raise error.RepoError(_('capability exchange failed'))
 
     return peer

mercurial/sslutil.py

@@ -1,877 +1,877
 # sslutil.py - SSL handling for mercurial
 #
 # Copyright 2005, 2006, 2007, 2008 Matt Mackall <[e-mail]>
 # Copyright 2006, 2007 Alexis S. L. Carvalho <[e-mail]>
 # Copyright 2006 Vadim Gelfer <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import hashlib
 import os
 import re
 import ssl
 
 from .i18n import _
 from . import (
     error,
     node,
     pycompat,
     util,
 )
 from .utils import (
     procutil,
     stringutil,
 )
 
 # Python 2.7.9+ overhauled the built-in SSL/TLS features of Python. It added
 # support for TLS 1.1, TLS 1.2, SNI, system CA stores, etc. These features are
 # all exposed via the "ssl" module.
 #
 # Depending on the version of Python being used, SSL/TLS support is either
 # modern/secure or legacy/insecure. Many operations in this module have
 # separate code paths depending on support in Python.
 
 configprotocols = {
     'tls1.0',
     'tls1.1',
     'tls1.2',
 }
 
 hassni = getattr(ssl, 'HAS_SNI', False)
 
 # TLS 1.1 and 1.2 may not be supported if the OpenSSL Python is compiled
 # against doesn't support them.
 supportedprotocols = {'tls1.0'}
 if util.safehasattr(ssl, 'PROTOCOL_TLSv1_1'):
     supportedprotocols.add('tls1.1')
 if util.safehasattr(ssl, 'PROTOCOL_TLSv1_2'):
     supportedprotocols.add('tls1.2')
 
 try:
     # ssl.SSLContext was added in 2.7.9 and presence indicates modern
     # SSL/TLS features are available.
     SSLContext = ssl.SSLContext
     modernssl = True
     _canloaddefaultcerts = util.safehasattr(SSLContext, 'load_default_certs')
 except AttributeError:
     modernssl = False
     _canloaddefaultcerts = False
 
     # We implement SSLContext using the interface from the standard library.
     class SSLContext(object):
         def __init__(self, protocol):
             # From the public interface of SSLContext
             self.protocol = protocol
             self.check_hostname = False
             self.options = 0
             self.verify_mode = ssl.CERT_NONE
 
             # Used by our implementation.
             self._certfile = None
             self._keyfile = None
             self._certpassword = None
             self._cacerts = None
             self._ciphers = None
 
         def load_cert_chain(self, certfile, keyfile=None, password=None):
             self._certfile = certfile
             self._keyfile = keyfile
             self._certpassword = password
 
         def load_default_certs(self, purpose=None):
             pass
 
         def load_verify_locations(self, cafile=None, capath=None, cadata=None):
             if capath:
                 raise error.Abort(_('capath not supported'))
             if cadata:
                 raise error.Abort(_('cadata not supported'))
 
             self._cacerts = cafile
 
         def set_ciphers(self, ciphers):
             self._ciphers = ciphers
 
         def wrap_socket(self, socket, server_hostname=None, server_side=False):
             # server_hostname is unique to SSLContext.wrap_socket and is used
             # for SNI in that context. So there's nothing for us to do with it
             # in this legacy code since we don't support SNI.
 
             args = {
                 r'keyfile': self._keyfile,
                 r'certfile': self._certfile,
                 r'server_side': server_side,
                 r'cert_reqs': self.verify_mode,
                 r'ssl_version': self.protocol,
                 r'ca_certs': self._cacerts,
                 r'ciphers': self._ciphers,
             }
 
             return ssl.wrap_socket(socket, **args)
 
 def _hostsettings(ui, hostname):
     """Obtain security settings for a hostname.
 
     Returns a dict of settings relevant to that hostname.
     """
     bhostname = pycompat.bytesurl(hostname)
     s = {
         # Whether we should attempt to load default/available CA certs
         # if an explicit ``cafile`` is not defined.
         'allowloaddefaultcerts': True,
         # List of 2-tuple of (hash algorithm, hash).
         'certfingerprints': [],
         # Path to file containing concatenated CA certs. Used by
         # SSLContext.load_verify_locations().
         'cafile': None,
         # Whether certificate verification should be disabled.
         'disablecertverification': False,
         # Whether the legacy [hostfingerprints] section has data for this host.
         'legacyfingerprint': False,
         # PROTOCOL_* constant to use for SSLContext.__init__.
         'protocol': None,
         # String representation of minimum protocol to be used for UI
         # presentation.
         'protocolui': None,
         # ssl.CERT_* constant used by SSLContext.verify_mode.
         'verifymode': None,
         # Defines extra ssl.OP* bitwise options to set.
         'ctxoptions': None,
         # OpenSSL Cipher List to use (instead of default).
         'ciphers': None,
     }
 
     # Allow minimum TLS protocol to be specified in the config.
     def validateprotocol(protocol, key):
         if protocol not in configprotocols:
             raise error.Abort(
                 _('unsupported protocol from hostsecurity.%s: %s') %
                 (key, protocol),
                 hint=_('valid protocols: %s') %
                      ' '.join(sorted(configprotocols)))
 
     # We default to TLS 1.1+ where we can because TLS 1.0 has known
     # vulnerabilities (like BEAST and POODLE). We allow users to downgrade to
     # TLS 1.0+ via config options in case a legacy server is encountered.
     if 'tls1.1' in supportedprotocols:
         defaultprotocol = 'tls1.1'
     else:
         # Let people know they are borderline secure.
         # We don't document this config option because we want people to see
         # the bold warnings on the web site.
         # internal config: hostsecurity.disabletls10warning
         if not ui.configbool('hostsecurity', 'disabletls10warning'):
             ui.warn(_('warning: connecting to %s using legacy security '
                       'technology (TLS 1.0); see '
                       'https://mercurial-scm.org/wiki/SecureConnections for '
                       'more info\n') % bhostname)
         defaultprotocol = 'tls1.0'
 
     key = 'minimumprotocol'
     protocol = ui.config('hostsecurity', key, defaultprotocol)
     validateprotocol(protocol, key)
 
     key = '%s:minimumprotocol' % bhostname
     protocol = ui.config('hostsecurity', key, protocol)
     validateprotocol(protocol, key)
 
     # If --insecure is used, we allow the use of TLS 1.0 despite config options.
     # We always print a "connection security to %s is disabled..." message when
     # --insecure is used. So no need to print anything more here.
     if ui.insecureconnections:
         protocol = 'tls1.0'
 
     s['protocol'], s['ctxoptions'], s['protocolui'] = protocolsettings(protocol)
 
     ciphers = ui.config('hostsecurity', 'ciphers')
     ciphers = ui.config('hostsecurity', '%s:ciphers' % bhostname, ciphers)
     s['ciphers'] = ciphers
 
     # Look for fingerprints in [hostsecurity] section. Value is a list
     # of <alg>:<fingerprint> strings.
     fingerprints = ui.configlist('hostsecurity', '%s:fingerprints' % bhostname)
     for fingerprint in fingerprints:
         if not (fingerprint.startswith(('sha1:', 'sha256:', 'sha512:'))):
             raise error.Abort(_('invalid fingerprint for %s: %s') % (
                                 bhostname, fingerprint),
                               hint=_('must begin with "sha1:", "sha256:", '
                                      'or "sha512:"'))
 
         alg, fingerprint = fingerprint.split(':', 1)
         fingerprint = fingerprint.replace(':', '').lower()
         s['certfingerprints'].append((alg, fingerprint))
 
     # Fingerprints from [hostfingerprints] are always SHA-1.
     for fingerprint in ui.configlist('hostfingerprints', bhostname):
         fingerprint = fingerprint.replace(':', '').lower()
         s['certfingerprints'].append(('sha1', fingerprint))
         s['legacyfingerprint'] = True
 
     # If a host cert fingerprint is defined, it is the only thing that
     # matters. No need to validate CA certs.
     if s['certfingerprints']:
         s['verifymode'] = ssl.CERT_NONE
         s['allowloaddefaultcerts'] = False
 
     # If --insecure is used, don't take CAs into consideration.
     elif ui.insecureconnections:
         s['disablecertverification'] = True
         s['verifymode'] = ssl.CERT_NONE
         s['allowloaddefaultcerts'] = False
 
     if ui.configbool('devel', 'disableloaddefaultcerts'):
         s['allowloaddefaultcerts'] = False
 
