sparse: make sure excluded subrepos don't break things

fix subprepo.subrepo - now provide DummySubrepo for subs out of sparse.
        This dummy sub is empty node, always clean and ready to get/remove
fix subrepoutil.submerge - now ignores changes for items out of sparce

hgext/sparse.py

@@ -1,440 +1,491 b''
 # 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]``.
 
 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
 
 from mercurial.i18n import _
 from mercurial.pycompat import setattr
 from mercurial import (
     commands,
     dirstate,
     error,
     extensions,
     logcmdutil,
     match as matchmod,
     merge as mergemod,
     pycompat,
     registrar,
     sparse,
+    subrepo,
+    subrepoutil,
     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 = b'ships-with-hg-core'
 
 cmdtable = {}
 command = registrar.command(cmdtable)
 
 
 def extsetup(ui):
     sparse.enabled = True
 
     _setupclone(ui)
     _setuplog(ui)
     _setupadd(ui)
     _setupdirstate(ui)
+    _setupsubrepo(ui)
 
 
 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(
             _(b"type '%s' has no property '%s'") % (origcls, propname)
         )
 
 
 def _setuplog(ui):
     entry = commands.table[b'log|history']
     entry[1].append(
         (
             b'',
             b'sparse',
             None,
             b"limit to changesets affecting the sparse checkout",
         )
     )
 
     def _initialrevs(orig, repo, wopts):
         revs = orig(repo, wopts)
         if wopts.opts.get(b'sparse'):
             sparsematch = sparse.matcher(repo)
 
             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(logcmdutil, b'_initialrevs', _initialrevs)
 
 
 def _clonesparsecmd(orig, ui, repo, *args, **opts):
     include_pat = opts.get('include')
     exclude_pat = opts.get('exclude')
     enableprofile_pat = opts.get('enable_profile')
     narrow_pat = opts.get('narrow')
     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(_(b"too many flags specified."))
     # if --narrow is passed, it means they are includes and excludes for narrow
     # clone
     if not narrow_pat and (include or exclude or enableprofile):
 
         def clonesparse(orig, ctx, *args, **kwargs):
             sparse.updateconfig(
                 ctx.repo().unfiltered(),
                 pat,
                 {},
                 include=include,
                 exclude=exclude,
                 enableprofile=enableprofile,
                 usereporootpaths=True,
             )
             return orig(ctx, *args, **kwargs)
 
         extensions.wrapfunction(mergemod, b'update', clonesparse)
     return orig(ui, repo, *args, **opts)
 
 
 def _setupclone(ui):
     entry = commands.table[b'clone']
     entry[1].append((b'', b'enable-profile', [], b'enable a sparse profile'))
     entry[1].append((b'', b'include', [], b'include sparse pattern'))
     entry[1].append((b'', b'exclude', [], b'exclude sparse pattern'))
     extensions.wrapcommand(commands.table, b'clone', _clonesparsecmd)
 
 
 def _setupadd(ui):
     entry = commands.table[b'add']
     entry[1].append(
         (
             b's',
             b'sparse',
             None,
             b'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)
             sparse.updateconfig(repo, list(dirs), opts, include=True)
         return orig(ui, repo, *pats, **opts)
 
     extensions.wrapcommand(commands.table, b'add', _add)
 
 
 def _setupdirstate(ui):
     """Modify the dirstate to prevent stat'ing excluded files,
     and to prevent modifications to files outside the checkout.
     """
 
     def walk(orig, self, match, subrepos, unknown, ignored, full=True):
         # hack to not exclude explicitly-specified paths so that they can
         # be warned later on e.g. dirstate.add()
         em = matchmod.exact(match.files())
         sm = matchmod.unionmatcher([self._sparsematcher, em])
         match = matchmod.intersectmatchers(match, sm)
         return orig(self, match, subrepos, unknown, ignored, full)
 
     extensions.wrapfunction(dirstate.dirstate, b'walk', walk)
 
