copies: move code into new manifestdict.filesnotin() method

copies._computeforwardmissing() finds files in one context that is not
in the other. Let's move this code into a new method on manifestdict,
so m1.filesnotin(m2) can be optimized for various types of manifests
(we expect more types of manifests soon).

mercurial/copies.py

@@ -1,487 +1,485 b''
 # copies.py - copy detection for Mercurial
 #
 # Copyright 2008 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.
 
 import util
 import heapq
 
 def _nonoverlap(d1, d2, d3):
     "Return list of elements in d1 not in d2 or d3"
     return sorted([d for d in d1 if d not in d3 and d not in d2])
 
 def _dirname(f):
     s = f.rfind("/")
     if s == -1:
         return ""
     return f[:s]
 
 def _findlimit(repo, a, b):
     """
     Find the last revision that needs to be checked to ensure that a full
     transitive closure for file copies can be properly calculated.
     Generally, this means finding the earliest revision number that's an
     ancestor of a or b but not both, except when a or b is a direct descendent
     of the other, in which case we can return the minimum revnum of a and b.
     None if no such revision exists.
     """
 
     # basic idea:
     # - mark a and b with different sides
     # - if a parent's children are all on the same side, the parent is
     #   on that side, otherwise it is on no side
     # - walk the graph in topological order with the help of a heap;
     #   - add unseen parents to side map
     #   - clear side of any parent that has children on different sides
     #   - track number of interesting revs that might still be on a side
     #   - track the lowest interesting rev seen
     #   - quit when interesting revs is zero
 
     cl = repo.changelog
     working = len(cl) # pseudo rev for the working directory
     if a is None:
         a = working
     if b is None:
         b = working
 
     side = {a: -1, b: 1}
     visit = [-a, -b]
     heapq.heapify(visit)
     interesting = len(visit)
     hascommonancestor = False
     limit = working
 
     while interesting:
         r = -heapq.heappop(visit)
         if r == working:
             parents = [cl.rev(p) for p in repo.dirstate.parents()]
         else:
             parents = cl.parentrevs(r)
         for p in parents:
             if p < 0:
                 continue
             if p not in side:
                 # first time we see p; add it to visit
                 side[p] = side[r]
                 if side[p]:
                     interesting += 1
                 heapq.heappush(visit, -p)
             elif side[p] and side[p] != side[r]:
                 # p was interesting but now we know better
                 side[p] = 0
                 interesting -= 1
                 hascommonancestor = True
         if side[r]:
             limit = r # lowest rev visited
             interesting -= 1
 
     if not hascommonancestor:
         return None
 
     # Consider the following flow (see test-commit-amend.t under issue4405):
     # 1/ File 'a0' committed
     # 2/ File renamed from 'a0' to 'a1' in a new commit (call it 'a1')
     # 3/ Move back to first commit
     # 4/ Create a new commit via revert to contents of 'a1' (call it 'a1-amend')
     # 5/ Rename file from 'a1' to 'a2' and commit --amend 'a1-msg'
     #
     # During the amend in step five, we will be in this state:
     #
     # @  3 temporary amend commit for a1-amend
     # |
     # o  2 a1-amend
     # |
     # | o  1 a1
     # |/
     # o  0 a0
     #
     # When _findlimit is called, a and b are revs 3 and 0, so limit will be 2,
     # yet the filelog has the copy information in rev 1 and we will not look
     # back far enough unless we also look at the a and b as candidates.
     # This only occurs when a is a descendent of b or visa-versa.
     return min(limit, a, b)
 
 def _chain(src, dst, a, b):
     '''chain two sets of copies a->b'''
     t = a.copy()
     for k, v in b.iteritems():
         if v in t:
             # found a chain
             if t[v] != k:
                 # file wasn't renamed back to itself
                 t[k] = t[v]
             if v not in dst:
                 # chain was a rename, not a copy
                 del t[v]
         if v in src:
             # file is a copy of an existing file
             t[k] = v
 
