mdiff/patch: fix bad hunk handling for unified diffs with zero context

Prior to this patch "hg diff -U0", i.e., zero lines of context, would
output hunk headers with a start line one greater than what GNU patch
and git output. Guido van Rossum documents the unified diff format[1]
as having a start line value "one lower than one would expect" for
zero length hunks.

Comparing the behaviour of the three systems prior to this patch in
transforming

  c1
  c3

to

  c1
  c2
  c3

- GNU "diff -U0" reports the hunk as "@@ -1,0 +2 @@"
- "git diff -U0" reports the hunk as "@@ -1,0 +2 @@"
- "hg diff -U0" reports the hunk as "@@ -2,0 +2,1 @@"

After this patch, "hg diff -U0" reports "@@ -1,0 +2,1 @@".

Since "hg export --config diff.unified=0" outputs zero-context unified
diffs, "hg import" has also been updated to account for start lines
one less than expected for zero length hunk ranges.

[1]: http://www.artima.com/weblogs/viewpost.jsp?thread=164293

mercurial/mdiff.py

@@ -1,281 +1,287
 # mdiff.py - diff and patch routines for mercurial
 #
 # Copyright 2005, 2006 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 bdiff, mpatch, util
 import re, struct
 
 def splitnewlines(text):
     '''like str.splitlines, but only split on newlines.'''
     lines = [l + '\n' for l in text.split('\n')]
     if lines:
         if lines[-1] == '\n':
             lines.pop()
         else:
             lines[-1] = lines[-1][:-1]
     return lines
 
 class diffopts(object):
     '''context is the number of context lines
     text treats all files as text
     showfunc enables diff -p output
     git enables the git extended patch format
     nodates removes dates from diff headers
     ignorews ignores all whitespace changes in the diff
     ignorewsamount ignores changes in the amount of whitespace
     ignoreblanklines ignores changes whose lines are all blank
     upgrade generates git diffs to avoid data loss
     '''
 
     defaults = {
         'context': 3,
         'text': False,
         'showfunc': False,
         'git': False,
         'nodates': False,
         'ignorews': False,
         'ignorewsamount': False,
         'ignoreblanklines': False,
         'upgrade': False,
         }
 
     __slots__ = defaults.keys()
 
     def __init__(self, **opts):
         for k in self.__slots__:
             v = opts.get(k)
             if v is None:
                 v = self.defaults[k]
             setattr(self, k, v)
 
         try:
             self.context = int(self.context)
         except ValueError:
             raise util.Abort(_('diff context lines count must be '
                                'an integer, not %r') % self.context)
 
     def copy(self, **kwargs):
         opts = dict((k, getattr(self, k)) for k in self.defaults)
         opts.update(kwargs)
         return diffopts(**opts)
 
 defaultopts = diffopts()
 
 def wsclean(opts, text, blank=True):
     if opts.ignorews:
         text = re.sub('[ \t\r]+', '', text)
     elif opts.ignorewsamount:
         text = re.sub('[ \t\r]+', ' ', text)
         text = text.replace(' \n', '\n')
     if blank and opts.ignoreblanklines:
         text = re.sub('\n+', '', text)
     return text
 
 def diffline(revs, a, b, opts):
     parts = ['diff']
     if opts.git:
         parts.append('--git')
     if revs and not opts.git:
         parts.append(' '.join(["-r %s" % rev for rev in revs]))
     if opts.git:
         parts.append('a/%s' % a)
         parts.append('b/%s' % b)
     else:
         parts.append(a)
     return ' '.join(parts) + '\n'
 
 def unidiff(a, ad, b, bd, fn1, fn2, r=None, opts=defaultopts):
     def datetag(date, addtab=True):
         if not opts.git and not opts.nodates:
             return '\t%s\n' % date
         if addtab and ' ' in fn1:
             return '\t\n'
         return '\n'
 
     if not a and not b:
         return ""
     epoch = util.datestr((0, 0))
 
     if not opts.text and (util.binary(a) or util.binary(b)):
         if a and b and len(a) == len(b) and a == b:
             return ""
         l = ['Binary file %s has changed\n' % fn1]
     elif not a:
         b = splitnewlines(b)
         if a is None:
             l1 = '--- /dev/null%s' % datetag(epoch, False)
         else:
             l1 = "--- %s%s" % ("a/" + fn1, datetag(ad))
         l2 = "+++ %s%s" % ("b/" + fn2, datetag(bd))
         l3 = "@@ -0,0 +1,%d @@\n" % len(b)
         l = [l1, l2, l3] + ["+" + e for e in b]
     elif not b:
         a = splitnewlines(a)
         l1 = "--- %s%s" % ("a/" + fn1, datetag(ad))
         if b is None:
             l2 = '+++ /dev/null%s' % datetag(epoch, False)
         else:
             l2 = "+++ %s%s" % ("b/" + fn2, datetag(bd))
         l3 = "@@ -1,%d +0,0 @@\n" % len(a)
         l = [l1, l2, l3] + ["-" + e for e in a]
     else:
         al = splitnewlines(a)
         bl = splitnewlines(b)
         l = list(_unidiff(a, b, al, bl, opts=opts))
         if not l:
             return ""
 
         l.insert(0, "--- a/%s%s" % (fn1, datetag(ad)))
         l.insert(1, "+++ b/%s%s" % (fn2, datetag(bd)))
 
     for ln in xrange(len(l)):
         if l[ln][-1] != '\n':
             l[ln] += "\n\ No newline at end of file\n"
 
     if r:
         l.insert(0, diffline(r, fn1, fn2, opts))
 
     return "".join(l)
 
 # creates a headerless unified diff
 # t1 and t2 are the text to be diffed
 # l1 and l2 are the text broken up into lines
 def _unidiff(t1, t2, l1, l2, opts=defaultopts):
     def contextend(l, len):
         ret = l + opts.context
         if ret > len:
             ret = len
         return ret
 
     def contextstart(l):
         ret = l - opts.context
         if ret < 0:
             return 0
         return ret
 
     lastfunc = [0, '']
     def yieldhunk(hunk):
         (astart, a2, bstart, b2, delta) = hunk
         aend = contextend(a2, len(l1))
         alen = aend - astart
         blen = b2 - bstart + aend - a2
 
         func = ""
         if opts.showfunc:
             lastpos, func = lastfunc
             # walk backwards from the start of the context up to the start of
             # the previous hunk context until we find a line starting with an
             # alphanumeric char.
             for i in xrange(astart - 1, lastpos - 1, -1):
                 if l1[i][0].isalnum():
                     func = ' ' + l1[i].rstrip()[:40]
                     lastfunc[1] = func
                     break
             # by recording this hunk's starting point as the next place to
             # start looking for function lines, we avoid reading any line in
             # the file more than once.
             lastfunc[0] = astart
 
-        yield "@@ -%d,%d +%d,%d @@%s\n" % (astart + 1, alen,
-                                           bstart + 1, blen, func)
+        # zero-length hunk ranges report their start line as one less
+        if alen:
+            astart += 1
+        if blen:
+            bstart += 1
+
+        yield "@@ -%d,%d +%d,%d @@%s\n" % (astart, alen,
+                                           bstart, blen, func)
         for x in delta:
             yield x
         for x in xrange(a2, aend):
             yield ' ' + l1[x]
 
     # bdiff.blocks gives us the matching sequences in the files.  The loop
     # below finds the spaces between those matching sequences and translates
     # them into diff output.
     #
     if opts.ignorews or opts.ignorewsamount:
         t1 = wsclean(opts, t1, False)
         t2 = wsclean(opts, t2, False)
 
     diff = bdiff.blocks(t1, t2)
     hunk = None
     for i, s1 in enumerate(diff):
         # The first match is special.
         # we've either found a match starting at line 0 or a match later
         # in the file.  If it starts later, old and new below will both be
         # empty and we'll continue to the next match.
         if i > 0:
             s = diff[i - 1]
         else:
             s = [0, 0, 0, 0]
         delta = []
         a1 = s[1]
         a2 = s1[0]
         b1 = s[3]
         b2 = s1[2]
 
         old = l1[a1:a2]
         new = l2[b1:b2]
 
         # bdiff sometimes gives huge matches past eof, this check eats them,
         # and deals with the special first match case described above
         if not old and not new:
             continue
 
         if opts.ignoreblanklines:
             if wsclean(opts, "".join(old)) == wsclean(opts, "".join(new)):
                 continue
 
         astart = contextstart(a1)
         bstart = contextstart(b1)
         prev = None
         if hunk:
             # join with the previous hunk if it falls inside the context
             if astart < hunk[1] + opts.context + 1:
                 prev = hunk
                 astart = hunk[1]
                 bstart = hunk[3]
             else:
                 for x in yieldhunk(hunk):
                     yield x
         if prev:
             # we've joined the previous hunk, record the new ending points.
             hunk[1] = a2
             hunk[3] = b2
             delta = hunk[4]
         else:
             # create a new hunk
             hunk = [astart, a2, bstart, b2, delta]
 
         delta[len(delta):] = [' ' + x for x in l1[astart:a1]]
         delta[len(delta):] = ['-' + x for x in old]
         delta[len(delta):] = ['+' + x for x in new]
 
     if hunk:
         for x in yieldhunk(hunk):
             yield x
 
 def patchtext(bin):
     pos = 0
     t = []
     while pos < len(bin):
         p1, p2, l = struct.unpack(">lll", bin[pos:pos + 12])
         pos += 12
         t.append(bin[pos:pos + l])
         pos += l
     return "".join(t)
 
 def patch(a, bin):
     if len(a) == 0:
         # skip over trivial delta header
         return buffer(bin, 12)
     return mpatch.patches(a, [bin])
 
 # similar to difflib.SequenceMatcher.get_matching_blocks
 def get_matching_blocks(a, b):
     return [(d[0], d[2], d[1] - d[0]) for d in bdiff.blocks(a, b)]
 
 def trivialdiffheader(length):
     return struct.pack(">lll", 0, 0, length)
 
 patches = mpatch.patches
 patchedsize = mpatch.patchedsize
 textdiff = bdiff.bdiff

mercurial/patch.py

@@ -1,1870 +1,1869
 # patch.py - patch file parsing routines
 #
 # Copyright 2006 Brendan Cully <[e-mail]>
 # Copyright 2007 Chris Mason <[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 cStringIO, email.Parser, os, errno, re
 import tempfile, zlib, shutil
 
 from i18n import _
 from node import hex, nullid, short
 import base85, mdiff, scmutil, util, diffhelpers, copies, encoding, error
 import context
 
 gitre = re.compile('diff --git a/(.*) b/(.*)')
 
 class PatchError(Exception):
     pass
 
 
 # public functions
 
 def split(stream):
     '''return an iterator of individual patches from a stream'''
     def isheader(line, inheader):
         if inheader and line[0] in (' ', '\t'):
             # continuation
             return True
         if line[0] in (' ', '-', '+'):
             # diff line - don't check for header pattern in there
             return False
         l = line.split(': ', 1)
         return len(l) == 2 and ' ' not in l[0]
 
     def chunk(lines):
         return cStringIO.StringIO(''.join(lines))
 
     def hgsplit(stream, cur):
         inheader = True
 
         for line in stream:
             if not line.strip():
                 inheader = False
             if not inheader and line.startswith('# HG changeset patch'):
                 yield chunk(cur)
                 cur = []
                 inheader = True
 
             cur.append(line)
 
         if cur:
             yield chunk(cur)
 
     def mboxsplit(stream, cur):
         for line in stream:
             if line.startswith('From '):
                 for c in split(chunk(cur[1:])):
                     yield c
                 cur = []
 
             cur.append(line)
 
         if cur:
             for c in split(chunk(cur[1:])):
                 yield c
 
     def mimesplit(stream, cur):
         def msgfp(m):
             fp = cStringIO.StringIO()
             g = email.Generator.Generator(fp, mangle_from_=False)
             g.flatten(m)
             fp.seek(0)
             return fp
 
         for line in stream:
             cur.append(line)
         c = chunk(cur)
 
         m = email.Parser.Parser().parse(c)
         if not m.is_multipart():
             yield msgfp(m)
         else:
             ok_types = ('text/plain', 'text/x-diff', 'text/x-patch')
             for part in m.walk():
                 ct = part.get_content_type()
                 if ct not in ok_types:
                     continue
                 yield msgfp(part)
 
     def headersplit(stream, cur):
         inheader = False
 
         for line in stream:
             if not inheader and isheader(line, inheader):
                 yield chunk(cur)
                 cur = []
                 inheader = True
             if inheader and not isheader(line, inheader):
                 inheader = False
 
             cur.append(line)
 
         if cur:
             yield chunk(cur)
 
     def remainder(cur):
         yield chunk(cur)
 
     class fiter(object):
         def __init__(self, fp):
             self.fp = fp
 
         def __iter__(self):
             return self
 
         def next(self):
             l = self.fp.readline()
             if not l:
                 raise StopIteration
             return l
 
     inheader = False
     cur = []
 
     mimeheaders = ['content-type']
 
     if not util.safehasattr(stream, 'next'):
         # http responses, for example, have readline but not next
         stream = fiter(stream)
 
     for line in stream:
         cur.append(line)
         if line.startswith('# HG changeset patch'):
             return hgsplit(stream, cur)
         elif line.startswith('From '):
             return mboxsplit(stream, cur)
         elif isheader(line, inheader):
             inheader = True
             if line.split(':', 1)[0].lower() in mimeheaders:
                 # let email parser handle this
                 return mimesplit(stream, cur)
         elif line.startswith('--- ') and inheader:
             # No evil headers seen by diff start, split by hand
             return headersplit(stream, cur)
         # Not enough info, keep reading
 
