match: move matchers from sparse into core

The sparse extension contains some matcher types that are
generic and can exist in core.

As part of the move, the classes now inherit from basematcher.
always(), files(), and isexact() have been dropped because
they match the default implementations in basematcher.

hgext/sparse.py

@@ -1,943 +1,864
 # sparse.py - allow sparse checkouts of the working directory
 #
 # Copyright 2014 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.
 
 """allow sparse checkouts of the working directory (EXPERIMENTAL)
 
 (This extension is not yet protected by backwards compatibility
 guarantees. Any aspect may break in future releases until this
 notice is removed.)
 
 This extension allows the working directory to only consist of a
 subset of files for the revision. This allows specific files or
 directories to be explicitly included or excluded. Many repository
 operations have performance proportional to the number of files in
 the working directory. So only realizing a subset of files in the
 working directory can improve performance.
 
 Sparse Config Files
 -------------------
 
 The set of files that are part of a sparse checkout are defined by
 a sparse config file. The file defines 3 things: includes (files to
 include in the sparse checkout), excludes (files to exclude from the
 sparse checkout), and profiles (links to other config files).
 
 The file format is newline delimited. Empty lines and lines beginning
 with ``#`` are ignored.
 
 Lines beginning with ``%include `` denote another sparse config file
 to include. e.g. ``%include tests.sparse``. The filename is relative
 to the repository root.
 
 The special lines ``[include]`` and ``[exclude]`` denote the section
 for includes and excludes that follow, respectively. It is illegal to
 have ``[include]`` after ``[exclude]``. If no sections are defined,
 entries are assumed to be in the ``[include]`` section.
 
 Non-special lines resemble file patterns to be added to either includes
 or excludes. The syntax of these lines is documented by :hg:`help patterns`.
 Patterns are interpreted as ``glob:`` by default and match against the
 root of the repository.
 
 Exclusion patterns take precedence over inclusion patterns. So even
 if a file is explicitly included, an ``[exclude]`` entry can remove it.
 
 For example, say you have a repository with 3 directories, ``frontend/``,
 ``backend/``, and ``tools/``. ``frontend/`` and ``backend/`` correspond
 to different projects and it is uncommon for someone working on one
 to need the files for the other. But ``tools/`` contains files shared
 between both projects. Your sparse config files may resemble::
 
   # frontend.sparse
   frontend/**
   tools/**
 
   # backend.sparse
   backend/**
   tools/**
 
 Say the backend grows in size. Or there's a directory with thousands
 of files you wish to exclude. You can modify the profile to exclude
 certain files::
 
   [include]
   backend/**
   tools/**
 
   [exclude]
   tools/tests/**
 """
 
 from __future__ import absolute_import
 
 import collections
 import os
 
 from mercurial.i18n import _
 from mercurial.node import nullid
 from mercurial import (
     cmdutil,
     commands,
     context,
     dirstate,
     error,
     extensions,
     hg,
     localrepo,
     match as matchmod,
     merge as mergemod,
     registrar,
     sparse,
     util,
 )
 
 # 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'
 
 cmdtable = {}
 command = registrar.command(cmdtable)
 
 def uisetup(ui):
     _setupupdates(ui)
     _setupcommit(ui)
 
 def extsetup(ui):
     sparse.enabled = True
 
     _setupclone(ui)
     _setuplog(ui)
     _setupadd(ui)
     _setupdirstate(ui)
 
 def reposetup(ui, repo):
     if not util.safehasattr(repo, 'dirstate'):
         return
 
     _wraprepo(ui, repo)
 
 def replacefilecache(cls, propname, replacement):
     """Replace a filecache property with a new class. This allows changing the
     cache invalidation condition."""
     origcls = cls
     assert callable(replacement)
     while cls is not object:
         if propname in cls.__dict__:
             orig = cls.__dict__[propname]
             setattr(cls, propname, replacement(orig))
             break
         cls = cls.__bases__[0]
 
     if cls is object:
         raise AttributeError(_("type '%s' has no property '%s'") % (origcls,
                              propname))
 
 def _setupupdates(ui):
     def _calculateupdates(orig, repo, wctx, mctx, ancestors, branchmerge, *arg,
                           **kwargs):
         """Filter updates to only lay out files that match the sparse rules.
         """
         actions, diverge, renamedelete = orig(repo, wctx, mctx, ancestors,
                                               branchmerge, *arg, **kwargs)
 
         if not util.safehasattr(repo, 'sparsematch'):
             return actions, diverge, renamedelete
 
         files = set()
         prunedactions = {}
         oldrevs = [pctx.rev() for pctx in wctx.parents()]
         oldsparsematch = repo.sparsematch(*oldrevs)
 
         if branchmerge:
             # If we're merging, use the wctx filter, since we're merging into
             # the wctx.
             sparsematch = repo.sparsematch(wctx.parents()[0].rev())
         else:
             # If we're updating, use the target context's filter, since we're
             # moving to the target context.
             sparsematch = repo.sparsematch(mctx.rev())
 
