filemerge: add internal:tagmerge merge tool

Add a new internal:tagmerge merge tool which implements an automatic merge
algorithm for mercurial's tag files

The tagmerge algorithm is able to resolve most merge conflicts that
currently would trigger a .hgtags merge conflict. The only case that
it does not (and cannot) handle is that in which two tags point to
different revisions on each merge parent _and_ their corresponding tag
histories have the same rank (i.e. the same length). In all other
cases the merge algorithm will choose the revision belonging to the
parent with the highest ranked tag history. The merged tag history is
the combination of both tag histories (special care is taken to try to
combine common tag histories where possible).

The algorithm also handles cases in which tags have been manually
removed from the .hgtags file and other similar corner cases.

In addition to actually merging the tags from two parents, taking into
account the base, the algorithm also tries to minimize the difference
between the merged tag file and the first parent's tag file (i.e. it
tries to make the merged tag order as as similar as possible to the
first parent's tag file order).

The algorithm works as follows:
1. read the tags from p1, p2 and the base
    - when reading the p1 tags, also get the line numbers associated to each
      tag node (these will be used to sort the merged tags in a way that
      minimizes the diff to p1). Ignore the file numbers when reading p2 and
      the base
2. recover the "lost tags" (i.e. those that are found in the base but not on p1
   or p2) and add them back to p1 and/or p2
    - at this point the only tags that are on p1 but not on p2 are those new
      tags that were introduced in p1. Same thing for the tags that are on p2
      but not on p2
3. take all tags that are only on p1 or only on p2 (but not on the base)
    - Note that these are the tags that were introduced between base and p1 and
      between base and p2, possibly on separate clones
4. for each tag found both on p1 and p2 perform the following merge algorithm:
    - the tags conflict if their tag "histories" have the same "rank" (i.e.
      length) _AND_ the last (current) tag is _NOT_ the same
    - for non conflicting tags:
        - choose which are the high and the low ranking nodes
            - the high ranking list of nodes is the one that is longer.
              In case of draw favor p1
            - the merged node list is made of 3 parts:
                - first the nodes that are common to the beginning of both the
                  low and the high ranking nodes
                - second the non common low ranking nodes
                - finally the non common high ranking nodes (with the last one
                  being the merged tag node)
            - note that this is equivalent to putting the whole low ranking node
              list first, followed by the non common high ranking nodes
    - note that during the merge we keep the "node line numbers", which will
      be used when writing the merged tags to the tag file
5. write the merged tags taking into account to their positions in the first
   parent (i.e. try to keep the relative ordering of the nodes that come
   from p1). This minimizes the diff between the merged and the p1 tag files
   This is done by using the following algorithm
    - group the nodes for a given tag that must be written next to each other
        - A: nodes that come from consecutive lines on p1
        - B: nodes that come from p2 (i.e. whose associated line number is None)
             and are next to one of the a nodes in A
        - each group is associated with a line number coming from p1
    - generate a "tag block" for each of the groups
        - a tag block is a set of consecutive "node tag" lines belonging to the
          same tag and which will be written next to each other on the merged
          tags file
    - sort the "tag blocks" according to their associated number line
        - put blocks whose nodes come all from p2 first
    - write the tag blocks in the sorted order

Notes:
- A few tests have been added to test-tag.t. These tests are very specific to
the new internal:tagmerge tool, so perhaps they should be moved to their own
test file.
- The merge algorithm was discussed in a thread on the mercurial mailing list.
In http://markmail.org/message/anqaxldup4tmgyrx a slightly different algorithm
was suggested. In it the p1 and p2 tags would have been interleaved instead of
put one before the other. It would be possible to implement that but my tests
suggest that the merge result would be more confusing and harder to understand.