     # If both fingerprints and a per-host ca file are specified, issue a warning
     # because users should not be surprised about what security is or isn't
     # being performed.
     cafile = ui.config('hostsecurity', '%s:verifycertsfile' % bhostname)
     if s['certfingerprints'] and cafile:
         ui.warn(_('(hostsecurity.%s:verifycertsfile ignored when host '
                   'fingerprints defined; using host fingerprints for '
                   'verification)\n') % bhostname)
 
     # Try to hook up CA certificate validation unless something above
     # makes it not necessary.
     if s['verifymode'] is None:
         # Look at per-host ca file first.
         if cafile:
             cafile = util.expandpath(cafile)
             if not os.path.exists(cafile):
                 raise error.Abort(_('path specified by %s does not exist: %s') %
                                   ('hostsecurity.%s:verifycertsfile' % (
                                       bhostname,), cafile))
             s['cafile'] = cafile
         else:
             # Find global certificates file in config.
             cafile = ui.config('web', 'cacerts')
 
             if cafile:
                 cafile = util.expandpath(cafile)
                 if not os.path.exists(cafile):
                     raise error.Abort(_('could not find web.cacerts: %s') %
                                       cafile)
             elif s['allowloaddefaultcerts']:
                 # CAs not defined in config. Try to find system bundles.
                 cafile = _defaultcacerts(ui)
                 if cafile:
                     ui.debug('using %s for CA file\n' % cafile)
 
             s['cafile'] = cafile
 
         # Require certificate validation if CA certs are being loaded and
         # verification hasn't been disabled above.
         if cafile or (_canloaddefaultcerts and s['allowloaddefaultcerts']):
             s['verifymode'] = ssl.CERT_REQUIRED
         else:
             # At this point we don't have a fingerprint, aren't being
             # explicitly insecure, and can't load CA certs. Connecting
             # is insecure. We allow the connection and abort during
             # validation (once we have the fingerprint to print to the
             # user).
             s['verifymode'] = ssl.CERT_NONE
 
     assert s['protocol'] is not None
     assert s['ctxoptions'] is not None
     assert s['verifymode'] is not None
 
     return s
 
 def protocolsettings(protocol):
     """Resolve the protocol for a config value.
 
     Returns a 3-tuple of (protocol, options, ui value) where the first
     2 items are values used by SSLContext and the last is a string value
     of the ``minimumprotocol`` config option equivalent.
     """
     if protocol not in configprotocols:
         raise ValueError('protocol value not supported: %s' % protocol)
 
     # Despite its name, PROTOCOL_SSLv23 selects the highest protocol
     # that both ends support, including TLS protocols. On legacy stacks,
     # the highest it likely goes is TLS 1.0. On modern stacks, it can
     # support TLS 1.2.
     #
     # The PROTOCOL_TLSv* constants select a specific TLS version
     # only (as opposed to multiple versions). So the method for
     # supporting multiple TLS versions is to use PROTOCOL_SSLv23 and
     # disable protocols via SSLContext.options and OP_NO_* constants.
     # However, SSLContext.options doesn't work unless we have the
     # full/real SSLContext available to us.
     if supportedprotocols == {'tls1.0'}:
         if protocol != 'tls1.0':
             raise error.Abort(_('current Python does not support protocol '
                                 'setting %s') % protocol,
                               hint=_('upgrade Python or disable setting since '
                                      'only TLS 1.0 is supported'))
 
         return ssl.PROTOCOL_TLSv1, 0, 'tls1.0'
 
     # WARNING: returned options don't work unless the modern ssl module
     # is available. Be careful when adding options here.
 
     # SSLv2 and SSLv3 are broken. We ban them outright.
     options = ssl.OP_NO_SSLv2 | ssl.OP_NO_SSLv3
 
     if protocol == 'tls1.0':
         # Defaults above are to use TLS 1.0+
         pass
     elif protocol == 'tls1.1':
         options |= ssl.OP_NO_TLSv1
     elif protocol == 'tls1.2':
         options |= ssl.OP_NO_TLSv1 | ssl.OP_NO_TLSv1_1
     else:
         raise error.Abort(_('this should not happen'))
 
     # Prevent CRIME.
     # There is no guarantee this attribute is defined on the module.
     options |= getattr(ssl, 'OP_NO_COMPRESSION', 0)
 
     return ssl.PROTOCOL_SSLv23, options, protocol
 
 def wrapsocket(sock, keyfile, certfile, ui, serverhostname=None):
     """Add SSL/TLS to a socket.
 
     This is a glorified wrapper for ``ssl.wrap_socket()``. It makes sane
     choices based on what security options are available.
 
     In addition to the arguments supported by ``ssl.wrap_socket``, we allow
     the following additional arguments:
 
     * serverhostname - The expected hostname of the remote server. If the
       server (and client) support SNI, this tells the server which certificate
       to use.
     """
     if not serverhostname:
         raise error.Abort(_('serverhostname argument is required'))
 
     for f in (keyfile, certfile):
         if f and not os.path.exists(f):
             raise error.Abort(
                 _('certificate file (%s) does not exist; cannot connect to %s')
                 % (f, pycompat.bytesurl(serverhostname)),
                 hint=_('restore missing file or fix references '
                        'in Mercurial config'))
 
     settings = _hostsettings(ui, serverhostname)
 
     # We can't use ssl.create_default_context() because it calls
     # load_default_certs() unless CA arguments are passed to it. We want to
     # have explicit control over CA loading because implicitly loading
     # CAs may undermine the user's intent. For example, a user may define a CA
     # bundle with a specific CA cert removed. If the system/default CA bundle
     # is loaded and contains that removed CA, you've just undone the user's
     # choice.
     sslcontext = SSLContext(settings['protocol'])
 
     # This is a no-op unless using modern ssl.
     sslcontext.options |= settings['ctxoptions']
 
     # This still works on our fake SSLContext.
     sslcontext.verify_mode = settings['verifymode']
 
     if settings['ciphers']:
         try:
             sslcontext.set_ciphers(pycompat.sysstr(settings['ciphers']))
         except ssl.SSLError as e:
             raise error.Abort(
                 _('could not set ciphers: %s')
                 % stringutil.forcebytestr(e.args[0]),
                 hint=_('change cipher string (%s) in config') %
                 settings['ciphers'])
 
     if certfile is not None:
         def password():
             f = keyfile or certfile
             return ui.getpass(_('passphrase for %s: ') % f, '')
         sslcontext.load_cert_chain(certfile, keyfile, password)
 
     if settings['cafile'] is not None:
         try:
             sslcontext.load_verify_locations(cafile=settings['cafile'])
         except ssl.SSLError as e:
             if len(e.args) == 1: # pypy has different SSLError args
                 msg = e.args[0]
             else:
                 msg = e.args[1]
             raise error.Abort(_('error loading CA file %s: %s') % (
                               settings['cafile'], stringutil.forcebytestr(msg)),
                               hint=_('file is empty or malformed?'))
         caloaded = True
     elif settings['allowloaddefaultcerts']:
         # This is a no-op on old Python.
         sslcontext.load_default_certs()
         caloaded = True
     else:
         caloaded = False
 