     # dirstate.rebuild should not add non-matching files
     def _rebuild(orig, self, parent, allfiles, changedfiles=None):
         matcher = self._sparsematcher
         if not matcher.always():
             allfiles = [f for f in allfiles if matcher(f)]
             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 = {f for f in self if not matcher(f)}
                 changedfiles = dirstatefilestoremove.union(changedfiles)
 
         return orig(self, parent, allfiles, changedfiles)
 
     extensions.wrapfunction(dirstate.dirstate, b'rebuild', _rebuild)
 
     # Prevent adding files that are outside the sparse checkout
     editfuncs = [
         b'normal',
         b'add',
         b'normallookup',
         b'copy',
         b'remove',
         b'merge',
     ]
     hint = _(
         b'include file with `hg debugsparse --include <pattern>` or use '
         + b'`hg add -s <file>` to include file directory while adding'
     )
     for func in editfuncs:
 
         def _wrapper(orig, self, *args, **kwargs):
             sparsematch = self._sparsematcher
             if not sparsematch.always():
                 for f in args:
                     if f is not None and not sparsematch(f) and f not in self:
                         raise error.Abort(
                             _(
                                 b"cannot add '%s' - it is outside "
                                 b"the sparse checkout"
                             )
                             % f,
                             hint=hint,
                         )
             return orig(self, *args, **kwargs)
 
         extensions.wrapfunction(dirstate.dirstate, func, _wrapper)
 
 
+class DummySubrepo(subrepo.abstractsubrepo):
+    """Dumy subrepo is replacement of subrepo, that should be filterout from sparce.
+    this subrepo acts as always clean and always get/remove well.
+    """
+
+    def dirty(self, ignoreupdate=False, missing=False):
+        return False
+
+    def get(self, state, overwrite=False):
+        return
+
+    def remove(self):
+        return
+
+
+def _setupsubrepo(ui):
+    """Modify the dirstate to prevent stat'ing excluded files,
+    and to prevent modifications to files outside the checkout.
+    """
+
+    def _state(orig, ctx, ui):
+        sparsematch = sparse.matcher(ctx.repo(), revs=[ctx.rev()])
+        if not sparsematch.always():
+            # filter allstate, leave only sparce pathes
+            allstate = orig(ctx, ui)
+            sparcestate = dict()
+            for (idx, item) in allstate.items():
+                if sparsematch(idx):
+                    sparcestate[idx] = item
+            return sparcestate
+        else:
+            return orig(ctx, ui)
+
+    # extensions.wrapfunction(subrepoutil, b'state', _state)
+
+    """ provide DummySubrepo for pathes out of sparse
+    """
+
+    def _subrepo(orig, ctx, path, allowwdir=False, allowcreate=True):
+        sparsematch = sparse.matcher(ctx.repo(), revs=[ctx.rev()])
+        if not sparsematch.always():
+            if not sparsematch(path):
+                return DummySubrepo(ctx, path)
+        return orig(ctx, path, allowwdir, allowcreate)
+
+    extensions.wrapfunction(subrepo, b'subrepo', _subrepo)
+
+
 @command(
     b'debugsparse',
     [
         (b'I', b'include', False, _(b'include files in the sparse checkout')),
         (b'X', b'exclude', False, _(b'exclude files in the sparse checkout')),
         (b'd', b'delete', False, _(b'delete an include/exclude rule')),
         (
             b'f',
             b'force',
             False,
             _(b'allow changing rules even with pending changes'),
         ),
         (b'', b'enable-profile', False, _(b'enables the specified profile')),
         (b'', b'disable-profile', False, _(b'disables the specified profile')),
         (b'', b'import-rules', False, _(b'imports rules from a file')),
         (b'', b'clear-rules', False, _(b'clears local include/exclude rules')),
         (
             b'',
             b'refresh',
             False,
             _(b'updates the working after sparseness changes'),
         ),
         (b'', b'reset', False, _(b'makes the repo full again')),
     ]
     + commands.templateopts,
     _(b'[--OPTION] PATTERN...'),
     helpbasic=True,
 )
 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.
     """
     opts = pycompat.byteskwargs(opts)
     include = opts.get(b'include')
     exclude = opts.get(b'exclude')
     force = opts.get(b'force')
     enableprofile = opts.get(b'enable_profile')
     disableprofile = opts.get(b'disable_profile')
     importrules = opts.get(b'import_rules')
     clearrules = opts.get(b'clear_rules')
     delete = opts.get(b'delete')
     refresh = opts.get(b'refresh')
     reset = opts.get(b'reset')
     count = sum(
         [
             include,
             exclude,
             enableprofile,
             disableprofile,
             delete,
             importrules,
             refresh,
             clearrules,
             reset,
         ]
     )
     if count > 1:
         raise error.Abort(_(b"too many flags specified"))
 