     # if we are here, we have a very plain patch
     return remainder(cur)
 
 def extract(ui, fileobj):
     '''extract patch from data read from fileobj.
 
     patch can be a normal patch or contained in an email message.
 
     return tuple (filename, message, user, date, branch, node, p1, p2).
     Any item in the returned tuple can be None. If filename is None,
     fileobj did not contain a patch. Caller must unlink filename when done.'''
 
     # attempt to detect the start of a patch
     # (this heuristic is borrowed from quilt)
     diffre = re.compile(r'^(?:Index:[ \t]|diff[ \t]|RCS file: |'
                         r'retrieving revision [0-9]+(\.[0-9]+)*$|'
                         r'---[ \t].*?^\+\+\+[ \t]|'
                         r'\*\*\*[ \t].*?^---[ \t])', re.MULTILINE|re.DOTALL)
 
     fd, tmpname = tempfile.mkstemp(prefix='hg-patch-')
     tmpfp = os.fdopen(fd, 'w')
     try:
         msg = email.Parser.Parser().parse(fileobj)
 
         subject = msg['Subject']
         user = msg['From']
         if not subject and not user:
             # Not an email, restore parsed headers if any
             subject = '\n'.join(': '.join(h) for h in msg.items()) + '\n'
 
         gitsendmail = 'git-send-email' in msg.get('X-Mailer', '')
         # should try to parse msg['Date']
         date = None
         nodeid = None
         branch = None
         parents = []
 
         if subject:
             if subject.startswith('[PATCH'):
                 pend = subject.find(']')
                 if pend >= 0:
                     subject = subject[pend + 1:].lstrip()
             subject = re.sub(r'\n[ \t]+', ' ', subject)
             ui.debug('Subject: %s\n' % subject)
         if user:
             ui.debug('From: %s\n' % user)
         diffs_seen = 0
         ok_types = ('text/plain', 'text/x-diff', 'text/x-patch')
         message = ''
         for part in msg.walk():
             content_type = part.get_content_type()
             ui.debug('Content-Type: %s\n' % content_type)
             if content_type not in ok_types:
                 continue
             payload = part.get_payload(decode=True)
             m = diffre.search(payload)
             if m:
                 hgpatch = False
                 hgpatchheader = False
                 ignoretext = False
 
                 ui.debug('found patch at byte %d\n' % m.start(0))
                 diffs_seen += 1
                 cfp = cStringIO.StringIO()
                 for line in payload[:m.start(0)].splitlines():
                     if line.startswith('# HG changeset patch') and not hgpatch:
                         ui.debug('patch generated by hg export\n')
                         hgpatch = True
                         hgpatchheader = True
                         # drop earlier commit message content
                         cfp.seek(0)
                         cfp.truncate()
                         subject = None
                     elif hgpatchheader:
                         if line.startswith('# User '):
                             user = line[7:]
                             ui.debug('From: %s\n' % user)
                         elif line.startswith("# Date "):
                             date = line[7:]
                         elif line.startswith("# Branch "):
                             branch = line[9:]
                         elif line.startswith("# Node ID "):
                             nodeid = line[10:]
                         elif line.startswith("# Parent "):
                             parents.append(line[10:])
                         elif not line.startswith("# "):
                             hgpatchheader = False
                     elif line == '---' and gitsendmail:
                         ignoretext = True
                     if not hgpatchheader and not ignoretext:
                         cfp.write(line)
                         cfp.write('\n')
                 message = cfp.getvalue()
                 if tmpfp:
                     tmpfp.write(payload)
                     if not payload.endswith('\n'):
                         tmpfp.write('\n')
             elif not diffs_seen and message and content_type == 'text/plain':
                 message += '\n' + payload
     except:
         tmpfp.close()
         os.unlink(tmpname)
         raise
 
     if subject and not message.startswith(subject):
         message = '%s\n%s' % (subject, message)
     tmpfp.close()
     if not diffs_seen:
         os.unlink(tmpname)
         return None, message, user, date, branch, None, None, None
     p1 = parents and parents.pop(0) or None
     p2 = parents and parents.pop(0) or None
     return tmpname, message, user, date, branch, nodeid, p1, p2
 
 class patchmeta(object):
     """Patched file metadata
 
     'op' is the performed operation within ADD, DELETE, RENAME, MODIFY
     or COPY.  'path' is patched file path. 'oldpath' is set to the
     origin file when 'op' is either COPY or RENAME, None otherwise. If
     file mode is changed, 'mode' is a tuple (islink, isexec) where
     'islink' is True if the file is a symlink and 'isexec' is True if
     the file is executable. Otherwise, 'mode' is None.
     """
     def __init__(self, path):
         self.path = path
         self.oldpath = None
         self.mode = None
         self.op = 'MODIFY'
         self.binary = False
 
     def setmode(self, mode):
         islink = mode & 020000
         isexec = mode & 0100
         self.mode = (islink, isexec)
 
     def copy(self):
         other = patchmeta(self.path)
         other.oldpath = self.oldpath
         other.mode = self.mode
         other.op = self.op
         other.binary = self.binary
         return other
 
     def __repr__(self):
         return "<patchmeta %s %r>" % (self.op, self.path)
 
 def readgitpatch(lr):
     """extract git-style metadata about patches from <patchname>"""
 
     # Filter patch for git information
     gp = None
     gitpatches = []
     for line in lr:
         line = line.rstrip(' \r\n')
         if line.startswith('diff --git'):
             m = gitre.match(line)
             if m:
                 if gp:
                     gitpatches.append(gp)
                 dst = m.group(2)
                 gp = patchmeta(dst)
         elif gp:
             if line.startswith('--- '):
                 gitpatches.append(gp)
                 gp = None
                 continue
             if line.startswith('rename from '):
                 gp.op = 'RENAME'
                 gp.oldpath = line[12:]
             elif line.startswith('rename to '):
                 gp.path = line[10:]
             elif line.startswith('copy from '):
                 gp.op = 'COPY'
                 gp.oldpath = line[10:]
             elif line.startswith('copy to '):
                 gp.path = line[8:]
             elif line.startswith('deleted file'):
                 gp.op = 'DELETE'
             elif line.startswith('new file mode '):
                 gp.op = 'ADD'
                 gp.setmode(int(line[-6:], 8))
             elif line.startswith('new mode '):
                 gp.setmode(int(line[-6:], 8))
             elif line.startswith('GIT binary patch'):
                 gp.binary = True
     if gp:
         gitpatches.append(gp)
 
     return gitpatches
 
 class linereader(object):
     # simple class to allow pushing lines back into the input stream
     def __init__(self, fp):
         self.fp = fp
         self.buf = []
 
     def push(self, line):
         if line is not None:
             self.buf.append(line)
 
     def readline(self):
         if self.buf:
             l = self.buf[0]
             del self.buf[0]
             return l
         return self.fp.readline()
 
     def __iter__(self):
         while True:
             l = self.readline()
             if not l:
                 break
             yield l
 
 class abstractbackend(object):
     def __init__(self, ui):
         self.ui = ui
 
     def getfile(self, fname):
         """Return target file data and flags as a (data, (islink,
         isexec)) tuple.
         """
         raise NotImplementedError
 
     def setfile(self, fname, data, mode, copysource):
         """Write data to target file fname and set its mode. mode is a
         (islink, isexec) tuple. If data is None, the file content should
         be left unchanged. If the file is modified after being copied,
         copysource is set to the original file name.
         """
         raise NotImplementedError
 
     def unlink(self, fname):
         """Unlink target file."""
         raise NotImplementedError
 
     def writerej(self, fname, failed, total, lines):
         """Write rejected lines for fname. total is the number of hunks
         which failed to apply and total the total number of hunks for this
         files.
         """
         pass
 
     def exists(self, fname):
         raise NotImplementedError
 
 class fsbackend(abstractbackend):
     def __init__(self, ui, basedir):
         super(fsbackend, self).__init__(ui)
         self.opener = scmutil.opener(basedir)
 
     def _join(self, f):
         return os.path.join(self.opener.base, f)
 
     def getfile(self, fname):
         path = self._join(fname)
         if os.path.islink(path):
             return (os.readlink(path), (True, False))
         isexec = False
         try:
             isexec = os.lstat(path).st_mode & 0100 != 0
         except OSError, e:
             if e.errno != errno.ENOENT:
                 raise
         return (self.opener.read(fname), (False, isexec))
 
     def setfile(self, fname, data, mode, copysource):
         islink, isexec = mode
         if data is None:
             util.setflags(self._join(fname), islink, isexec)
             return
         if islink:
             self.opener.symlink(data, fname)
         else:
             self.opener.write(fname, data)
             if isexec:
                 util.setflags(self._join(fname), False, True)
 
     def unlink(self, fname):
         try:
             util.unlinkpath(self._join(fname))
         except OSError, inst:
             if inst.errno != errno.ENOENT:
                 raise
 
     def writerej(self, fname, failed, total, lines):
         fname = fname + ".rej"
         self.ui.warn(
             _("%d out of %d hunks FAILED -- saving rejects to file %s\n") %
             (failed, total, fname))
         fp = self.opener(fname, 'w')
         fp.writelines(lines)
         fp.close()
 
     def exists(self, fname):
         return os.path.lexists(self._join(fname))
 
 class workingbackend(fsbackend):
     def __init__(self, ui, repo, similarity):
         super(workingbackend, self).__init__(ui, repo.root)
         self.repo = repo
         self.similarity = similarity
         self.removed = set()
         self.changed = set()
         self.copied = []
 
     def _checkknown(self, fname):
         if self.repo.dirstate[fname] == '?' and self.exists(fname):
             raise PatchError(_('cannot patch %s: file is not tracked') % fname)
 
     def setfile(self, fname, data, mode, copysource):
         self._checkknown(fname)
         super(workingbackend, self).setfile(fname, data, mode, copysource)
         if copysource is not None:
             self.copied.append((copysource, fname))
         self.changed.add(fname)
 
     def unlink(self, fname):
         self._checkknown(fname)
         super(workingbackend, self).unlink(fname)
         self.removed.add(fname)
         self.changed.add(fname)
 
