config: pass both relative and absolute paths to `include` callback

The `include` callback is responsible for loading configs from
`%include` statements. The callback currently gets passed the absolute
path [1] to the config to read. That is created by joining the dirname
of the file that contains the `%include` statement. For PyOxidizer
support, I'm trying to reduce dependence on paths. This patch helps
with that by passing the relative path found in the `%include`
statement (but with username expansion, etc.) to the `include`
callback. It also turns out that the existing callers can easily adapt
to using the relative path. Coming patches will clean that up and then
we'll remove the absolute path from the callback.

 [1] The "absolute path" bit is a bit of a lie -- it's going to be an
 absolute path if the path that was passed into `config.parse()` was
 absolute.

Differential Revision: https://phab.mercurial-scm.org/D8790

mercurial/config.py

@@ -1,311 +1,315
 # config.py - configuration parsing for Mercurial
 #
 #  Copyright 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 errno
 import os
 
 from .i18n import _
 from .pycompat import getattr
 from . import (
     encoding,
     error,
     pycompat,
     util,
 )
 
 
 class config(object):
     def __init__(self, data=None, includepaths=None):
         self._data = {}
         self._unset = []
         self._includepaths = includepaths or []
         if data:
             for k in data._data:
                 self._data[k] = data[k].copy()
             self._source = data._source.copy()
         else:
             self._source = util.cowdict()
 
     def copy(self):
         return config(self)
 
     def __contains__(self, section):
         return section in self._data
 
     def hasitem(self, section, item):
         return item in self._data.get(section, {})
 
     def __getitem__(self, section):
         return self._data.get(section, {})
 
     def __iter__(self):
         for d in self.sections():
             yield d
 
     def update(self, src):
         self._source = self._source.preparewrite()
         for s, n in src._unset:
             ds = self._data.get(s, None)
             if ds is not None and n in ds:
                 self._data[s] = ds.preparewrite()
                 del self._data[s][n]
                 del self._source[(s, n)]
         for s in src:
             ds = self._data.get(s, None)
             if ds:
                 self._data[s] = ds.preparewrite()
             else:
                 self._data[s] = util.cowsortdict()
             self._data[s].update(src._data[s])
         self._source.update(src._source)
 
     def get(self, section, item, default=None):
         return self._data.get(section, {}).get(item, default)
 
     def backup(self, section, item):
         """return a tuple allowing restore to reinstall a previous value
 
         The main reason we need it is because it handles the "no data" case.
         """
         try:
             value = self._data[section][item]
             source = self.source(section, item)
             return (section, item, value, source)
         except KeyError:
             return (section, item)
 
     def source(self, section, item):
         return self._source.get((section, item), b"")
 
     def sections(self):
         return sorted(self._data.keys())
 
     def items(self, section):
         return list(pycompat.iteritems(self._data.get(section, {})))
 
     def set(self, section, item, value, source=b""):
         if pycompat.ispy3:
             assert not isinstance(
                 section, str
             ), b'config section may not be unicode strings on Python 3'
             assert not isinstance(
                 item, str
             ), b'config item may not be unicode strings on Python 3'
             assert not isinstance(
                 value, str
             ), b'config values may not be unicode strings on Python 3'
         if section not in self:
             self._data[section] = util.cowsortdict()
         else:
             self._data[section] = self._data[section].preparewrite()
         self._data[section][item] = value
         if source:
             self._source = self._source.preparewrite()
             self._source[(section, item)] = source
 
     def restore(self, data):
         """restore data returned by self.backup"""
         self._source = self._source.preparewrite()
         if len(data) == 4:
             # restore old data
             section, item, value, source = data
             self._data[section] = self._data[section].preparewrite()
             self._data[section][item] = value
             self._source[(section, item)] = source
         else:
             # no data before, remove everything
             section, item = data
             if section in self._data:
                 self._data[section].pop(item, None)
             self._source.pop((section, item), None)
 
     def parse(self, src, data, sections=None, remap=None, include=None):
         sectionre = util.re.compile(br'\[([^\[]+)\]')
         itemre = util.re.compile(br'([^=\s][^=]*?)\s*=\s*(.*\S|)')
         contre = util.re.compile(br'\s+(\S|\S.*\S)\s*$')
         emptyre = util.re.compile(br'(;|#|\s*$)')
         commentre = util.re.compile(br'(;|#)')
         unsetre = util.re.compile(br'%unset\s+(\S+)')
         includere = util.re.compile(br'%include\s+(\S|\S.*\S)\s*$')
         section = b""
         item = None
         line = 0
         cont = False
 
         if remap:
             section = remap.get(section, section)
 
         for l in data.splitlines(True):
             line += 1
             if line == 1 and l.startswith(b'\xef\xbb\xbf'):
                 # Someone set us up the BOM
                 l = l[3:]
             if cont:
                 if commentre.match(l):
                     continue
                 m = contre.match(l)
                 if m:
                     if sections and section not in sections:
                         continue
                     v = self.get(section, item) + b"\n" + m.group(1)
                     self.set(section, item, v, b"%s:%d" % (src, line))
                     continue
                 item = None
                 cont = False
             m = includere.match(l)
 
             if m and include:
                 expanded = util.expandpath(m.group(1))
                 includepaths = [os.path.dirname(src)] + self._includepaths
 
                 for base in includepaths:
                     inc = os.path.normpath(os.path.join(base, expanded))
 