mercurial/tagmerge.py

@@ -0,0 +1,265
+# tagmerge.py - merge .hgtags files
+#
+# Copyright 2014 Angel Ezquerra <[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.
+
+# This module implements an automatic merge algorithm for mercurial's tag files
+#
+# The tagmerge algorithm implemented in this module is able to resolve most
+# merge conflicts that currently would trigger a .hgtags merge conflict. The
+# only case that it does not (and cannot) handle is that in which two tags point
+# to different revisions on each merge parent _and_ their corresponding tag
+# histories have the same rank (i.e. the same length). In all other cases the
+# merge algorithm will choose the revision belonging to the parent with the
+# highest ranked tag history. The merged tag history is the combination of both
+# tag histories (special care is taken to try to combine common tag histories
+# where possible).
+#
+# In addition to actually merging the tags from two parents, taking into
+# account the base, the algorithm also tries to minimize the difference
+# between the merged tag file and the first parent's tag file (i.e. it tries to
+# make the merged tag order as as similar as possible to the first parent's tag
+# file order).
+#
+# The algorithm works as follows:
+# 1. read the tags from p1, p2 and the base
+#     - when reading the p1 tags, also get the line numbers associated to each
+#       tag node (these will be used to sort the merged tags in a way that
+#       minimizes the diff to p1). Ignore the file numbers when reading p2 and
+#       the base
+# 2. recover the "lost tags" (i.e. those that are found in the base but not on
+#    p1 or p2) and add them back to p1 and/or p2
+#     - at this point the only tags that are on p1 but not on p2 are those new
+#       tags that were introduced in p1. Same thing for the tags that are on p2
+#       but not on p2
+# 3. take all tags that are only on p1 or only on p2 (but not on the base)
+#     - Note that these are the tags that were introduced between base and p1
+#       and between base and p2, possibly on separate clones
+# 4. for each tag found both on p1 and p2 perform the following merge algorithm:
+#     - the tags conflict if their tag "histories" have the same "rank" (i.e.
+#       length) _AND_ the last (current) tag is _NOT_ the same
+#     - for non conflicting tags:
+#         - choose which are the high and the low ranking nodes
+#             - the high ranking list of nodes is the one that is longer.
+#               In case of draw favor p1
+#             - the merged node list is made of 3 parts:
+#                 - first the nodes that are common to the beginning of both
+#                   the low and the high ranking nodes
+#                 - second the non common low ranking nodes
+#                 - finally the non common high ranking nodes (with the last
+#                   one being the merged tag node)
+#             - note that this is equivalent to putting the whole low ranking
+#               node list first, followed by the non common high ranking nodes
+#     - note that during the merge we keep the "node line numbers", which will
+#       be used when writing the merged tags to the tag file
+# 5. write the merged tags taking into account to their positions in the first
+#    parent (i.e. try to keep the relative ordering of the nodes that come
+#    from p1). This minimizes the diff between the merged and the p1 tag files
+#    This is donw by using the following algorithm
+#     - group the nodes for a given tag that must be written next to each other
+#         - A: nodes that come from consecutive lines on p1
+#         - B: nodes that come from p2 (i.e. whose associated line number is
+#              None) and are next to one of the a nodes in A
+#         - each group is associated with a line number coming from p1
+#     - generate a "tag block" for each of the groups