         temporaryfiles = []
         for file, action in actions.iteritems():
             type, args, msg = action
             files.add(file)
             if sparsematch(file):
                 prunedactions[file] = action
             elif type == 'm':
                 temporaryfiles.append(file)
                 prunedactions[file] = action
             elif branchmerge:
                 if type != 'k':
                     temporaryfiles.append(file)
                     prunedactions[file] = action
             elif type == 'f':
                 prunedactions[file] = action
             elif file in wctx:
                 prunedactions[file] = ('r', args, msg)
 
         if len(temporaryfiles) > 0:
             ui.status(_("temporarily included %d file(s) in the sparse checkout"
                 " for merging\n") % len(temporaryfiles))
             sparse.addtemporaryincludes(repo, temporaryfiles)
 
             # Add the new files to the working copy so they can be merged, etc
             actions = []
             message = 'temporarily adding to sparse checkout'
             wctxmanifest = repo[None].manifest()
             for file in temporaryfiles:
                 if file in wctxmanifest:
                     fctx = repo[None][file]
                     actions.append((file, (fctx.flags(), False), message))
 
             typeactions = collections.defaultdict(list)
             typeactions['g'] = actions
             mergemod.applyupdates(repo, typeactions, repo[None], repo['.'],
                                   False)
 
             dirstate = repo.dirstate
             for file, flags, msg in actions:
                 dirstate.normal(file)
 
         profiles = sparse.activeprofiles(repo)
         changedprofiles = profiles & files
         # If an active profile changed during the update, refresh the checkout.
         # Don't do this during a branch merge, since all incoming changes should
         # have been handled by the temporary includes above.
         if changedprofiles and not branchmerge:
             mf = mctx.manifest()
             for file in mf:
                 old = oldsparsematch(file)
                 new = sparsematch(file)
                 if not old and new:
                     flags = mf.flags(file)
                     prunedactions[file] = ('g', (flags, False), '')
                 elif old and not new:
                     prunedactions[file] = ('r', [], '')
 
         return prunedactions, diverge, renamedelete
 
     extensions.wrapfunction(mergemod, 'calculateupdates', _calculateupdates)
 
     def _update(orig, repo, node, branchmerge, *args, **kwargs):
         results = orig(repo, node, branchmerge, *args, **kwargs)
 
         # If we're updating to a location, clean up any stale temporary includes
         # (ex: this happens during hg rebase --abort).
         if not branchmerge and util.safehasattr(repo, 'sparsematch'):
             repo.prunetemporaryincludes()
         return results
 
     extensions.wrapfunction(mergemod, 'update', _update)
 
 def _setupcommit(ui):
     def _refreshoncommit(orig, self, node):
         """Refresh the checkout when commits touch .hgsparse
         """
         orig(self, node)
         repo = self._repo
 
         ctx = repo[node]
         profiles = sparse.patternsforrev(repo, ctx.rev())[2]
 
         # profiles will only have data if sparse is enabled.
         if set(profiles) & set(ctx.files()):
             origstatus = repo.status()
             origsparsematch = repo.sparsematch()
             _refresh(repo.ui, repo, origstatus, origsparsematch, True)
 
         if util.safehasattr(repo, 'prunetemporaryincludes'):
             repo.prunetemporaryincludes()
 
     extensions.wrapfunction(context.committablectx, 'markcommitted',
         _refreshoncommit)
 
 def _setuplog(ui):
     entry = commands.table['^log|history']
     entry[1].append(('', 'sparse', None,
         "limit to changesets affecting the sparse checkout"))
 
     def _logrevs(orig, repo, opts):
         revs = orig(repo, opts)
         if opts.get('sparse'):
             sparsematch = repo.sparsematch()
             def ctxmatch(rev):
                 ctx = repo[rev]
                 return any(f for f in ctx.files() if sparsematch(f))
             revs = revs.filter(ctxmatch)
         return revs
     extensions.wrapfunction(cmdutil, '_logrevs', _logrevs)
 
 def _clonesparsecmd(orig, ui, repo, *args, **opts):
     include_pat = opts.get('include')
     exclude_pat = opts.get('exclude')
     enableprofile_pat = opts.get('enable_profile')
     include = exclude = enableprofile = False
     if include_pat:
         pat = include_pat
         include = True
     if exclude_pat:
         pat = exclude_pat
         exclude = True
     if enableprofile_pat:
         pat = enableprofile_pat
         enableprofile = True
     if sum([include, exclude, enableprofile]) > 1:
         raise error.Abort(_("too many flags specified."))
     if include or exclude or enableprofile:
         def clonesparse(orig, self, node, overwrite, *args, **kwargs):
             _config(self.ui, self.unfiltered(), pat, {}, include=include,
                     exclude=exclude, enableprofile=enableprofile)
             return orig(self, node, overwrite, *args, **kwargs)
         extensions.wrapfunction(hg, 'updaterepo', clonesparse)
     return orig(ui, repo, *args, **opts)
 
