manifest: add match argument to diff and filesnotin

As part of removing manifest.matches (since it is O(manifest)), let's start by
adding match arguments to diff and filesnotin. As we'll see in later patches,
these are the only flows that actually use matchers, so by moving the matching
into the actual functions, other manifest implementations can make more efficient
algorithsm.

For instance, this will allow treemanifest diff's to only iterate over the files
that are different AND meet the match criteria.

No consumers are changed in this patches, but the code is fairly easy to verify
visually. Future patches will convert consumers to use it.

One test was affected because it did not use the kwargs version of the clean
parameter.

mercurial/manifest.py

@@ -1,1582 +1,1599 b''
 # manifest.py - manifest revision class for mercurial
 #
 # Copyright 2005-2007 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from __future__ import absolute_import
 
 import array
 import heapq
 import os
 import struct
 
 from .i18n import _
 from . import (
     error,
     mdiff,
     parsers,
     revlog,
     util,
 )
 
 propertycache = util.propertycache
 
 def _parsev1(data):
     # This method does a little bit of excessive-looking
     # precondition checking. This is so that the behavior of this
     # class exactly matches its C counterpart to try and help
     # prevent surprise breakage for anyone that develops against
     # the pure version.
     if data and data[-1] != '\n':
         raise ValueError('Manifest did not end in a newline.')
     prev = None
     for l in data.splitlines():
         if prev is not None and prev > l:
             raise ValueError('Manifest lines not in sorted order.')
         prev = l
         f, n = l.split('\0')
         if len(n) > 40:
             yield f, revlog.bin(n[:40]), n[40:]
         else:
             yield f, revlog.bin(n), ''
 
 def _parsev2(data):
     metadataend = data.find('\n')
     # Just ignore metadata for now
     pos = metadataend + 1
     prevf = ''
     while pos < len(data):
         end = data.find('\n', pos + 1) # +1 to skip stem length byte
         if end == -1:
             raise ValueError('Manifest ended with incomplete file entry.')
         stemlen = ord(data[pos])
         items = data[pos + 1:end].split('\0')
         f = prevf[:stemlen] + items[0]
         if prevf > f:
             raise ValueError('Manifest entries not in sorted order.')
         fl = items[1]
         # Just ignore metadata (items[2:] for now)
         n = data[end + 1:end + 21]
         yield f, n, fl
         pos = end + 22
         prevf = f
 
 def _parse(data):
     """Generates (path, node, flags) tuples from a manifest text"""
     if data.startswith('\0'):
         return iter(_parsev2(data))
     else:
         return iter(_parsev1(data))
 
 def _text(it, usemanifestv2):
     """Given an iterator over (path, node, flags) tuples, returns a manifest
     text"""
     if usemanifestv2:
         return _textv2(it)
     else:
         return _textv1(it)
 
 def _textv1(it):
     files = []
     lines = []
     _hex = revlog.hex
     for f, n, fl in it:
         files.append(f)
         # if this is changed to support newlines in filenames,
         # be sure to check the templates/ dir again (especially *-raw.tmpl)
         lines.append("%s\0%s%s\n" % (f, _hex(n), fl))
 
     _checkforbidden(files)
     return ''.join(lines)
 
 def _textv2(it):
     files = []
     lines = ['\0\n']
     prevf = ''
     for f, n, fl in it:
         files.append(f)
         stem = os.path.commonprefix([prevf, f])
         stemlen = min(len(stem), 255)
         lines.append("%c%s\0%s\n%s\n" % (stemlen, f[stemlen:], fl, n))
         prevf = f
     _checkforbidden(files)
     return ''.join(lines)
 
 class lazymanifestiter(object):
     def __init__(self, lm):
         self.pos = 0
         self.lm = lm
 
     def __iter__(self):
         return self
 
     def next(self):
         try:
             data, pos = self.lm._get(self.pos)
         except IndexError:
             raise StopIteration
         if pos == -1:
             self.pos += 1
             return data[0]
         self.pos += 1
         zeropos = data.find('\x00', pos)
         return data[pos:zeropos]
 
 class lazymanifestiterentries(object):
     def __init__(self, lm):
         self.lm = lm
         self.pos = 0
 
     def __iter__(self):
         return self
 
     def next(self):
         try:
             data, pos = self.lm._get(self.pos)
         except IndexError:
             raise StopIteration
         if pos == -1:
             self.pos += 1
             return data
         zeropos = data.find('\x00', pos)
         hashval = unhexlify(data, self.lm.extrainfo[self.pos],
                             zeropos + 1, 40)
         flags = self.lm._getflags(data, self.pos, zeropos)
         self.pos += 1
         return (data[pos:zeropos], hashval, flags)
 
 def unhexlify(data, extra, pos, length):
     s = data[pos:pos + length].decode('hex')
     if extra:
         s += chr(extra & 0xff)
     return s
 
 def _cmp(a, b):
     return (a > b) - (a < b)
 
 class _lazymanifest(object):
     def __init__(self, data, positions=None, extrainfo=None, extradata=None):
         if positions is None:
             self.positions = self.findlines(data)
             self.extrainfo = [0] * len(self.positions)
             self.data = data
             self.extradata = []
         else:
             self.positions = positions[:]
             self.extrainfo = extrainfo[:]
             self.extradata = extradata[:]
             self.data = data
 
     def findlines(self, data):
         if not data:
             return []
         pos = data.find("\n")
         if pos == -1 or data[-1] != '\n':
             raise ValueError("Manifest did not end in a newline.")
         positions = [0]
         prev = data[:data.find('\x00')]
         while pos < len(data) - 1 and pos != -1:
             positions.append(pos + 1)
             nexts = data[pos + 1:data.find('\x00', pos + 1)]
             if nexts < prev:
                 raise ValueError("Manifest lines not in sorted order.")
             prev = nexts
             pos = data.find("\n", pos + 1)
         return positions
 
     def _get(self, index):
         # get the position encoded in pos:
         #   positive number is an index in 'data'
         #   negative number is in extrapieces
         pos = self.positions[index]
         if pos >= 0:
             return self.data, pos
         return self.extradata[-pos - 1], -1
 
     def _getkey(self, pos):
         if pos >= 0:
             return self.data[pos:self.data.find('\x00', pos + 1)]
         return self.extradata[-pos - 1][0]
 
     def bsearch(self, key):
         first = 0
         last = len(self.positions) - 1
 
         while first <= last:
             midpoint = (first + last)//2
             nextpos = self.positions[midpoint]
             candidate = self._getkey(nextpos)
             r = _cmp(key, candidate)
             if r == 0:
                 return midpoint
             else:
                 if r < 0:
                     last = midpoint - 1
                 else:
                     first = midpoint + 1
         return -1
 
     def bsearch2(self, key):
         # same as the above, but will always return the position
         # done for performance reasons
         first = 0
         last = len(self.positions) - 1
 
         while first <= last:
             midpoint = (first + last)//2
             nextpos = self.positions[midpoint]
             candidate = self._getkey(nextpos)
             r = _cmp(key, candidate)
             if r == 0:
                 return (midpoint, True)
             else:
                 if r < 0:
                     last = midpoint - 1
                 else:
                     first = midpoint + 1
         return (first, False)
 
     def __contains__(self, key):
         return self.bsearch(key) != -1
 
     def _getflags(self, data, needle, pos):
         start = pos + 41
         end = data.find("\n", start)
         if end == -1:
             end = len(data) - 1
         if start == end:
             return ''
         return self.data[start:end]
 
     def __getitem__(self, key):
         if not isinstance(key, str):
             raise TypeError("getitem: manifest keys must be a string.")
         needle = self.bsearch(key)
         if needle == -1:
             raise KeyError
         data, pos = self._get(needle)
         if pos == -1:
             return (data[1], data[2])
         zeropos = data.find('\x00', pos)
         assert 0 <= needle <= len(self.positions)
         assert len(self.extrainfo) == len(self.positions)
         hashval = unhexlify(data, self.extrainfo[needle], zeropos + 1, 40)
         flags = self._getflags(data, needle, zeropos)
         return (hashval, flags)
 
     def __delitem__(self, key):
         needle, found = self.bsearch2(key)
         if not found:
             raise KeyError
         cur = self.positions[needle]
         self.positions = self.positions[:needle] + self.positions[needle + 1:]
         self.extrainfo = self.extrainfo[:needle] + self.extrainfo[needle + 1:]
         if cur >= 0:
             self.data = self.data[:cur] + '\x00' + self.data[cur + 1:]
 