     def close(self):
         wctx = self.repo[None]
         addremoved = set(self.changed)
         for src, dst in self.copied:
             scmutil.dirstatecopy(self.ui, self.repo, wctx, src, dst)
             addremoved.discard(src)
         if (not self.similarity) and self.removed:
             wctx.forget(sorted(self.removed))
         if addremoved:
             cwd = self.repo.getcwd()
             if cwd:
                 addremoved = [util.pathto(self.repo.root, cwd, f)
                               for f in addremoved]
             scmutil.addremove(self.repo, addremoved, similarity=self.similarity)
         return sorted(self.changed)
 
 class filestore(object):
     def __init__(self, maxsize=None):
         self.opener = None
         self.files = {}
         self.created = 0
         self.maxsize = maxsize
         if self.maxsize is None:
             self.maxsize = 4*(2**20)
         self.size = 0
         self.data = {}
 
     def setfile(self, fname, data, mode, copied=None):
         if self.maxsize < 0 or (len(data) + self.size) <= self.maxsize:
             self.data[fname] = (data, mode, copied)
             self.size += len(data)
         else:
             if self.opener is None:
                 root = tempfile.mkdtemp(prefix='hg-patch-')
                 self.opener = scmutil.opener(root)
             # Avoid filename issues with these simple names
             fn = str(self.created)
             self.opener.write(fn, data)
             self.created += 1
             self.files[fname] = (fn, mode, copied)
 
     def getfile(self, fname):
         if fname in self.data:
             return self.data[fname]
         if not self.opener or fname not in self.files:
             raise IOError()
         fn, mode, copied = self.files[fname]
         return self.opener.read(fn), mode, copied
 
     def close(self):
         if self.opener:
             shutil.rmtree(self.opener.base)
 
 class repobackend(abstractbackend):
     def __init__(self, ui, repo, ctx, store):
         super(repobackend, self).__init__(ui)
         self.repo = repo
         self.ctx = ctx
         self.store = store
         self.changed = set()
         self.removed = set()
         self.copied = {}
 
     def _checkknown(self, fname):
         if fname not in self.ctx:
             raise PatchError(_('cannot patch %s: file is not tracked') % fname)
 
     def getfile(self, fname):
         try:
             fctx = self.ctx[fname]
         except error.LookupError:
             raise IOError()
         flags = fctx.flags()
         return fctx.data(), ('l' in flags, 'x' in flags)
 
     def setfile(self, fname, data, mode, copysource):
         if copysource:
             self._checkknown(copysource)
         if data is None:
             data = self.ctx[fname].data()
         self.store.setfile(fname, data, mode, copysource)
         self.changed.add(fname)
         if copysource:
             self.copied[fname] = copysource
 
     def unlink(self, fname):
         self._checkknown(fname)
         self.removed.add(fname)
 
     def exists(self, fname):
         return fname in self.ctx
 
     def close(self):
         return self.changed | self.removed
 
 # @@ -start,len +start,len @@ or @@ -start +start @@ if len is 1
 unidesc = re.compile('@@ -(\d+)(,(\d+))? \+(\d+)(,(\d+))? @@')
 contextdesc = re.compile('(---|\*\*\*) (\d+)(,(\d+))? (---|\*\*\*)')
 eolmodes = ['strict', 'crlf', 'lf', 'auto']
 
 class patchfile(object):
     def __init__(self, ui, gp, backend, store, eolmode='strict'):
         self.fname = gp.path
         self.eolmode = eolmode
         self.eol = None
         self.backend = backend
         self.ui = ui
         self.lines = []
         self.exists = False
         self.missing = True
         self.mode = gp.mode
         self.copysource = gp.oldpath
         self.create = gp.op in ('ADD', 'COPY', 'RENAME')
         self.remove = gp.op == 'DELETE'
         try:
             if self.copysource is None:
                 data, mode = backend.getfile(self.fname)
                 self.exists = True
             else:
                 data, mode = store.getfile(self.copysource)[:2]
                 self.exists = backend.exists(self.fname)
             self.missing = False
             if data:
                 self.lines = mdiff.splitnewlines(data)
             if self.mode is None:
                 self.mode = mode
             if self.lines:
                 # Normalize line endings
                 if self.lines[0].endswith('\r\n'):
                     self.eol = '\r\n'
                 elif self.lines[0].endswith('\n'):
                     self.eol = '\n'
                 if eolmode != 'strict':
                     nlines = []
                     for l in self.lines:
                         if l.endswith('\r\n'):
                             l = l[:-2] + '\n'
                         nlines.append(l)
                     self.lines = nlines
         except IOError:
             if self.create:
                 self.missing = False
             if self.mode is None:
                 self.mode = (False, False)
         if self.missing:
              self.ui.warn(_("unable to find '%s' for patching\n") % self.fname)
 
         self.hash = {}
         self.dirty = 0
         self.offset = 0
         self.skew = 0
         self.rej = []
         self.fileprinted = False
         self.printfile(False)
         self.hunks = 0
 
     def writelines(self, fname, lines, mode):
         if self.eolmode == 'auto':
             eol = self.eol
         elif self.eolmode == 'crlf':
             eol = '\r\n'
         else:
             eol = '\n'
 
         if self.eolmode != 'strict' and eol and eol != '\n':
             rawlines = []
             for l in lines:
                 if l and l[-1] == '\n':
                     l = l[:-1] + eol
                 rawlines.append(l)
             lines = rawlines
 
         self.backend.setfile(fname, ''.join(lines), mode, self.copysource)
 
     def printfile(self, warn):
         if self.fileprinted:
             return
         if warn or self.ui.verbose:
             self.fileprinted = True
         s = _("patching file %s\n") % self.fname
         if warn:
             self.ui.warn(s)
         else:
             self.ui.note(s)
 
 
     def findlines(self, l, linenum):
         # looks through the hash and finds candidate lines.  The
         # result is a list of line numbers sorted based on distance
         # from linenum
 
         cand = self.hash.get(l, [])
         if len(cand) > 1:
             # resort our list of potentials forward then back.
             cand.sort(key=lambda x: abs(x - linenum))
         return cand
 
     def write_rej(self):
         # our rejects are a little different from patch(1).  This always
         # creates rejects in the same form as the original patch.  A file
         # header is inserted so that you can run the reject through patch again
         # without having to type the filename.
         if not self.rej:
             return
         base = os.path.basename(self.fname)
         lines = ["--- %s\n+++ %s\n" % (base, base)]
         for x in self.rej:
             for l in x.hunk:
                 lines.append(l)
                 if l[-1] != '\n':
                     lines.append("\n\ No newline at end of file\n")
         self.backend.writerej(self.fname, len(self.rej), self.hunks, lines)
 
     def apply(self, h):
         if not h.complete():
             raise PatchError(_("bad hunk #%d %s (%d %d %d %d)") %
                             (h.number, h.desc, len(h.a), h.lena, len(h.b),
                             h.lenb))
 
         self.hunks += 1
 
         if self.missing:
             self.rej.append(h)
             return -1
 
         if self.exists and self.create:
             if self.copysource:
                 self.ui.warn(_("cannot create %s: destination already "
                                "exists\n" % self.fname))
             else:
                 self.ui.warn(_("file %s already exists\n") % self.fname)
             self.rej.append(h)
             return -1
 
         if isinstance(h, binhunk):
             if self.remove:
                 self.backend.unlink(self.fname)
             else:
                 self.lines[:] = h.new()
                 self.offset += len(h.new())
                 self.dirty = True
             return 0
 
         horig = h
         if (self.eolmode in ('crlf', 'lf')
             or self.eolmode == 'auto' and self.eol):
             # If new eols are going to be normalized, then normalize
             # hunk data before patching. Otherwise, preserve input
             # line-endings.
             h = h.getnormalized()
 
         # fast case first, no offsets, no fuzz
         old = h.old()
-        # patch starts counting at 1 unless we are adding the file
-        if h.starta == 0:
-            start = 0
-        else:
-            start = h.starta + self.offset - 1
+        start = h.starta + self.offset
+        # zero length hunk ranges already have their start decremented
+        if h.lena:
+            start -= 1
         orig_start = start
         # if there's skew we want to emit the "(offset %d lines)" even
         # when the hunk cleanly applies at start + skew, so skip the
         # fast case code
         if self.skew == 0 and diffhelpers.testhunk(old, self.lines, start) == 0:
             if self.remove:
                 self.backend.unlink(self.fname)
             else:
                 self.lines[start : start + h.lena] = h.new()
                 self.offset += h.lenb - h.lena
                 self.dirty = True
             return 0
 
         # ok, we couldn't match the hunk. Lets look for offsets and fuzz it
         self.hash = {}
         for x, s in enumerate(self.lines):
             self.hash.setdefault(s, []).append(x)
         if h.hunk[-1][0] != ' ':
             # if the hunk tried to put something at the bottom of the file
             # override the start line and use eof here
             search_start = len(self.lines)
         else:
             search_start = orig_start + self.skew
 
         for fuzzlen in xrange(3):
             for toponly in [True, False]:
                 old = h.old(fuzzlen, toponly)
 
                 cand = self.findlines(old[0][1:], search_start)
                 for l in cand:
                     if diffhelpers.testhunk(old, self.lines, l) == 0:
                         newlines = h.new(fuzzlen, toponly)
                         self.lines[l : l + len(old)] = newlines
                         self.offset += len(newlines) - len(old)
                         self.skew = l - orig_start
                         self.dirty = True
                         offset = l - orig_start - fuzzlen
                         if fuzzlen:
                             msg = _("Hunk #%d succeeded at %d "
                                     "with fuzz %d "
                                     "(offset %d lines).\n")
                             self.printfile(True)
                             self.ui.warn(msg %
                                 (h.number, l + 1, fuzzlen, offset))
                         else:
                             msg = _("Hunk #%d succeeded at %d "
                                     "(offset %d lines).\n")
                             self.ui.note(msg % (h.number, l + 1, offset))
                         return fuzzlen
         self.printfile(True)
         self.ui.warn(_("Hunk #%d FAILED at %d\n") % (h.number, orig_start))
         self.rej.append(horig)
         return -1
 
     def close(self):
         if self.dirty:
             self.writelines(self.fname, self.lines, self.mode)
         self.write_rej()
         return len(self.rej)
 
 class hunk(object):
     def __init__(self, desc, num, lr, context):
         self.number = num
         self.desc = desc
         self.hunk = [desc]
         self.a = []
         self.b = []
         self.starta = self.lena = None
         self.startb = self.lenb = None
         if lr is not None:
             if context:
                 self.read_context_hunk(lr)
             else:
                 self.read_unified_hunk(lr)
 
     def getnormalized(self):
         """Return a copy with line endings normalized to LF."""
 
         def normalize(lines):
             nlines = []
             for line in lines:
                 if line.endswith('\r\n'):
                     line = line[:-2] + '\n'
                 nlines.append(line)
             return nlines
 