     try:
         sslsocket = sslcontext.wrap_socket(sock, server_hostname=serverhostname)
     except ssl.SSLError as e:
         # If we're doing certificate verification and no CA certs are loaded,
         # that is almost certainly the reason why verification failed. Provide
         # a hint to the user.
         # Only modern ssl module exposes SSLContext.get_ca_certs() so we can
         # only show this warning if modern ssl is available.
         # The exception handler is here to handle bugs around cert attributes:
         # https://bugs.python.org/issue20916#msg213479.  (See issues5313.)
         # When the main 20916 bug occurs, 'sslcontext.get_ca_certs()' is a
         # non-empty list, but the following conditional is otherwise True.
         try:
             if (caloaded and settings['verifymode'] == ssl.CERT_REQUIRED and
                 modernssl and not sslcontext.get_ca_certs()):
                 ui.warn(_('(an attempt was made to load CA certificates but '
                           'none were loaded; see '
                           'https://mercurial-scm.org/wiki/SecureConnections '
                           'for how to configure Mercurial to avoid this '
                           'error)\n'))
         except ssl.SSLError:
             pass
         # Try to print more helpful error messages for known failures.
         if util.safehasattr(e, 'reason'):
             # This error occurs when the client and server don't share a
             # common/supported SSL/TLS protocol. We've disabled SSLv2 and SSLv3
             # outright. Hopefully the reason for this error is that we require
             # TLS 1.1+ and the server only supports TLS 1.0. Whatever the
             # reason, try to emit an actionable warning.
             if e.reason == 'UNSUPPORTED_PROTOCOL':
                 # We attempted TLS 1.0+.
                 if settings['protocolui'] == 'tls1.0':
                     # We support more than just TLS 1.0+. If this happens,
                     # the likely scenario is either the client or the server
                     # is really old. (e.g. server doesn't support TLS 1.0+ or
                     # client doesn't support modern TLS versions introduced
                     # several years from when this comment was written).
                     if supportedprotocols != {'tls1.0'}:
                         ui.warn(_(
                             '(could not communicate with %s using security '
                             'protocols %s; if you are using a modern Mercurial '
                             'version, consider contacting the operator of this '
                             'server; see '
                             'https://mercurial-scm.org/wiki/SecureConnections '
                             'for more info)\n') % (
                                 serverhostname,
                                 ', '.join(sorted(supportedprotocols))))
                     else:
                         ui.warn(_(
                             '(could not communicate with %s using TLS 1.0; the '
                             'likely cause of this is the server no longer '
                             'supports TLS 1.0 because it has known security '
                             'vulnerabilities; see '
                             'https://mercurial-scm.org/wiki/SecureConnections '
                             'for more info)\n') % serverhostname)
                 else:
                     # We attempted TLS 1.1+. We can only get here if the client
                     # supports the configured protocol. So the likely reason is
                     # the client wants better security than the server can
                     # offer.
                     ui.warn(_(
                         '(could not negotiate a common security protocol (%s+) '
                         'with %s; the likely cause is Mercurial is configured '
                         'to be more secure than the server can support)\n') % (
                         settings['protocolui'], serverhostname))
                     ui.warn(_('(consider contacting the operator of this '
                               'server and ask them to support modern TLS '
                               'protocol versions; or, set '
                               'hostsecurity.%s:minimumprotocol=tls1.0 to allow '
                               'use of legacy, less secure protocols when '
                               'communicating with this server)\n') %
                             serverhostname)
                     ui.warn(_(
                         '(see https://mercurial-scm.org/wiki/SecureConnections '
                         'for more info)\n'))
 
             elif (e.reason == 'CERTIFICATE_VERIFY_FAILED' and
                 pycompat.iswindows):
 
                 ui.warn(_('(the full certificate chain may not be available '
                           'locally; see "hg help debugssl")\n'))
         raise
 
     # check if wrap_socket failed silently because socket had been
     # closed
     # - see http://bugs.python.org/issue13721
     if not sslsocket.cipher():
         raise error.Abort(_('ssl connection failed'))
 
     sslsocket._hgstate = {
         'caloaded': caloaded,
         'hostname': serverhostname,
         'settings': settings,
         'ui': ui,
     }
 
     return sslsocket
 
 def wrapserversocket(sock, ui, certfile=None, keyfile=None, cafile=None,
                      requireclientcert=False):
     """Wrap a socket for use by servers.
 
     ``certfile`` and ``keyfile`` specify the files containing the certificate's
     public and private keys, respectively. Both keys can be defined in the same
     file via ``certfile`` (the private key must come first in the file).
 
     ``cafile`` defines the path to certificate authorities.
 
     ``requireclientcert`` specifies whether to require client certificates.
 
     Typically ``cafile`` is only defined if ``requireclientcert`` is true.
     """
     # This function is not used much by core Mercurial, so the error messaging
     # doesn't have to be as detailed as for wrapsocket().
     for f in (certfile, keyfile, cafile):
         if f and not os.path.exists(f):
             raise error.Abort(_('referenced certificate file (%s) does not '
                                 'exist') % f)
 
     protocol, options, _protocolui = protocolsettings('tls1.0')
 
     # This config option is intended for use in tests only. It is a giant
     # footgun to kill security. Don't define it.
     exactprotocol = ui.config('devel', 'serverexactprotocol')
     if exactprotocol == 'tls1.0':
         protocol = ssl.PROTOCOL_TLSv1
     elif exactprotocol == 'tls1.1':
         if 'tls1.1' not in supportedprotocols:
             raise error.Abort(_('TLS 1.1 not supported by this Python'))
         protocol = ssl.PROTOCOL_TLSv1_1
     elif exactprotocol == 'tls1.2':
         if 'tls1.2' not in supportedprotocols:
             raise error.Abort(_('TLS 1.2 not supported by this Python'))
         protocol = ssl.PROTOCOL_TLSv1_2
     elif exactprotocol:
         raise error.Abort(_('invalid value for serverexactprotocol: %s') %
                           exactprotocol)
 
     if modernssl:
         # We /could/ use create_default_context() here since it doesn't load
         # CAs when configured for client auth. However, it is hard-coded to
         # use ssl.PROTOCOL_SSLv23 which may not be appropriate here.
         sslcontext = SSLContext(protocol)
         sslcontext.options |= options
 
         # Improve forward secrecy.
         sslcontext.options |= getattr(ssl, 'OP_SINGLE_DH_USE', 0)
         sslcontext.options |= getattr(ssl, 'OP_SINGLE_ECDH_USE', 0)
 
         # Use the list of more secure ciphers if found in the ssl module.
         if util.safehasattr(ssl, '_RESTRICTED_SERVER_CIPHERS'):
             sslcontext.options |= getattr(ssl, 'OP_CIPHER_SERVER_PREFERENCE', 0)
             sslcontext.set_ciphers(ssl._RESTRICTED_SERVER_CIPHERS)
     else:
         sslcontext = SSLContext(ssl.PROTOCOL_TLSv1)
 
     if requireclientcert:
         sslcontext.verify_mode = ssl.CERT_REQUIRED
     else:
         sslcontext.verify_mode = ssl.CERT_NONE
 
     if certfile or keyfile:
         sslcontext.load_cert_chain(certfile=certfile, keyfile=keyfile)
 
     if cafile:
         sslcontext.load_verify_locations(cafile=cafile)
 
     return sslcontext.wrap_socket(sock, server_side=True)
 
 class wildcarderror(Exception):
     """Represents an error parsing wildcards in DNS name."""
 
 def _dnsnamematch(dn, hostname, maxwildcards=1):
     """Match DNS names according RFC 6125 section 6.4.3.
 
     This code is effectively copied from CPython's ssl._dnsname_match.
 
     Returns a bool indicating whether the expected hostname matches
     the value in ``dn``.
     """
     pats = []
     if not dn:
         return False
     dn = pycompat.bytesurl(dn)
     hostname = pycompat.bytesurl(hostname)
 
     pieces = dn.split('.')
     leftmost = pieces[0]
     remainder = pieces[1:]
     wildcards = leftmost.count('*')
     if wildcards > maxwildcards:
         raise wildcarderror(
             _('too many wildcards in certificate DNS name: %s') % dn)
 
     # speed up common case w/o wildcards
     if not wildcards:
         return dn.lower() == hostname.lower()
 
     # RFC 6125, section 6.4.3, subitem 1.
     # The client SHOULD NOT attempt to match a presented identifier in which
     # the wildcard character comprises a label other than the left-most label.
     if leftmost == '*':
         # When '*' is a fragment by itself, it matches a non-empty dotless
         # fragment.
         pats.append('[^.]+')
     elif leftmost.startswith('xn--') or hostname.startswith('xn--'):
         # RFC 6125, section 6.4.3, subitem 3.
         # The client SHOULD NOT attempt to match a presented identifier
         # where the wildcard character is embedded within an A-label or
         # U-label of an internationalized domain name.
-        pats.append(re.escape(leftmost))
+        pats.append(stringutil.reescape(leftmost))
     else:
         # Otherwise, '*' matches any dotless string, e.g. www*
-        pats.append(re.escape(leftmost).replace(br'\*', '[^.]*'))
+        pats.append(stringutil.reescape(leftmost).replace(br'\*', '[^.]*'))
 
     # add the remaining fragments, ignore any wildcards
     for frag in remainder:
-        pats.append(re.escape(frag))
+        pats.append(stringutil.reescape(frag))
 
     pat = re.compile(br'\A' + br'\.'.join(pats) + br'\Z', re.IGNORECASE)
     return pat.match(hostname) is not None
 
 def _verifycert(cert, hostname):
     '''Verify that cert (in socket.getpeercert() format) matches hostname.
     CRLs is not handled.
 