     def __setitem__(self, key, value):
         if not isinstance(key, str):
             raise TypeError("setitem: manifest keys must be a string.")
         if not isinstance(value, tuple) or len(value) != 2:
             raise TypeError("Manifest values must be a tuple of (node, flags).")
         hashval = value[0]
         if not isinstance(hashval, str) or not 20 <= len(hashval) <= 22:
             raise TypeError("node must be a 20-byte string")
         flags = value[1]
         if len(hashval) == 22:
             hashval = hashval[:-1]
         if not isinstance(flags, str) or len(flags) > 1:
             raise TypeError("flags must a 0 or 1 byte string, got %r", flags)
         needle, found = self.bsearch2(key)
         if found:
             # put the item
             pos = self.positions[needle]
             if pos < 0:
                 self.extradata[-pos - 1] = (key, hashval, value[1])
             else:
                 # just don't bother
                 self.extradata.append((key, hashval, value[1]))
                 self.positions[needle] = -len(self.extradata)
         else:
             # not found, put it in with extra positions
             self.extradata.append((key, hashval, value[1]))
             self.positions = (self.positions[:needle] + [-len(self.extradata)]
                               + self.positions[needle:])
             self.extrainfo = (self.extrainfo[:needle] + [0] +
                               self.extrainfo[needle:])
 
     def copy(self):
         # XXX call _compact like in C?
         return _lazymanifest(self.data, self.positions, self.extrainfo,
             self.extradata)
 
     def _compact(self):
         # hopefully not called TOO often
         if len(self.extradata) == 0:
             return
         l = []
         last_cut = 0
         i = 0
         offset = 0
         self.extrainfo = [0] * len(self.positions)
         while i < len(self.positions):
             if self.positions[i] >= 0:
                 cur = self.positions[i]
                 last_cut = cur
                 while True:
                     self.positions[i] = offset
                     i += 1
                     if i == len(self.positions) or self.positions[i] < 0:
                         break
                     offset += self.positions[i] - cur
                     cur = self.positions[i]
                 end_cut = self.data.find('\n', cur)
                 if end_cut != -1:
                     end_cut += 1
                 offset += end_cut - cur
                 l.append(self.data[last_cut:end_cut])
             else:
                 while i < len(self.positions) and self.positions[i] < 0:
                     cur = self.positions[i]
                     t = self.extradata[-cur - 1]
                     l.append(self._pack(t))
                     self.positions[i] = offset
                     if len(t[1]) > 20:
                         self.extrainfo[i] = ord(t[1][21])
                     offset += len(l[-1])
                     i += 1
         self.data = ''.join(l)
         self.extradata = []
 
     def _pack(self, d):
         return d[0] + '\x00' + d[1][:20].encode('hex') + d[2] + '\n'
 
     def text(self):
         self._compact()
         return self.data
 
     def diff(self, m2, clean=False):
         '''Finds changes between the current manifest and m2.'''
         # XXX think whether efficiency matters here
         diff = {}
 
         for fn, e1, flags in self.iterentries():
             if fn not in m2:
                 diff[fn] = (e1, flags), (None, '')
             else:
                 e2 = m2[fn]
                 if (e1, flags) != e2:
                     diff[fn] = (e1, flags), e2
                 elif clean:
                     diff[fn] = None
 
         for fn, e2, flags in m2.iterentries():
             if fn not in self:
                 diff[fn] = (None, ''), (e2, flags)
 
         return diff
 
     def iterentries(self):
         return lazymanifestiterentries(self)
 
     def iterkeys(self):
         return lazymanifestiter(self)
 
     def __iter__(self):
         return lazymanifestiter(self)
 
     def __len__(self):
         return len(self.positions)
 
     def filtercopy(self, filterfn):
         # XXX should be optimized
         c = _lazymanifest('')
         for f, n, fl in self.iterentries():
             if filterfn(f):
                 c[f] = n, fl
         return c
 
 try:
     _lazymanifest = parsers.lazymanifest
 except AttributeError:
     pass
 
 class manifestdict(object):
     def __init__(self, data=''):
         if data.startswith('\0'):
             #_lazymanifest can not parse v2
             self._lm = _lazymanifest('')
             for f, n, fl in _parsev2(data):
                 self._lm[f] = n, fl
         else:
             self._lm = _lazymanifest(data)
 
     def __getitem__(self, key):
         return self._lm[key][0]
 
     def find(self, key):
         return self._lm[key]
 
     def __len__(self):
         return len(self._lm)
 
     def __nonzero__(self):
         # nonzero is covered by the __len__ function, but implementing it here
         # makes it easier for extensions to override.
         return len(self._lm) != 0
 
     def __setitem__(self, key, node):
         self._lm[key] = node, self.flags(key, '')
 
     def __contains__(self, key):
         return key in self._lm
 
     def __delitem__(self, key):
         del self._lm[key]
 
     def __iter__(self):
         return self._lm.__iter__()
 
     def iterkeys(self):
         return self._lm.iterkeys()
 
     def keys(self):
         return list(self.iterkeys())
 
-    def filesnotin(self, m2):
+    def filesnotin(self, m2, match=None):
         '''Set of files in this manifest that are not in the other'''
+        if match:
+            m1 = self.matches(match)
+            m2 = m2.matches(match)
+            return m1.filesnotin(m2)
         diff = self.diff(m2)
         files = set(filepath
                     for filepath, hashflags in diff.iteritems()
                     if hashflags[1][0] is None)
         return files
 
     @propertycache
     def _dirs(self):
         return util.dirs(self)
 
     def dirs(self):
         return self._dirs
 
     def hasdir(self, dir):
         return dir in self._dirs
 
     def _filesfastpath(self, match):
         '''Checks whether we can correctly and quickly iterate over matcher
         files instead of over manifest files.'''
         files = match.files()
         return (len(files) < 100 and (match.isexact() or
             (match.prefix() and all(fn in self for fn in files))))
 
     def walk(self, match):
         '''Generates matching file names.
 
         Equivalent to manifest.matches(match).iterkeys(), but without creating
         an entirely new manifest.
 
         It also reports nonexistent files by marking them bad with match.bad().
         '''
         if match.always():
             for f in iter(self):
                 yield f
             return
 
         fset = set(match.files())
 
         # avoid the entire walk if we're only looking for specific files
         if self._filesfastpath(match):
             for fn in sorted(fset):
                 yield fn
             return
 
         for fn in self:
             if fn in fset:
                 # specified pattern is the exact name
                 fset.remove(fn)
             if match(fn):
                 yield fn
 
         # for dirstate.walk, files=['.'] means "walk the whole tree".
         # follow that here, too
         fset.discard('.')
 
         for fn in sorted(fset):
             if not self.hasdir(fn):
                 match.bad(fn, None)
 
     def matches(self, match):
         '''generate a new manifest filtered by the match argument'''
         if match.always():
             return self.copy()
 
         if self._filesfastpath(match):
             m = manifestdict()
             lm = self._lm
             for fn in match.files():
                 if fn in lm:
                     m._lm[fn] = lm[fn]
             return m
 
         m = manifestdict()
         m._lm = self._lm.filtercopy(match)
         return m
 
-    def diff(self, m2, clean=False):
+    def diff(self, m2, match=None, clean=False):
         '''Finds changes between the current manifest and m2.
 
         Args:
           m2: the manifest to which this manifest should be compared.
           clean: if true, include files unchanged between these manifests
                  with a None value in the returned dictionary.
 
         The result is returned as a dict with filename as key and
         values of the form ((n1,fl1),(n2,fl2)), where n1/n2 is the
         nodeid in the current/other manifest and fl1/fl2 is the flag
         in the current/other manifest. Where the file does not exist,
         the nodeid will be None and the flags will be the empty
         string.
         '''
+        if match:
+            m1 = self.matches(match)
+            m2 = m2.matches(match)
+            return m1.diff(m2, clean=clean)
         return self._lm.diff(m2._lm, clean)
 
     def setflag(self, key, flag):
         self._lm[key] = self[key], flag
 
     def get(self, key, default=None):
         try:
             return self._lm[key][0]
         except KeyError:
             return default
 
     def flags(self, key, default=''):
         try:
             return self._lm[key][1]
         except KeyError:
             return default
 
     def copy(self):
         c = manifestdict()
         c._lm = self._lm.copy()
         return c
 
     def iteritems(self):
         return (x[:2] for x in self._lm.iterentries())
 
     def iterentries(self):
         return self._lm.iterentries()
 
     def text(self, usemanifestv2=False):
         if usemanifestv2:
             return _textv2(self._lm.iterentries())
         else:
             # use (probably) native version for v1
             return self._lm.text()
 
     def fastdelta(self, base, changes):
         """Given a base manifest text as an array.array and a list of changes
         relative to that text, compute a delta that can be used by revlog.
         """
         delta = []
         dstart = None
         dend = None
         dline = [""]
         start = 0
         # zero copy representation of base as a buffer
         addbuf = util.buffer(base)
 