 def _setupclone(ui):
     entry = commands.table['^clone']
     entry[1].append(('', 'enable-profile', [],
                     'enable a sparse profile'))
     entry[1].append(('', 'include', [],
                     'include sparse pattern'))
     entry[1].append(('', 'exclude', [],
                     'exclude sparse pattern'))
     extensions.wrapcommand(commands.table, 'clone', _clonesparsecmd)
 
 def _setupadd(ui):
     entry = commands.table['^add']
     entry[1].append(('s', 'sparse', None,
                     'also include directories of added files in sparse config'))
 
     def _add(orig, ui, repo, *pats, **opts):
         if opts.get('sparse'):
             dirs = set()
             for pat in pats:
                 dirname, basename = util.split(pat)
                 dirs.add(dirname)
             _config(ui, repo, list(dirs), opts, include=True)
         return orig(ui, repo, *pats, **opts)
 
     extensions.wrapcommand(commands.table, 'add', _add)
 
 def _setupdirstate(ui):
     """Modify the dirstate to prevent stat'ing excluded files,
     and to prevent modifications to files outside the checkout.
     """
 
     def _dirstate(orig, repo):
         dirstate = orig(repo)
         dirstate.repo = repo
         return dirstate
     extensions.wrapfunction(
         localrepo.localrepository.dirstate, 'func', _dirstate)
 
     # The atrocity below is needed to wrap dirstate._ignore. It is a cached
     # property, which means normal function wrapping doesn't work.
     class ignorewrapper(object):
         def __init__(self, orig):
             self.orig = orig
             self.origignore = None
             self.func = None
             self.sparsematch = None
 
         def __get__(self, obj, type=None):
             repo = obj.repo
             origignore = self.orig.__get__(obj)
             if not util.safehasattr(repo, 'sparsematch'):
                 return origignore
 
             sparsematch = repo.sparsematch()
             if self.sparsematch != sparsematch or self.origignore != origignore:
-                self.func = unionmatcher([origignore,
-                                          negatematcher(sparsematch)])
+                self.func = matchmod.unionmatcher([
+                    origignore, matchmod.negatematcher(sparsematch)])
                 self.sparsematch = sparsematch
                 self.origignore = origignore
             return self.func
 
         def __set__(self, obj, value):
             return self.orig.__set__(obj, value)
 
         def __delete__(self, obj):
             return self.orig.__delete__(obj)
 
     replacefilecache(dirstate.dirstate, '_ignore', ignorewrapper)
 
     # dirstate.rebuild should not add non-matching files
     def _rebuild(orig, self, parent, allfiles, changedfiles=None):
         if util.safehasattr(self.repo, 'sparsematch'):
             matcher = self.repo.sparsematch()
             allfiles = allfiles.matches(matcher)
             if changedfiles:
                 changedfiles = [f for f in changedfiles if matcher(f)]
 
             if changedfiles is not None:
                 # In _rebuild, these files will be deleted from the dirstate
                 # when they are not found to be in allfiles
                 dirstatefilestoremove = set(f for f in self if not matcher(f))
                 changedfiles = dirstatefilestoremove.union(changedfiles)
 
         return orig(self, parent, allfiles, changedfiles)
     extensions.wrapfunction(dirstate.dirstate, 'rebuild', _rebuild)
 
     # Prevent adding files that are outside the sparse checkout
     editfuncs = ['normal', 'add', 'normallookup', 'copy', 'remove', 'merge']
     hint = _('include file with `hg debugsparse --include <pattern>` or use ' +
              '`hg add -s <file>` to include file directory while adding')
     for func in editfuncs:
         def _wrapper(orig, self, *args):
             repo = self.repo
             if util.safehasattr(repo, 'sparsematch'):
                 dirstate = repo.dirstate
                 sparsematch = repo.sparsematch()
                 for f in args:
                     if (f is not None and not sparsematch(f) and
                         f not in dirstate):
                         raise error.Abort(_("cannot add '%s' - it is outside "
                                             "the sparse checkout") % f,
                                           hint=hint)
             return orig(self, *args)
         extensions.wrapfunction(dirstate.dirstate, func, _wrapper)
 
 def _wraprepo(ui, repo):
     class SparseRepo(repo.__class__):
         def sparsematch(self, *revs, **kwargs):
             """Returns the sparse match function for the given revs.
 
             If multiple revs are specified, the match function is the union
             of all the revs.
 