     # remove criss-crossed copies
     for k, v in t.items():
         if k in src and v in dst:
             del t[k]
 
     return t
 
 def _tracefile(fctx, am, limit=-1):
     '''return file context that is the ancestor of fctx present in ancestor
     manifest am, stopping after the first ancestor lower than limit'''
 
     for f in fctx.ancestors():
         if am.get(f.path(), None) == f.filenode():
             return f
         if limit >= 0 and f.linkrev() < limit and f.rev() < limit:
             return None
 
 def _dirstatecopies(d):
     ds = d._repo.dirstate
     c = ds.copies().copy()
     for k in c.keys():
         if ds[k] not in 'anm':
             del c[k]
     return c
 
 def _computeforwardmissing(a, b):
     """Computes which files are in b but not a.
     This is its own function so extensions can easily wrap this call to see what
     files _forwardcopies is about to process.
     """
-    missing = set(b.manifest().iterkeys())
-    missing.difference_update(a.manifest().iterkeys())
-    return missing
+    return b.manifest().filesnotin(a.manifest())
 
 def _forwardcopies(a, b):
     '''find {dst@b: src@a} copy mapping where a is an ancestor of b'''
 
     # check for working copy
     w = None
     if b.rev() is None:
         w = b
         b = w.p1()
         if a == b:
             # short-circuit to avoid issues with merge states
             return _dirstatecopies(w)
 
     # files might have to be traced back to the fctx parent of the last
     # one-side-only changeset, but not further back than that
     limit = _findlimit(a._repo, a.rev(), b.rev())
     if limit is None:
         limit = -1
     am = a.manifest()
 
     # find where new files came from
     # we currently don't try to find where old files went, too expensive
     # this means we can miss a case like 'hg rm b; hg cp a b'
     cm = {}
     missing = _computeforwardmissing(a, b)
     ancestrycontext = a._repo.changelog.ancestors([b.rev()], inclusive=True)
     for f in missing:
         fctx = b[f]
         fctx._ancestrycontext = ancestrycontext
         ofctx = _tracefile(fctx, am, limit)
         if ofctx:
             cm[f] = ofctx.path()
 
     # combine copies from dirstate if necessary
     if w is not None:
         cm = _chain(a, w, cm, _dirstatecopies(w))
 
     return cm
 
 def _backwardrenames(a, b):
     # Even though we're not taking copies into account, 1:n rename situations
     # can still exist (e.g. hg cp a b; hg mv a c). In those cases we
     # arbitrarily pick one of the renames.
     f = _forwardcopies(b, a)
     r = {}
     for k, v in sorted(f.iteritems()):
         # remove copies
         if v in a:
             continue
         r[v] = k
     return r
 
 def pathcopies(x, y):
     '''find {dst@y: src@x} copy mapping for directed compare'''
     if x == y or not x or not y:
         return {}
     a = y.ancestor(x)
     if a == x:
         return _forwardcopies(x, y)
     if a == y:
         return _backwardrenames(x, y)
     return _chain(x, y, _backwardrenames(x, a), _forwardcopies(a, y))
 
 def _computenonoverlap(repo, m1, m2, ma):
     """Computes the files exclusive to m1 and m2.
     This is its own function so extensions can easily wrap this call to see what
     files mergecopies is about to process.
     """
     u1 = _nonoverlap(m1, m2, ma)
     u2 = _nonoverlap(m2, m1, ma)
 
     if u1:
         repo.ui.debug("  unmatched files in local:\n   %s\n"
                       % "\n   ".join(u1))
     if u2:
         repo.ui.debug("  unmatched files in other:\n   %s\n"
                       % "\n   ".join(u2))
     return u1, u2
 
 def mergecopies(repo, c1, c2, ca):
     """
     Find moves and copies between context c1 and c2 that are relevant
     for merging.
 
     Returns four dicts: "copy", "movewithdir", "diverge", and
     "renamedelete".
 
     "copy" is a mapping from destination name -> source name,
     where source is in c1 and destination is in c2 or vice-versa.
 
     "movewithdir" is a mapping from source name -> destination name,
     where the file at source present in one context but not the other
     needs to be moved to destination by the merge process, because the
     other context moved the directory it is in.
 
     "diverge" is a mapping of source name -> list of destination names
     for divergent renames.
 