         changes = list(changes)
         if len(changes) < 1000:
             # start with a readonly loop that finds the offset of
             # each line and creates the deltas
             for f, todelete in changes:
                 # bs will either be the index of the item or the insert point
                 start, end = _msearch(addbuf, f, start)
                 if not todelete:
                     h, fl = self._lm[f]
                     l = "%s\0%s%s\n" % (f, revlog.hex(h), fl)
                 else:
                     if start == end:
                         # item we want to delete was not found, error out
                         raise AssertionError(
                                 _("failed to remove %s from manifest") % f)
                     l = ""
                 if dstart is not None and dstart <= start and dend >= start:
                     if dend < end:
                         dend = end
                     if l:
                         dline.append(l)
                 else:
                     if dstart is not None:
                         delta.append([dstart, dend, "".join(dline)])
                     dstart = start
                     dend = end
                     dline = [l]
 
             if dstart is not None:
                 delta.append([dstart, dend, "".join(dline)])
             # apply the delta to the base, and get a delta for addrevision
             deltatext, arraytext = _addlistdelta(base, delta)
         else:
             # For large changes, it's much cheaper to just build the text and
             # diff it.
             arraytext = array.array('c', self.text())
             deltatext = mdiff.textdiff(base, arraytext)
 
         return arraytext, deltatext
 
 def _msearch(m, s, lo=0, hi=None):
     '''return a tuple (start, end) that says where to find s within m.
 
     If the string is found m[start:end] are the line containing
     that string.  If start == end the string was not found and
     they indicate the proper sorted insertion point.
 
     m should be a buffer or a string
     s is a string'''
     def advance(i, c):
         while i < lenm and m[i] != c:
             i += 1
         return i
     if not s:
         return (lo, lo)
     lenm = len(m)
     if not hi:
         hi = lenm
     while lo < hi:
         mid = (lo + hi) // 2
         start = mid
         while start > 0 and m[start - 1] != '\n':
             start -= 1
         end = advance(start, '\0')
         if m[start:end] < s:
             # we know that after the null there are 40 bytes of sha1
             # this translates to the bisect lo = mid + 1
             lo = advance(end + 40, '\n') + 1
         else:
             # this translates to the bisect hi = mid
             hi = start
     end = advance(lo, '\0')
     found = m[lo:end]
     if s == found:
         # we know that after the null there are 40 bytes of sha1
         end = advance(end + 40, '\n')
         return (lo, end + 1)
     else:
         return (lo, lo)
 
 def _checkforbidden(l):
     """Check filenames for illegal characters."""
     for f in l:
         if '\n' in f or '\r' in f:
             raise error.RevlogError(
                 _("'\\n' and '\\r' disallowed in filenames: %r") % f)
 
 
 # apply the changes collected during the bisect loop to our addlist
 # return a delta suitable for addrevision
 def _addlistdelta(addlist, x):
     # for large addlist arrays, building a new array is cheaper
     # than repeatedly modifying the existing one
     currentposition = 0
     newaddlist = array.array('c')
 
     for start, end, content in x:
         newaddlist += addlist[currentposition:start]
         if content:
             newaddlist += array.array('c', content)
 
         currentposition = end
 
     newaddlist += addlist[currentposition:]
 
     deltatext = "".join(struct.pack(">lll", start, end, len(content))
                    + content for start, end, content in x)
     return deltatext, newaddlist
 
 def _splittopdir(f):
     if '/' in f:
         dir, subpath = f.split('/', 1)
         return dir + '/', subpath
     else:
         return '', f
 
 _noop = lambda s: None
 
 class treemanifest(object):
     def __init__(self, dir='', text=''):
         self._dir = dir
         self._node = revlog.nullid
         self._loadfunc = _noop
         self._copyfunc = _noop
         self._dirty = False
         self._dirs = {}
         # Using _lazymanifest here is a little slower than plain old dicts
         self._files = {}
         self._flags = {}
         if text:
             def readsubtree(subdir, subm):
                 raise AssertionError('treemanifest constructor only accepts '
                                      'flat manifests')
             self.parse(text, readsubtree)
             self._dirty = True # Mark flat manifest dirty after parsing
 
     def _subpath(self, path):
         return self._dir + path
 
     def __len__(self):
         self._load()
         size = len(self._files)
         for m in self._dirs.values():
             size += m.__len__()
         return size
 
     def _isempty(self):
         self._load() # for consistency; already loaded by all callers
         return (not self._files and (not self._dirs or
                 all(m._isempty() for m in self._dirs.values())))
 
     def __repr__(self):
         return ('<treemanifest dir=%s, node=%s, loaded=%s, dirty=%s at 0x%x>' %
                 (self._dir, revlog.hex(self._node),
                  bool(self._loadfunc is _noop),
                  self._dirty, id(self)))
 
     def dir(self):
         '''The directory that this tree manifest represents, including a
         trailing '/'. Empty string for the repo root directory.'''
         return self._dir
 
     def node(self):
         '''This node of this instance. nullid for unsaved instances. Should
         be updated when the instance is read or written from a revlog.
         '''
         assert not self._dirty
         return self._node
 
     def setnode(self, node):
         self._node = node
         self._dirty = False
 
     def iterentries(self):
         self._load()
         for p, n in sorted(self._dirs.items() + self._files.items()):
             if p in self._files:
                 yield self._subpath(p), n, self._flags.get(p, '')
             else:
                 for x in n.iterentries():
                     yield x
 
     def iteritems(self):
         self._load()
         for p, n in sorted(self._dirs.items() + self._files.items()):
             if p in self._files:
                 yield self._subpath(p), n
             else:
                 for f, sn in n.iteritems():
                     yield f, sn
 
     def iterkeys(self):
         self._load()
         for p in sorted(self._dirs.keys() + self._files.keys()):
             if p in self._files:
                 yield self._subpath(p)
             else:
                 for f in self._dirs[p].iterkeys():
                     yield f
 
     def keys(self):
         return list(self.iterkeys())
 
     def __iter__(self):
         return self.iterkeys()
 
     def __contains__(self, f):
         if f is None:
             return False
         self._load()
         dir, subpath = _splittopdir(f)
         if dir:
             if dir not in self._dirs:
                 return False
             return self._dirs[dir].__contains__(subpath)
         else:
             return f in self._files
 
     def get(self, f, default=None):
         self._load()
         dir, subpath = _splittopdir(f)
         if dir:
             if dir not in self._dirs:
                 return default
             return self._dirs[dir].get(subpath, default)
         else:
             return self._files.get(f, default)
 
     def __getitem__(self, f):
         self._load()
         dir, subpath = _splittopdir(f)
         if dir:
             return self._dirs[dir].__getitem__(subpath)
         else:
             return self._files[f]
 
     def flags(self, f):
         self._load()
         dir, subpath = _splittopdir(f)
         if dir:
             if dir not in self._dirs:
                 return ''
             return self._dirs[dir].flags(subpath)
         else:
             if f in self._dirs:
                 return ''
             return self._flags.get(f, '')
 
     def find(self, f):
         self._load()
         dir, subpath = _splittopdir(f)
         if dir:
             return self._dirs[dir].find(subpath)
         else:
             return self._files[f], self._flags.get(f, '')
 
     def __delitem__(self, f):
         self._load()
         dir, subpath = _splittopdir(f)
         if dir:
             self._dirs[dir].__delitem__(subpath)
             # If the directory is now empty, remove it
             if self._dirs[dir]._isempty():
                 del self._dirs[dir]
         else:
             del self._files[f]
             if f in self._flags:
                 del self._flags[f]
         self._dirty = True
 
     def __setitem__(self, f, n):
         assert n is not None
         self._load()
         dir, subpath = _splittopdir(f)
         if dir:
             if dir not in self._dirs:
                 self._dirs[dir] = treemanifest(self._subpath(dir))
             self._dirs[dir].__setitem__(subpath, n)
         else:
             self._files[f] = n[:21] # to match manifestdict's behavior
         self._dirty = True
 
     def _load(self):
         if self._loadfunc is not _noop:
             lf, self._loadfunc = self._loadfunc, _noop
             lf(self)
         elif self._copyfunc is not _noop:
             cf, self._copyfunc = self._copyfunc, _noop
             cf(self)
 
     def setflag(self, f, flags):
         """Set the flags (symlink, executable) for path f."""
         self._load()
         dir, subpath = _splittopdir(f)
         if dir:
             if dir not in self._dirs:
                 self._dirs[dir] = treemanifest(self._subpath(dir))
             self._dirs[dir].setflag(subpath, flags)
         else:
             self._flags[f] = flags
         self._dirty = True
 
     def copy(self):
         copy = treemanifest(self._dir)
         copy._node = self._node
         copy._dirty = self._dirty
         if self._copyfunc is _noop:
             def _copyfunc(s):
                 self._load()
                 for d in self._dirs:
                     s._dirs[d] = self._dirs[d].copy()
                 s._files = dict.copy(self._files)
                 s._flags = dict.copy(self._flags)
             if self._loadfunc is _noop:
                 _copyfunc(copy)
             else:
                 copy._copyfunc = _copyfunc
         else:
             copy._copyfunc = self._copyfunc
         return copy
 