         # Dummy object, it is rebuilt manually
         nh = hunk(self.desc, self.number, None, None)
         nh.number = self.number
         nh.desc = self.desc
         nh.hunk = self.hunk
         nh.a = normalize(self.a)
         nh.b = normalize(self.b)
         nh.starta = self.starta
         nh.startb = self.startb
         nh.lena = self.lena
         nh.lenb = self.lenb
         return nh
 
     def read_unified_hunk(self, lr):
         m = unidesc.match(self.desc)
         if not m:
             raise PatchError(_("bad hunk #%d") % self.number)
         self.starta, foo, self.lena, self.startb, foo2, self.lenb = m.groups()
         if self.lena is None:
             self.lena = 1
         else:
             self.lena = int(self.lena)
         if self.lenb is None:
             self.lenb = 1
         else:
             self.lenb = int(self.lenb)
         self.starta = int(self.starta)
         self.startb = int(self.startb)
         diffhelpers.addlines(lr, self.hunk, self.lena, self.lenb, self.a, self.b)
         # if we hit eof before finishing out the hunk, the last line will
         # be zero length.  Lets try to fix it up.
         while len(self.hunk[-1]) == 0:
             del self.hunk[-1]
             del self.a[-1]
             del self.b[-1]
             self.lena -= 1
             self.lenb -= 1
         self._fixnewline(lr)
 
     def read_context_hunk(self, lr):
         self.desc = lr.readline()
         m = contextdesc.match(self.desc)
         if not m:
             raise PatchError(_("bad hunk #%d") % self.number)
         foo, self.starta, foo2, aend, foo3 = m.groups()
         self.starta = int(self.starta)
         if aend is None:
             aend = self.starta
         self.lena = int(aend) - self.starta
         if self.starta:
             self.lena += 1
         for x in xrange(self.lena):
             l = lr.readline()
             if l.startswith('---'):
                 # lines addition, old block is empty
                 lr.push(l)
                 break
             s = l[2:]
             if l.startswith('- ') or l.startswith('! '):
                 u = '-' + s
             elif l.startswith('  '):
                 u = ' ' + s
             else:
                 raise PatchError(_("bad hunk #%d old text line %d") %
                                  (self.number, x))
             self.a.append(u)
             self.hunk.append(u)
 
         l = lr.readline()
         if l.startswith('\ '):
             s = self.a[-1][:-1]
             self.a[-1] = s
             self.hunk[-1] = s
             l = lr.readline()
         m = contextdesc.match(l)
         if not m:
             raise PatchError(_("bad hunk #%d") % self.number)
         foo, self.startb, foo2, bend, foo3 = m.groups()
         self.startb = int(self.startb)
         if bend is None:
             bend = self.startb
         self.lenb = int(bend) - self.startb
         if self.startb:
             self.lenb += 1
         hunki = 1
         for x in xrange(self.lenb):
             l = lr.readline()
             if l.startswith('\ '):
                 # XXX: the only way to hit this is with an invalid line range.
                 # The no-eol marker is not counted in the line range, but I
                 # guess there are diff(1) out there which behave differently.
                 s = self.b[-1][:-1]
                 self.b[-1] = s
                 self.hunk[hunki - 1] = s
                 continue
             if not l:
                 # line deletions, new block is empty and we hit EOF
                 lr.push(l)
                 break
             s = l[2:]
             if l.startswith('+ ') or l.startswith('! '):
                 u = '+' + s
             elif l.startswith('  '):
                 u = ' ' + s
             elif len(self.b) == 0:
                 # line deletions, new block is empty
                 lr.push(l)
                 break
             else:
                 raise PatchError(_("bad hunk #%d old text line %d") %
                                  (self.number, x))
             self.b.append(s)
             while True:
                 if hunki >= len(self.hunk):
                     h = ""
                 else:
                     h = self.hunk[hunki]
                 hunki += 1
                 if h == u:
                     break
                 elif h.startswith('-'):
                     continue
                 else:
                     self.hunk.insert(hunki - 1, u)
                     break
 
         if not self.a:
             # this happens when lines were only added to the hunk
             for x in self.hunk:
                 if x.startswith('-') or x.startswith(' '):
                     self.a.append(x)
         if not self.b:
             # this happens when lines were only deleted from the hunk
             for x in self.hunk:
                 if x.startswith('+') or x.startswith(' '):
                     self.b.append(x[1:])
         # @@ -start,len +start,len @@
         self.desc = "@@ -%d,%d +%d,%d @@\n" % (self.starta, self.lena,
                                              self.startb, self.lenb)
         self.hunk[0] = self.desc
         self._fixnewline(lr)
 
     def _fixnewline(self, lr):
         l = lr.readline()
         if l.startswith('\ '):
             diffhelpers.fix_newline(self.hunk, self.a, self.b)
         else:
             lr.push(l)
 
     def complete(self):
         return len(self.a) == self.lena and len(self.b) == self.lenb
 
     def fuzzit(self, l, fuzz, toponly):
         # this removes context lines from the top and bottom of list 'l'.  It
         # checks the hunk to make sure only context lines are removed, and then
         # returns a new shortened list of lines.
         fuzz = min(fuzz, len(l)-1)
         if fuzz:
             top = 0
             bot = 0
             hlen = len(self.hunk)
             for x in xrange(hlen - 1):
                 # the hunk starts with the @@ line, so use x+1
                 if self.hunk[x + 1][0] == ' ':
                     top += 1
                 else:
                     break
             if not toponly:
                 for x in xrange(hlen - 1):
                     if self.hunk[hlen - bot - 1][0] == ' ':
                         bot += 1
                     else:
                         break
 
             # top and bot now count context in the hunk
             # adjust them if either one is short
             context = max(top, bot, 3)
             if bot < context:
                 bot = max(0, fuzz - (context - bot))
             else:
                 bot = min(fuzz, bot)
             if top < context:
                 top = max(0, fuzz - (context - top))
             else:
                 top = min(fuzz, top)
 
             return l[top:len(l)-bot]
         return l
 
     def old(self, fuzz=0, toponly=False):
         return self.fuzzit(self.a, fuzz, toponly)
 
     def new(self, fuzz=0, toponly=False):
         return self.fuzzit(self.b, fuzz, toponly)
 
 class binhunk(object):
     'A binary patch file. Only understands literals so far.'
     def __init__(self, lr):
         self.text = None
         self.hunk = ['GIT binary patch\n']
         self._read(lr)
 
     def complete(self):
         return self.text is not None
 
     def new(self):
         return [self.text]
 
     def _read(self, lr):
         line = lr.readline()
         self.hunk.append(line)
         while line and not line.startswith('literal '):
             line = lr.readline()
             self.hunk.append(line)
         if not line:
             raise PatchError(_('could not extract binary patch'))
         size = int(line[8:].rstrip())
         dec = []
         line = lr.readline()
         self.hunk.append(line)
         while len(line) > 1:
             l = line[0]
             if l <= 'Z' and l >= 'A':
                 l = ord(l) - ord('A') + 1
             else:
                 l = ord(l) - ord('a') + 27
             dec.append(base85.b85decode(line[1:-1])[:l])
             line = lr.readline()
             self.hunk.append(line)
         text = zlib.decompress(''.join(dec))
         if len(text) != size:
             raise PatchError(_('binary patch is %d bytes, not %d') %
                              len(text), size)
         self.text = text
 
 def parsefilename(str):
     # --- filename \t|space stuff
     s = str[4:].rstrip('\r\n')
     i = s.find('\t')
     if i < 0:
         i = s.find(' ')
         if i < 0:
             return s
     return s[:i]
 
 def pathstrip(path, strip):
     pathlen = len(path)
     i = 0
     if strip == 0:
         return '', path.rstrip()
     count = strip
     while count > 0:
         i = path.find('/', i)
         if i == -1:
             raise PatchError(_("unable to strip away %d of %d dirs from %s") %
                              (count, strip, path))
         i += 1
         # consume '//' in the path
         while i < pathlen - 1 and path[i] == '/':
             i += 1
         count -= 1
     return path[:i].lstrip(), path[i:].rstrip()
 
 def makepatchmeta(backend, afile_orig, bfile_orig, hunk, strip):
     nulla = afile_orig == "/dev/null"
     nullb = bfile_orig == "/dev/null"
     create = nulla and hunk.starta == 0 and hunk.lena == 0
     remove = nullb and hunk.startb == 0 and hunk.lenb == 0
     abase, afile = pathstrip(afile_orig, strip)
     gooda = not nulla and backend.exists(afile)
     bbase, bfile = pathstrip(bfile_orig, strip)
     if afile == bfile:
         goodb = gooda
     else:
         goodb = not nullb and backend.exists(bfile)
     missing = not goodb and not gooda and not create
 
     # some diff programs apparently produce patches where the afile is
     # not /dev/null, but afile starts with bfile
     abasedir = afile[:afile.rfind('/') + 1]
     bbasedir = bfile[:bfile.rfind('/') + 1]
     if (missing and abasedir == bbasedir and afile.startswith(bfile)
         and hunk.starta == 0 and hunk.lena == 0):
         create = True
         missing = False
 
     # If afile is "a/b/foo" and bfile is "a/b/foo.orig" we assume the
     # diff is between a file and its backup. In this case, the original
     # file should be patched (see original mpatch code).
     isbackup = (abase == bbase and bfile.startswith(afile))
     fname = None
     if not missing:
         if gooda and goodb:
             fname = isbackup and afile or bfile
         elif gooda:
             fname = afile
 
     if not fname:
         if not nullb:
             fname = isbackup and afile or bfile
         elif not nulla:
             fname = afile
         else:
             raise PatchError(_("undefined source and destination files"))
 
     gp = patchmeta(fname)
     if create:
         gp.op = 'ADD'
     elif remove:
         gp.op = 'DELETE'
     return gp
 
 def scangitpatch(lr, firstline):
     """
     Git patches can emit:
     - rename a to b
     - change b
     - copy a to c
     - change c
 
     We cannot apply this sequence as-is, the renamed 'a' could not be
     found for it would have been renamed already. And we cannot copy
     from 'b' instead because 'b' would have been changed already. So
     we scan the git patch for copy and rename commands so we can
     perform the copies ahead of time.
     """
     pos = 0
     try:
         pos = lr.fp.tell()
         fp = lr.fp
     except IOError:
         fp = cStringIO.StringIO(lr.fp.read())
     gitlr = linereader(fp)
     gitlr.push(firstline)
     gitpatches = readgitpatch(gitlr)
     fp.seek(pos)
     return gitpatches
 
 def iterhunks(fp):
     """Read a patch and yield the following events:
     - ("file", afile, bfile, firsthunk): select a new target file.
     - ("hunk", hunk): a new hunk is ready to be applied, follows a
     "file" event.
     - ("git", gitchanges): current diff is in git format, gitchanges
     maps filenames to gitpatch records. Unique event.
     """
     afile = ""
     bfile = ""
     state = None
     hunknum = 0
     emitfile = newfile = False
     gitpatches = None
 
     # our states
     BFILE = 1
     context = None
     lr = linereader(fp)
 
     while True:
         x = lr.readline()
         if not x:
             break
         if state == BFILE and (
             (not context and x[0] == '@')
             or (context is not False and x.startswith('***************'))
             or x.startswith('GIT binary patch')):
             gp = None
             if (gitpatches and
                 (gitpatches[-1][0] == afile or gitpatches[-1][1] == bfile)):
                 gp = gitpatches.pop()[2]
             if x.startswith('GIT binary patch'):
                 h = binhunk(lr)
             else:
                 if context is None and x.startswith('***************'):
                     context = True
                 h = hunk(x, hunknum + 1, lr, context)
             hunknum += 1
             if emitfile:
                 emitfile = False
                 yield 'file', (afile, bfile, h, gp and gp.copy() or None)
             yield 'hunk', h
         elif x.startswith('diff --git'):
             m = gitre.match(x)
             if not m:
                 continue
             if not gitpatches:
                 # scan whole input for git metadata
                 gitpatches = [('a/' + gp.path, 'b/' + gp.path, gp) for gp
                               in scangitpatch(lr, x)]
                 yield 'git', [g[2].copy() for g in gitpatches
                               if g[2].op in ('COPY', 'RENAME')]
                 gitpatches.reverse()
             afile = 'a/' + m.group(1)
             bfile = 'b/' + m.group(2)
             while afile != gitpatches[-1][0] and bfile != gitpatches[-1][1]:
                 gp = gitpatches.pop()[2]
                 yield 'file', ('a/' + gp.path, 'b/' + gp.path, None, gp.copy())
             gp = gitpatches[-1][2]
             # copy/rename + modify should modify target, not source
             if gp.op in ('COPY', 'DELETE', 'RENAME', 'ADD') or gp.mode:
                 afile = bfile
             newfile = True
         elif x.startswith('---'):
             # check for a unified diff
             l2 = lr.readline()
             if not l2.startswith('+++'):
                 lr.push(l2)
                 continue
             newfile = True
             context = False
             afile = parsefilename(x)
             bfile = parsefilename(l2)
         elif x.startswith('***'):
             # check for a context diff
             l2 = lr.readline()
             if not l2.startswith('---'):
                 lr.push(l2)
                 continue
             l3 = lr.readline()
             lr.push(l3)
             if not l3.startswith("***************"):
                 lr.push(l2)
                 continue
             newfile = True
             context = True
             afile = parsefilename(x)
             bfile = parsefilename(l2)
 
         if newfile:
             newfile = False
             emitfile = True
             state = BFILE
             hunknum = 0
 
     while gitpatches:
         gp = gitpatches.pop()[2]
         yield 'file', ('a/' + gp.path, 'b/' + gp.path, None, gp.copy())
 
 def applydiff(ui, fp, backend, store, strip=1, eolmode='strict'):
     """Reads a patch from fp and tries to apply it.
 