     "renamedelete" is a mapping of source name -> list of destination
     names for files deleted in c1 that were renamed in c2 or vice-versa.
     """
     # avoid silly behavior for update from empty dir
     if not c1 or not c2 or c1 == c2:
         return {}, {}, {}, {}
 
     # avoid silly behavior for parent -> working dir
     if c2.node() is None and c1.node() == repo.dirstate.p1():
         return repo.dirstate.copies(), {}, {}, {}
 
     limit = _findlimit(repo, c1.rev(), c2.rev())
     if limit is None:
         # no common ancestor, no copies
         return {}, {}, {}, {}
     m1 = c1.manifest()
     m2 = c2.manifest()
     ma = ca.manifest()
 
     def makectx(f, n):
         if len(n) != 20: # in a working context?
             if c1.rev() is None:
                 return c1.filectx(f)
             return c2.filectx(f)
         return repo.filectx(f, fileid=n)
 
     ctx = util.lrucachefunc(makectx)
     copy = {}
     movewithdir = {}
     fullcopy = {}
     diverge = {}
 
     repo.ui.debug("  searching for copies back to rev %d\n" % limit)
 
     u1, u2 = _computenonoverlap(repo, m1, m2, ma)
 
     for f in u1:
         checkcopies(ctx, f, m1, m2, ca, limit, diverge, copy, fullcopy)
 
     for f in u2:
         checkcopies(ctx, f, m2, m1, ca, limit, diverge, copy, fullcopy)
 
     renamedelete = {}
     renamedelete2 = set()
     diverge2 = set()
     for of, fl in diverge.items():
         if len(fl) == 1 or of in c1 or of in c2:
             del diverge[of] # not actually divergent, or not a rename
             if of not in c1 and of not in c2:
                 # renamed on one side, deleted on the other side, but filter
                 # out files that have been renamed and then deleted
                 renamedelete[of] = [f for f in fl if f in c1 or f in c2]
                 renamedelete2.update(fl) # reverse map for below
         else:
             diverge2.update(fl) # reverse map for below
 
     bothnew = sorted([d for d in m1 if d in m2 and d not in ma])
     if bothnew:
         repo.ui.debug("  unmatched files new in both:\n   %s\n"
                       % "\n   ".join(bothnew))
     bothdiverge, _copy, _fullcopy = {}, {}, {}
     for f in bothnew:
         checkcopies(ctx, f, m1, m2, ca, limit, bothdiverge, _copy, _fullcopy)
         checkcopies(ctx, f, m2, m1, ca, limit, bothdiverge, _copy, _fullcopy)
     for of, fl in bothdiverge.items():
         if len(fl) == 2 and fl[0] == fl[1]:
             copy[fl[0]] = of # not actually divergent, just matching renames
 
     if fullcopy and repo.ui.debugflag:
         repo.ui.debug("  all copies found (* = to merge, ! = divergent, "
                       "% = renamed and deleted):\n")
         for f in sorted(fullcopy):
             note = ""
             if f in copy:
                 note += "*"
             if f in diverge2:
                 note += "!"
             if f in renamedelete2:
                 note += "%"
             repo.ui.debug("   src: '%s' -> dst: '%s' %s\n" % (fullcopy[f], f,
                                                               note))
     del diverge2
 
     if not fullcopy:
         return copy, movewithdir, diverge, renamedelete
 
     repo.ui.debug("  checking for directory renames\n")
 
     # generate a directory move map
     d1, d2 = c1.dirs(), c2.dirs()
     d1.addpath('/')
     d2.addpath('/')
     invalid = set()
     dirmove = {}
 
     # examine each file copy for a potential directory move, which is
     # when all the files in a directory are moved to a new directory
     for dst, src in fullcopy.iteritems():
         dsrc, ddst = _dirname(src), _dirname(dst)
         if dsrc in invalid:
             # already seen to be uninteresting
             continue
         elif dsrc in d1 and ddst in d1:
             # directory wasn't entirely moved locally
             invalid.add(dsrc)
         elif dsrc in d2 and ddst in d2:
             # directory wasn't entirely moved remotely
             invalid.add(dsrc)
         elif dsrc in dirmove and dirmove[dsrc] != ddst:
             # files from the same directory moved to two different places
             invalid.add(dsrc)
         else:
             # looks good so far
             dirmove[dsrc + "/"] = ddst + "/"
 
     for i in invalid:
         if i in dirmove:
             del dirmove[i]
     del d1, d2, invalid
 
     if not dirmove:
         return copy, movewithdir, diverge, renamedelete
 
     for d in dirmove:
         repo.ui.debug("   discovered dir src: '%s' -> dst: '%s'\n" %
                       (d, dirmove[d]))
 