     if count == 0:
         if repo.vfs.exists(b'sparse'):
             ui.status(repo.vfs.read(b"sparse") + b"\n")
             temporaryincludes = sparse.readtemporaryincludes(repo)
             if temporaryincludes:
                 ui.status(
                     _(b"Temporarily Included Files (for merge/rebase):\n")
                 )
                 ui.status((b"\n".join(temporaryincludes) + b"\n"))
             return
         else:
             raise error.Abort(
                 _(
                     b'the debugsparse command is only supported on'
                     b' sparse repositories'
                 )
             )
 
     if include or exclude or delete or reset or enableprofile or disableprofile:
         sparse.updateconfig(
             repo,
             pats,
             opts,
             include=include,
             exclude=exclude,
             reset=reset,
             delete=delete,
             enableprofile=enableprofile,
             disableprofile=disableprofile,
             force=force,
         )
 
     if importrules:
         sparse.importfromfiles(repo, opts, pats, force=force)
 
     if clearrules:
         sparse.clearrules(repo, force=force)
 
     if refresh:
         try:
             wlock = repo.wlock()
             fcounts = map(
                 len,
                 sparse.refreshwdir(
                     repo, repo.status(), sparse.matcher(repo), force=force
                 ),
             )
             sparse.printchanges(
                 ui,
                 opts,
                 added=fcounts[0],
                 dropped=fcounts[1],
                 conflicting=fcounts[2],
             )
         finally:
             wlock.release()

mercurial/subrepoutil.py

@@ -1,510 +1,528 b''
 # subrepoutil.py - sub-repository operations and substate handling
 #
 # Copyright 2009-2010 Olivia 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 errno
 import os
 import posixpath
 import re
 
 from .i18n import _
 from .pycompat import getattr
 from . import (
     config,
     error,
     filemerge,
     pathutil,
     phases,
     pycompat,
+    sparse,
     util,
 )
 from .utils import (
     stringutil,
     urlutil,
 )
 
 nullstate = (b'', b'', b'empty')
 
 if pycompat.TYPE_CHECKING:
     from typing import (
         Any,
         Dict,
         List,
         Optional,
         Set,
         Tuple,
     )
     from . import (
         context,
         localrepo,
         match as matchmod,
         scmutil,
         subrepo,
         ui as uimod,
     )
 
     Substate = Dict[bytes, Tuple[bytes, bytes, bytes]]
 
 
 def state(ctx, ui):
     # type: (context.changectx, uimod.ui) -> Substate
     """return a state dict, mapping subrepo paths configured in .hgsub
     to tuple: (source from .hgsub, revision from .hgsubstate, kind
     (key in types dict))
     """
     p = config.config()
     repo = ctx.repo()
 
     def read(f, sections=None, remap=None):
         if f in ctx:
             try:
                 data = ctx[f].data()
             except IOError as err:
                 if err.errno != errno.ENOENT:
                     raise
                 # handle missing subrepo spec files as removed
                 ui.warn(
                     _(b"warning: subrepo spec file \'%s\' not found\n")
                     % repo.pathto(f)
                 )
                 return
             p.parse(f, data, sections, remap, read)
         else:
             raise error.Abort(
                 _(b"subrepo spec file \'%s\' not found") % repo.pathto(f)
             )
 