             `includetemp` is used to indicate if the temporarily included file
             should be part of the matcher.
             """
             if not revs or revs == (None,):
                 revs = [self.changelog.rev(node) for node in
                     self.dirstate.parents() if node != nullid]
 
             includetemp = kwargs.get('includetemp', True)
             signature = sparse.configsignature(self, includetemp=includetemp)
 
             key = '%s %s' % (str(signature), ' '.join([str(r) for r in revs]))
 
             result = self._sparsematchercache.get(key, None)
             if result:
                 return result
 
             matchers = []
             for rev in revs:
                 try:
                     includes, excludes, profiles = sparse.patternsforrev(
                         self, rev)
 
                     if includes or excludes:
                         # Explicitly include subdirectories of includes so
                         # status will walk them down to the actual include.
                         subdirs = set()
                         for include in includes:
                             dirname = os.path.dirname(include)
                             # basename is used to avoid issues with absolute
                             # paths (which on Windows can include the drive).
                             while os.path.basename(dirname):
                                 subdirs.add(dirname)
                                 dirname = os.path.dirname(dirname)
 
                         matcher = matchmod.match(self.root, '', [],
                             include=includes, exclude=excludes,
                             default='relpath')
                         if subdirs:
-                            matcher = forceincludematcher(matcher, subdirs)
+                            matcher = matchmod.forceincludematcher(matcher,
+                                                                   subdirs)
                         matchers.append(matcher)
                 except IOError:
                     pass
 
             result = None
             if not matchers:
                 result = matchmod.always(self.root, '')
             elif len(matchers) == 1:
                 result = matchers[0]
             else:
-                result = unionmatcher(matchers)
+                result = matchmod.unionmatcher(matchers)
 
             if kwargs.get('includetemp', True):
                 tempincludes = sparse.readtemporaryincludes(self)
-                result = forceincludematcher(result, tempincludes)
+                result = matchmod.forceincludematcher(result, tempincludes)
 
             self._sparsematchercache[key] = result
 
             return result
 
         def prunetemporaryincludes(self):
             if repo.vfs.exists('tempsparse'):
                 origstatus = self.status()
                 modified, added, removed, deleted, a, b, c = origstatus
                 if modified or added or removed or deleted:
                     # Still have pending changes. Don't bother trying to prune.
                     return
 
                 sparsematch = self.sparsematch(includetemp=False)
                 dirstate = self.dirstate
                 actions = []
                 dropped = []
                 tempincludes = sparse.readtemporaryincludes(self)
                 for file in tempincludes:
                     if file in dirstate and not sparsematch(file):
                         message = 'dropping temporarily included sparse files'
                         actions.append((file, None, message))
                         dropped.append(file)
 
                 typeactions = collections.defaultdict(list)
                 typeactions['r'] = actions
                 mergemod.applyupdates(self, typeactions, self[None], self['.'],
                                       False)
 
                 # Fix dirstate
                 for file in dropped:
                     dirstate.drop(file)
 
                 self.vfs.unlink('tempsparse')
                 sparse.invalidatesignaturecache(self)
                 msg = _("cleaned up %d temporarily added file(s) from the "
                         "sparse checkout\n")
                 ui.status(msg % len(tempincludes))
 
     if 'dirstate' in repo._filecache:
         repo.dirstate.repo = repo
 
     repo.__class__ = SparseRepo
 
 @command('^debugsparse', [
     ('I', 'include', False, _('include files in the sparse checkout')),
     ('X', 'exclude', False, _('exclude files in the sparse checkout')),
     ('d', 'delete', False, _('delete an include/exclude rule')),
     ('f', 'force', False, _('allow changing rules even with pending changes')),
     ('', 'enable-profile', False, _('enables the specified profile')),
     ('', 'disable-profile', False, _('disables the specified profile')),
     ('', 'import-rules', False, _('imports rules from a file')),
     ('', 'clear-rules', False, _('clears local include/exclude rules')),
     ('', 'refresh', False, _('updates the working after sparseness changes')),
     ('', 'reset', False, _('makes the repo full again')),
     ] + commands.templateopts,
     _('[--OPTION] PATTERN...'))
 def debugsparse(ui, repo, *pats, **opts):
     """make the current checkout sparse, or edit the existing checkout
 
     The sparse command is used to make the current checkout sparse.
     This means files that don't meet the sparse condition will not be
     written to disk, or show up in any working copy operations. It does
     not affect files in history in any way.
 
     Passing no arguments prints the currently applied sparse rules.
 
     --include and --exclude are used to add and remove files from the sparse
     checkout. The effects of adding an include or exclude rule are applied
     immediately. If applying the new rule would cause a file with pending
     changes to be added or removed, the command will fail. Pass --force to
     force a rule change even with pending changes (the changes on disk will
     be preserved).
 
     --delete removes an existing include/exclude rule. The effects are
     immediate.
 
     --refresh refreshes the files on disk based on the sparse rules. This is
     only necessary if .hg/sparse was changed by hand.
 
     --enable-profile and --disable-profile accept a path to a .hgsparse file.
     This allows defining sparse checkouts and tracking them inside the
     repository. This is useful for defining commonly used sparse checkouts for
     many people to use. As the profile definition changes over time, the sparse
     checkout will automatically be updated appropriately, depending on which
     changeset is checked out. Changes to .hgsparse are not applied until they
     have been committed.
 
     --import-rules accepts a path to a file containing rules in the .hgsparse
     format, allowing you to add --include, --exclude and --enable-profile rules
     in bulk. Like the --include, --exclude and --enable-profile switches, the
     changes are applied immediately.
 