     # check unaccounted nonoverlapping files against directory moves
     for f in u1 + u2:
         if f not in fullcopy:
             for d in dirmove:
                 if f.startswith(d):
                     # new file added in a directory that was moved, move it
                     df = dirmove[d] + f[len(d):]
                     if df not in copy:
                         movewithdir[f] = df
                         repo.ui.debug(("   pending file src: '%s' -> "
                                        "dst: '%s'\n") % (f, df))
                     break
 
     return copy, movewithdir, diverge, renamedelete
 
 def checkcopies(ctx, f, m1, m2, ca, limit, diverge, copy, fullcopy):
     """
     check possible copies of f from m1 to m2
 
     ctx = function accepting (filename, node) that returns a filectx.
     f = the filename to check
     m1 = the source manifest
     m2 = the destination manifest
     ca = the changectx of the common ancestor
     limit = the rev number to not search beyond
     diverge = record all diverges in this dict
     copy = record all non-divergent copies in this dict
     fullcopy = record all copies in this dict
     """
 
     ma = ca.manifest()
 
     def _related(f1, f2, limit):
         # Walk back to common ancestor to see if the two files originate
         # from the same file. Since workingfilectx's rev() is None it messes
         # up the integer comparison logic, hence the pre-step check for
         # None (f1 and f2 can only be workingfilectx's initially).
 
         if f1 == f2:
             return f1 # a match
 
         g1, g2 = f1.ancestors(), f2.ancestors()
         try:
             f1r, f2r = f1.rev(), f2.rev()
 
             if f1r is None:
                 f1 = g1.next()
             if f2r is None:
                 f2 = g2.next()
 
             while True:
                 f1r, f2r = f1.rev(), f2.rev()
                 if f1r > f2r:
                     f1 = g1.next()
                 elif f2r > f1r:
                     f2 = g2.next()
                 elif f1 == f2:
                     return f1 # a match
                 elif f1r == f2r or f1r < limit or f2r < limit:
                     return False # copy no longer relevant
         except StopIteration:
             return False
 
     of = None
     seen = set([f])
     for oc in ctx(f, m1[f]).ancestors():
         ocr = oc.rev()
         of = oc.path()
         if of in seen:
             # check limit late - grab last rename before
             if ocr < limit:
                 break
             continue
         seen.add(of)
 
         fullcopy[f] = of # remember for dir rename detection
         if of not in m2:
             continue # no match, keep looking
         if m2[of] == ma.get(of):
             break # no merge needed, quit early
         c2 = ctx(of, m2[of])
         cr = _related(oc, c2, ca.rev())
         if cr and (of == f or of == c2.path()): # non-divergent
             copy[f] = of
             of = None
             break
 
     if of in ma:
         diverge.setdefault(of, []).append(f)
 
 def duplicatecopies(repo, rev, fromrev, skiprev=None):
     '''reproduce copies from fromrev to rev in the dirstate
 
     If skiprev is specified, it's a revision that should be used to
     filter copy records. Any copies that occur between fromrev and
     skiprev will not be duplicated, even if they appear in the set of
     copies between fromrev and rev.
     '''
     exclude = {}
     if skiprev is not None:
         exclude = pathcopies(repo[fromrev], repo[skiprev])
     for dst, src in pathcopies(repo[fromrev], repo[rev]).iteritems():
         # copies.pathcopies returns backward renames, so dst might not
         # actually be in the dirstate
         if dst in exclude:
             continue
         if repo.dirstate[dst] in "nma":
             repo.dirstate.copy(src, dst)

mercurial/manifest.py

@@ -1,298 +1,304 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 i18n import _
 import mdiff, parsers, error, revlog, util
 import array, struct
 
 class manifestdict(dict):
     def __init__(self):
         self._flags = {}
     def __setitem__(self, k, v):
         assert v is not None
         dict.__setitem__(self, k, v)
     def flags(self, f):
         return self._flags.get(f, "")
     def setflag(self, f, flags):
         """Set the flags (symlink, executable) for path f."""
         self._flags[f] = flags
     def copy(self):
         copy = manifestdict()
         dict.__init__(copy, self)
         copy._flags = dict.copy(self._flags)
         return copy
     def intersectfiles(self, files):
         '''make a new manifestdict with the intersection of self with files
 
         The algorithm assumes that files is much smaller than self.'''
         ret = manifestdict()
         for fn in files:
             if fn in self:
                 ret[fn] = self[fn]
                 flags = self._flags.get(fn, None)
                 if flags:
                     ret._flags[fn] = flags
         return ret
 