     if b'.hgsub' in ctx:
         read(b'.hgsub')
 
     for path, src in ui.configitems(b'subpaths'):
         p.set(b'subpaths', path, src, ui.configsource(b'subpaths', path))
 
     rev = {}
     if b'.hgsubstate' in ctx:
         try:
             for i, l in enumerate(ctx[b'.hgsubstate'].data().splitlines()):
                 l = l.lstrip()
                 if not l:
                     continue
                 try:
                     revision, path = l.split(b" ", 1)
                 except ValueError:
                     raise error.Abort(
                         _(
                             b"invalid subrepository revision "
                             b"specifier in \'%s\' line %d"
                         )
                         % (repo.pathto(b'.hgsubstate'), (i + 1))
                     )
                 rev[path] = revision
         except IOError as err:
             if err.errno != errno.ENOENT:
                 raise
 
     def remap(src):
         # type: (bytes) -> bytes
         for pattern, repl in p.items(b'subpaths'):
             # Turn r'C:\foo\bar' into r'C:\\foo\\bar' since re.sub
             # does a string decode.
             repl = stringutil.escapestr(repl)
             # However, we still want to allow back references to go
             # through unharmed, so we turn r'\\1' into r'\1'. Again,
             # extra escapes are needed because re.sub string decodes.
             repl = re.sub(br'\\\\([0-9]+)', br'\\\1', repl)
             try:
                 src = re.sub(pattern, repl, src, 1)
             except re.error as e:
                 raise error.Abort(
                     _(b"bad subrepository pattern in %s: %s")
                     % (
                         p.source(b'subpaths', pattern),
                         stringutil.forcebytestr(e),
                     )
                 )
         return src
 
     state = {}
     for path, src in p.items(b''):  # type: bytes
         kind = b'hg'
         if src.startswith(b'['):
             if b']' not in src:
                 raise error.Abort(_(b'missing ] in subrepository source'))
             kind, src = src.split(b']', 1)
             kind = kind[1:]
             src = src.lstrip()  # strip any extra whitespace after ']'
 
         if not urlutil.url(src).isabs():
             parent = _abssource(repo, abort=False)
             if parent:
                 parent = urlutil.url(parent)
                 parent.path = posixpath.join(parent.path or b'', src)
                 parent.path = posixpath.normpath(parent.path)
                 joined = bytes(parent)
                 # Remap the full joined path and use it if it changes,
                 # else remap the original source.
                 remapped = remap(joined)
                 if remapped == joined:
                     src = remap(src)
                 else:
                     src = remapped
 
         src = remap(src)
         state[util.pconvert(path)] = (src.strip(), rev.get(path, b''), kind)
 
     return state
 
 
 def writestate(repo, state):
     # type: (localrepo.localrepository, Substate) -> None
     """rewrite .hgsubstate in (outer) repo with these subrepo states"""
     lines = [
         b'%s %s\n' % (state[s][1], s)
         for s in sorted(state)
         if state[s][1] != nullstate[1]
     ]
     repo.wwrite(b'.hgsubstate', b''.join(lines), b'')
 
 
 def submerge(repo, wctx, mctx, actx, overwrite, labels=None):
     # type: (localrepo.localrepository, context.workingctx, context.changectx, context.changectx, bool, Optional[Any]) -> Substate
     # TODO: type the `labels` arg
     """delegated from merge.applyupdates: merging of .hgsubstate file
     in working context, merging context and ancestor context"""
     if mctx == actx:  # backwards?
         actx = wctx.p1()
     s1 = wctx.substate
     s2 = mctx.substate
     sa = actx.substate
     sm = {}
 