     --clear-rules removes all local include and exclude rules, while leaving
     any enabled profiles in place.
 
     Returns 0 if editing the sparse checkout succeeds.
     """
     include = opts.get('include')
     exclude = opts.get('exclude')
     force = opts.get('force')
     enableprofile = opts.get('enable_profile')
     disableprofile = opts.get('disable_profile')
     importrules = opts.get('import_rules')
     clearrules = opts.get('clear_rules')
     delete = opts.get('delete')
     refresh = opts.get('refresh')
     reset = opts.get('reset')
     count = sum([include, exclude, enableprofile, disableprofile, delete,
                  importrules, refresh, clearrules, reset])
     if count > 1:
         raise error.Abort(_("too many flags specified"))
 
     if count == 0:
         if repo.vfs.exists('sparse'):
             ui.status(repo.vfs.read("sparse") + "\n")
             temporaryincludes = sparse.readtemporaryincludes(repo)
             if temporaryincludes:
                 ui.status(_("Temporarily Included Files (for merge/rebase):\n"))
                 ui.status(("\n".join(temporaryincludes) + "\n"))
         else:
             ui.status(_('repo is not sparse\n'))
         return
 
     if include or exclude or delete or reset or enableprofile or disableprofile:
         _config(ui, repo, pats, opts, include=include, exclude=exclude,
                 reset=reset, delete=delete, enableprofile=enableprofile,
                 disableprofile=disableprofile, force=force)
 
     if importrules:
         _import(ui, repo, pats, opts, force=force)
 
     if clearrules:
         _clear(ui, repo, pats, force=force)
 
     if refresh:
         try:
             wlock = repo.wlock()
             fcounts = map(
                 len,
                 _refresh(ui, repo, repo.status(), repo.sparsematch(), force))
             _verbose_output(ui, opts, 0, 0, 0, *fcounts)
         finally:
             wlock.release()
 
 def _config(ui, repo, pats, opts, include=False, exclude=False, reset=False,
             delete=False, enableprofile=False, disableprofile=False,
             force=False):
     """
     Perform a sparse config update. Only one of the kwargs may be specified.
     """
     wlock = repo.wlock()
     try:
         oldsparsematch = repo.sparsematch()
 
         raw = repo.vfs.tryread('sparse')
         if raw:
             oldinclude, oldexclude, oldprofiles = map(
                 set, sparse.parseconfig(ui, raw))
         else:
             oldinclude = set()
             oldexclude = set()
             oldprofiles = set()
 
         try:
             if reset:
                 newinclude = set()
                 newexclude = set()
                 newprofiles = set()
             else:
                 newinclude = set(oldinclude)
                 newexclude = set(oldexclude)
                 newprofiles = set(oldprofiles)
 
             oldstatus = repo.status()
 
             if any(pat.startswith('/') for pat in pats):
                 ui.warn(_('warning: paths cannot start with /, ignoring: %s\n')
                           % ([pat for pat in pats if pat.startswith('/')]))
             elif include:
                 newinclude.update(pats)
             elif exclude:
                 newexclude.update(pats)
             elif enableprofile:
                 newprofiles.update(pats)
             elif disableprofile:
                 newprofiles.difference_update(pats)
             elif delete:
                 newinclude.difference_update(pats)
                 newexclude.difference_update(pats)
 
             sparse.writeconfig(repo, newinclude, newexclude, newprofiles)
 
             fcounts = map(
                 len, _refresh(ui, repo, oldstatus, oldsparsematch, force))
 
             profilecount = (len(newprofiles - oldprofiles) -
                             len(oldprofiles - newprofiles))
             includecount = (len(newinclude - oldinclude) -
                             len(oldinclude - newinclude))
             excludecount = (len(newexclude - oldexclude) -
                             len(oldexclude - newexclude))
             _verbose_output(
                 ui, opts, profilecount, includecount, excludecount, *fcounts)
         except Exception:
             sparse.writeconfig(repo, oldinclude, oldexclude, oldprofiles)
             raise
     finally:
         wlock.release()
 
 def _import(ui, repo, files, opts, force=False):
     with repo.wlock():
         # load union of current active profile
         revs = [repo.changelog.rev(node) for node in
                 repo.dirstate.parents() if node != nullid]
 
         # read current configuration
         raw = repo.vfs.tryread('sparse')
         oincludes, oexcludes, oprofiles = sparse.parseconfig(ui, raw)
         includes, excludes, profiles = map(
                 set, (oincludes, oexcludes, oprofiles))
 
         # all active rules
         aincludes, aexcludes, aprofiles = set(), set(), set()
         for rev in revs:
             rincludes, rexcludes, rprofiles = sparse.patternsforrev(repo, rev)
             aincludes.update(rincludes)
             aexcludes.update(rexcludes)
             aprofiles.update(rprofiles)
 