-    def filesnotin(self, m2):
+    def filesnotin(self, m2, match=None):
         '''Set of files in this manifest that are not in the other'''
+        if match:
+            m1 = self.matches(match)
+            m2 = m2.matches(match)
+            return m1.filesnotin(m2)
+
         files = set()
         def _filesnotin(t1, t2):
             if t1._node == t2._node and not t1._dirty and not t2._dirty:
                 return
             t1._load()
             t2._load()
             for d, m1 in t1._dirs.iteritems():
                 if d in t2._dirs:
                     m2 = t2._dirs[d]
                     _filesnotin(m1, m2)
                 else:
                     files.update(m1.iterkeys())
 
             for fn in t1._files.iterkeys():
                 if fn not in t2._files:
                     files.add(t1._subpath(fn))
 
         _filesnotin(self, m2)
         return files
 
     @propertycache
     def _alldirs(self):
         return util.dirs(self)
 
     def dirs(self):
         return self._alldirs
 
     def hasdir(self, dir):
         self._load()
         topdir, subdir = _splittopdir(dir)
         if topdir:
             if topdir in self._dirs:
                 return self._dirs[topdir].hasdir(subdir)
             return False
         return (dir + '/') in self._dirs
 
     def walk(self, match):
         '''Generates matching file names.
 
         Equivalent to manifest.matches(match).iterkeys(), but without creating
         an entirely new manifest.
 
         It also reports nonexistent files by marking them bad with match.bad().
         '''
         if match.always():
             for f in iter(self):
                 yield f
             return
 
         fset = set(match.files())
 
         for fn in self._walk(match):
             if fn in fset:
                 # specified pattern is the exact name
                 fset.remove(fn)
             yield fn
 
         # for dirstate.walk, files=['.'] means "walk the whole tree".
         # follow that here, too
         fset.discard('.')
 
         for fn in sorted(fset):
             if not self.hasdir(fn):
                 match.bad(fn, None)
 
     def _walk(self, match):
         '''Recursively generates matching file names for walk().'''
         if not match.visitdir(self._dir[:-1] or '.'):
             return
 
         # yield this dir's files and walk its submanifests
         self._load()
         for p in sorted(self._dirs.keys() + self._files.keys()):
             if p in self._files:
                 fullp = self._subpath(p)
                 if match(fullp):
                     yield fullp
             else:
                 for f in self._dirs[p]._walk(match):
                     yield f
 
     def matches(self, match):
         '''generate a new manifest filtered by the match argument'''
         if match.always():
             return self.copy()
 
         return self._matches(match)
 
     def _matches(self, match):
         '''recursively generate a new manifest filtered by the match argument.
         '''
 
         visit = match.visitdir(self._dir[:-1] or '.')
         if visit == 'all':
             return self.copy()
         ret = treemanifest(self._dir)
         if not visit:
             return ret
 
         self._load()
         for fn in self._files:
             fullp = self._subpath(fn)
             if not match(fullp):
                 continue
             ret._files[fn] = self._files[fn]
             if fn in self._flags:
                 ret._flags[fn] = self._flags[fn]
 
         for dir, subm in self._dirs.iteritems():
             m = subm._matches(match)
             if not m._isempty():
                 ret._dirs[dir] = m
 
         if not ret._isempty():
             ret._dirty = True
         return ret
 
-    def diff(self, m2, clean=False):
+    def diff(self, m2, match=None, clean=False):
         '''Finds changes between the current manifest and m2.
 
         Args:
           m2: the manifest to which this manifest should be compared.
           clean: if true, include files unchanged between these manifests
                  with a None value in the returned dictionary.
 
         The result is returned as a dict with filename as key and
         values of the form ((n1,fl1),(n2,fl2)), where n1/n2 is the
         nodeid in the current/other manifest and fl1/fl2 is the flag
         in the current/other manifest. Where the file does not exist,
         the nodeid will be None and the flags will be the empty
         string.
         '''
+        if match:
+            m1 = self.matches(match)
+            m2 = m2.matches(match)
+            return m1.diff(m2, clean=clean)
         result = {}
         emptytree = treemanifest()
         def _diff(t1, t2):
             if t1._node == t2._node and not t1._dirty and not t2._dirty:
                 return
             t1._load()
             t2._load()
             for d, m1 in t1._dirs.iteritems():
                 m2 = t2._dirs.get(d, emptytree)
                 _diff(m1, m2)
 
             for d, m2 in t2._dirs.iteritems():
                 if d not in t1._dirs:
                     _diff(emptytree, m2)
 
             for fn, n1 in t1._files.iteritems():
                 fl1 = t1._flags.get(fn, '')
                 n2 = t2._files.get(fn, None)
                 fl2 = t2._flags.get(fn, '')
                 if n1 != n2 or fl1 != fl2:
                     result[t1._subpath(fn)] = ((n1, fl1), (n2, fl2))
                 elif clean:
                     result[t1._subpath(fn)] = None
 
             for fn, n2 in t2._files.iteritems():
                 if fn not in t1._files:
                     fl2 = t2._flags.get(fn, '')
                     result[t2._subpath(fn)] = ((None, ''), (n2, fl2))
 
         _diff(self, m2)
         return result
 
     def unmodifiedsince(self, m2):
         return not self._dirty and not m2._dirty and self._node == m2._node
 
     def parse(self, text, readsubtree):
         for f, n, fl in _parse(text):
             if fl == 't':
                 f = f + '/'
                 self._dirs[f] = readsubtree(self._subpath(f), n)
             elif '/' in f:
                 # This is a flat manifest, so use __setitem__ and setflag rather
                 # than assigning directly to _files and _flags, so we can
                 # assign a path in a subdirectory, and to mark dirty (compared
                 # to nullid).
                 self[f] = n
                 if fl:
                     self.setflag(f, fl)
             else:
                 # Assigning to _files and _flags avoids marking as dirty,
                 # and should be a little faster.
                 self._files[f] = n
                 if fl:
                     self._flags[f] = fl
 
     def text(self, usemanifestv2=False):
         """Get the full data of this manifest as a bytestring."""
         self._load()
         return _text(self.iterentries(), usemanifestv2)
 
     def dirtext(self, usemanifestv2=False):
         """Get the full data of this directory as a bytestring. Make sure that
         any submanifests have been written first, so their nodeids are correct.
         """
         self._load()
         flags = self.flags
         dirs = [(d[:-1], self._dirs[d]._node, 't') for d in self._dirs]
         files = [(f, self._files[f], flags(f)) for f in self._files]
         return _text(sorted(dirs + files), usemanifestv2)
 
     def read(self, gettext, readsubtree):
         def _load_for_read(s):
             s.parse(gettext(), readsubtree)
             s._dirty = False
         self._loadfunc = _load_for_read
 
     def writesubtrees(self, m1, m2, writesubtree):
         self._load() # for consistency; should never have any effect here
         m1._load()
         m2._load()
         emptytree = treemanifest()
         for d, subm in self._dirs.iteritems():
             subp1 = m1._dirs.get(d, emptytree)._node
             subp2 = m2._dirs.get(d, emptytree)._node
             if subp1 == revlog.nullid:
                 subp1, subp2 = subp2, subp1
             writesubtree(subm, subp1, subp2)
 
 class manifestrevlog(revlog.revlog):
     '''A revlog that stores manifest texts. This is responsible for caching the
     full-text manifest contents.
     '''
     def __init__(self, opener, dir='', dirlogcache=None, indexfile=None):
         """Constructs a new manifest revlog
 