     Returns 0 for a clean patch, -1 if any rejects were found and 1 if
     there was any fuzz.
 
     If 'eolmode' is 'strict', the patch content and patched file are
     read in binary mode. Otherwise, line endings are ignored when
     patching then normalized according to 'eolmode'.
     """
     return _applydiff(ui, fp, patchfile, backend, store, strip=strip,
                       eolmode=eolmode)
 
 def _applydiff(ui, fp, patcher, backend, store, strip=1,
                eolmode='strict'):
 
     def pstrip(p):
         return pathstrip(p, strip - 1)[1]
 
     rejects = 0
     err = 0
     current_file = None
 
     for state, values in iterhunks(fp):
         if state == 'hunk':
             if not current_file:
                 continue
             ret = current_file.apply(values)
             if ret > 0:
                 err = 1
         elif state == 'file':
             if current_file:
                 rejects += current_file.close()
                 current_file = None
             afile, bfile, first_hunk, gp = values
             if gp:
                 path = pstrip(gp.path)
                 gp.path = pstrip(gp.path)
                 if gp.oldpath:
                     gp.oldpath = pstrip(gp.oldpath)
             else:
                 gp = makepatchmeta(backend, afile, bfile, first_hunk, strip)
             if gp.op == 'RENAME':
                 backend.unlink(gp.oldpath)
             if not first_hunk:
                 if gp.op == 'DELETE':
                     backend.unlink(gp.path)
                     continue
                 data, mode = None, None
                 if gp.op in ('RENAME', 'COPY'):
                     data, mode = store.getfile(gp.oldpath)[:2]
                 if gp.mode:
                     mode = gp.mode
                     if gp.op == 'ADD':
                         # Added files without content have no hunk and
                         # must be created
                         data = ''
                 if data or mode:
                     if (gp.op in ('ADD', 'RENAME', 'COPY')
                         and backend.exists(gp.path)):
                         raise PatchError(_("cannot create %s: destination "
                                            "already exists") % gp.path)
                     backend.setfile(gp.path, data, mode, gp.oldpath)
                 continue
             try:
                 current_file = patcher(ui, gp, backend, store,
                                        eolmode=eolmode)
             except PatchError, inst:
                 ui.warn(str(inst) + '\n')
                 current_file = None
                 rejects += 1
                 continue
         elif state == 'git':
             for gp in values:
                 path = pstrip(gp.oldpath)
                 data, mode = backend.getfile(path)
                 store.setfile(path, data, mode)
         else:
             raise util.Abort(_('unsupported parser state: %s') % state)
 
     if current_file:
         rejects += current_file.close()
 
     if rejects:
         return -1
     return err
 
 def _externalpatch(ui, repo, patcher, patchname, strip, files,
                    similarity):
     """use <patcher> to apply <patchname> to the working directory.
     returns whether patch was applied with fuzz factor."""
 
     fuzz = False
     args = []
     cwd = repo.root
     if cwd:
         args.append('-d %s' % util.shellquote(cwd))
     fp = util.popen('%s %s -p%d < %s' % (patcher, ' '.join(args), strip,
                                        util.shellquote(patchname)))
     try:
         for line in fp:
             line = line.rstrip()
             ui.note(line + '\n')
             if line.startswith('patching file '):
                 pf = util.parsepatchoutput(line)
                 printed_file = False
                 files.add(pf)
             elif line.find('with fuzz') >= 0:
                 fuzz = True
                 if not printed_file:
                     ui.warn(pf + '\n')
                     printed_file = True
                 ui.warn(line + '\n')
             elif line.find('saving rejects to file') >= 0:
                 ui.warn(line + '\n')
             elif line.find('FAILED') >= 0:
                 if not printed_file:
                     ui.warn(pf + '\n')
                     printed_file = True
                 ui.warn(line + '\n')
     finally:
         if files:
             cfiles = list(files)
             cwd = repo.getcwd()
             if cwd:
                 cfiles = [util.pathto(repo.root, cwd, f)
                           for f in cfiles]
             scmutil.addremove(repo, cfiles, similarity=similarity)
     code = fp.close()
     if code:
         raise PatchError(_("patch command failed: %s") %
                          util.explainexit(code)[0])
     return fuzz
 
 def patchbackend(ui, backend, patchobj, strip, files=None, eolmode='strict'):
     if files is None:
         files = set()
     if eolmode is None:
         eolmode = ui.config('patch', 'eol', 'strict')
     if eolmode.lower() not in eolmodes:
         raise util.Abort(_('unsupported line endings type: %s') % eolmode)
     eolmode = eolmode.lower()
 
     store = filestore()
     try:
         fp = open(patchobj, 'rb')
     except TypeError:
         fp = patchobj
     try:
         ret = applydiff(ui, fp, backend, store, strip=strip,
                         eolmode=eolmode)
     finally:
         if fp != patchobj:
             fp.close()
         files.update(backend.close())
         store.close()
     if ret < 0:
         raise PatchError(_('patch failed to apply'))
     return ret > 0
 
 def internalpatch(ui, repo, patchobj, strip, files=None, eolmode='strict',
                   similarity=0):
     """use builtin patch to apply <patchobj> to the working directory.
     returns whether patch was applied with fuzz factor."""
     backend = workingbackend(ui, repo, similarity)
     return patchbackend(ui, backend, patchobj, strip, files, eolmode)
 
 def patchrepo(ui, repo, ctx, store, patchobj, strip, files=None,
               eolmode='strict'):
     backend = repobackend(ui, repo, ctx, store)
     return patchbackend(ui, backend, patchobj, strip, files, eolmode)
 
 def makememctx(repo, parents, text, user, date, branch, files, store,
                editor=None):
     def getfilectx(repo, memctx, path):
         data, (islink, isexec), copied = store.getfile(path)
         return context.memfilectx(path, data, islink=islink, isexec=isexec,
                                   copied=copied)
     extra = {}
     if branch:
         extra['branch'] = encoding.fromlocal(branch)
     ctx =  context.memctx(repo, parents, text, files, getfilectx, user,
                           date, extra)
     if editor:
         ctx._text = editor(repo, ctx, [])
     return ctx
 
 def patch(ui, repo, patchname, strip=1, files=None, eolmode='strict',
           similarity=0):
     """Apply <patchname> to the working directory.
 
     'eolmode' specifies how end of lines should be handled. It can be:
     - 'strict': inputs are read in binary mode, EOLs are preserved
     - 'crlf': EOLs are ignored when patching and reset to CRLF
     - 'lf': EOLs are ignored when patching and reset to LF
     - None: get it from user settings, default to 'strict'
     'eolmode' is ignored when using an external patcher program.
 
     Returns whether patch was applied with fuzz factor.
     """
     patcher = ui.config('ui', 'patch')
     if files is None:
         files = set()
     try:
         if patcher:
             return _externalpatch(ui, repo, patcher, patchname, strip,
                                   files, similarity)
         return internalpatch(ui, repo, patchname, strip, files, eolmode,
                              similarity)
     except PatchError, err:
         raise util.Abort(str(err))
 
 def changedfiles(ui, repo, patchpath, strip=1):
     backend = fsbackend(ui, repo.root)
     fp = open(patchpath, 'rb')
     try:
         changed = set()
         for state, values in iterhunks(fp):
             if state == 'file':
                 afile, bfile, first_hunk, gp = values
                 if gp:
                     gp.path = pathstrip(gp.path, strip - 1)[1]
                     if gp.oldpath:
                         gp.oldpath = pathstrip(gp.oldpath, strip - 1)[1]
                 else:
                     gp = makepatchmeta(backend, afile, bfile, first_hunk, strip)
                 changed.add(gp.path)
                 if gp.op == 'RENAME':
                     changed.add(gp.oldpath)
             elif state not in ('hunk', 'git'):
                 raise util.Abort(_('unsupported parser state: %s') % state)
         return changed
     finally:
         fp.close()
 
 def b85diff(to, tn):
     '''print base85-encoded binary diff'''
     def gitindex(text):
         if not text:
             return hex(nullid)
         l = len(text)
         s = util.sha1('blob %d\0' % l)
         s.update(text)
         return s.hexdigest()
 
     def fmtline(line):
         l = len(line)
         if l <= 26:
             l = chr(ord('A') + l - 1)
         else:
             l = chr(l - 26 + ord('a') - 1)
         return '%c%s\n' % (l, base85.b85encode(line, True))
 
     def chunk(text, csize=52):
         l = len(text)
         i = 0
         while i < l:
             yield text[i:i + csize]
             i += csize
 
     tohash = gitindex(to)
     tnhash = gitindex(tn)
     if tohash == tnhash:
         return ""
 
     # TODO: deltas
     ret = ['index %s..%s\nGIT binary patch\nliteral %s\n' %
            (tohash, tnhash, len(tn))]
     for l in chunk(zlib.compress(tn)):
         ret.append(fmtline(l))
     ret.append('\n')
     return ''.join(ret)
 
 class GitDiffRequired(Exception):
     pass
 
 def diffopts(ui, opts=None, untrusted=False):
     def get(key, name=None, getter=ui.configbool):
         return ((opts and opts.get(key)) or
                 getter('diff', name or key, None, untrusted=untrusted))
     return mdiff.diffopts(
         text=opts and opts.get('text'),
         git=get('git'),
         nodates=get('nodates'),
         showfunc=get('show_function', 'showfunc'),
         ignorews=get('ignore_all_space', 'ignorews'),
         ignorewsamount=get('ignore_space_change', 'ignorewsamount'),
         ignoreblanklines=get('ignore_blank_lines', 'ignoreblanklines'),
         context=get('unified', getter=ui.config))
 
 def diff(repo, node1=None, node2=None, match=None, changes=None, opts=None,
          losedatafn=None, prefix=''):
     '''yields diff of changes to files between two nodes, or node and
     working directory.
 
     if node1 is None, use first dirstate parent instead.
     if node2 is None, compare node1 with working directory.
 
     losedatafn(**kwarg) is a callable run when opts.upgrade=True and
     every time some change cannot be represented with the current
     patch format. Return False to upgrade to git patch format, True to
     accept the loss or raise an exception to abort the diff. It is
     called with the name of current file being diffed as 'fn'. If set
     to None, patches will always be upgraded to git format when
     necessary.
 