         # import rules on top; only take in rules that are not yet
         # part of the active rules.
         changed = False
         for file in files:
             with util.posixfile(util.expandpath(file)) as importfile:
                 iincludes, iexcludes, iprofiles = sparse.parseconfig(
                     ui, importfile.read())
                 oldsize = len(includes) + len(excludes) + len(profiles)
                 includes.update(iincludes - aincludes)
                 excludes.update(iexcludes - aexcludes)
                 profiles.update(set(iprofiles) - aprofiles)
                 if len(includes) + len(excludes) + len(profiles) > oldsize:
                     changed = True
 
         profilecount = includecount = excludecount = 0
         fcounts = (0, 0, 0)
 
         if changed:
             profilecount = len(profiles - aprofiles)
             includecount = len(includes - aincludes)
             excludecount = len(excludes - aexcludes)
 
             oldstatus = repo.status()
             oldsparsematch = repo.sparsematch()
             sparse.writeconfig(repo, includes, excludes, profiles)
 
             try:
                 fcounts = map(
                     len, _refresh(ui, repo, oldstatus, oldsparsematch, force))
             except Exception:
                 sparse.writeconfig(repo, oincludes, oexcludes, oprofiles)
                 raise
 
         _verbose_output(ui, opts, profilecount, includecount, excludecount,
                         *fcounts)
 
 def _clear(ui, repo, files, force=False):
     with repo.wlock():
         raw = repo.vfs.tryread('sparse')
         includes, excludes, profiles = sparse.parseconfig(ui, raw)
 
         if includes or excludes:
             oldstatus = repo.status()
             oldsparsematch = repo.sparsematch()
             sparse.writeconfig(repo, set(), set(), profiles)
             _refresh(ui, repo, oldstatus, oldsparsematch, force)
 
 def _refresh(ui, repo, origstatus, origsparsematch, force):
     """Refreshes which files are on disk by comparing the old status and
     sparsematch with the new sparsematch.
 
     Will raise an exception if a file with pending changes is being excluded
     or included (unless force=True).
     """
     modified, added, removed, deleted, unknown, ignored, clean = origstatus
 
     # Verify there are no pending changes
     pending = set()
     pending.update(modified)
     pending.update(added)
     pending.update(removed)
     sparsematch = repo.sparsematch()
     abort = False
     for file in pending:
         if not sparsematch(file):
             ui.warn(_("pending changes to '%s'\n") % file)
             abort = not force
     if abort:
         raise error.Abort(_("could not update sparseness due to " +
             "pending changes"))
 
     # Calculate actions
     dirstate = repo.dirstate
     ctx = repo['.']
     added = []
     lookup = []
     dropped = []
     mf = ctx.manifest()
     files = set(mf)
 
     actions = {}
 
     for file in files:
         old = origsparsematch(file)
         new = sparsematch(file)
         # Add files that are newly included, or that don't exist in
         # the dirstate yet.
         if (new and not old) or (old and new and not file in dirstate):
             fl = mf.flags(file)
             if repo.wvfs.exists(file):
                 actions[file] = ('e', (fl,), '')
                 lookup.append(file)
             else:
                 actions[file] = ('g', (fl, False), '')
                 added.append(file)
         # Drop files that are newly excluded, or that still exist in
         # the dirstate.
         elif (old and not new) or (not old and not new and file in dirstate):
             dropped.append(file)
             if file not in pending:
                 actions[file] = ('r', [], '')
 
     # Verify there are no pending changes in newly included files
     abort = False
     for file in lookup:
         ui.warn(_("pending changes to '%s'\n") % file)
         abort = not force
     if abort:
         raise error.Abort(_("cannot change sparseness due to " +
             "pending changes (delete the files or use --force " +
             "to bring them back dirty)"))
 
     # Check for files that were only in the dirstate.
     for file, state in dirstate.iteritems():
         if not file in files:
             old = origsparsematch(file)
             new = sparsematch(file)
             if old and not new:
                 dropped.append(file)
 
     # Apply changes to disk
     typeactions = dict((m, []) for m in 'a f g am cd dc r dm dg m e k'.split())
     for f, (m, args, msg) in actions.iteritems():
         if m not in typeactions:
             typeactions[m] = []
         typeactions[m].append((f, args, msg))
     mergemod.applyupdates(repo, typeactions, repo[None], repo['.'], False)
 
     # Fix dirstate
     for file in added:
         dirstate.normal(file)
 
     for file in dropped:
         dirstate.drop(file)
 
     for file in lookup:
         # File exists on disk, and we're bringing it back in an unknown state.
         dirstate.normallookup(file)
 
     return added, dropped, lookup
 
 def _verbose_output(ui, opts, profilecount, includecount, excludecount, added,
                     dropped, lookup):
     """Produce --verbose and templatable output
 
     This specifically enables -Tjson, providing machine-readable stats on how
     the sparse profile changed.
 