         `indexfile` - used by extensions to have two manifests at once, like
         when transitioning between flatmanifeset and treemanifests.
         """
         # During normal operations, we expect to deal with not more than four
         # revs at a time (such as during commit --amend). When rebasing large
         # stacks of commits, the number can go up, hence the config knob below.
         cachesize = 4
         usetreemanifest = False
         usemanifestv2 = False
         opts = getattr(opener, 'options', None)
         if opts is not None:
             cachesize = opts.get('manifestcachesize', cachesize)
             usetreemanifest = opts.get('treemanifest', usetreemanifest)
             usemanifestv2 = opts.get('manifestv2', usemanifestv2)
 
         self._treeondisk = usetreemanifest
         self._usemanifestv2 = usemanifestv2
 
         self._fulltextcache = util.lrucachedict(cachesize)
 
         if dir:
             assert self._treeondisk, 'opts is %r' % opts
             if not dir.endswith('/'):
                 dir = dir + '/'
 
         if indexfile is None:
             indexfile = '00manifest.i'
             if dir:
                 indexfile = "meta/" + dir + indexfile
 
         self._dir = dir
         # The dirlogcache is kept on the root manifest log
         if dir:
             self._dirlogcache = dirlogcache
         else:
             self._dirlogcache = {'': self}
 
         super(manifestrevlog, self).__init__(opener, indexfile,
                                              checkambig=bool(dir))
 
     @property
     def fulltextcache(self):
         return self._fulltextcache
 
     def clearcaches(self):
         super(manifestrevlog, self).clearcaches()
         self._fulltextcache.clear()
         self._dirlogcache = {'': self}
 
     def dirlog(self, dir):
         if dir:
             assert self._treeondisk
         if dir not in self._dirlogcache:
             self._dirlogcache[dir] = manifestrevlog(self.opener, dir,
                                                     self._dirlogcache)
         return self._dirlogcache[dir]
 
     def add(self, m, transaction, link, p1, p2, added, removed, readtree=None):
         if (p1 in self.fulltextcache and util.safehasattr(m, 'fastdelta')
             and not self._usemanifestv2):
             # If our first parent is in the manifest cache, we can
             # compute a delta here using properties we know about the
             # manifest up-front, which may save time later for the
             # revlog layer.
 
             _checkforbidden(added)
             # combine the changed lists into one sorted iterator
             work = heapq.merge([(x, False) for x in added],
                                [(x, True) for x in removed])
 
             arraytext, deltatext = m.fastdelta(self.fulltextcache[p1], work)
             cachedelta = self.rev(p1), deltatext
             text = util.buffer(arraytext)
             n = self.addrevision(text, transaction, link, p1, p2, cachedelta)
         else:
             # The first parent manifest isn't already loaded, so we'll
             # just encode a fulltext of the manifest and pass that
             # through to the revlog layer, and let it handle the delta
             # process.
             if self._treeondisk:
                 assert readtree, "readtree must be set for treemanifest writes"
                 m1 = readtree(self._dir, p1)
                 m2 = readtree(self._dir, p2)
                 n = self._addtree(m, transaction, link, m1, m2, readtree)
                 arraytext = None
             else:
                 text = m.text(self._usemanifestv2)
                 n = self.addrevision(text, transaction, link, p1, p2)
                 arraytext = array.array('c', text)
 
         if arraytext is not None:
             self.fulltextcache[n] = arraytext
 
         return n
 
     def _addtree(self, m, transaction, link, m1, m2, readtree):
         # If the manifest is unchanged compared to one parent,
         # don't write a new revision
         if m.unmodifiedsince(m1) or m.unmodifiedsince(m2):
             return m.node()
         def writesubtree(subm, subp1, subp2):
             sublog = self.dirlog(subm.dir())
             sublog.add(subm, transaction, link, subp1, subp2, None, None,
                        readtree=readtree)
         m.writesubtrees(m1, m2, writesubtree)
         text = m.dirtext(self._usemanifestv2)
         # Double-check whether contents are unchanged to one parent
         if text == m1.dirtext(self._usemanifestv2):
             n = m1.node()
         elif text == m2.dirtext(self._usemanifestv2):
             n = m2.node()
         else:
             n = self.addrevision(text, transaction, link, m1.node(), m2.node())
         # Save nodeid so parent manifest can calculate its nodeid
         m.setnode(n)
         return n
 
 class manifestlog(object):
     """A collection class representing the collection of manifest snapshots
     referenced by commits in the repository.
 
     In this situation, 'manifest' refers to the abstract concept of a snapshot
     of the list of files in the given commit. Consumers of the output of this
     class do not care about the implementation details of the actual manifests
     they receive (i.e. tree or flat or lazily loaded, etc)."""
     def __init__(self, opener, repo):
         usetreemanifest = False
         cachesize = 4
 
         opts = getattr(opener, 'options', None)
         if opts is not None:
             usetreemanifest = opts.get('treemanifest', usetreemanifest)
             cachesize = opts.get('manifestcachesize', cachesize)
         self._treeinmem = usetreemanifest
 
         self._oldmanifest = repo._constructmanifest()
         self._revlog = self._oldmanifest
 
         # A cache of the manifestctx or treemanifestctx for each directory
         self._dirmancache = {}
         self._dirmancache[''] = util.lrucachedict(cachesize)
 
         self.cachesize = cachesize
 
     def __getitem__(self, node):
         """Retrieves the manifest instance for the given node. Throws a
         LookupError if not found.
         """
         return self.get('', node)
 
     def get(self, dir, node, verify=True):
         """Retrieves the manifest instance for the given node. Throws a
         LookupError if not found.
 
         `verify` - if True an exception will be thrown if the node is not in
                    the revlog
         """
         if node in self._dirmancache.get(dir, ()):
             cachemf = self._dirmancache[dir][node]
             # The old manifest may put non-ctx manifests in the cache, so
             # skip those since they don't implement the full api.
             if (isinstance(cachemf, manifestctx) or
                 isinstance(cachemf, treemanifestctx)):
                 return cachemf
 
         if dir:
             if self._revlog._treeondisk:
                 if verify:
                     dirlog = self._revlog.dirlog(dir)
                     if node not in dirlog.nodemap:
                         raise LookupError(node, dirlog.indexfile,
                                           _('no node'))
                 m = treemanifestctx(self, dir, node)
             else:
                 raise error.Abort(
                         _("cannot ask for manifest directory '%s' in a flat "
                           "manifest") % dir)
         else:
             if verify:
                 if node not in self._revlog.nodemap:
                     raise LookupError(node, self._revlog.indexfile,
                                       _('no node'))
             if self._treeinmem:
                 m = treemanifestctx(self, '', node)
             else:
                 m = manifestctx(self, node)
 
         if node != revlog.nullid:
             mancache = self._dirmancache.get(dir)
             if not mancache:
                 mancache = util.lrucachedict(self.cachesize)
                 self._dirmancache[dir] = mancache
             mancache[node] = m
         return m
 
     def clearcaches(self):
         self._dirmancache.clear()
         self._revlog.clearcaches()
 
 class memmanifestctx(object):
     def __init__(self, manifestlog):
         self._manifestlog = manifestlog
         self._manifestdict = manifestdict()
 
     def _revlog(self):
         return self._manifestlog._revlog
 
     def new(self):
         return memmanifestctx(self._manifestlog)
 
     def copy(self):
         memmf = memmanifestctx(self._manifestlog)
         memmf._manifestdict = self.read().copy()
         return memmf
 
     def read(self):
         return self._manifestdict
 
     def write(self, transaction, link, p1, p2, added, removed):
         return self._revlog().add(self._manifestdict, transaction, link, p1, p2,
                                   added, removed)
 
 class manifestctx(object):
     """A class representing a single revision of a manifest, including its
     contents, its parent revs, and its linkrev.
     """
     def __init__(self, manifestlog, node):
         self._manifestlog = manifestlog
         self._data = None
 
         self._node = node
 
         # TODO: We eventually want p1, p2, and linkrev exposed on this class,
         # but let's add it later when something needs it and we can load it
         # lazily.
         #self.p1, self.p2 = revlog.parents(node)
         #rev = revlog.rev(node)
         #self.linkrev = revlog.linkrev(rev)
 
     def _revlog(self):
         return self._manifestlog._revlog
 
     def node(self):
         return self._node
 
     def new(self):
         return memmanifestctx(self._manifestlog)
 
     def copy(self):
         memmf = memmanifestctx(self._manifestlog)
         memmf._manifestdict = self.read().copy()
         return memmf
 
     @propertycache
     def parents(self):
         return self._revlog().parents(self._node)
 
     def read(self):
         if self._data is None:
             if self._node == revlog.nullid:
                 self._data = manifestdict()
             else:
                 rl = self._revlog()
                 text = rl.revision(self._node)
                 arraytext = array.array('c', text)
                 rl._fulltextcache[self._node] = arraytext
                 self._data = manifestdict(text)
         return self._data
 
     def readfast(self, shallow=False):
         '''Calls either readdelta or read, based on which would be less work.
         readdelta is called if the delta is against the p1, and therefore can be
         read quickly.
 