     Returns error message if any problems are found and None on success.
     '''
     if not cert:
         return _('no certificate received')
 
     dnsnames = []
     san = cert.get(r'subjectAltName', [])
     for key, value in san:
         if key == r'DNS':
             try:
                 if _dnsnamematch(value, hostname):
                     return
             except wildcarderror as e:
                 return stringutil.forcebytestr(e.args[0])
 
             dnsnames.append(value)
 
     if not dnsnames:
         # The subject is only checked when there is no DNS in subjectAltName.
         for sub in cert.get(r'subject', []):
             for key, value in sub:
                 # According to RFC 2818 the most specific Common Name must
                 # be used.
                 if key == r'commonName':
                     # 'subject' entries are unicode.
                     try:
                         value = value.encode('ascii')
                     except UnicodeEncodeError:
                         return _('IDN in certificate not supported')
 
                     try:
                         if _dnsnamematch(value, hostname):
                             return
                     except wildcarderror as e:
                         return stringutil.forcebytestr(e.args[0])
 
                     dnsnames.append(value)
 
     dnsnames = [pycompat.bytesurl(d) for d in dnsnames]
     if len(dnsnames) > 1:
         return _('certificate is for %s') % ', '.join(dnsnames)
     elif len(dnsnames) == 1:
         return _('certificate is for %s') % dnsnames[0]
     else:
         return _('no commonName or subjectAltName found in certificate')
 
 def _plainapplepython():
     """return true if this seems to be a pure Apple Python that
     * is unfrozen and presumably has the whole mercurial module in the file
       system
     * presumably is an Apple Python that uses Apple OpenSSL which has patches
       for using system certificate store CAs in addition to the provided
       cacerts file
     """
     if (not pycompat.isdarwin or procutil.mainfrozen() or
         not pycompat.sysexecutable):
         return False
     exe = os.path.realpath(pycompat.sysexecutable).lower()
     return (exe.startswith('/usr/bin/python') or
             exe.startswith('/system/library/frameworks/python.framework/'))
 
 _systemcacertpaths = [
     # RHEL, CentOS, and Fedora
     '/etc/pki/tls/certs/ca-bundle.trust.crt',
     # Debian, Ubuntu, Gentoo
     '/etc/ssl/certs/ca-certificates.crt',
 ]
 
 def _defaultcacerts(ui):
     """return path to default CA certificates or None.
 
     It is assumed this function is called when the returned certificates
     file will actually be used to validate connections. Therefore this
     function may print warnings or debug messages assuming this usage.
 
     We don't print a message when the Python is able to load default
     CA certs because this scenario is detected at socket connect time.
     """
     # The "certifi" Python package provides certificates. If it is installed
     # and usable, assume the user intends it to be used and use it.
     try:
         import certifi
         certs = certifi.where()
         if os.path.exists(certs):
             ui.debug('using ca certificates from certifi\n')
             return certs
     except (ImportError, AttributeError):
         pass
 
     # On Windows, only the modern ssl module is capable of loading the system
     # CA certificates. If we're not capable of doing that, emit a warning
     # because we'll get a certificate verification error later and the lack
     # of loaded CA certificates will be the reason why.
     # Assertion: this code is only called if certificates are being verified.
     if pycompat.iswindows:
         if not _canloaddefaultcerts:
             ui.warn(_('(unable to load Windows CA certificates; see '
                       'https://mercurial-scm.org/wiki/SecureConnections for '
                       'how to configure Mercurial to avoid this message)\n'))
 
         return None
 
     # Apple's OpenSSL has patches that allow a specially constructed certificate
     # to load the system CA store. If we're running on Apple Python, use this
     # trick.
     if _plainapplepython():
         dummycert = os.path.join(
             os.path.dirname(pycompat.fsencode(__file__)), 'dummycert.pem')
         if os.path.exists(dummycert):
             return dummycert
 
     # The Apple OpenSSL trick isn't available to us. If Python isn't able to
     # load system certs, we're out of luck.
     if pycompat.isdarwin:
         # FUTURE Consider looking for Homebrew or MacPorts installed certs
         # files. Also consider exporting the keychain certs to a file during
         # Mercurial install.
         if not _canloaddefaultcerts:
             ui.warn(_('(unable to load CA certificates; see '
                       'https://mercurial-scm.org/wiki/SecureConnections for '
                       'how to configure Mercurial to avoid this message)\n'))
         return None
 
     # / is writable on Windows. Out of an abundance of caution make sure
     # we're not on Windows because paths from _systemcacerts could be installed
     # by non-admin users.
     assert not pycompat.iswindows
 
     # Try to find CA certificates in well-known locations. We print a warning
     # when using a found file because we don't want too much silent magic
     # for security settings. The expectation is that proper Mercurial
     # installs will have the CA certs path defined at install time and the
     # installer/packager will make an appropriate decision on the user's
     # behalf. We only get here and perform this setting as a feature of
     # last resort.
     if not _canloaddefaultcerts:
         for path in _systemcacertpaths:
             if os.path.isfile(path):
                 ui.warn(_('(using CA certificates from %s; if you see this '
                           'message, your Mercurial install is not properly '
                           'configured; see '
                           'https://mercurial-scm.org/wiki/SecureConnections '
                           'for how to configure Mercurial to avoid this '
                           'message)\n') % path)
                 return path
 
         ui.warn(_('(unable to load CA certificates; see '
                   'https://mercurial-scm.org/wiki/SecureConnections for '
                   'how to configure Mercurial to avoid this message)\n'))
 
     return None
 
 def validatesocket(sock):
     """Validate a socket meets security requirements.
 
     The passed socket must have been created with ``wrapsocket()``.
     """
     shost = sock._hgstate['hostname']
     host = pycompat.bytesurl(shost)
     ui = sock._hgstate['ui']
     settings = sock._hgstate['settings']
 
     try:
         peercert = sock.getpeercert(True)
         peercert2 = sock.getpeercert()
     except AttributeError:
         raise error.Abort(_('%s ssl connection error') % host)
 
     if not peercert:
         raise error.Abort(_('%s certificate error: '
                            'no certificate received') % host)
 
     if settings['disablecertverification']:
         # We don't print the certificate fingerprint because it shouldn't
         # be necessary: if the user requested certificate verification be
         # disabled, they presumably already saw a message about the inability
         # to verify the certificate and this message would have printed the
         # fingerprint. So printing the fingerprint here adds little to no
         # value.
         ui.warn(_('warning: connection security to %s is disabled per current '
                   'settings; communication is susceptible to eavesdropping '
                   'and tampering\n') % host)
         return
 
     # If a certificate fingerprint is pinned, use it and only it to
     # validate the remote cert.
     peerfingerprints = {
         'sha1': node.hex(hashlib.sha1(peercert).digest()),
         'sha256': node.hex(hashlib.sha256(peercert).digest()),
         'sha512': node.hex(hashlib.sha512(peercert).digest()),
     }
 
     def fmtfingerprint(s):
         return ':'.join([s[x:x + 2] for x in range(0, len(s), 2)])
 
     nicefingerprint = 'sha256:%s' % fmtfingerprint(peerfingerprints['sha256'])
 
     if settings['certfingerprints']:
         for hash, fingerprint in settings['certfingerprints']:
             if peerfingerprints[hash].lower() == fingerprint:
                 ui.debug('%s certificate matched fingerprint %s:%s\n' %
                          (host, hash, fmtfingerprint(fingerprint)))
                 if settings['legacyfingerprint']:
                     ui.warn(_('(SHA-1 fingerprint for %s found in legacy '
                               '[hostfingerprints] section; '
                               'if you trust this fingerprint, remove the old '
                               'SHA-1 fingerprint from [hostfingerprints] and '
                               'add the following entry to the new '
                               '[hostsecurity] section: %s:fingerprints=%s)\n') %
                             (host, host, nicefingerprint))
                 return
 
         # Pinned fingerprint didn't match. This is a fatal error.
         if settings['legacyfingerprint']:
             section = 'hostfingerprint'
             nice = fmtfingerprint(peerfingerprints['sha1'])
         else:
             section = 'hostsecurity'
             nice = '%s:%s' % (hash, fmtfingerprint(peerfingerprints[hash]))
         raise error.Abort(_('certificate for %s has unexpected '
                             'fingerprint %s') % (host, nice),
                           hint=_('check %s configuration') % section)
 