     """
     with ui.formatter('sparse', opts) as fm:
         fm.startitem()
         fm.condwrite(ui.verbose, 'profiles_added', 'Profile # change: %d\n',
                      profilecount)
         fm.condwrite(ui.verbose, 'include_rules_added',
                      'Include rule # change: %d\n', includecount)
         fm.condwrite(ui.verbose, 'exclude_rules_added',
                      'Exclude rule # change: %d\n', excludecount)
         # In 'plain' verbose mode, mergemod.applyupdates already outputs what
         # files are added or removed outside of the templating formatter
         # framework. No point in repeating ourselves in that case.
         if not fm.isplain():
             fm.condwrite(ui.verbose, 'files_added', 'Files added: %d\n',
                          added)
             fm.condwrite(ui.verbose, 'files_dropped', 'Files dropped: %d\n',
                          dropped)
             fm.condwrite(ui.verbose, 'files_conflicting',
                          'Files conflicting: %d\n', lookup)
-
-class forceincludematcher(object):
-    """A matcher that returns true for any of the forced includes before testing
-    against the actual matcher."""
-    def __init__(self, matcher, includes):
-        self._matcher = matcher
-        self._includes = includes
-
-    def __call__(self, value):
-        return value in self._includes or self._matcher(value)
-
-    def always(self):
-        return False
-
-    def files(self):
-        return []
-
-    def isexact(self):
-        return False
-
-    def anypats(self):
-        return True
-
-    def prefix(self):
-        return False
-
-    def __repr__(self):
-        return ('<forceincludematcher matcher=%r, includes=%r>' %
-                (self._matcher, sorted(self._includes)))
-
-class unionmatcher(object):
-    """A matcher that is the union of several matchers."""
-    def __init__(self, matchers):
-        self._matchers = matchers
-
-    def __call__(self, value):
-        for match in self._matchers:
-            if match(value):
-                return True
-        return False
-
-    def always(self):
-        return False
-
-    def files(self):
-        return []
-
-    def isexact(self):
-        return False
-
-    def anypats(self):
-        return True
-
-    def prefix(self):
-        return False
-
-    def __repr__(self):
-        return ('<unionmatcher matchers=%r>' % self._matchers)
-
-class negatematcher(object):
-    def __init__(self, matcher):
-        self._matcher = matcher
-
-    def __call__(self, value):
-        return not self._matcher(value)
-
-    def always(self):
-        return False
-
-    def files(self):
-        return []
-
-    def isexact(self):
-        return False
-
-    def anypats(self):
-        return True
-
-    def __repr__(self):
-        return ('<negatematcher matcher=%r>' % self._matcher)

mercurial/match.py

@@ -1,983 +1,1036
 # 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
 
 import copy
 import os
 import re
 
 from .i18n import _
 from . import (
     error,
     pathutil,
     util,
 )
 
 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 ('glob', 'relpath'):
             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, 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.'''
         pass
 
     # 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.
 
         This function's behavior is undefined if it has returned False for
         one of the dir's parent directories.
         '''
         return False
 
     def anypats(self):
         '''Matcher uses patterns or include/exclude.'''
         return False
 
     def always(self):
         '''Matcher will match everything and .files() will be empty
         - optimization might be possible and necessary.'''
         return False
 
     def isexact(self):
         return False
 
     def prefix(self):
         return not self.always() and not self.isexact() and not self.anypats()
 
 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 '<alwaysmatcher>'
 
 class nevermatcher(basematcher):
     '''Matches nothing.'''
 
     def __init__(self, root, cwd, badfn=None):
         super(nevermatcher, self).__init__(root, cwd, badfn)
 
     def __repr__(self):
         return '<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._anypats = _anypats(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 anypats(self):
         return self._anypats
 
     def __repr__(self):
         return ('<patternmatcher patterns=%r>' % 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._anypats = _anypats(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 not self._anypats and dir in self._roots:
             # The condition above is essentially self.prefix() for includes
             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)))
 
     def anypats(self):
         return True
 
     def __repr__(self):
         return ('<includematcher includes=%r>' % 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
 
     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. Well, almost... If the user provides a pattern like "-X foo foo",
     Mercurial actually does match "foo" against that. That's because exact
     matches are treated specially. So, since this differencematcher is used for
     excludes, it needs to special-case exact matching.
 
     The second matcher's non-matching-attributes (root, cwd, bad, explicitdir,
     traversedir) are ignored.
 