+#         - a tag block is a set of consecutive "node tag" lines belonging to
+#           the same tag and which will be written next to each other on the
+#           merged tags file
+#     - sort the "tag blocks" according to their associated number line
+#         - put blocks whose nodes come all from p2 first
+#     - write the tag blocks in the sorted order
+
+import tags
+import util
+from node import nullid, hex
+from i18n import _
+import operator
+hexnullid = hex(nullid)
+
+def readtagsformerge(ui, repo, lines, fn='', keeplinenums=False):
+    '''read the .hgtags file into a structure that is suitable for merging
+
+    Sepending on the keeplinenumbers flag, clear the line numbers associated
+    with each tag. Rhis is done because only the line numbers of the first
+    parent are useful for merging
+    '''
+    filetags = tags._readtaghist(ui, repo, lines, fn=fn, recode=None,
+                              calcnodelines=True)[1]
+    for tagname, taginfo in filetags.items():
+        if not keeplinenums:
+            for el in taginfo:
+                el[1] = None
+    return filetags
+
+def grouptagnodesbyline(tagnodes):
+    '''
+    Group nearby nodes (i.e. those that must be written next to each other)
+
+    The input is a list of [node, position] pairs, corresponding to a given tag
+    The position is the line number where the node was found on the first parent
+    .hgtags file, or None for those nodes that came from the base or the second
+    parent .hgtags files.
+
+    This function groups those [node, position] pairs, returning a list of
+    groups of nodes that must be written next to each other because their
+    positions are consecutive or have no position preference (because their
+    position is None).
+
+    The result is a list of [position, [consecutive node list]]
+    '''
+    firstlinenum = None
+    for hexnode, linenum in tagnodes:
+        firstlinenum = linenum
+        if firstlinenum is not None:
+            break
+    if firstlinenum is None:
+        return [[None, [el[0] for el in tagnodes]]]
+    tagnodes[0][1] = firstlinenum
+    groupednodes = [[firstlinenum, []]]
+    prevlinenum = firstlinenum
+    for hexnode, linenum in tagnodes:
+        if linenum is not None and linenum - prevlinenum > 1:
+            groupednodes.append([linenum, []])
+        groupednodes[-1][1].append(hexnode)
+        if linenum is not None:
+            prevlinenum = linenum
+    return groupednodes
+
+def writemergedtags(repo, mergedtags):
+    '''
+    write the merged tags while trying to minimize the diff to the first parent
+
+    This function uses the ordering info stored on the merged tags dict to
+    generate an .hgtags file which is correct (in the sense that its contents
+    correspond to the result of the tag merge) while also being as close as
+    possible to the first parent's .hgtags file.
+    '''
+    # group the node-tag pairs that must be written next to each other
+    for tname, taglist in mergedtags.items():
+        mergedtags[tname] = grouptagnodesbyline(taglist)
+
+    # convert the grouped merged tags dict into a format that resembles the
+    # final .hgtags file (i.e. a list of blocks of 'node tag' pairs)
+    def taglist2string(tlist, tname):
+        return '\n'.join(['%s %s' % (hexnode, tname) for hexnode in tlist])
+
+    finaltags = []
+    for tname, tags in mergedtags.items():
+        for block in tags:
+            block[1] = taglist2string(block[1], tname)
+        finaltags += tags
+
+    # the tag groups are linked to a "position" that can be used to sort them
+    # before writing them
+    # the position is calculated to ensure that the diff of the merged .hgtags
+    # file to the first parent's .hgtags file is as small as possible
+    finaltags.sort(key=operator.itemgetter(0))
+
+    # finally we can join the sorted groups to get the final contents of the