     # Security is enabled but no CAs are loaded. We can't establish trust
     # for the cert so abort.
     if not sock._hgstate['caloaded']:
         raise error.Abort(
             _('unable to verify security of %s (no loaded CA certificates); '
               'refusing to connect') % host,
             hint=_('see https://mercurial-scm.org/wiki/SecureConnections for '
                    'how to configure Mercurial to avoid this error or set '
                    'hostsecurity.%s:fingerprints=%s to trust this server') %
                    (host, nicefingerprint))
 
     msg = _verifycert(peercert2, shost)
     if msg:
         raise error.Abort(_('%s certificate error: %s') % (host, msg),
                          hint=_('set hostsecurity.%s:certfingerprints=%s '
                                 'config setting or use --insecure to connect '
                                 'insecurely') %
                               (host, nicefingerprint))

tests/test-walk.t

@@ -1,652 +1,652
   $ hg init t
   $ cd t
   $ mkdir -p beans
   $ for b in kidney navy turtle borlotti black pinto; do
   >     echo $b > beans/$b
   > done
   $ mkdir -p mammals/Procyonidae
   $ for m in cacomistle coatimundi raccoon; do
   >     echo $m > mammals/Procyonidae/$m
   > done
   $ echo skunk > mammals/skunk
   $ echo fennel > fennel
   $ echo fenugreek > fenugreek
   $ echo fiddlehead > fiddlehead
   $ hg addremove
   adding beans/black
   adding beans/borlotti
   adding beans/kidney
   adding beans/navy
   adding beans/pinto
   adding beans/turtle
   adding fennel
   adding fenugreek
   adding fiddlehead
   adding mammals/Procyonidae/cacomistle
   adding mammals/Procyonidae/coatimundi
   adding mammals/Procyonidae/raccoon
   adding mammals/skunk
   $ hg commit -m "commit #0"
 
   $ hg debugwalk -v
   * matcher:
   <alwaysmatcher>
   f  beans/black                     beans/black
   f  beans/borlotti                  beans/borlotti
   f  beans/kidney                    beans/kidney
   f  beans/navy                      beans/navy
   f  beans/pinto                     beans/pinto
   f  beans/turtle                    beans/turtle
   f  fennel                          fennel
   f  fenugreek                       fenugreek
   f  fiddlehead                      fiddlehead
   f  mammals/Procyonidae/cacomistle  mammals/Procyonidae/cacomistle
   f  mammals/Procyonidae/coatimundi  mammals/Procyonidae/coatimundi
   f  mammals/Procyonidae/raccoon     mammals/Procyonidae/raccoon
   f  mammals/skunk                   mammals/skunk
   $ hg debugwalk -v -I.
   * matcher:
   <includematcher includes='(?:)'>
   f  beans/black                     beans/black
   f  beans/borlotti                  beans/borlotti
   f  beans/kidney                    beans/kidney
   f  beans/navy                      beans/navy
   f  beans/pinto                     beans/pinto
   f  beans/turtle                    beans/turtle
   f  fennel                          fennel
   f  fenugreek                       fenugreek
   f  fiddlehead                      fiddlehead
   f  mammals/Procyonidae/cacomistle  mammals/Procyonidae/cacomistle
   f  mammals/Procyonidae/coatimundi  mammals/Procyonidae/coatimundi
   f  mammals/Procyonidae/raccoon     mammals/Procyonidae/raccoon
   f  mammals/skunk                   mammals/skunk
 
   $ cd mammals
   $ hg debugwalk -v
   * matcher:
   <alwaysmatcher>
   f  beans/black                     ../beans/black
   f  beans/borlotti                  ../beans/borlotti
   f  beans/kidney                    ../beans/kidney
   f  beans/navy                      ../beans/navy
   f  beans/pinto                     ../beans/pinto
   f  beans/turtle                    ../beans/turtle
   f  fennel                          ../fennel
   f  fenugreek                       ../fenugreek
   f  fiddlehead                      ../fiddlehead
   f  mammals/Procyonidae/cacomistle  Procyonidae/cacomistle
   f  mammals/Procyonidae/coatimundi  Procyonidae/coatimundi
   f  mammals/Procyonidae/raccoon     Procyonidae/raccoon
   f  mammals/skunk                   skunk
   $ hg debugwalk -v -X ../beans
   * matcher:
   <differencematcher
     m1=<alwaysmatcher>,
     m2=<includematcher includes='(?:beans(?:/|$))'>>
   f  fennel                          ../fennel
   f  fenugreek                       ../fenugreek
   f  fiddlehead                      ../fiddlehead
   f  mammals/Procyonidae/cacomistle  Procyonidae/cacomistle
   f  mammals/Procyonidae/coatimundi  Procyonidae/coatimundi
   f  mammals/Procyonidae/raccoon     Procyonidae/raccoon
   f  mammals/skunk                   skunk
   $ hg debugwalk -v -I '*k'
   * matcher:
-  <includematcher includes='(?:mammals\\/[^/]*k(?:/|$))'>
+  <includematcher includes='(?:mammals/[^/]*k(?:/|$))'>
   f  mammals/skunk  skunk
   $ hg debugwalk -v -I 'glob:*k'
   * matcher:
-  <includematcher includes='(?:mammals\\/[^/]*k(?:/|$))'>
+  <includematcher includes='(?:mammals/[^/]*k(?:/|$))'>
   f  mammals/skunk  skunk
   $ hg debugwalk -v -I 'relglob:*k'
   * matcher:
   <includematcher includes='(?:(?:|.*/)[^/]*k(?:/|$))'>
   f  beans/black    ../beans/black
   f  fenugreek      ../fenugreek
   f  mammals/skunk  skunk
   $ hg debugwalk -v -I 'relglob:*k' .
   * matcher:
   <intersectionmatcher
     m1=<patternmatcher patterns='(?:mammals(?:/|$))'>,
     m2=<includematcher includes='(?:(?:|.*/)[^/]*k(?:/|$))'>>
   f  mammals/skunk  skunk
   $ hg debugwalk -v -I 're:.*k$'
   * matcher:
   <includematcher includes='(?:.*k$)'>
   f  beans/black    ../beans/black
   f  fenugreek      ../fenugreek
   f  mammals/skunk  skunk
   $ hg debugwalk -v -I 'relre:.*k$'
   * matcher:
   <includematcher includes='(?:.*.*k$)'>
   f  beans/black    ../beans/black
   f  fenugreek      ../fenugreek
   f  mammals/skunk  skunk
   $ hg debugwalk -v -I 'path:beans'
   * matcher:
   <includematcher includes='(?:beans(?:/|$))'>
   f  beans/black     ../beans/black
   f  beans/borlotti  ../beans/borlotti
   f  beans/kidney    ../beans/kidney
   f  beans/navy      ../beans/navy
   f  beans/pinto     ../beans/pinto
   f  beans/turtle    ../beans/turtle
   $ hg debugwalk -v -I 'relpath:detour/../../beans'
   * matcher:
   <includematcher includes='(?:beans(?:/|$))'>
   f  beans/black     ../beans/black
   f  beans/borlotti  ../beans/borlotti
   f  beans/kidney    ../beans/kidney
   f  beans/navy      ../beans/navy
   f  beans/pinto     ../beans/pinto
   f  beans/turtle    ../beans/turtle
 