     TODO: If we want to keep the behavior described above for exact matches, we
     should consider instead treating the above case something like this:
     union(exact(foo), difference(pattern(foo), include(foo)))
     '''
     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) or self._m1.exact(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':
             # There's a bug here: If m1 matches file 'dir/file' and m2 excludes
             # 'dir' (recursively), we should still visit 'dir' due to the
             # exception we have for exact matches.
             return False
         return bool(self._m1.visitdir(dir))
 
     def isexact(self):
         return self._m1.isexact()
 
     def anypats(self):
         return self._m1.anypats() or self._m2.anypats()
 
     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()
 
     def anypats(self):
         return self._m1.anypats() or self._m2.anypats()
 
     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:
 
     >>> m1 = match('root', '', ['a.txt', 'sub/b.txt'])
     >>> m2 = subdirmatcher('sub', m1)
     >>> bool(m2('a.txt'))
     False
     >>> bool(m2('b.txt'))
     True
     >>> bool(m2.matchfn('a.txt'))
     False
     >>> bool(m2.matchfn('b.txt'))
     True
     >>> m2.files()
     ['b.txt']
     >>> m2.exact('b.txt')
     True
     >>> util.pconvert(m2.rel('b.txt'))
     'sub/b.txt'
     >>> def bad(f, msg):
     ...     print "%s: %s" % (f, msg)
     >>> m1.bad = bad
     >>> m2.bad('x.txt', 'No such file')
     sub/x.txt: No such file
     >>> m2.abs('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 anypats(self):
         return self._matcher.anypats()
 
     def __repr__(self):
         return ('<subdirmatcher path=%r, matcher=%r>' %
                 (self._path, self._matcher))
 
+class forceincludematcher(basematcher):
+    """A matcher that returns true for any of the forced includes before testing
+    against the actual matcher."""
+    def __init__(self, matcher, includes):
+        self._matcher = matcher
+        self._includes = includes
+
+    def __call__(self, value):
+        return value in self._includes or self._matcher(value)
+
+    def anypats(self):
+        return True
+
+    def prefix(self):
+        return False
+
+    def __repr__(self):
+        return ('<forceincludematcher matcher=%r, includes=%r>' %
+                (self._matcher, sorted(self._includes)))
+
+class unionmatcher(basematcher):
+    """A matcher that is the union of several matchers."""
+    def __init__(self, matchers):
+        self._matchers = matchers
+
+    def __call__(self, value):
+        for match in self._matchers:
+            if match(value):
+                return True
+        return False
+
+    def anypats(self):
+        return True
+
+    def prefix(self):
+        return False
+
+    def __repr__(self):
+        return ('<unionmatcher matchers=%r>' % self._matchers)
+
+class negatematcher(basematcher):
+    def __init__(self, matcher):
+        self._matcher = matcher
+
+    def __call__(self, value):
+        return not self._matcher(value)
+
+    def anypats(self):
+        return True
+
+    def __repr__(self):
+        return ('<negatematcher matcher=%r>' % self._matcher)
+
 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 ('re', 'glob', 'path', 'relglob', 'relpath', 'relre',
                     'listfile', 'listfile0', 'set', 'include', 'subinclude',
                     'rootfilesin'):
             return kind, pat
     return default, pattern
 
 def _globre(pat):
     r'''Convert an extended glob string to a regexp string.
 
     >>> print _globre(r'?')
     .
     >>> print _globre(r'*')
     [^/]*
     >>> print _globre(r'**')
     .*
     >>> print _globre(r'**/a')
     (?:.*/)?a
     >>> print _globre(r'a/**/b')
     a\/(?:.*/)?b
     >>> print _globre(r'[a*?!^][^b][!c]')
     [a*?!^][\^b][^c]
     >>> print _globre(r'{a,b}')
     (?:a|b)
     >>> print _globre(r'.\*\?')
     \.\*\?
     '''
     i, n = 0, len(pat)
     res = ''
     group = 0
     escape = util.re.escape
     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 == 'path':
         if pat == '.':
             return ''
         return '^' + util.re.escape(pat) + '(?:/|$)'
     if kind == 'rootfilesin':
         if pat == '.':
             escaped = ''
         else:
             # Pattern is a directory name.
             escaped = util.re.escape(pat) + '/'
         # Anything after the pattern must be a non-directory.
         return '^' + escaped + '[^/]+$'
     if kind == 'relglob':
         return '(?:|.*/)' + _globre(pat) + globsuffix
     if kind == 'relpath':
         return util.re.escape(pat) + '(?:/|$)'
     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(\
         [('glob', 'g/h/*', ''), ('glob', 'g/h', ''), ('glob', 'g*', '')])
     (['g/h', 'g/h', '.'], ['g', '.'])
     >>> _rootsanddirs(\
         [('rootfilesin', 'g/h', ''), ('rootfilesin', '', '')])
     ([], ['g/h', '.', 'g', '.'])
     >>> _rootsanddirs(\
         [('relpath', 'r', ''), ('path', 'p/p', ''), ('path', '', '')])
     (['r', 'p/p', '.'], ['p', '.'])
     >>> _rootsanddirs(\
         [('relglob', 'rg*', ''), ('re', 're/', ''), ('relre', 'rr', '')])
     (['.', '.', '.'], ['.'])
     '''
     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([('path', 'foo/bar', '')])
     ['foo/bar']
     >>> _explicitfiles([('rootfilesin', 'foo/bar', '')])
     []
     '''
     # 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 _anypats(kindpats):
     for kind, pat, source in kindpats:
         if kind in ('glob', 're', 'relglob', 'relre', 'set', 'rootfilesin'):
             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