+    # merged .hgtags file, and then write it to disk
+    mergedtagstring = '\n'.join([tags for rank, tags in finaltags if tags])
+    fp = repo.wfile('.hgtags', 'wb')
+    fp.write(mergedtagstring + '\n')
+    fp.close()
+
+def singletagmerge(p1nodes, p2nodes):
+    '''
+    merge the nodes corresponding to a single tag
+
+    Note that the inputs are lists of node-linenum pairs (i.e. not just lists
+    of nodes)
+    '''
+    if not p2nodes:
+        return p1nodes
+    if not p1nodes:
+        return p2nodes
+
+    # there is no conflict unless both tags point to different revisions
+    # and have a non identical tag history
+    p1currentnode = p1nodes[-1][0]
+    p2currentnode = p2nodes[-1][0]
+    if p1currentnode != p2currentnode and len(p1nodes) == len(p2nodes):
+        # cannot merge two tags with same rank pointing to different nodes
+        return None
+
+    # which are the highest ranking (hr) / lowest ranking (lr) nodes?
+    if len(p1nodes) >= len(p2nodes):
+        hrnodes, lrnodes = p1nodes, p2nodes
+    else:
+        hrnodes, lrnodes = p2nodes, p1nodes
+
+    # the lowest ranking nodes will be written first, followed by the highest
+    # ranking nodes
+    # to avoid unwanted tag rank explosion we try to see if there are some
+    # common nodes that can be written only once
+    commonidx = len(lrnodes)
+    for n in range(len(lrnodes)):
+        if hrnodes[n][0] != lrnodes[n][0]:
+            commonidx = n
+            break
+        lrnodes[n][1] = p1nodes[n][1]
+
+    # the merged node list has 3 parts:
+    # - common nodes
+    # - non common lowest ranking nodes
+    # - non common highest ranking nodes
+    # note that the common nodes plus the non common lowest ranking nodes is the
+    # whole list of lr nodes
+    return lrnodes + hrnodes[commonidx:]
+
+def merge(repo, fcd, fco, fca):
+    '''
+    Merge the tags of two revisions, taking into account the base tags
+    Try to minimize the diff between the merged tags and the first parent tags
+    '''
+    ui = repo.ui
+    # read the p1, p2 and base tags
+    # only keep the line numbers for the p1 tags
+    p1tags = readtagsformerge(
+        ui, repo, fcd.data().splitlines(), fn="p1 tags",
+        keeplinenums=True)
+    p2tags = readtagsformerge(
+        ui, repo, fco.data().splitlines(), fn="p2 tags",
+        keeplinenums=False)
+    basetags = readtagsformerge(
+        ui, repo, fca.data().splitlines(), fn="base tags",
+        keeplinenums=False)
+
+    # recover the list of "lost tags" (i.e. those that were found on the base
+    # revision but not on one of the revisions being merged)
+    basetagset = set(basetags)
+    for n, pntags in enumerate((p1tags, p2tags)):
+        pntagset = set(pntags)
+        pnlosttagset = basetagset - pntagset
+        for t in pnlosttagset:
+            pntags[t] = basetags[t]
+            if pntags[t][-1][0] != hexnullid:
+                pntags[t].append([hexnullid, None])
+
+    conflictedtags = []  # for reporting purposes
+    mergedtags = util.sortdict(p1tags)
+    # sortdict does not implement iteritems()
+    for tname, p2nodes in p2tags.items():
+        if tname not in mergedtags:
+            mergedtags[tname] = p2nodes
+            continue
+        p1nodes = mergedtags[tname]
+        mergednodes = singletagmerge(p1nodes, p2nodes)
+        if mergednodes is None:
+            conflictedtags.append(tname)
+            continue
+        mergedtags[tname] = mergednodes
+
+    if conflictedtags:
+        numconflicts = len(conflictedtags)
+        ui.warn(_('automatic .hgtags merge failed\n'
+            'the following %d tags are in conflict: %s\n')
+            % (numconflicts, ', '.join(sorted(conflictedtags))))
+        return True, 1
+
+    writemergedtags(repo, mergedtags)
+    ui.note(_('.hgtags merged successfully\n'))
+    return False, 0
+