         If `shallow` is True, nothing changes since this is a flat manifest.
         '''
         rl = self._revlog()
         r = rl.rev(self._node)
         deltaparent = rl.deltaparent(r)
         if deltaparent != revlog.nullrev and deltaparent in rl.parentrevs(r):
             return self.readdelta()
         return self.read()
 
     def readdelta(self, shallow=False):
         '''Returns a manifest containing just the entries that are present
         in this manifest, but not in its p1 manifest. This is efficient to read
         if the revlog delta is already p1.
 
         Changing the value of `shallow` has no effect on flat manifests.
         '''
         revlog = self._revlog()
         if revlog._usemanifestv2:
             # Need to perform a slow delta
             r0 = revlog.deltaparent(revlog.rev(self._node))
             m0 = self._manifestlog[revlog.node(r0)].read()
             m1 = self.read()
             md = manifestdict()
             for f, ((n0, fl0), (n1, fl1)) in m0.diff(m1).iteritems():
                 if n1:
                     md[f] = n1
                     if fl1:
                         md.setflag(f, fl1)
             return md
 
         r = revlog.rev(self._node)
         d = mdiff.patchtext(revlog.revdiff(revlog.deltaparent(r), r))
         return manifestdict(d)
 
     def find(self, key):
         return self.read().find(key)
 
 class memtreemanifestctx(object):
     def __init__(self, manifestlog, dir=''):
         self._manifestlog = manifestlog
         self._dir = dir
         self._treemanifest = treemanifest()
 
     def _revlog(self):
         return self._manifestlog._revlog
 
     def new(self, dir=''):
         return memtreemanifestctx(self._manifestlog, dir=dir)
 
     def copy(self):
         memmf = memtreemanifestctx(self._manifestlog, dir=self._dir)
         memmf._treemanifest = self._treemanifest.copy()
         return memmf
 
     def read(self):
         return self._treemanifest
 
     def write(self, transaction, link, p1, p2, added, removed):
         def readtree(dir, node):
             return self._manifestlog.get(dir, node).read()
         return self._revlog().add(self._treemanifest, transaction, link, p1, p2,
                                   added, removed, readtree=readtree)
 
 class treemanifestctx(object):
     def __init__(self, manifestlog, dir, node):
         self._manifestlog = manifestlog
         self._dir = dir
         self._data = None
 
         self._node = node
 
         # TODO: Load p1/p2/linkrev lazily. They need to be lazily loaded so that
         # we can instantiate treemanifestctx objects for directories we don't
         # have on disk.
         #self.p1, self.p2 = revlog.parents(node)
         #rev = revlog.rev(node)
         #self.linkrev = revlog.linkrev(rev)
 
     def _revlog(self):
         return self._manifestlog._revlog.dirlog(self._dir)
 
     def read(self):
         if self._data is None:
             rl = self._revlog()
             if self._node == revlog.nullid:
                 self._data = treemanifest()
             elif rl._treeondisk:
                 m = treemanifest(dir=self._dir)
                 def gettext():
                     return rl.revision(self._node)
                 def readsubtree(dir, subm):
                     # Set verify to False since we need to be able to create
                     # subtrees for trees that don't exist on disk.
                     return self._manifestlog.get(dir, subm, verify=False).read()
                 m.read(gettext, readsubtree)
                 m.setnode(self._node)
                 self._data = m
             else:
                 text = rl.revision(self._node)
                 arraytext = array.array('c', text)
                 rl.fulltextcache[self._node] = arraytext
                 self._data = treemanifest(dir=self._dir, text=text)
 
         return self._data
 
     def node(self):
         return self._node
 
     def new(self, dir=''):
         return memtreemanifestctx(self._manifestlog, dir=dir)
 
     def copy(self):
         memmf = memtreemanifestctx(self._manifestlog, dir=self._dir)
         memmf._treemanifest = self.read().copy()
         return memmf
 
     @propertycache
     def parents(self):
         return self._revlog().parents(self._node)
 
     def readdelta(self, shallow=False):
         '''Returns a manifest containing just the entries that are present
         in this manifest, but not in its p1 manifest. This is efficient to read
         if the revlog delta is already p1.
 
         If `shallow` is True, this will read the delta for this directory,
         without recursively reading subdirectory manifests. Instead, any
         subdirectory entry will be reported as it appears in the manifest, i.e.
         the subdirectory will be reported among files and distinguished only by
         its 't' flag.
         '''
         revlog = self._revlog()
         if shallow and not revlog._usemanifestv2:
             r = revlog.rev(self._node)
             d = mdiff.patchtext(revlog.revdiff(revlog.deltaparent(r), r))
             return manifestdict(d)
         else:
             # Need to perform a slow delta
             r0 = revlog.deltaparent(revlog.rev(self._node))
             m0 = self._manifestlog.get(self._dir, revlog.node(r0)).read()
             m1 = self.read()
             md = treemanifest(dir=self._dir)
             for f, ((n0, fl0), (n1, fl1)) in m0.diff(m1).iteritems():
                 if n1:
                     md[f] = n1
                     if fl1:
                         md.setflag(f, fl1)
             return md
 
     def readfast(self, shallow=False):
         '''Calls either readdelta or read, based on which would be less work.
         readdelta is called if the delta is against the p1, and therefore can be
         read quickly.
 
         If `shallow` is True, it only returns the entries from this manifest,
         and not any submanifests.
         '''
         rl = self._revlog()
         r = rl.rev(self._node)
         deltaparent = rl.deltaparent(r)
         if (deltaparent != revlog.nullrev and
             deltaparent in rl.parentrevs(r)):
             return self.readdelta(shallow=shallow)
 
         if shallow:
             return manifestdict(rl.revision(self._node))
         else:
             return self.read()
 
     def find(self, key):
         return self.read().find(key)

tests/test-manifest.py

@@ -1,471 +1,471 b''
 from __future__ import absolute_import
 
 import binascii
 import itertools
 import silenttestrunner
 import unittest
 
 from mercurial import (
     manifest as manifestmod,
     match as matchmod,
 )
 
 EMTPY_MANIFEST = ''
 EMTPY_MANIFEST_V2 = '\0\n'
 
 HASH_1 = '1' * 40
 BIN_HASH_1 = binascii.unhexlify(HASH_1)
 HASH_2 = 'f' * 40
 BIN_HASH_2 = binascii.unhexlify(HASH_2)
 HASH_3 = '1234567890abcdef0987654321deadbeef0fcafe'
 BIN_HASH_3 = binascii.unhexlify(HASH_3)
 A_SHORT_MANIFEST = (
     'bar/baz/qux.py\0%(hash2)s%(flag2)s\n'
     'foo\0%(hash1)s%(flag1)s\n'
     ) % {'hash1': HASH_1,
          'flag1': '',
          'hash2': HASH_2,
          'flag2': 'l',
          }
 
 # Same data as A_SHORT_MANIFEST
 A_SHORT_MANIFEST_V2 = (
     '\0\n'
     '\x00bar/baz/qux.py\0%(flag2)s\n%(hash2)s\n'
     '\x00foo\0%(flag1)s\n%(hash1)s\n'
     ) % {'hash1': BIN_HASH_1,
          'flag1': '',
          'hash2': BIN_HASH_2,
          'flag2': 'l',
          }
 
 # Same data as A_SHORT_MANIFEST
 A_METADATA_MANIFEST = (
     '\0foo\0bar\n'
     '\x00bar/baz/qux.py\0%(flag2)s\0foo\0bar\n%(hash2)s\n' # flag and metadata
     '\x00foo\0%(flag1)s\0foo\n%(hash1)s\n' # no flag, but metadata
     ) % {'hash1': BIN_HASH_1,
          'flag1': '',
          'hash2': BIN_HASH_2,
          'flag2': 'l',
          }
 
 A_STEM_COMPRESSED_MANIFEST = (
     '\0\n'
     '\x00bar/baz/qux.py\0%(flag2)s\n%(hash2)s\n'
     '\x04qux/foo.py\0%(flag1)s\n%(hash1)s\n' # simple case of 4 stem chars
     '\x0az.py\0%(flag1)s\n%(hash1)s\n' # tricky newline = 10 stem characters
     '\x00%(verylongdir)sx/x\0\n%(hash1)s\n'
     '\xffx/y\0\n%(hash2)s\n' # more than 255 stem chars
     ) % {'hash1': BIN_HASH_1,
          'flag1': '',
          'hash2': BIN_HASH_2,
          'flag2': 'l',
          'verylongdir': 255 * 'x',
          }
 