   $ hg debugwalk -v 'rootfilesin:'
   * matcher:
   <patternmatcher patterns='(?:[^/]+$)'>
   f  fennel      ../fennel
   f  fenugreek   ../fenugreek
   f  fiddlehead  ../fiddlehead
   $ hg debugwalk -v -I 'rootfilesin:'
   * matcher:
   <includematcher includes='(?:[^/]+$)'>
   f  fennel      ../fennel
   f  fenugreek   ../fenugreek
   f  fiddlehead  ../fiddlehead
   $ hg debugwalk -v 'rootfilesin:.'
   * matcher:
   <patternmatcher patterns='(?:[^/]+$)'>
   f  fennel      ../fennel
   f  fenugreek   ../fenugreek
   f  fiddlehead  ../fiddlehead
   $ hg debugwalk -v -I 'rootfilesin:.'
   * matcher:
   <includematcher includes='(?:[^/]+$)'>
   f  fennel      ../fennel
   f  fenugreek   ../fenugreek
   f  fiddlehead  ../fiddlehead
   $ hg debugwalk -v -X 'rootfilesin:'
   * matcher:
   <differencematcher
     m1=<alwaysmatcher>,
     m2=<includematcher includes='(?:[^/]+$)'>>
   f  beans/black                     ../beans/black
   f  beans/borlotti                  ../beans/borlotti
   f  beans/kidney                    ../beans/kidney
   f  beans/navy                      ../beans/navy
   f  beans/pinto                     ../beans/pinto
   f  beans/turtle                    ../beans/turtle
   f  mammals/Procyonidae/cacomistle  Procyonidae/cacomistle
   f  mammals/Procyonidae/coatimundi  Procyonidae/coatimundi
   f  mammals/Procyonidae/raccoon     Procyonidae/raccoon
   f  mammals/skunk                   skunk
   $ hg debugwalk -v 'rootfilesin:fennel'
   * matcher:
   <patternmatcher patterns='(?:fennel/[^/]+$)'>
   $ hg debugwalk -v -I 'rootfilesin:fennel'
   * matcher:
   <includematcher includes='(?:fennel/[^/]+$)'>
   $ hg debugwalk -v 'rootfilesin:skunk'
   * matcher:
   <patternmatcher patterns='(?:skunk/[^/]+$)'>
   $ hg debugwalk -v -I 'rootfilesin:skunk'
   * matcher:
   <includematcher includes='(?:skunk/[^/]+$)'>
   $ hg debugwalk -v 'rootfilesin:beans'
   * matcher:
   <patternmatcher patterns='(?:beans/[^/]+$)'>
   f  beans/black     ../beans/black
   f  beans/borlotti  ../beans/borlotti
   f  beans/kidney    ../beans/kidney
   f  beans/navy      ../beans/navy
   f  beans/pinto     ../beans/pinto
   f  beans/turtle    ../beans/turtle
   $ hg debugwalk -v -I 'rootfilesin:beans'
   * matcher:
   <includematcher includes='(?:beans/[^/]+$)'>
   f  beans/black     ../beans/black
   f  beans/borlotti  ../beans/borlotti
   f  beans/kidney    ../beans/kidney
   f  beans/navy      ../beans/navy
   f  beans/pinto     ../beans/pinto
   f  beans/turtle    ../beans/turtle
   $ hg debugwalk -v 'rootfilesin:mammals'
   * matcher:
   <patternmatcher patterns='(?:mammals/[^/]+$)'>
   f  mammals/skunk  skunk
   $ hg debugwalk -v -I 'rootfilesin:mammals'
   * matcher:
   <includematcher includes='(?:mammals/[^/]+$)'>
   f  mammals/skunk  skunk
   $ hg debugwalk -v 'rootfilesin:mammals/'
   * matcher:
   <patternmatcher patterns='(?:mammals/[^/]+$)'>
   f  mammals/skunk  skunk
   $ hg debugwalk -v -I 'rootfilesin:mammals/'
   * matcher:
   <includematcher includes='(?:mammals/[^/]+$)'>
   f  mammals/skunk  skunk
   $ hg debugwalk -v -X 'rootfilesin:mammals'
   * matcher:
   <differencematcher
     m1=<alwaysmatcher>,
     m2=<includematcher includes='(?:mammals/[^/]+$)'>>
   f  beans/black                     ../beans/black
   f  beans/borlotti                  ../beans/borlotti
   f  beans/kidney                    ../beans/kidney
   f  beans/navy                      ../beans/navy
   f  beans/pinto                     ../beans/pinto
   f  beans/turtle                    ../beans/turtle
   f  fennel                          ../fennel
   f  fenugreek                       ../fenugreek
   f  fiddlehead                      ../fiddlehead
   f  mammals/Procyonidae/cacomistle  Procyonidae/cacomistle
   f  mammals/Procyonidae/coatimundi  Procyonidae/coatimundi
   f  mammals/Procyonidae/raccoon     Procyonidae/raccoon
 
   $ hg debugwalk -v .
   * matcher:
   <patternmatcher patterns='(?:mammals(?:/|$))'>
   f  mammals/Procyonidae/cacomistle  Procyonidae/cacomistle
   f  mammals/Procyonidae/coatimundi  Procyonidae/coatimundi
   f  mammals/Procyonidae/raccoon     Procyonidae/raccoon
   f  mammals/skunk                   skunk
   $ hg debugwalk -v -I.
   * matcher:
   <includematcher includes='(?:mammals(?:/|$))'>
   f  mammals/Procyonidae/cacomistle  Procyonidae/cacomistle
   f  mammals/Procyonidae/coatimundi  Procyonidae/coatimundi
   f  mammals/Procyonidae/raccoon     Procyonidae/raccoon
   f  mammals/skunk                   skunk
   $ hg debugwalk -v Procyonidae
   * matcher:
-  <patternmatcher patterns='(?:mammals\\/Procyonidae(?:/|$))'>
+  <patternmatcher patterns='(?:mammals/Procyonidae(?:/|$))'>
   f  mammals/Procyonidae/cacomistle  Procyonidae/cacomistle
   f  mammals/Procyonidae/coatimundi  Procyonidae/coatimundi
   f  mammals/Procyonidae/raccoon     Procyonidae/raccoon
 
   $ cd Procyonidae
   $ hg debugwalk -v .
   * matcher:
-  <patternmatcher patterns='(?:mammals\\/Procyonidae(?:/|$))'>
+  <patternmatcher patterns='(?:mammals/Procyonidae(?:/|$))'>
   f  mammals/Procyonidae/cacomistle  cacomistle
   f  mammals/Procyonidae/coatimundi  coatimundi
   f  mammals/Procyonidae/raccoon     raccoon
   $ hg debugwalk -v ..
   * matcher:
   <patternmatcher patterns='(?:mammals(?:/|$))'>
   f  mammals/Procyonidae/cacomistle  cacomistle
   f  mammals/Procyonidae/coatimundi  coatimundi
   f  mammals/Procyonidae/raccoon     raccoon
   f  mammals/skunk                   ../skunk
   $ cd ..
 
   $ hg debugwalk -v ../beans
   * matcher:
   <patternmatcher patterns='(?:beans(?:/|$))'>
   f  beans/black     ../beans/black
   f  beans/borlotti  ../beans/borlotti
   f  beans/kidney    ../beans/kidney
   f  beans/navy      ../beans/navy
   f  beans/pinto     ../beans/pinto
   f  beans/turtle    ../beans/turtle
   $ hg debugwalk -v .
   * matcher:
   <patternmatcher patterns='(?:mammals(?:/|$))'>
   f  mammals/Procyonidae/cacomistle  Procyonidae/cacomistle
   f  mammals/Procyonidae/coatimundi  Procyonidae/coatimundi
   f  mammals/Procyonidae/raccoon     Procyonidae/raccoon
   f  mammals/skunk                   skunk
   $ hg debugwalk -v .hg
   abort: path 'mammals/.hg' is inside nested repo 'mammals'
   [255]
   $ hg debugwalk -v ../.hg
   abort: path contains illegal component: .hg
   [255]
   $ cd ..
 