mercurial/filemerge.py

@@ -1,436 +1,447
 # filemerge.py - file-level merge handling for Mercurial
 #
 # Copyright 2006, 2007, 2008 Matt Mackall <[e-mail]>
 #
 # This software may be used and distributed according to the terms of the
 # GNU General Public License version 2 or any later version.
 
 from node import short
 from i18n import _
 import util, simplemerge, match, error, templater, templatekw
 import os, tempfile, re, filecmp
+import tagmerge
 
 def _toolstr(ui, tool, part, default=""):
     return ui.config("merge-tools", tool + "." + part, default)
 
 def _toolbool(ui, tool, part, default=False):
     return ui.configbool("merge-tools", tool + "." + part, default)
 
 def _toollist(ui, tool, part, default=[]):
     return ui.configlist("merge-tools", tool + "." + part, default)
 
 internals = {}
 
 def internaltool(name, trymerge, onfailure=None):
     '''return a decorator for populating internal merge tool table'''
     def decorator(func):
         fullname = 'internal:' + name
         func.__doc__ = "``%s``\n" % fullname + func.__doc__.strip()
         internals[fullname] = func
         func.trymerge = trymerge
         func.onfailure = onfailure
         return func
     return decorator
 
 def _findtool(ui, tool):
     if tool in internals:
         return tool
     for kn in ("regkey", "regkeyalt"):
         k = _toolstr(ui, tool, kn)
         if not k:
             continue
         p = util.lookupreg(k, _toolstr(ui, tool, "regname"))
         if p:
             p = util.findexe(p + _toolstr(ui, tool, "regappend"))
             if p:
                 return p
     exe = _toolstr(ui, tool, "executable", tool)
     return util.findexe(util.expandpath(exe))
 
 def _picktool(repo, ui, path, binary, symlink):
     def check(tool, pat, symlink, binary):
         tmsg = tool
         if pat:
             tmsg += " specified for " + pat
         if not _findtool(ui, tool):
             if pat: # explicitly requested tool deserves a warning
                 ui.warn(_("couldn't find merge tool %s\n") % tmsg)
             else: # configured but non-existing tools are more silent
                 ui.note(_("couldn't find merge tool %s\n") % tmsg)
         elif symlink and not _toolbool(ui, tool, "symlink"):
             ui.warn(_("tool %s can't handle symlinks\n") % tmsg)
         elif binary and not _toolbool(ui, tool, "binary"):
             ui.warn(_("tool %s can't handle binary\n") % tmsg)
         elif not util.gui() and _toolbool(ui, tool, "gui"):
             ui.warn(_("tool %s requires a GUI\n") % tmsg)
         else:
             return True
         return False
 
     # forcemerge comes from command line arguments, highest priority
     force = ui.config('ui', 'forcemerge')
     if force:
         toolpath = _findtool(ui, force)
         if toolpath:
             return (force, util.shellquote(toolpath))
         else:
             # mimic HGMERGE if given tool not found
             return (force, force)
 
     # HGMERGE takes next precedence
     hgmerge = os.environ.get("HGMERGE")
     if hgmerge:
         return (hgmerge, hgmerge)
 
     # then patterns
     for pat, tool in ui.configitems("merge-patterns"):
         mf = match.match(repo.root, '', [pat])
         if mf(path) and check(tool, pat, symlink, False):
             toolpath = _findtool(ui, tool)
             return (tool, util.shellquote(toolpath))
 
     # then merge tools
     tools = {}
     for k, v in ui.configitems("merge-tools"):
         t = k.split('.')[0]
         if t not in tools:
             tools[t] = int(_toolstr(ui, t, "priority", "0"))
     names = tools.keys()
     tools = sorted([(-p, t) for t, p in tools.items()])
     uimerge = ui.config("ui", "merge")
     if uimerge:
         if uimerge not in names:
             return (uimerge, uimerge)
         tools.insert(0, (None, uimerge)) # highest priority
     tools.append((None, "hgmerge")) # the old default, if found
     for p, t in tools:
         if check(t, None, symlink, binary):
             toolpath = _findtool(ui, t)
             return (t, util.shellquote(toolpath))
 
     # internal merge or prompt as last resort
     if symlink or binary:
         return "internal:prompt", None
     return "internal:merge", None
 
 def _eoltype(data):
     "Guess the EOL type of a file"
     if '\0' in data: # binary
         return None
     if '\r\n' in data: # Windows
         return '\r\n'
     if '\r' in data: # Old Mac
         return '\r'
     if '\n' in data: # UNIX
         return '\n'
     return None # unknown
 
 def _matcheol(file, origfile):
     "Convert EOL markers in a file to match origfile"
     tostyle = _eoltype(util.readfile(origfile))
     if tostyle:
         data = util.readfile(file)
         style = _eoltype(data)
         if style:
             newdata = data.replace(style, tostyle)
             if newdata != data:
                 util.writefile(file, newdata)
 