 A_DEEPER_MANIFEST = (
     'a/b/c/bar.py\0%(hash3)s%(flag1)s\n'
     'a/b/c/bar.txt\0%(hash1)s%(flag1)s\n'
     'a/b/c/foo.py\0%(hash3)s%(flag1)s\n'
     'a/b/c/foo.txt\0%(hash2)s%(flag2)s\n'
     'a/b/d/baz.py\0%(hash3)s%(flag1)s\n'
     'a/b/d/qux.py\0%(hash1)s%(flag2)s\n'
     'a/b/d/ten.txt\0%(hash3)s%(flag2)s\n'
     'a/b/dog.py\0%(hash3)s%(flag1)s\n'
     'a/b/fish.py\0%(hash2)s%(flag1)s\n'
     'a/c/london.py\0%(hash3)s%(flag2)s\n'
     'a/c/paper.txt\0%(hash2)s%(flag2)s\n'
     'a/c/paris.py\0%(hash2)s%(flag1)s\n'
     'a/d/apple.py\0%(hash3)s%(flag1)s\n'
     'a/d/pizza.py\0%(hash3)s%(flag2)s\n'
     'a/green.py\0%(hash1)s%(flag2)s\n'
     'a/purple.py\0%(hash2)s%(flag1)s\n'
     'app.py\0%(hash3)s%(flag1)s\n'
     'readme.txt\0%(hash2)s%(flag1)s\n'
     ) % {'hash1': HASH_1,
          'flag1': '',
          'hash2': HASH_2,
          'flag2': 'l',
          'hash3': HASH_3,
          }
 
 HUGE_MANIFEST_ENTRIES = 200001
 
 A_HUGE_MANIFEST = ''.join(sorted(
     'file%d\0%s%s\n' % (i, h, f) for i, h, f in
     itertools.izip(xrange(200001),
                    itertools.cycle((HASH_1, HASH_2)),
                    itertools.cycle(('', 'x', 'l')))))
 
 class basemanifesttests(object):
     def parsemanifest(self, text):
         raise NotImplementedError('parsemanifest not implemented by test case')
 
     def assertIn(self, thing, container, msg=None):
         # assertIn new in 2.7, use it if available, otherwise polyfill
         sup = getattr(unittest.TestCase, 'assertIn', False)
         if sup:
             return sup(self, thing, container, msg=msg)
         if not msg:
             msg = 'Expected %r in %r' % (thing, container)
         self.assert_(thing in container, msg)
 
     def testEmptyManifest(self):
         m = self.parsemanifest(EMTPY_MANIFEST)
         self.assertEqual(0, len(m))
         self.assertEqual([], list(m))
 
     def testEmptyManifestv2(self):
         m = self.parsemanifest(EMTPY_MANIFEST_V2)
         self.assertEqual(0, len(m))
         self.assertEqual([], list(m))
 
     def testManifest(self):
         m = self.parsemanifest(A_SHORT_MANIFEST)
         self.assertEqual(['bar/baz/qux.py', 'foo'], list(m))
         self.assertEqual(BIN_HASH_2, m['bar/baz/qux.py'])
         self.assertEqual('l', m.flags('bar/baz/qux.py'))
         self.assertEqual(BIN_HASH_1, m['foo'])
         self.assertEqual('', m.flags('foo'))
         self.assertRaises(KeyError, lambda : m['wat'])
 
     def testParseManifestV2(self):
         m1 = self.parsemanifest(A_SHORT_MANIFEST)
         m2 = self.parsemanifest(A_SHORT_MANIFEST_V2)
         # Should have same content as A_SHORT_MANIFEST
         self.assertEqual(m1.text(), m2.text())
 
     def testParseManifestMetadata(self):
         # Metadata is for future-proofing and should be accepted but ignored
         m = self.parsemanifest(A_METADATA_MANIFEST)
         self.assertEqual(A_SHORT_MANIFEST, m.text())
 
     def testParseManifestStemCompression(self):
         m = self.parsemanifest(A_STEM_COMPRESSED_MANIFEST)
         self.assertIn('bar/baz/qux.py', m)
         self.assertIn('bar/qux/foo.py', m)
         self.assertIn('bar/qux/foz.py', m)
         self.assertIn(256 * 'x' + '/x', m)
         self.assertIn(256 * 'x' + '/y', m)
         self.assertEqual(A_STEM_COMPRESSED_MANIFEST, m.text(usemanifestv2=True))
 
     def testTextV2(self):
         m1 = self.parsemanifest(A_SHORT_MANIFEST)
         v2text = m1.text(usemanifestv2=True)
         self.assertEqual(A_SHORT_MANIFEST_V2, v2text)
 
     def testSetItem(self):
         want = BIN_HASH_1
 
         m = self.parsemanifest(EMTPY_MANIFEST)
         m['a'] = want
         self.assertIn('a', m)
         self.assertEqual(want, m['a'])
         self.assertEqual('a\0' + HASH_1 + '\n', m.text())
 
         m = self.parsemanifest(A_SHORT_MANIFEST)
         m['a'] = want
         self.assertEqual(want, m['a'])
         self.assertEqual('a\0' + HASH_1 + '\n' + A_SHORT_MANIFEST,
                          m.text())
 
     def testSetFlag(self):
         want = 'x'
 
         m = self.parsemanifest(EMTPY_MANIFEST)
         # first add a file; a file-less flag makes no sense
         m['a'] = BIN_HASH_1
         m.setflag('a', want)
         self.assertEqual(want, m.flags('a'))
         self.assertEqual('a\0' + HASH_1 + want + '\n', m.text())
 
         m = self.parsemanifest(A_SHORT_MANIFEST)
         # first add a file; a file-less flag makes no sense
         m['a'] = BIN_HASH_1
         m.setflag('a', want)
         self.assertEqual(want, m.flags('a'))
         self.assertEqual('a\0' + HASH_1 + want + '\n' + A_SHORT_MANIFEST,
                          m.text())
 
     def testCopy(self):
         m = self.parsemanifest(A_SHORT_MANIFEST)
         m['a'] = BIN_HASH_1
         m2 = m.copy()
         del m
         del m2 # make sure we don't double free() anything
 
     def testCompaction(self):
         unhex = binascii.unhexlify
         h1, h2 = unhex(HASH_1), unhex(HASH_2)
         m = self.parsemanifest(A_SHORT_MANIFEST)
         m['alpha'] = h1
         m['beta'] = h2
         del m['foo']
         want = 'alpha\0%s\nbar/baz/qux.py\0%sl\nbeta\0%s\n' % (
             HASH_1, HASH_2, HASH_2)
         self.assertEqual(want, m.text())
         self.assertEqual(3, len(m))
         self.assertEqual(['alpha', 'bar/baz/qux.py', 'beta'], list(m))
         self.assertEqual(h1, m['alpha'])
         self.assertEqual(h2, m['bar/baz/qux.py'])
         self.assertEqual(h2, m['beta'])
         self.assertEqual('', m.flags('alpha'))
         self.assertEqual('l', m.flags('bar/baz/qux.py'))
         self.assertEqual('', m.flags('beta'))
         self.assertRaises(KeyError, lambda : m['foo'])
 
     def testSetGetNodeSuffix(self):
         clean = self.parsemanifest(A_SHORT_MANIFEST)
         m = self.parsemanifest(A_SHORT_MANIFEST)
         h = m['foo']
         f = m.flags('foo')
         want = h + 'a'
         # Merge code wants to set 21-byte fake hashes at times
         m['foo'] = want
         self.assertEqual(want, m['foo'])
         self.assertEqual([('bar/baz/qux.py', BIN_HASH_2),
                           ('foo', BIN_HASH_1 + 'a')],
                          list(m.iteritems()))
         # Sometimes it even tries a 22-byte fake hash, but we can
         # return 21 and it'll work out
         m['foo'] = want + '+'
         self.assertEqual(want, m['foo'])
         # make sure the suffix survives a copy
         match = matchmod.match('', '', ['re:foo'])
         m2 = m.matches(match)
         self.assertEqual(want, m2['foo'])
         self.assertEqual(1, len(m2))
         m2 = m.copy()
         self.assertEqual(want, m2['foo'])
         # suffix with iteration
         self.assertEqual([('bar/baz/qux.py', BIN_HASH_2),
                           ('foo', want)],
                          list(m.iteritems()))
 