   $ hg debugwalk -v -Ibeans
   * matcher:
   <includematcher includes='(?:beans(?:/|$))'>
   f  beans/black     beans/black
   f  beans/borlotti  beans/borlotti
   f  beans/kidney    beans/kidney
   f  beans/navy      beans/navy
   f  beans/pinto     beans/pinto
   f  beans/turtle    beans/turtle
   $ hg debugwalk -v -I '{*,{b,m}*/*}k'
   * matcher:
-  <includematcher includes='(?:(?:[^/]*|(?:b|m)[^/]*\\/[^/]*)k(?:/|$))'>
+  <includematcher includes='(?:(?:[^/]*|(?:b|m)[^/]*/[^/]*)k(?:/|$))'>
   f  beans/black    beans/black
   f  fenugreek      fenugreek
   f  mammals/skunk  mammals/skunk
   $ hg debugwalk -v -Ibeans mammals
   * matcher:
   <intersectionmatcher
     m1=<patternmatcher patterns='(?:mammals(?:/|$))'>,
     m2=<includematcher includes='(?:beans(?:/|$))'>>
   $ hg debugwalk -v -Inon-existent
   * matcher:
   <includematcher includes='(?:non\\-existent(?:/|$))'>
   $ hg debugwalk -v -Inon-existent -Ibeans/black
   * matcher:
-  <includematcher includes='(?:non\\-existent(?:/|$)|beans\\/black(?:/|$))'>
+  <includematcher includes='(?:non\\-existent(?:/|$)|beans/black(?:/|$))'>
   f  beans/black  beans/black
   $ hg debugwalk -v -Ibeans beans/black
   * matcher:
   <intersectionmatcher
-    m1=<patternmatcher patterns='(?:beans\\/black(?:/|$))'>,
+    m1=<patternmatcher patterns='(?:beans/black(?:/|$))'>,
     m2=<includematcher includes='(?:beans(?:/|$))'>>
   f  beans/black  beans/black  exact
   $ hg debugwalk -v -Ibeans/black beans
   * matcher:
   <intersectionmatcher
     m1=<patternmatcher patterns='(?:beans(?:/|$))'>,
-    m2=<includematcher includes='(?:beans\\/black(?:/|$))'>>
+    m2=<includematcher includes='(?:beans/black(?:/|$))'>>
   f  beans/black  beans/black
   $ hg debugwalk -v -Xbeans/black beans
   * matcher:
   <differencematcher
     m1=<patternmatcher patterns='(?:beans(?:/|$))'>,
-    m2=<includematcher includes='(?:beans\\/black(?:/|$))'>>
+    m2=<includematcher includes='(?:beans/black(?:/|$))'>>
   f  beans/borlotti  beans/borlotti
   f  beans/kidney    beans/kidney
   f  beans/navy      beans/navy
   f  beans/pinto     beans/pinto
   f  beans/turtle    beans/turtle
   $ hg debugwalk -v -Xbeans/black -Ibeans
   * matcher:
   <differencematcher
     m1=<includematcher includes='(?:beans(?:/|$))'>,
-    m2=<includematcher includes='(?:beans\\/black(?:/|$))'>>
+    m2=<includematcher includes='(?:beans/black(?:/|$))'>>
   f  beans/borlotti  beans/borlotti
   f  beans/kidney    beans/kidney
   f  beans/navy      beans/navy
   f  beans/pinto     beans/pinto
   f  beans/turtle    beans/turtle
   $ hg debugwalk -v -Xbeans/black beans/black
   * matcher:
   <differencematcher
-    m1=<patternmatcher patterns='(?:beans\\/black(?:/|$))'>,
-    m2=<includematcher includes='(?:beans\\/black(?:/|$))'>>
+    m1=<patternmatcher patterns='(?:beans/black(?:/|$))'>,
+    m2=<includematcher includes='(?:beans/black(?:/|$))'>>
   $ hg debugwalk -v -Xbeans/black -Ibeans/black
   * matcher:
   <differencematcher
-    m1=<includematcher includes='(?:beans\\/black(?:/|$))'>,
-    m2=<includematcher includes='(?:beans\\/black(?:/|$))'>>
+    m1=<includematcher includes='(?:beans/black(?:/|$))'>,
+    m2=<includematcher includes='(?:beans/black(?:/|$))'>>
   $ hg debugwalk -v -Xbeans beans/black
   * matcher:
   <differencematcher
-    m1=<patternmatcher patterns='(?:beans\\/black(?:/|$))'>,
+    m1=<patternmatcher patterns='(?:beans/black(?:/|$))'>,
     m2=<includematcher includes='(?:beans(?:/|$))'>>
   $ hg debugwalk -v -Xbeans -Ibeans/black
   * matcher:
   <differencematcher
-    m1=<includematcher includes='(?:beans\\/black(?:/|$))'>,
+    m1=<includematcher includes='(?:beans/black(?:/|$))'>,
     m2=<includematcher includes='(?:beans(?:/|$))'>>
   $ hg debugwalk -v 'glob:mammals/../beans/b*'
   * matcher:
-  <patternmatcher patterns='(?:beans\\/b[^/]*$)'>
+  <patternmatcher patterns='(?:beans/b[^/]*$)'>
   f  beans/black     beans/black
   f  beans/borlotti  beans/borlotti
   $ hg debugwalk -v '-X*/Procyonidae' mammals
   * matcher:
   <differencematcher
     m1=<patternmatcher patterns='(?:mammals(?:/|$))'>,
-    m2=<includematcher includes='(?:[^/]*\\/Procyonidae(?:/|$))'>>
+    m2=<includematcher includes='(?:[^/]*/Procyonidae(?:/|$))'>>
   f  mammals/skunk  mammals/skunk
   $ hg debugwalk -v path:mammals
   * matcher:
   <patternmatcher patterns='(?:mammals(?:/|$))'>
   f  mammals/Procyonidae/cacomistle  mammals/Procyonidae/cacomistle
   f  mammals/Procyonidae/coatimundi  mammals/Procyonidae/coatimundi
   f  mammals/Procyonidae/raccoon     mammals/Procyonidae/raccoon
   f  mammals/skunk                   mammals/skunk
   $ hg debugwalk -v ..
   abort: .. not under root '$TESTTMP/t'
   [255]
   $ hg debugwalk -v beans/../..
   abort: beans/../.. not under root '$TESTTMP/t'
   [255]
   $ hg debugwalk -v .hg
   abort: path contains illegal component: .hg
   [255]
   $ hg debugwalk -v beans/../.hg
   abort: path contains illegal component: .hg
   [255]
   $ hg debugwalk -v beans/../.hg/data
   abort: path contains illegal component: .hg/data
   [255]
   $ hg debugwalk -v beans/.hg
   abort: path 'beans/.hg' is inside nested repo 'beans'
   [255]
 
 Test explicit paths and excludes:
 
   $ hg debugwalk -v fennel -X fennel
   * matcher:
   <differencematcher
     m1=<patternmatcher patterns='(?:fennel(?:/|$))'>,
     m2=<includematcher includes='(?:fennel(?:/|$))'>>
   $ hg debugwalk -v fennel -X 'f*'
   * matcher:
   <differencematcher
     m1=<patternmatcher patterns='(?:fennel(?:/|$))'>,
     m2=<includematcher includes='(?:f[^/]*(?:/|$))'>>
   $ hg debugwalk -v beans/black -X 'path:beans'
   * matcher:
   <differencematcher
-    m1=<patternmatcher patterns='(?:beans\\/black(?:/|$))'>,
+    m1=<patternmatcher patterns='(?:beans/black(?:/|$))'>,
     m2=<includematcher includes='(?:beans(?:/|$))'>>
   $ hg debugwalk -v -I 'path:beans/black' -X 'path:beans'
   * matcher:
   <differencematcher
-    m1=<includematcher includes='(?:beans\\/black(?:/|$))'>,
+    m1=<includematcher includes='(?:beans/black(?:/|$))'>,
     m2=<includematcher includes='(?:beans(?:/|$))'>>
 
 Test absolute paths:
 
   $ hg debugwalk -v `pwd`/beans
   * matcher:
   <patternmatcher patterns='(?:beans(?:/|$))'>
   f  beans/black     beans/black
   f  beans/borlotti  beans/borlotti
   f  beans/kidney    beans/kidney
   f  beans/navy      beans/navy
   f  beans/pinto     beans/pinto
   f  beans/turtle    beans/turtle
   $ hg debugwalk -v `pwd`/..
   abort: $TESTTMP/t/.. not under root '$TESTTMP/t'
   [255]
 
 Test patterns:
 
   $ hg debugwalk -v glob:\*
   * matcher:
   <patternmatcher patterns='(?:[^/]*$)'>
   f  fennel      fennel
   f  fenugreek   fenugreek
   f  fiddlehead  fiddlehead
 #if eol-in-paths
   $ echo glob:glob > glob:glob
   $ hg addremove
   adding glob:glob
   warning: filename contains ':', which is reserved on Windows: 'glob:glob'
   $ hg debugwalk -v glob:\*
   * matcher:
   <patternmatcher patterns='(?:[^/]*$)'>
   f  fennel      fennel
   f  fenugreek   fenugreek
   f  fiddlehead  fiddlehead
   f  glob:glob   glob:glob
   $ hg debugwalk -v glob:glob
   * matcher:
   <patternmatcher patterns='(?:glob$)'>
   glob: $ENOENT$
   $ hg debugwalk -v glob:glob:glob
   * matcher:
-  <patternmatcher patterns='(?:glob\\:glob$)'>
+  <patternmatcher patterns='(?:glob:glob$)'>
   f  glob:glob  glob:glob  exact
   $ hg debugwalk -v path:glob:glob
   * matcher:
-  <patternmatcher patterns='(?:glob\\:glob(?:/|$))'>
+  <patternmatcher patterns='(?:glob:glob(?:/|$))'>
   f  glob:glob  glob:glob  exact
   $ rm glob:glob
   $ hg addremove
   removing glob:glob
 #endif
 