+    s1match = sparse.matcher(repo, revs=[wctx.rev()])
+    s2match = sparse.matcher(repo, revs=[mctx.rev()])
+
     repo.ui.debug(b"subrepo merge %s %s %s\n" % (wctx, mctx, actx))
 
     def debug(s, msg, r=b""):
         if r:
             r = b"%s:%s:%s" % r
         repo.ui.debug(b"  subrepo %s: %s %s\n" % (s, msg, r))
 
     promptssrc = filemerge.partextras(labels)
     for s, l in sorted(pycompat.iteritems(s1)):
+        if not s1match(s):
+            sm[s] = l  # ignore changes out of sparse
+            continue
         a = sa.get(s, nullstate)
         ld = l  # local state with possible dirty flag for compares
         if wctx.sub(s).dirty():
             ld = (l[0], l[1] + b"+")
         if wctx == actx:  # overwrite
             a = ld
 
         prompts = promptssrc.copy()
         prompts[b's'] = s
         if s in s2:
+            if not s2match(s):
+                sm[s] = l  # ignore changes out of sparse
+                continue
             r = s2[s]
             if ld == r or r == a:  # no change or local is newer
                 sm[s] = l
                 continue
             elif ld == a:  # other side changed
                 debug(s, b"other changed, get", r)
                 wctx.sub(s).get(r, overwrite)
                 sm[s] = r
             elif ld[0] != r[0]:  # sources differ
                 prompts[b'lo'] = l[0]
                 prompts[b'ro'] = r[0]
                 if repo.ui.promptchoice(
                     _(
                         b' subrepository sources for %(s)s differ\n'
                         b'you can use (l)ocal%(l)s source (%(lo)s)'
                         b' or (r)emote%(o)s source (%(ro)s).\n'
                         b'what do you want to do?'
                         b'$$ &Local $$ &Remote'
                     )
                     % prompts,
                     0,
                 ):
                     debug(s, b"prompt changed, get", r)
                     wctx.sub(s).get(r, overwrite)
                     sm[s] = r
             elif ld[1] == a[1]:  # local side is unchanged
                 debug(s, b"other side changed, get", r)
                 wctx.sub(s).get(r, overwrite)
                 sm[s] = r
             else:
                 debug(s, b"both sides changed")
                 srepo = wctx.sub(s)
                 prompts[b'sl'] = srepo.shortid(l[1])
                 prompts[b'sr'] = srepo.shortid(r[1])
                 option = repo.ui.promptchoice(
                     _(
                         b' subrepository %(s)s diverged (local revision: %(sl)s, '
                         b'remote revision: %(sr)s)\n'
                         b'you can (m)erge, keep (l)ocal%(l)s or keep '
                         b'(r)emote%(o)s.\n'
                         b'what do you want to do?'
                         b'$$ &Merge $$ &Local $$ &Remote'
                     )
                     % prompts,
                     0,
                 )
                 if option == 0:
                     wctx.sub(s).merge(r)
                     sm[s] = l
                     debug(s, b"merge with", r)
                 elif option == 1:
                     sm[s] = l
                     debug(s, b"keep local subrepo revision", l)
                 else:
                     wctx.sub(s).get(r, overwrite)
                     sm[s] = r
                     debug(s, b"get remote subrepo revision", r)
         elif ld == a:  # remote removed, local unchanged
             debug(s, b"remote removed, remove")
             wctx.sub(s).remove()
         elif a == nullstate:  # not present in remote or ancestor
             debug(s, b"local added, keep")
             sm[s] = l
             continue
         else:
             if repo.ui.promptchoice(
                 _(
                     b' local%(l)s changed subrepository %(s)s'
                     b' which remote%(o)s removed\n'
                     b'use (c)hanged version or (d)elete?'
                     b'$$ &Changed $$ &Delete'
                 )
                 % prompts,
                 0,
             ):
                 debug(s, b"prompt remove")
                 wctx.sub(s).remove()
 