         # shows up in diff
         self.assertEqual({'foo': ((want, f), (h, ''))}, m.diff(clean))
         self.assertEqual({'foo': ((h, ''), (want, f))}, clean.diff(m))
 
     def testMatchException(self):
         m = self.parsemanifest(A_SHORT_MANIFEST)
         match = matchmod.match('', '', ['re:.*'])
         def filt(path):
             if path == 'foo':
                 assert False
             return True
         match.matchfn = filt
         self.assertRaises(AssertionError, m.matches, match)
 
     def testRemoveItem(self):
         m = self.parsemanifest(A_SHORT_MANIFEST)
         del m['foo']
         self.assertRaises(KeyError, lambda : m['foo'])
         self.assertEqual(1, len(m))
         self.assertEqual(1, len(list(m)))
         # now restore and make sure everything works right
         m['foo'] = 'a' * 20
         self.assertEqual(2, len(m))
         self.assertEqual(2, len(list(m)))
 
     def testManifestDiff(self):
         MISSING = (None, '')
         addl = 'z-only-in-left\0' + HASH_1 + '\n'
         addr = 'z-only-in-right\0' + HASH_2 + 'x\n'
         left = self.parsemanifest(
             A_SHORT_MANIFEST.replace(HASH_1, HASH_3 + 'x') + addl)
         right = self.parsemanifest(A_SHORT_MANIFEST + addr)
         want = {
             'foo': ((BIN_HASH_3, 'x'),
                     (BIN_HASH_1, '')),
             'z-only-in-left': ((BIN_HASH_1, ''), MISSING),
             'z-only-in-right': (MISSING, (BIN_HASH_2, 'x')),
             }
         self.assertEqual(want, left.diff(right))
 
         want = {
             'bar/baz/qux.py': (MISSING, (BIN_HASH_2, 'l')),
             'foo': (MISSING, (BIN_HASH_3, 'x')),
             'z-only-in-left': (MISSING, (BIN_HASH_1, '')),
             }
         self.assertEqual(want, self.parsemanifest(EMTPY_MANIFEST).diff(left))
 
         want = {
             'bar/baz/qux.py': ((BIN_HASH_2, 'l'), MISSING),
             'foo': ((BIN_HASH_3, 'x'), MISSING),
             'z-only-in-left': ((BIN_HASH_1, ''), MISSING),
             }
         self.assertEqual(want, left.diff(self.parsemanifest(EMTPY_MANIFEST)))
         copy = right.copy()
         del copy['z-only-in-right']
         del right['foo']
         want = {
             'foo': (MISSING, (BIN_HASH_1, '')),
             'z-only-in-right': ((BIN_HASH_2, 'x'), MISSING),
             }
         self.assertEqual(want, right.diff(copy))
 
         short = self.parsemanifest(A_SHORT_MANIFEST)
         pruned = short.copy()
         del pruned['foo']
         want = {
             'foo': ((BIN_HASH_1, ''), MISSING),
             }
         self.assertEqual(want, short.diff(pruned))
         want = {
             'foo': (MISSING, (BIN_HASH_1, '')),
             }
         self.assertEqual(want, pruned.diff(short))
         want = {
             'bar/baz/qux.py': None,
             'foo': (MISSING, (BIN_HASH_1, '')),
             }
-        self.assertEqual(want, pruned.diff(short, True))
+        self.assertEqual(want, pruned.diff(short, clean=True))
 
     def testReversedLines(self):
         backwards = ''.join(
             l + '\n' for l in reversed(A_SHORT_MANIFEST.split('\n')) if l)
         try:
             self.parsemanifest(backwards)
             self.fail('Should have raised ValueError')
         except ValueError as v:
             self.assertIn('Manifest lines not in sorted order.', str(v))
 
     def testNoTerminalNewline(self):
         try:
             self.parsemanifest(A_SHORT_MANIFEST + 'wat')
             self.fail('Should have raised ValueError')
         except ValueError as v:
             self.assertIn('Manifest did not end in a newline.', str(v))
 
     def testNoNewLineAtAll(self):
         try:
             self.parsemanifest('wat')
             self.fail('Should have raised ValueError')
         except ValueError as v:
             self.assertIn('Manifest did not end in a newline.', str(v))
 
     def testHugeManifest(self):
         m = self.parsemanifest(A_HUGE_MANIFEST)
         self.assertEqual(HUGE_MANIFEST_ENTRIES, len(m))
         self.assertEqual(len(m), len(list(m)))
 
     def testMatchesMetadata(self):
         '''Tests matches() for a few specific files to make sure that both
         the set of files as well as their flags and nodeids are correct in
         the resulting manifest.'''
         m = self.parsemanifest(A_HUGE_MANIFEST)
 
         match = matchmod.match('/', '',
                 ['file1', 'file200', 'file300'], exact=True)
         m2 = m.matches(match)
 
         w = ('file1\0%sx\n'
              'file200\0%sl\n'
              'file300\0%s\n') % (HASH_2, HASH_1, HASH_1)
         self.assertEqual(w, m2.text())
 
     def testMatchesNonexistentFile(self):
         '''Tests matches() for a small set of specific files, including one
         nonexistent file to make sure in only matches against existing files.
         '''
         m = self.parsemanifest(A_DEEPER_MANIFEST)
 
         match = matchmod.match('/', '',
                 ['a/b/c/bar.txt', 'a/b/d/qux.py', 'readme.txt', 'nonexistent'],
                 exact=True)
         m2 = m.matches(match)
 
         self.assertEqual(
                 ['a/b/c/bar.txt', 'a/b/d/qux.py', 'readme.txt'],
                 m2.keys())
 
     def testMatchesNonexistentDirectory(self):
         '''Tests matches() for a relpath match on a directory that doesn't
         actually exist.'''
         m = self.parsemanifest(A_DEEPER_MANIFEST)
 
         match = matchmod.match('/', '', ['a/f'], default='relpath')
         m2 = m.matches(match)
 
         self.assertEqual([], m2.keys())
 
     def testMatchesExactLarge(self):
         '''Tests matches() for files matching a large list of exact files.
         '''
         m = self.parsemanifest(A_HUGE_MANIFEST)
 
         flist = m.keys()[80:300]
         match = matchmod.match('/', '', flist, exact=True)
         m2 = m.matches(match)
 
         self.assertEqual(flist, m2.keys())
 
     def testMatchesFull(self):
         '''Tests matches() for what should be a full match.'''
         m = self.parsemanifest(A_DEEPER_MANIFEST)
 
         match = matchmod.match('/', '', [''])
         m2 = m.matches(match)
 
         self.assertEqual(m.keys(), m2.keys())
 
     def testMatchesDirectory(self):
         '''Tests matches() on a relpath match on a directory, which should
         match against all files within said directory.'''
         m = self.parsemanifest(A_DEEPER_MANIFEST)
 
         match = matchmod.match('/', '', ['a/b'], default='relpath')
         m2 = m.matches(match)
 
         self.assertEqual([
             'a/b/c/bar.py', 'a/b/c/bar.txt', 'a/b/c/foo.py', 'a/b/c/foo.txt',
             'a/b/d/baz.py', 'a/b/d/qux.py', 'a/b/d/ten.txt', 'a/b/dog.py',
             'a/b/fish.py'], m2.keys())
 
     def testMatchesExactPath(self):
         '''Tests matches() on an exact match on a directory, which should
         result in an empty manifest because you can't perform an exact match
         against a directory.'''
         m = self.parsemanifest(A_DEEPER_MANIFEST)
 
         match = matchmod.match('/', '', ['a/b'], exact=True)
         m2 = m.matches(match)
 
         self.assertEqual([], m2.keys())
 
     def testMatchesCwd(self):
         '''Tests matches() on a relpath match with the current directory ('.')
         when not in the root directory.'''
         m = self.parsemanifest(A_DEEPER_MANIFEST)
 
         match = matchmod.match('/', 'a/b', ['.'], default='relpath')
         m2 = m.matches(match)
 
         self.assertEqual([
             'a/b/c/bar.py', 'a/b/c/bar.txt', 'a/b/c/foo.py', 'a/b/c/foo.txt',
             'a/b/d/baz.py', 'a/b/d/qux.py', 'a/b/d/ten.txt', 'a/b/dog.py',
             'a/b/fish.py'], m2.keys())
 
     def testMatchesWithPattern(self):
         '''Tests matches() for files matching a pattern that reside
         deeper than the specified directory.'''
         m = self.parsemanifest(A_DEEPER_MANIFEST)
 
         match = matchmod.match('/', '', ['a/b/*/*.txt'])
         m2 = m.matches(match)
 
         self.assertEqual(
                 ['a/b/c/bar.txt', 'a/b/c/foo.txt', 'a/b/d/ten.txt'],
                 m2.keys())
 
 class testmanifestdict(unittest.TestCase, basemanifesttests):
     def parsemanifest(self, text):
         return manifestmod.manifestdict(text)
 
 class testtreemanifest(unittest.TestCase, basemanifesttests):
     def parsemanifest(self, text):
         return manifestmod.treemanifest('', text)
 
 if __name__ == '__main__':
     silenttestrunner.main(__name__)