   $ hg debugwalk -v 'glob:**e'
   * matcher:
   <patternmatcher patterns='(?:.*e$)'>
   f  beans/turtle                    beans/turtle
   f  mammals/Procyonidae/cacomistle  mammals/Procyonidae/cacomistle
 
   $ hg debugwalk -v 're:.*[kb]$'
   * matcher:
   <patternmatcher patterns='(?:.*[kb]$)'>
   f  beans/black    beans/black
   f  fenugreek      fenugreek
   f  mammals/skunk  mammals/skunk
 
   $ hg debugwalk -v path:beans/black
   * matcher:
-  <patternmatcher patterns='(?:beans\\/black(?:/|$))'>
+  <patternmatcher patterns='(?:beans/black(?:/|$))'>
   f  beans/black  beans/black  exact
   $ hg debugwalk -v path:beans//black
   * matcher:
-  <patternmatcher patterns='(?:beans\\/black(?:/|$))'>
+  <patternmatcher patterns='(?:beans/black(?:/|$))'>
   f  beans/black  beans/black  exact
 
   $ hg debugwalk -v relglob:Procyonidae
   * matcher:
   <patternmatcher patterns='(?:(?:|.*/)Procyonidae$)'>
   $ hg debugwalk -v 'relglob:Procyonidae/**'
   * matcher:
-  <patternmatcher patterns='(?:(?:|.*/)Procyonidae\\/.*$)'>
+  <patternmatcher patterns='(?:(?:|.*/)Procyonidae/.*$)'>
   f  mammals/Procyonidae/cacomistle  mammals/Procyonidae/cacomistle
   f  mammals/Procyonidae/coatimundi  mammals/Procyonidae/coatimundi
   f  mammals/Procyonidae/raccoon     mammals/Procyonidae/raccoon
   $ hg debugwalk -v 'relglob:Procyonidae/**' fennel
   * matcher:
-  <patternmatcher patterns='(?:(?:|.*/)Procyonidae\\/.*$|fennel(?:/|$))'>
+  <patternmatcher patterns='(?:(?:|.*/)Procyonidae/.*$|fennel(?:/|$))'>
   f  fennel                          fennel                          exact
   f  mammals/Procyonidae/cacomistle  mammals/Procyonidae/cacomistle
   f  mammals/Procyonidae/coatimundi  mammals/Procyonidae/coatimundi
   f  mammals/Procyonidae/raccoon     mammals/Procyonidae/raccoon
   $ hg debugwalk -v beans 'glob:beans/*'
   * matcher:
-  <patternmatcher patterns='(?:beans(?:/|$)|beans\\/[^/]*$)'>
+  <patternmatcher patterns='(?:beans(?:/|$)|beans/[^/]*$)'>
   f  beans/black     beans/black
   f  beans/borlotti  beans/borlotti
   f  beans/kidney    beans/kidney
   f  beans/navy      beans/navy
   f  beans/pinto     beans/pinto
   f  beans/turtle    beans/turtle
   $ hg debugwalk -v 'glob:mamm**'
   * matcher:
   <patternmatcher patterns='(?:mamm.*$)'>
   f  mammals/Procyonidae/cacomistle  mammals/Procyonidae/cacomistle
   f  mammals/Procyonidae/coatimundi  mammals/Procyonidae/coatimundi
   f  mammals/Procyonidae/raccoon     mammals/Procyonidae/raccoon
   f  mammals/skunk                   mammals/skunk
   $ hg debugwalk -v 'glob:mamm**' fennel
   * matcher:
   <patternmatcher patterns='(?:mamm.*$|fennel(?:/|$))'>
   f  fennel                          fennel                          exact
   f  mammals/Procyonidae/cacomistle  mammals/Procyonidae/cacomistle
   f  mammals/Procyonidae/coatimundi  mammals/Procyonidae/coatimundi
   f  mammals/Procyonidae/raccoon     mammals/Procyonidae/raccoon
   f  mammals/skunk                   mammals/skunk
   $ hg debugwalk -v 'glob:j*'
   * matcher:
   <patternmatcher patterns='(?:j[^/]*$)'>
   $ hg debugwalk -v NOEXIST
   * matcher:
   <patternmatcher patterns='(?:NOEXIST(?:/|$))'>
   NOEXIST: * (glob)
 
 #if fifo
   $ mkfifo fifo
   $ hg debugwalk -v fifo
   * matcher:
   <patternmatcher patterns='(?:fifo(?:/|$))'>
   fifo: unsupported file type (type is fifo)
 #endif
 
   $ rm fenugreek
   $ hg debugwalk -v fenugreek
   * matcher:
   <patternmatcher patterns='(?:fenugreek(?:/|$))'>
   f  fenugreek  fenugreek  exact
   $ hg rm fenugreek
   $ hg debugwalk -v fenugreek
   * matcher:
   <patternmatcher patterns='(?:fenugreek(?:/|$))'>
   f  fenugreek  fenugreek  exact
   $ touch new
   $ hg debugwalk -v new
   * matcher:
   <patternmatcher patterns='(?:new(?:/|$))'>
   f  new  new  exact
 
   $ mkdir ignored
   $ touch ignored/file
   $ echo '^ignored$' > .hgignore
   $ hg debugwalk -v ignored
   * matcher:
   <patternmatcher patterns='(?:ignored(?:/|$))'>
   $ hg debugwalk -v ignored/file
   * matcher:
-  <patternmatcher patterns='(?:ignored\\/file(?:/|$))'>
+  <patternmatcher patterns='(?:ignored/file(?:/|$))'>
   f  ignored/file  ignored/file  exact
 
 Test listfile and listfile0
 
   $ $PYTHON -c "open('listfile0', 'wb').write(b'fenugreek\0new\0')"
   $ hg debugwalk -v -I 'listfile0:listfile0'
   * matcher:
   <includematcher includes='(?:fenugreek(?:/|$)|new(?:/|$))'>
   f  fenugreek  fenugreek
   f  new        new
   $ $PYTHON -c "open('listfile', 'wb').write(b'fenugreek\nnew\r\nmammals/skunk\n')"
   $ hg debugwalk -v -I 'listfile:listfile'
   * matcher:
-  <includematcher includes='(?:fenugreek(?:/|$)|new(?:/|$)|mammals\\/skunk(?:/|$))'>
+  <includematcher includes='(?:fenugreek(?:/|$)|new(?:/|$)|mammals/skunk(?:/|$))'>
   f  fenugreek      fenugreek
   f  mammals/skunk  mammals/skunk
   f  new            new
 
   $ cd ..
   $ hg debugwalk -v -R t t/mammals/skunk
   * matcher:
-  <patternmatcher patterns='(?:mammals\\/skunk(?:/|$))'>
+  <patternmatcher patterns='(?:mammals/skunk(?:/|$))'>
   f  mammals/skunk  t/mammals/skunk  exact
   $ mkdir t2
   $ cd t2
   $ hg debugwalk -v -R ../t ../t/mammals/skunk
   * matcher:
-  <patternmatcher patterns='(?:mammals\\/skunk(?:/|$))'>
+  <patternmatcher patterns='(?:mammals/skunk(?:/|$))'>
   f  mammals/skunk  ../t/mammals/skunk  exact
   $ hg debugwalk -v --cwd ../t mammals/skunk
   * matcher:
-  <patternmatcher patterns='(?:mammals\\/skunk(?:/|$))'>
+  <patternmatcher patterns='(?:mammals/skunk(?:/|$))'>
   f  mammals/skunk  mammals/skunk  exact
 
   $ cd ..
 
 Test split patterns on overflow
 
   $ cd t
   $ echo fennel > overflow.list
   $ cat >> printnum.py <<EOF
   > from __future__ import print_function
   > for i in range(20000 // 100):
   >   print('x' * 100)
   > EOF
   $ $PYTHON printnum.py >> overflow.list
   $ echo fenugreek >> overflow.list
   $ hg debugwalk 'listfile:overflow.list' 2>&1 | egrep -v '^xxx'
   f  fennel     fennel     exact
   f  fenugreek  fenugreek  exact
   $ cd ..