                     try:
-                        include(inc, remap=remap, sections=sections)
+                        include(expanded, inc, remap=remap, sections=sections)
                         break
                     except IOError as inst:
                         if inst.errno != errno.ENOENT:
                             raise error.ParseError(
                                 _(b"cannot include %s (%s)")
                                 % (inc, encoding.strtolocal(inst.strerror)),
                                 b"%s:%d" % (src, line),
                             )
                 continue
             if emptyre.match(l):
                 continue
             m = sectionre.match(l)
             if m:
                 section = m.group(1)
                 if remap:
                     section = remap.get(section, section)
                 if section not in self:
                     self._data[section] = util.cowsortdict()
                 continue
             m = itemre.match(l)
             if m:
                 item = m.group(1)
                 cont = True
                 if sections and section not in sections:
                     continue
                 self.set(section, item, m.group(2), b"%s:%d" % (src, line))
                 continue
             m = unsetre.match(l)
             if m:
                 name = m.group(1)
                 if sections and section not in sections:
                     continue
                 if self.get(section, name) is not None:
                     self._data[section] = self._data[section].preparewrite()
                     del self._data[section][name]
                 self._unset.append((section, name))
                 continue
 
             raise error.ParseError(l.rstrip(), (b"%s:%d" % (src, line)))
 
     def read(self, path, fp=None, sections=None, remap=None):
         if not fp:
             fp = util.posixfile(path, b'rb')
         assert getattr(fp, 'mode', 'rb') == 'rb', (
             b'config files must be opened in binary mode, got fp=%r mode=%r'
             % (fp, fp.mode,)
         )
+
+        def include(rel, abs, remap, sections):
+            self.read(abs, remap=remap, sections=sections)
+
         self.parse(
-            path, fp.read(), sections=sections, remap=remap, include=self.read
+            path, fp.read(), sections=sections, remap=remap, include=include
         )
 
 
 def parselist(value):
     """parse a configuration value as a list of comma/space separated strings
 
     >>> parselist(b'this,is "a small" ,test')
     ['this', 'is', 'a small', 'test']
     """
 
     def _parse_plain(parts, s, offset):
         whitespace = False
         while offset < len(s) and (
             s[offset : offset + 1].isspace() or s[offset : offset + 1] == b','
         ):
             whitespace = True
             offset += 1
         if offset >= len(s):
             return None, parts, offset
         if whitespace:
             parts.append(b'')
         if s[offset : offset + 1] == b'"' and not parts[-1]:
             return _parse_quote, parts, offset + 1
         elif s[offset : offset + 1] == b'"' and parts[-1][-1:] == b'\\':
             parts[-1] = parts[-1][:-1] + s[offset : offset + 1]
             return _parse_plain, parts, offset + 1
         parts[-1] += s[offset : offset + 1]
         return _parse_plain, parts, offset + 1
 
     def _parse_quote(parts, s, offset):
         if offset < len(s) and s[offset : offset + 1] == b'"':  # ""
             parts.append(b'')
             offset += 1
             while offset < len(s) and (
                 s[offset : offset + 1].isspace()
                 or s[offset : offset + 1] == b','
             ):
                 offset += 1
             return _parse_plain, parts, offset
 
         while offset < len(s) and s[offset : offset + 1] != b'"':
             if (
                 s[offset : offset + 1] == b'\\'
                 and offset + 1 < len(s)
                 and s[offset + 1 : offset + 2] == b'"'
             ):
                 offset += 1
                 parts[-1] += b'"'
             else:
                 parts[-1] += s[offset : offset + 1]
             offset += 1
 
         if offset >= len(s):
             real_parts = _configlist(parts[-1])
             if not real_parts:
                 parts[-1] = b'"'
             else:
                 real_parts[0] = b'"' + real_parts[0]
                 parts = parts[:-1]
                 parts.extend(real_parts)
             return None, parts, offset
 
         offset += 1
         while offset < len(s) and s[offset : offset + 1] in [b' ', b',']:
             offset += 1
 
         if offset < len(s):
             if offset + 1 == len(s) and s[offset : offset + 1] == b'"':
                 parts[-1] += b'"'
                 offset += 1
             else:
                 parts.append(b'')
         else:
             return None, parts, offset
 
         return _parse_plain, parts, offset
 
     def _configlist(s):
         s = s.rstrip(b' ,')
         if not s:
             return []
         parser, parts, offset = _parse_plain, [b''], 0
         while parser:
             parser, parts, offset = parser(parts, s, offset)
         return parts
 
     if value is not None and isinstance(value, bytes):
         result = _configlist(value.lstrip(b' ,\n'))
     else:
         result = value
     return result or []

mercurial/subrepoutil.py

@@ -1,459 +1,459
 # subrepoutil.py - sub-repository operations and substate handling
 #
 # Copyright 2009-2010 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import errno
 import os
 import posixpath
 import re
 
 from .i18n import _
 from .pycompat import getattr
 from . import (
     config,
     error,
     filemerge,
     pathutil,
     phases,
     pycompat,
     util,
 )
 from .utils import stringutil
 
 nullstate = (b'', b'', b'empty')
 
 
 def state(ctx, ui):
     """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):
+    def read(rel, 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')
+        read(b'.hgsub', 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):
         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[b''].items():
         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 util.url(src).isabs():
             parent = _abssource(repo, abort=False)
             if parent:
                 parent = util.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):
     """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):
     """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 = {}
 
     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)):
         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:
             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]:
             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):
     """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 reporelpath(repo):
     """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):
     """return path to this subrepo as seen from outermost repo"""
     return sub._relpath
 
 
 def _abssource(repo, push=False, abort=True):
     """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 = util.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 = util.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 util.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):
     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