     prefix is a filename prefix that is prepended to all filenames on
     display (used for subrepos).
     '''
 
     if opts is None:
         opts = mdiff.defaultopts
 
     if not node1 and not node2:
         node1 = repo.dirstate.p1()
 
     def lrugetfilectx():
         cache = {}
         order = []
         def getfilectx(f, ctx):
             fctx = ctx.filectx(f, filelog=cache.get(f))
             if f not in cache:
                 if len(cache) > 20:
                     del cache[order.pop(0)]
                 cache[f] = fctx.filelog()
             else:
                 order.remove(f)
             order.append(f)
             return fctx
         return getfilectx
     getfilectx = lrugetfilectx()
 
     ctx1 = repo[node1]
     ctx2 = repo[node2]
 
     if not changes:
         changes = repo.status(ctx1, ctx2, match=match)
     modified, added, removed = changes[:3]
 
     if not modified and not added and not removed:
         return []
 
     revs = None
     if not repo.ui.quiet:
         hexfunc = repo.ui.debugflag and hex or short
         revs = [hexfunc(node) for node in [node1, node2] if node]
 
     copy = {}
     if opts.git or opts.upgrade:
         copy = copies.copies(repo, ctx1, ctx2, repo[nullid])[0]
 
     difffn = lambda opts, losedata: trydiff(repo, revs, ctx1, ctx2,
                  modified, added, removed, copy, getfilectx, opts, losedata, prefix)
     if opts.upgrade and not opts.git:
         try:
             def losedata(fn):
                 if not losedatafn or not losedatafn(fn=fn):
                     raise GitDiffRequired()
             # Buffer the whole output until we are sure it can be generated
             return list(difffn(opts.copy(git=False), losedata))
         except GitDiffRequired:
             return difffn(opts.copy(git=True), None)
     else:
         return difffn(opts, None)
 
 def difflabel(func, *args, **kw):
     '''yields 2-tuples of (output, label) based on the output of func()'''
     headprefixes = [('diff', 'diff.diffline'),
                     ('copy', 'diff.extended'),
                     ('rename', 'diff.extended'),
                     ('old', 'diff.extended'),
                     ('new', 'diff.extended'),
                     ('deleted', 'diff.extended'),
                     ('---', 'diff.file_a'),
                     ('+++', 'diff.file_b')]
     textprefixes = [('@', 'diff.hunk'),
                     ('-', 'diff.deleted'),
                     ('+', 'diff.inserted')]
     head = False
     for chunk in func(*args, **kw):
         lines = chunk.split('\n')
         for i, line in enumerate(lines):
             if i != 0:
                 yield ('\n', '')
             if head:
                 if line.startswith('@'):
                     head = False
             else:
                 if line and not line[0] in ' +-@':
                     head = True
             stripline = line
             if not head and line and line[0] in '+-':
                 # highlight trailing whitespace, but only in changed lines
                 stripline = line.rstrip()
             prefixes = textprefixes
             if head:
                 prefixes = headprefixes
             for prefix, label in prefixes:
                 if stripline.startswith(prefix):
                     yield (stripline, label)
                     break
             else:
                 yield (line, '')
             if line != stripline:
                 yield (line[len(stripline):], 'diff.trailingwhitespace')
 
 def diffui(*args, **kw):
     '''like diff(), but yields 2-tuples of (output, label) for ui.write()'''
     return difflabel(diff, *args, **kw)
 
 
 def _addmodehdr(header, omode, nmode):
     if omode != nmode:
         header.append('old mode %s\n' % omode)
         header.append('new mode %s\n' % nmode)
 
 def trydiff(repo, revs, ctx1, ctx2, modified, added, removed,
             copy, getfilectx, opts, losedatafn, prefix):
 
     def join(f):
         return os.path.join(prefix, f)
 
     date1 = util.datestr(ctx1.date())
     man1 = ctx1.manifest()
 
     gone = set()
     gitmode = {'l': '120000', 'x': '100755', '': '100644'}
 
     copyto = dict([(v, k) for k, v in copy.items()])
 
     if opts.git:
         revs = None
 
     for f in sorted(modified + added + removed):
         to = None
         tn = None
         dodiff = True
         header = []
         if f in man1:
             to = getfilectx(f, ctx1).data()
         if f not in removed:
             tn = getfilectx(f, ctx2).data()
         a, b = f, f
         if opts.git or losedatafn:
             if f in added:
                 mode = gitmode[ctx2.flags(f)]
                 if f in copy or f in copyto:
                     if opts.git:
                         if f in copy:
                             a = copy[f]
                         else:
                             a = copyto[f]
                         omode = gitmode[man1.flags(a)]
                         _addmodehdr(header, omode, mode)
                         if a in removed and a not in gone:
                             op = 'rename'
                             gone.add(a)
                         else:
                             op = 'copy'
                         header.append('%s from %s\n' % (op, join(a)))
                         header.append('%s to %s\n' % (op, join(f)))
                         to = getfilectx(a, ctx1).data()
                     else:
                         losedatafn(f)
                 else:
                     if opts.git:
                         header.append('new file mode %s\n' % mode)
                     elif ctx2.flags(f):
                         losedatafn(f)
                 # In theory, if tn was copied or renamed we should check
                 # if the source is binary too but the copy record already
                 # forces git mode.
                 if util.binary(tn):
                     if opts.git:
                         dodiff = 'binary'
                     else:
                         losedatafn(f)
                 if not opts.git and not tn:
                     # regular diffs cannot represent new empty file
                     losedatafn(f)
             elif f in removed:
                 if opts.git:
                     # have we already reported a copy above?
                     if ((f in copy and copy[f] in added
                          and copyto[copy[f]] == f) or
                         (f in copyto and copyto[f] in added
                          and copy[copyto[f]] == f)):
                         dodiff = False
                     else:
                         header.append('deleted file mode %s\n' %
                                       gitmode[man1.flags(f)])
                 elif not to or util.binary(to):
                     # regular diffs cannot represent empty file deletion
                     losedatafn(f)
             else:
                 oflag = man1.flags(f)
                 nflag = ctx2.flags(f)
                 binary = util.binary(to) or util.binary(tn)
                 if opts.git:
                     _addmodehdr(header, gitmode[oflag], gitmode[nflag])
                     if binary:
                         dodiff = 'binary'
                 elif binary or nflag != oflag:
                     losedatafn(f)
             if opts.git:
                 header.insert(0, mdiff.diffline(revs, join(a), join(b), opts))
 
         if dodiff:
             if dodiff == 'binary':
                 text = b85diff(to, tn)
             else:
                 text = mdiff.unidiff(to, date1,
                                     # ctx2 date may be dynamic
                                     tn, util.datestr(ctx2.date()),
                                     join(a), join(b), revs, opts=opts)
             if header and (text or len(header) > 1):
                 yield ''.join(header)
             if text:
                 yield text
 
 def diffstatsum(stats):
     maxfile, maxtotal, addtotal, removetotal, binary = 0, 0, 0, 0, False
     for f, a, r, b in stats:
         maxfile = max(maxfile, encoding.colwidth(f))
         maxtotal = max(maxtotal, a + r)
         addtotal += a
         removetotal += r
         binary = binary or b
 
     return maxfile, maxtotal, addtotal, removetotal, binary
 
 def diffstatdata(lines):
     diffre = re.compile('^diff .*-r [a-z0-9]+\s(.*)$')
 
     results = []
     filename, adds, removes, isbinary = None, 0, 0, False
 
     def addresult():
         if filename:
             results.append((filename, adds, removes, isbinary))
 
     for line in lines:
         if line.startswith('diff'):
             addresult()
             # set numbers to 0 anyway when starting new file
             adds, removes, isbinary = 0, 0, False
             if line.startswith('diff --git'):
                 filename = gitre.search(line).group(1)
             elif line.startswith('diff -r'):
                 # format: "diff -r ... -r ... filename"
                 filename = diffre.search(line).group(1)
         elif line.startswith('+') and not line.startswith('+++'):
             adds += 1
         elif line.startswith('-') and not line.startswith('---'):
             removes += 1
         elif (line.startswith('GIT binary patch') or
               line.startswith('Binary file')):
             isbinary = True
     addresult()
     return results
 
 def diffstat(lines, width=80, git=False):
     output = []
     stats = diffstatdata(lines)
     maxname, maxtotal, totaladds, totalremoves, hasbinary = diffstatsum(stats)
 
     countwidth = len(str(maxtotal))
     if hasbinary and countwidth < 3:
         countwidth = 3
     graphwidth = width - countwidth - maxname - 6
     if graphwidth < 10:
         graphwidth = 10
 
     def scale(i):
         if maxtotal <= graphwidth:
             return i
         # If diffstat runs out of room it doesn't print anything,
         # which isn't very useful, so always print at least one + or -
         # if there were at least some changes.
         return max(i * graphwidth // maxtotal, int(bool(i)))
 
     for filename, adds, removes, isbinary in stats:
         if isbinary:
             count = 'Bin'
         else:
             count = adds + removes
         pluses = '+' * scale(adds)
         minuses = '-' * scale(removes)
         output.append(' %s%s |  %*s %s%s\n' %
                       (filename, ' ' * (maxname - encoding.colwidth(filename)),
                        countwidth, count, pluses, minuses))
 
     if stats:
         output.append(_(' %d files changed, %d insertions(+), %d deletions(-)\n')
                       % (len(stats), totaladds, totalremoves))
 
     return ''.join(output)
 
 def diffstatui(*args, **kw):
     '''like diffstat(), but yields 2-tuples of (output, label) for
     ui.write()
     '''
 
     for line in diffstat(*args, **kw).splitlines():
         if line and line[-1] in '+-':
             name, graph = line.rsplit(' ', 1)
             yield (name + ' ', '')
             m = re.search(r'\++', graph)
             if m:
                 yield (m.group(0), 'diffstat.inserted')
             m = re.search(r'-+', graph)
             if m:
                 yield (m.group(0), 'diffstat.deleted')
         else:
             yield (line, '')
         yield ('\n', '')

tests/test-diff-unified.t

@@ -1,111 +1,153
   $ hg init repo
   $ cd repo
   $ cat > a <<EOF
   > c
   > c
   > a
   > a
   > b
   > a
   > a
   > c
   > c
   > EOF
   $ hg ci -Am adda
   adding a
 
   $ cat > a <<EOF
   > c
   > c
   > a
   > a
   > dd
   > a
   > a
   > c
   > c
   > EOF
 
 default context
 
   $ hg diff --nodates
   diff -r cf9f4ba66af2 a
   --- a/a
   +++ b/a
   @@ -2,7 +2,7 @@
    c
    a
    a
   -b
   +dd
    a
    a
    c
 
 invalid --unified
 
   $ hg diff --nodates -U foo
   abort: diff context lines count must be an integer, not 'foo'
   [255]
 
 
   $ hg diff --nodates -U 2
   diff -r cf9f4ba66af2 a
   --- a/a
   +++ b/a
   @@ -3,5 +3,5 @@
    a
    a
   -b
   +dd
    a
    a
 
   $ hg --config diff.unified=2 diff --nodates
   diff -r cf9f4ba66af2 a
   --- a/a
   +++ b/a
   @@ -3,5 +3,5 @@
    a
    a
   -b
   +dd
    a
    a
 
   $ hg diff --nodates -U 1
   diff -r cf9f4ba66af2 a
   --- a/a
   +++ b/a
   @@ -4,3 +4,3 @@
    a
   -b
   +dd
    a
 
 invalid diff.unified
 
   $ hg --config diff.unified=foo diff --nodates
   abort: diff context lines count must be an integer, not 'foo'
   [255]
 
-test off-by-one error with diff -p
+0 lines of context hunk header matches gnu diff hunk header
+
+  $ hg init diffzero
+  $ cd diffzero
+  $ cat > f1 << EOF
+  > c2
+  > c4
+  > c5
+  > EOF
+  $ hg commit -Am0
+  adding f1
+
+  $ cat > f2 << EOF
+  > c1
+  > c2
+  > c3
+  > c4
+  > EOF
+  $ diff -U0 f1 f2
+  --- f1	* (glob)
+  +++ f2	* (glob)
+  @@ -0,0 +1 @@
+  +c1
+  @@ -1,0 +3 @@
+  +c3
+  @@ -3 +4,0 @@
+  -c5
+  [1]
 
-  $ hg init diffp
-  $ cd diffp
-  $ echo a > a
-  $ hg ci -Ama
-  adding a
-  $ rm a
-  $ echo b > a
-  $ echo a >> a
-  $ echo c >> a
-  $ hg diff -U0 -p --nodates
-  diff -r cb9a9f314b8b a
-  --- a/a
-  +++ b/a
-  @@ -1,0 +1,1 @@
-  +b
-  @@ -2,0 +3,1 @@ a
-  +c
+  $ mv f2 f1
+  $ hg diff -U0 --nodates
+  diff -r 55d8ff78db23 f1
+  --- a/f1
+  +++ b/f1
+  @@ -0,0 +1,1 @@
+  +c1
+  @@ -1,0 +3,1 @@
+  +c3
+  @@ -3,1 +4,0 @@
+  -c5
 