     for s, r in sorted(s2.items()):
         if s in s1:
             continue
         elif s not in sa:
             debug(s, b"remote added, get", r)
             mctx.sub(s).get(r)
             sm[s] = r
         elif r != sa[s]:
+            if not s2match(s):
+                # ignore changes out of sparse,
+                continue
+            elif not s1match(s):
+                # recreate changes out of sparse,
+                # sm[s] = r
+                continue
+
             prompts = promptssrc.copy()
             prompts[b's'] = s
             if (
                 repo.ui.promptchoice(
                     _(
                         b' remote%(o)s changed subrepository %(s)s'
                         b' which local%(l)s removed\n'
                         b'use (c)hanged version or (d)elete?'
                         b'$$ &Changed $$ &Delete'
                     )
                     % prompts,
                     0,
                 )
                 == 0
             ):
                 debug(s, b"prompt recreate", r)
                 mctx.sub(s).get(r)
                 sm[s] = r
 
     # record merged .hgsubstate
     writestate(repo, sm)
     return sm
 
 
 def precommit(ui, wctx, status, match, force=False):
     # type: (uimod.ui, context.workingcommitctx, scmutil.status, matchmod.basematcher, bool) -> Tuple[List[bytes], Set[bytes], Substate]
     """Calculate .hgsubstate changes that should be applied before committing
 
     Returns (subs, commitsubs, newstate) where
     - subs: changed subrepos (including dirty ones)
     - commitsubs: dirty subrepos which the caller needs to commit recursively
     - newstate: new state dict which the caller must write to .hgsubstate
 
     This also updates the given status argument.
     """
     subs = []
     commitsubs = set()
     newstate = wctx.substate.copy()
 
     # only manage subrepos and .hgsubstate if .hgsub is present
     if b'.hgsub' in wctx:
         # we'll decide whether to track this ourselves, thanks
         for c in status.modified, status.added, status.removed:
             if b'.hgsubstate' in c:
                 c.remove(b'.hgsubstate')
 
         # compare current state to last committed state
         # build new substate based on last committed state
         oldstate = wctx.p1().substate
         for s in sorted(newstate.keys()):
             if not match(s):
                 # ignore working copy, use old state if present
                 if s in oldstate:
                     newstate[s] = oldstate[s]
                     continue
                 if not force:
                     raise error.Abort(
                         _(b"commit with new subrepo %s excluded") % s
                     )
             dirtyreason = wctx.sub(s).dirtyreason(True)
             if dirtyreason:
                 if not ui.configbool(b'ui', b'commitsubrepos'):
                     raise error.Abort(
                         dirtyreason,
                         hint=_(b"use --subrepos for recursive commit"),
                     )
                 subs.append(s)
                 commitsubs.add(s)
             else:
                 bs = wctx.sub(s).basestate()
                 newstate[s] = (newstate[s][0], bs, newstate[s][2])
                 if oldstate.get(s, (None, None, None))[1] != bs:
                     subs.append(s)
 
         # check for removed subrepos
         for p in wctx.parents():
             r = [s for s in p.substate if s not in newstate]
             subs += [s for s in r if match(s)]
         if subs:
             if not match(b'.hgsub') and b'.hgsub' in (
                 wctx.modified() + wctx.added()
             ):
                 raise error.Abort(_(b"can't commit subrepos without .hgsub"))
             status.modified.insert(0, b'.hgsubstate')
 
     elif b'.hgsub' in status.removed:
         # clean up .hgsubstate when .hgsub is removed
         if b'.hgsubstate' in wctx and b'.hgsubstate' not in (
             status.modified + status.added + status.removed
         ):
             status.removed.insert(0, b'.hgsubstate')
 
     return subs, commitsubs, newstate
 
 
 def repo_rel_or_abs_source(repo):
     """return the source of this repo
 
     Either absolute or relative the outermost repo"""
     parent = repo
     chunks = []
     while util.safehasattr(parent, b'_subparent'):
         source = urlutil.url(parent._subsource)
         chunks.append(bytes(source))
         if source.isabs():
             break
         parent = parent._subparent
 