+    def filesnotin(self, m2):
+        '''Set of files in this manifest that are not in the other'''
+        files = set(self.iterkeys())
+        files.difference_update(m2.iterkeys())
+        return files
+
     def matches(self, match):
         '''generate a new manifest filtered by the match argument'''
         if match.always():
             return self.copy()
 
         files = match.files()
         if (match.matchfn == match.exact or
             (not match.anypats() and util.all(fn in self for fn in files))):
             return self.intersectfiles(files)
 
         m = self.copy()
         for fn in m.keys():
             if not match(fn):
                 del m[fn]
         return m
 
     def diff(self, m2, 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.
         '''
         diff = {}
 
         for fn, n1 in self.iteritems():
             fl1 = self._flags.get(fn, '')
             n2 = m2.get(fn, None)
             fl2 = m2._flags.get(fn, '')
             if n2 is None:
                 fl2 = ''
             if n1 != n2 or fl1 != fl2:
                 diff[fn] = ((n1, fl1), (n2, fl2))
             elif clean:
                 diff[fn] = None
 
         for fn, n2 in m2.iteritems():
             if fn not in self:
                 fl2 = m2._flags.get(fn, '')
                 diff[fn] = ((None, ''), (n2, fl2))
 
         return diff
 
     def text(self):
         """Get the full data of this manifest as a bytestring."""
         fl = sorted(self)
         _checkforbidden(fl)
 
         hex, flags = revlog.hex, self.flags
         # if this is changed to support newlines in filenames,
         # be sure to check the templates/ dir again (especially *-raw.tmpl)
         return ''.join("%s\0%s%s\n" % (f, hex(self[f]), flags(f)) for f in fl)
 
     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)
 
         # 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:
                 l = "%s\0%s%s\n" % (f, revlog.hex(self[f]), self.flags(f))
             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)
         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 _parse(lines):
     mfdict = manifestdict()
     parsers.parse_manifest(mfdict, mfdict._flags, lines)
     return mfdict
 
 class manifest(revlog.revlog):
     def __init__(self, opener):
         # 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
         opts = getattr(opener, 'options', None)
         if opts is not None:
             cachesize = opts.get('manifestcachesize', cachesize)
         self._mancache = util.lrucachedict(cachesize)
         revlog.revlog.__init__(self, opener, "00manifest.i")
 
     def readdelta(self, node):
         r = self.rev(node)
         return _parse(mdiff.patchtext(self.revdiff(self.deltaparent(r), r)))
 
     def readfast(self, node):
         '''use the faster of readdelta or read'''
         r = self.rev(node)
         deltaparent = self.deltaparent(r)
         if deltaparent != revlog.nullrev and deltaparent in self.parentrevs(r):
             return self.readdelta(node)
         return self.read(node)
 
     def read(self, node):
         if node == revlog.nullid:
             return manifestdict() # don't upset local cache
         if node in self._mancache:
             return self._mancache[node][0]
         text = self.revision(node)
         arraytext = array.array('c', text)
         m = _parse(text)
         self._mancache[node] = (m, arraytext)
         return m
 
     def find(self, node, f):
         '''look up entry for a single file efficiently.
         return (node, flags) pair if found, (None, None) if not.'''
         if node in self._mancache:
             m = self._mancache[node][0]
             return m.get(f), m.flags(f)
         text = self.revision(node)
         start, end = _msearch(text, f)
         if start == end:
             return None, None
         l = text[start:end]
         f, n = l.split('\0')
         return revlog.bin(n[:40]), n[40:-1]
 
     def add(self, m, transaction, link, p1, p2, added, removed):
         if p1 in self._mancache:
             # 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 list for sorting
             work = [(x, False) for x in added]
             work.extend((x, True) for x in removed)
             # this could use heapq.merge() (from Python 2.6+) or equivalent
             # since the lists are already sorted
             work.sort()
 
             arraytext, deltatext = m.fastdelta(self._mancache[p1][1], work)
             cachedelta = self.rev(p1), deltatext
             text = util.buffer(arraytext)
         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.
             text = m.text()
             arraytext = array.array('c', text)
             cachedelta = None
 
         n = self.addrevision(text, transaction, link, p1, p2, cachedelta)
         self._mancache[n] = (m, arraytext)
 
         return n