+  $ hg diff -U0 --nodates --git
+  diff --git a/f1 b/f1
+  --- a/f1
+  +++ b/f1
+  @@ -0,0 +1,1 @@
+  +c1
+  @@ -1,0 +3,1 @@
+  +c3
+  @@ -3,1 +4,0 @@
+  -c5
+
+  $ hg diff -U0 --nodates -p
+  diff -r 55d8ff78db23 f1
+  --- a/f1
+  +++ b/f1
+  @@ -0,0 +1,1 @@
+  +c1
+  @@ -1,0 +3,1 @@ c2
+  +c3
+  @@ -3,1 +4,0 @@ c4
+  -c5

tests/test-import.t

@@ -1,960 +1,996
   $ hg init a
   $ mkdir a/d1
   $ mkdir a/d1/d2
   $ echo line 1 > a/a
   $ echo line 1 > a/d1/d2/a
   $ hg --cwd a ci -Ama
   adding a
   adding d1/d2/a
 
   $ echo line 2 >> a/a
   $ hg --cwd a ci -u someone -d '1 0' -m'second change'
 
 import with no args:
 
   $ hg --cwd a import
   abort: need at least one patch to import
   [255]
 
 generate patches for the test
 
   $ hg --cwd a export tip > exported-tip.patch
   $ hg --cwd a diff -r0:1 > diffed-tip.patch
 
 
 import exported patch
 
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg --cwd b import ../exported-tip.patch
   applying ../exported-tip.patch
 
 message and committer should be same
 
   $ hg --cwd b tip
   changeset:   1:1d4bd90af0e4
   tag:         tip
   user:        someone
   date:        Thu Jan 01 00:00:01 1970 +0000
   summary:     second change
   
   $ rm -r b
 
 
 import exported patch with external patcher
 
   $ cat > dummypatch.py <<EOF
   > print 'patching file a'
   > file('a', 'wb').write('line2\n')
   > EOF
   $ chmod +x dummypatch.py
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg --config ui.patch='python ../dummypatch.py' --cwd b import ../exported-tip.patch
   applying ../exported-tip.patch
   $ cat b/a
   line2
   $ rm -r b
 
 
 import of plain diff should fail without message
 
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg --cwd b import ../diffed-tip.patch
   applying ../diffed-tip.patch
   abort: empty commit message
   [255]
   $ rm -r b
 
 
 import of plain diff should be ok with message
 
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg --cwd b import -mpatch ../diffed-tip.patch
   applying ../diffed-tip.patch
   $ rm -r b
 
 
 import of plain diff with specific date and user
 
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg --cwd b import -mpatch -d '1 0' -u '[e-mail]' ../diffed-tip.patch
   applying ../diffed-tip.patch
   $ hg -R b tip -pv
   changeset:   1:ca68f19f3a40
   tag:         tip
   user:        [e-mail]
   date:        Thu Jan 01 00:00:01 1970 +0000
   files:       a
   description:
   patch
   
   
   diff -r 80971e65b431 -r ca68f19f3a40 a
   --- a/a	Thu Jan 01 00:00:00 1970 +0000
   +++ b/a	Thu Jan 01 00:00:01 1970 +0000
   @@ -1,1 +1,2 @@
    line 1
   +line 2
   
   $ rm -r b
 
 
 import of plain diff should be ok with --no-commit
 
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg --cwd b import --no-commit ../diffed-tip.patch
   applying ../diffed-tip.patch
   $ hg --cwd b diff --nodates
   diff -r 80971e65b431 a
   --- a/a
   +++ b/a
   @@ -1,1 +1,2 @@
    line 1
   +line 2
   $ rm -r b
 
 
 import of malformed plain diff should fail
 
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ sed 's/1,1/foo/' < diffed-tip.patch > broken.patch
   $ hg --cwd b import -mpatch ../broken.patch
   applying ../broken.patch
   abort: bad hunk #1
   [255]
   $ rm -r b
 
 
 hg -R repo import
 put the clone in a subdir - having a directory named "a"
 used to hide a bug.
 
   $ mkdir dir
   $ hg clone -r0 a dir/b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ cd dir
   $ hg -R b import ../exported-tip.patch
   applying ../exported-tip.patch
   $ cd ..
   $ rm -r dir
 
 
 import from stdin
 
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg --cwd b import - < exported-tip.patch
   applying patch from stdin
   $ rm -r b
 
 
 import two patches in one stream
 
   $ hg init b
   $ hg --cwd a export 0:tip | hg --cwd b import -
   applying patch from stdin
   $ hg --cwd a id
   1d4bd90af0e4 tip
   $ hg --cwd b id
   1d4bd90af0e4 tip
   $ rm -r b
 
 
 override commit message
 
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg --cwd b import -m 'override' - < exported-tip.patch
   applying patch from stdin
   $ hg --cwd b tip | grep override
   summary:     override
   $ rm -r b
 
   $ cat > mkmsg.py <<EOF
   > import email.Message, sys
   > msg = email.Message.Message()
   > patch = open(sys.argv[1], 'rb').read()
   > msg.set_payload('email commit message\n' + patch)
   > msg['Subject'] = 'email patch'
   > msg['From'] = 'email patcher'
   > sys.stdout.write(msg.as_string())
   > EOF
 
 
 plain diff in email, subject, message body
 
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ python mkmsg.py diffed-tip.patch > msg.patch
   $ hg --cwd b import ../msg.patch
   applying ../msg.patch
   $ hg --cwd b tip | grep email
   user:        email patcher
   summary:     email patch
   $ rm -r b
 
 
 plain diff in email, no subject, message body
 
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ grep -v '^Subject:' msg.patch | hg --cwd b import -
   applying patch from stdin
   $ rm -r b
 
 
 plain diff in email, subject, no message body
 
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ grep -v '^email ' msg.patch | hg --cwd b import -
   applying patch from stdin
   $ rm -r b
 
 
 plain diff in email, no subject, no message body, should fail
 
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ egrep -v '^(Subject|email)' msg.patch | hg --cwd b import -
   applying patch from stdin
   abort: empty commit message
   [255]
   $ rm -r b
 
 
 hg export in email, should use patch header
 
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ python mkmsg.py exported-tip.patch | hg --cwd b import -
   applying patch from stdin
   $ hg --cwd b tip | grep second
   summary:     second change
   $ rm -r b
 
 
 subject: duplicate detection, removal of [PATCH]
 The '---' tests the gitsendmail handling without proper mail headers
 
   $ cat > mkmsg2.py <<EOF
   > import email.Message, sys
   > msg = email.Message.Message()
   > patch = open(sys.argv[1], 'rb').read()
   > msg.set_payload('email patch\n\nnext line\n---\n' + patch)
   > msg['Subject'] = '[PATCH] email patch'
   > msg['From'] = 'email patcher'
   > sys.stdout.write(msg.as_string())
   > EOF
 
 
 plain diff in email, [PATCH] subject, message body with subject
 
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ python mkmsg2.py diffed-tip.patch | hg --cwd b import -
   applying patch from stdin
   $ hg --cwd b tip --template '{desc}\n'
   email patch
   
   next line
   ---
   $ rm -r b
 
 
 Issue963: Parent of working dir incorrect after import of multiple
 patches and rollback
 
 We weren't backing up the correct dirstate file when importing many
 patches: import patch1 patch2; rollback
 
   $ echo line 3 >> a/a
   $ hg --cwd a ci -m'third change'
   $ hg --cwd a export -o '../patch%R' 1 2
   $ hg clone -qr0 a b
   $ hg --cwd b parents --template 'parent: {rev}\n'
   parent: 0
   $ hg --cwd b import -v ../patch1 ../patch2
   applying ../patch1
   patching file a
   a
   created 1d4bd90af0e4
   applying ../patch2
   patching file a
   a
   created 6d019af21222
   $ hg --cwd b rollback
   repository tip rolled back to revision 0 (undo import)
   working directory now based on revision 0
   $ hg --cwd b parents --template 'parent: {rev}\n'
   parent: 0
   $ rm -r b
 
 
 importing a patch in a subdirectory failed at the commit stage
 
   $ echo line 2 >> a/d1/d2/a
   $ hg --cwd a ci -u someoneelse -d '1 0' -m'subdir change'
 
 hg import in a subdirectory
 
   $ hg clone -r0 a b
   adding changesets
   adding manifests
   adding file changes
   added 1 changesets with 2 changes to 2 files
   updating to branch default
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg --cwd a export tip > tmp
   $ sed -e 's/d1\/d2\///' < tmp > subdir-tip.patch
   $ dir=`pwd`
   $ cd b/d1/d2 2>&1 > /dev/null
   $ hg import  ../../../subdir-tip.patch
   applying ../../../subdir-tip.patch
   $ cd "$dir"
 
 message should be 'subdir change'
 committer should be 'someoneelse'
 
   $ hg --cwd b tip
   changeset:   1:3577f5aea227
   tag:         tip
   user:        someoneelse
   date:        Thu Jan 01 00:00:01 1970 +0000
   summary:     subdir change
   
 
 should be empty
 
   $ hg --cwd b status
 
 
 Test fuzziness (ambiguous patch location, fuzz=2)
 
   $ hg init fuzzy
   $ cd fuzzy
   $ echo line1 > a
   $ echo line0 >> a
   $ echo line3 >> a
   $ hg ci -Am adda
   adding a
   $ echo line1 > a
   $ echo line2 >> a
   $ echo line0 >> a
   $ echo line3 >> a
   $ hg ci -m change a
   $ hg export tip > fuzzy-tip.patch
   $ hg up -C 0
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ echo line1 > a
   $ echo line0 >> a
   $ echo line1 >> a
   $ echo line0 >> a
   $ hg ci -m brancha
   created new head
   $ hg import --no-commit -v fuzzy-tip.patch
   applying fuzzy-tip.patch
   patching file a
   Hunk #1 succeeded at 1 with fuzz 2 (offset -2 lines).
   applied to working directory
   $ hg revert -a
   reverting a
 
 
 import with --no-commit should have written .hg/last-message.txt
 
   $ cat .hg/last-message.txt
   change (no-eol)
 
 
 test fuzziness with eol=auto
 
   $ hg --config patch.eol=auto import --no-commit -v fuzzy-tip.patch
   applying fuzzy-tip.patch
   patching file a
   Hunk #1 succeeded at 1 with fuzz 2 (offset -2 lines).
   applied to working directory
   $ cd ..
 
 
 Test hunk touching empty files (issue906)
 
   $ hg init empty
   $ cd empty
   $ touch a
   $ touch b1
   $ touch c1
   $ echo d > d
   $ hg ci -Am init
   adding a
   adding b1
   adding c1
   adding d
   $ echo a > a
   $ echo b > b1
   $ hg mv b1 b2
   $ echo c > c1
   $ hg copy c1 c2
   $ rm d
   $ touch d
   $ hg diff --git
   diff --git a/a b/a
   --- a/a
   +++ b/a
   @@ -0,0 +1,1 @@
   +a
   diff --git a/b1 b/b2
   rename from b1
   rename to b2
   --- a/b1
   +++ b/b2
   @@ -0,0 +1,1 @@
   +b
   diff --git a/c1 b/c1
   --- a/c1
   +++ b/c1
   @@ -0,0 +1,1 @@
   +c
   diff --git a/c1 b/c2
   copy from c1
   copy to c2
   --- a/c1
   +++ b/c2
   @@ -0,0 +1,1 @@
   +c
   diff --git a/d b/d
   --- a/d
   +++ b/d
   @@ -1,1 +0,0 @@
   -d
   $ hg ci -m empty
   $ hg export --git tip > empty.diff
   $ hg up -C 0
   4 files updated, 0 files merged, 2 files removed, 0 files unresolved
   $ hg import empty.diff
   applying empty.diff
   $ for name in a b1 b2 c1 c2 d; do
   >   echo % $name file
   >   test -f $name && cat $name
   >   done
   % a file
   a
   % b1 file
   % b2 file
   b
   % c1 file
   c
   % c2 file
   c
   % d file
   $ cd ..
 