     chunks.reverse()
     path = posixpath.join(*chunks)
     return posixpath.normpath(path)
 
 
 def reporelpath(repo):
     # type: (localrepo.localrepository) -> bytes
     """return path to this (sub)repo as seen from outermost repo"""
     parent = repo
     while util.safehasattr(parent, b'_subparent'):
         parent = parent._subparent
     return repo.root[len(pathutil.normasprefix(parent.root)) :]
 
 
 def subrelpath(sub):
     # type: (subrepo.abstractsubrepo) -> bytes
     """return path to this subrepo as seen from outermost repo"""
     return sub._relpath
 
 
 def _abssource(repo, push=False, abort=True):
     # type: (localrepo.localrepository, bool, bool) -> Optional[bytes]
     """return pull/push path of repo - either based on parent repo .hgsub info
     or on the top repo config. Abort or return None if no source found."""
     if util.safehasattr(repo, b'_subparent'):
         source = urlutil.url(repo._subsource)
         if source.isabs():
             return bytes(source)
         source.path = posixpath.normpath(source.path)
         parent = _abssource(repo._subparent, push, abort=False)
         if parent:
             parent = urlutil.url(util.pconvert(parent))
             parent.path = posixpath.join(parent.path or b'', source.path)
             parent.path = posixpath.normpath(parent.path)
             return bytes(parent)
     else:  # recursion reached top repo
         path = None
         if util.safehasattr(repo, b'_subtoppath'):
             path = repo._subtoppath
         elif push and repo.ui.config(b'paths', b'default-push'):
             path = repo.ui.config(b'paths', b'default-push')
         elif repo.ui.config(b'paths', b'default'):
             path = repo.ui.config(b'paths', b'default')
         elif repo.shared():
             # chop off the .hg component to get the default path form.  This has
             # already run through vfsmod.vfs(..., realpath=True), so it doesn't
             # have problems with 'C:'
             return os.path.dirname(repo.sharedpath)
         if path:
             # issue5770: 'C:\' and 'C:' are not equivalent paths.  The former is
             # as expected: an absolute path to the root of the C: drive.  The
             # latter is a relative path, and works like so:
             #
             #   C:\>cd C:\some\path
             #   C:\>D:
             #   D:\>python -c "import os; print os.path.abspath('C:')"
             #   C:\some\path
             #
             #   D:\>python -c "import os; print os.path.abspath('C:relative')"
             #   C:\some\path\relative
             if urlutil.hasdriveletter(path):
                 if len(path) == 2 or path[2:3] not in br'\/':
                     path = os.path.abspath(path)
             return path
 
     if abort:
         raise error.Abort(_(b"default path for subrepository not found"))
 
 
 def newcommitphase(ui, ctx):
     # type: (uimod.ui, context.changectx) -> int
     commitphase = phases.newcommitphase(ui)
     substate = getattr(ctx, "substate", None)
     if not substate:
         return commitphase
     check = ui.config(b'phases', b'checksubrepos')
     if check not in (b'ignore', b'follow', b'abort'):
         raise error.Abort(
             _(b'invalid phases.checksubrepos configuration: %s') % check
         )
     if check == b'ignore':
         return commitphase
     maxphase = phases.public
     maxsub = None
     for s in sorted(substate):
         sub = ctx.sub(s)
         subphase = sub.phase(substate[s][1])
         if maxphase < subphase:
             maxphase = subphase
             maxsub = s
     if commitphase < maxphase:
         if check == b'abort':
             raise error.Abort(
                 _(
                     b"can't commit in %s phase"
                     b" conflicting %s from subrepository %s"
                 )
                 % (
                     phases.phasenames[commitphase],
                     phases.phasenames[maxphase],
                     maxsub,
                 )
             )
         ui.warn(
             _(
                 b"warning: changes are committed in"
                 b" %s phase from subrepository %s\n"
             )
             % (phases.phasenames[maxphase], maxsub)
         )
         return maxphase
     return commitphase