 
 Test importing a patch ending with a binary file removal
 
   $ hg init binaryremoval
   $ cd binaryremoval
   $ echo a > a
   $ python -c "file('b', 'wb').write('a\x00b')"
   $ hg ci -Am addall
   adding a
   adding b
   $ hg rm a
   $ hg rm b
   $ hg st
   R a
   R b
   $ hg ci -m remove
   $ hg export --git . > remove.diff
   $ cat remove.diff | grep git
   diff --git a/a b/a
   diff --git a/b b/b
   $ hg up -C 0
   2 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg import remove.diff
   applying remove.diff
   $ hg manifest
   $ cd ..
 
 
 Issue927: test update+rename with common name
 
   $ hg init t
   $ cd t
   $ touch a
   $ hg ci -Am t
   adding a
   $ echo a > a
 
 Here, bfile.startswith(afile)
 
   $ hg copy a a2
   $ hg ci -m copya
   $ hg export --git tip > copy.diff
   $ hg up -C 0
   1 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg import copy.diff
   applying copy.diff
 
 a should contain an 'a'
 
   $ cat a
   a
 
 and a2 should have duplicated it
 
   $ cat a2
   a
   $ cd ..
 
 
 test -p0
 
   $ hg init p0
   $ cd p0
   $ echo a > a
   $ hg ci -Am t
   adding a
   $ hg import -p0 - << EOF
   > foobar
   > --- a	Sat Apr 12 22:43:58 2008 -0400
   > +++ a	Sat Apr 12 22:44:05 2008 -0400
   > @@ -1,1 +1,1 @@
   > -a
   > +bb
   > EOF
   applying patch from stdin
   $ hg status
   $ cat a
   bb
   $ cd ..
 
 
 test paths outside repo root
 
   $ mkdir outside
   $ touch outside/foo
   $ hg init inside
   $ cd inside
   $ hg import - <<EOF
   > diff --git a/a b/b
   > rename from ../outside/foo
   > rename to bar
   > EOF
   applying patch from stdin
   abort: path contains illegal component: ../outside/foo
   [255]
   $ cd ..
 
 
 test import with similarity and git and strip (issue295 et al.)
 
   $ hg init sim
   $ cd sim
   $ echo 'this is a test' > a
   $ hg ci -Ama
   adding a
   $ cat > ../rename.diff <<EOF
   > diff --git a/foo/a b/foo/a
   > deleted file mode 100644
   > --- a/foo/a
   > +++ /dev/null
   > @@ -1,1 +0,0 @@
   > -this is a test
   > diff --git a/foo/b b/foo/b
   > new file mode 100644
   > --- /dev/null
   > +++ b/foo/b
   > @@ -0,0 +1,2 @@
   > +this is a test
   > +foo
   > EOF
   $ hg import --no-commit -v -s 1 ../rename.diff -p2
   applying ../rename.diff
   patching file a
   patching file b
   removing a
   adding b
   recording removal of a as rename to b (88% similar)
   applied to working directory
   $ hg st -C
   A b
     a
   R a
   $ hg revert -a
   undeleting a
   forgetting b
   $ rm b
   $ hg import --no-commit -v -s 100 ../rename.diff -p2
   applying ../rename.diff
   patching file a
   patching file b
   removing a
   adding b
   applied to working directory
   $ hg st -C
   A b
   R a
   $ cd ..
 
 
 Issue1495: add empty file from the end of patch
 
   $ hg init addemptyend
   $ cd addemptyend
   $ touch a
   $ hg addremove
   adding a
   $ hg ci -m "commit"
   $ cat > a.patch <<EOF
   > add a, b
   > diff --git a/a b/a
   > --- a/a
   > +++ b/a
   > @@ -0,0 +1,1 @@
   > +a
   > diff --git a/b b/b
   > new file mode 100644
   > EOF
   $ hg import --no-commit a.patch
   applying a.patch
 
 apply a good patch followed by an empty patch (mainly to ensure
 that dirstate is *not* updated when import crashes)
   $ hg update -q -C .
   $ rm b
   $ touch empty.patch
   $ hg import a.patch empty.patch
   applying a.patch
   applying empty.patch
   transaction abort!
   rollback completed
   abort: empty.patch: no diffs found
   [255]
   $ hg tip --template '{rev}  {desc|firstline}\n'
   0  commit
   $ hg -q status
   M a
   $ cd ..
 
 create file when source is not /dev/null
 
   $ cat > create.patch <<EOF
   > diff -Naur proj-orig/foo proj-new/foo
   > --- proj-orig/foo       1969-12-31 16:00:00.000000000 -0800
   > +++ proj-new/foo        2009-07-17 16:50:45.801368000 -0700
   > @@ -0,0 +1,1 @@
   > +a
   > EOF
 
 some people have patches like the following too
 
   $ cat > create2.patch <<EOF
   > diff -Naur proj-orig/foo proj-new/foo
   > --- proj-orig/foo.orig  1969-12-31 16:00:00.000000000 -0800
   > +++ proj-new/foo        2009-07-17 16:50:45.801368000 -0700
   > @@ -0,0 +1,1 @@
   > +a
   > EOF
   $ hg init oddcreate
   $ cd oddcreate
   $ hg import --no-commit ../create.patch
   applying ../create.patch
   $ cat foo
   a
   $ rm foo
   $ hg revert foo
   $ hg import --no-commit ../create2.patch
   applying ../create2.patch
   $ cat foo
   a
 
 
 Issue1859: first line mistaken for email headers
 
   $ hg init emailconfusion
   $ cd emailconfusion
   $ cat > a.patch <<EOF
   > module: summary
   > 
   > description
   > 
   > 
   > diff -r 000000000000 -r 9b4c1e343b55 test.txt
   > --- /dev/null
   > +++ b/a
   > @@ -0,0 +1,1 @@
   > +a
   > EOF
   $ hg import -d '0 0' a.patch
   applying a.patch
   $ hg parents -v
   changeset:   0:5a681217c0ad
   tag:         tip
   user:        test
   date:        Thu Jan 01 00:00:00 1970 +0000
   files:       a
   description:
   module: summary
   
   description
   
   
   $ cd ..
 
 
 --- in commit message
 
   $ hg init commitconfusion
   $ cd commitconfusion
   $ cat > a.patch <<EOF
   > module: summary
   > 
   > --- description
   > 
   > diff --git a/a b/a
   > new file mode 100644
   > --- /dev/null
   > +++ b/a
   > @@ -0,0 +1,1 @@
   > +a
   > EOF
   > hg import -d '0 0' a.patch
   > hg parents -v
   > cd ..
   > 
   > echo '% tricky header splitting'
   > cat > trickyheaders.patch <<EOF
   > From: User A <user@a>
   > Subject: [PATCH] from: tricky!
   > 
   > # HG changeset patch
   > # User User B
   > # Date 1266264441 18000
   > # Branch stable
   > # Node ID f2be6a1170ac83bf31cb4ae0bad00d7678115bc0
   > # Parent  0000000000000000000000000000000000000000
   > from: tricky!
   > 
   > That is not a header.
   > 
   > diff -r 000000000000 -r f2be6a1170ac foo
   > --- /dev/null
   > +++ b/foo
   > @@ -0,0 +1,1 @@
   > +foo
   > EOF
   applying a.patch
   changeset:   0:f34d9187897d
   tag:         tip
   user:        test
   date:        Thu Jan 01 00:00:00 1970 +0000
   files:       a
   description:
   module: summary
   
   
   % tricky header splitting
 
   $ hg init trickyheaders
   $ cd trickyheaders
   $ hg import -d '0 0' ../trickyheaders.patch
   applying ../trickyheaders.patch
   $ hg export --git tip
   # HG changeset patch
   # User User B
   # Date 0 0
   # Node ID eb56ab91903632294ac504838508cb370c0901d2
   # Parent  0000000000000000000000000000000000000000
   from: tricky!
   
   That is not a header.
   
   diff --git a/foo b/foo
   new file mode 100644
   --- /dev/null
   +++ b/foo
   @@ -0,0 +1,1 @@
   +foo
   $ cd ..
 
 
 Issue2102: hg export and hg import speak different languages
 
   $ hg init issue2102
   $ cd issue2102
   $ mkdir -p src/cmd/gc
   $ touch src/cmd/gc/mksys.bash
   $ hg ci -Am init
   adding src/cmd/gc/mksys.bash
   $ hg import - <<EOF
   > # HG changeset patch
   > # User Rob Pike
   > # Date 1216685449 25200
   > # Node ID 03aa2b206f499ad6eb50e6e207b9e710d6409c98
   > # Parent  93d10138ad8df586827ca90b4ddb5033e21a3a84
   > help management of empty pkg and lib directories in perforce
   > 
   > R=gri
   > DELTA=4  (4 added, 0 deleted, 0 changed)
   > OCL=13328
   > CL=13328
   > 
   > diff --git a/lib/place-holder b/lib/place-holder
   > new file mode 100644
   > --- /dev/null
   > +++ b/lib/place-holder
   > @@ -0,0 +1,2 @@
   > +perforce does not maintain empty directories.
   > +this file helps.
   > diff --git a/pkg/place-holder b/pkg/place-holder
   > new file mode 100644
   > --- /dev/null
   > +++ b/pkg/place-holder
   > @@ -0,0 +1,2 @@
   > +perforce does not maintain empty directories.
   > +this file helps.
   > diff --git a/src/cmd/gc/mksys.bash b/src/cmd/gc/mksys.bash
   > old mode 100644
   > new mode 100755
   > EOF
   applying patch from stdin
   $ hg sum
   parent: 1:d59915696727 tip
    help management of empty pkg and lib directories in perforce
   branch: default
   commit: (clean)
   update: (current)
   $ hg diff --git -c tip
   diff --git a/lib/place-holder b/lib/place-holder
   new file mode 100644
   --- /dev/null
   +++ b/lib/place-holder
   @@ -0,0 +1,2 @@
   +perforce does not maintain empty directories.
   +this file helps.
   diff --git a/pkg/place-holder b/pkg/place-holder
   new file mode 100644
   --- /dev/null
   +++ b/pkg/place-holder
   @@ -0,0 +1,2 @@
   +perforce does not maintain empty directories.
   +this file helps.
   diff --git a/src/cmd/gc/mksys.bash b/src/cmd/gc/mksys.bash
   old mode 100644
   new mode 100755
   $ cd ..
 
 
 diff lines looking like headers
 
   $ hg init difflineslikeheaders
   $ cd difflineslikeheaders
   $ echo a >a
   $ echo b >b
   $ echo c >c
   $ hg ci -Am1
   adding a
   adding b
   adding c
 
   $ echo "key: value" >>a
   $ echo "key: value" >>b
   $ echo "foo" >>c
   $ hg ci -m2
 
   $ hg up -C 0
   3 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg diff --git -c1 >want
   $ hg diff -c1 | hg import --no-commit -
   applying patch from stdin
   $ hg diff --git >have
   $ diff want have
   $ cd ..
 
+import a unified diff with no lines of context (diff -U0)
+
+  $ hg init diffzero
+  $ cd diffzero
+  $ cat > f << EOF
+  > c2
+  > c4
+  > c5
+  > EOF
+  $ hg commit -Am0
+  adding f
+
+  $ hg import --no-commit - << EOF
+  > # HG changeset patch
+  > # User test
+  > # Date 0 0
+  > # Node ID f4974ab632f3dee767567b0576c0ec9a4508575c
+  > # Parent  8679a12a975b819fae5f7ad3853a2886d143d794
+  > 1
+  > diff -r 8679a12a975b -r f4974ab632f3 f
+  > --- a/f	Thu Jan 01 00:00:00 1970 +0000
+  > +++ b/f	Thu Jan 01 00:00:00 1970 +0000
+  > @@ -0,0 +1,1 @@
+  > +c1
+  > @@ -1,0 +3,1 @@
+  > +c3
+  > @@ -3,1 +4,0 @@
+  > -c5
+  > EOF
+  applying patch from stdin
+
+  $ cat f
+  c1
+  c2
+  c3
+  c4