 @internaltool('prompt', False)
 def _iprompt(repo, mynode, orig, fcd, fco, fca, toolconf):
     """Asks the user which of the local or the other version to keep as
     the merged version."""
     ui = repo.ui
     fd = fcd.path()
 
     if ui.promptchoice(_(" no tool found to merge %s\n"
                          "keep (l)ocal or take (o)ther?"
                          "$$ &Local $$ &Other") % fd, 0):
         return _iother(repo, mynode, orig, fcd, fco, fca, toolconf)
     else:
         return _ilocal(repo, mynode, orig, fcd, fco, fca, toolconf)
 
 @internaltool('local', False)
 def _ilocal(repo, mynode, orig, fcd, fco, fca, toolconf):
     """Uses the local version of files as the merged version."""
     return 0
 
 @internaltool('other', False)
 def _iother(repo, mynode, orig, fcd, fco, fca, toolconf):
     """Uses the other version of files as the merged version."""
     repo.wwrite(fcd.path(), fco.data(), fco.flags())
     return 0
 
 @internaltool('fail', False)
 def _ifail(repo, mynode, orig, fcd, fco, fca, toolconf):
     """
     Rather than attempting to merge files that were modified on both
     branches, it marks them as unresolved. The resolve command must be
     used to resolve these conflicts."""
     return 1
 
 def _premerge(repo, toolconf, files, labels=None):
     tool, toolpath, binary, symlink = toolconf
     if symlink:
         return 1
     a, b, c, back = files
 
     ui = repo.ui
 
     # do we attempt to simplemerge first?
     try:
         premerge = _toolbool(ui, tool, "premerge", not binary)
     except error.ConfigError:
         premerge = _toolstr(ui, tool, "premerge").lower()
         valid = 'keep'.split()
         if premerge not in valid:
             _valid = ', '.join(["'" + v + "'" for v in valid])
             raise error.ConfigError(_("%s.premerge not valid "
                                       "('%s' is neither boolean nor %s)") %
                                     (tool, premerge, _valid))
 
     if premerge:
         r = simplemerge.simplemerge(ui, a, b, c, quiet=True, label=labels)
         if not r:
             ui.debug(" premerge successful\n")
             return 0
         if premerge != 'keep':
             util.copyfile(back, a) # restore from backup and try again
     return 1 # continue merging
 
 @internaltool('merge', True,
               _("merging %s incomplete! "
                 "(edit conflicts, then use 'hg resolve --mark')\n"))
 def _imerge(repo, mynode, orig, fcd, fco, fca, toolconf, files, labels=None):
     """
     Uses the internal non-interactive simple merge algorithm for merging
     files. It will fail if there are any conflicts and leave markers in
     the partially merged file."""
     tool, toolpath, binary, symlink = toolconf
     if symlink:
         repo.ui.warn(_('warning: internal:merge cannot merge symlinks '
                        'for %s\n') % fcd.path())
         return False, 1
     r = _premerge(repo, toolconf, files, labels=labels)
     if r:
         a, b, c, back = files
 
         ui = repo.ui
 
         r = simplemerge.simplemerge(ui, a, b, c, label=labels, no_minimal=True)
         return True, r
     return False, 0
 
+@internaltool('tagmerge', True,
+              _("automatic tag merging of %s failed! "
+                "(use 'hg resolve --tool internal:merge' or another merge "
+                "tool of your choice)\n"))
+def _itagmerge(repo, mynode, orig, fcd, fco, fca, toolconf, files, labels=None):
+    """
+    Uses the internal tag merge algorithm (experimental).
+    """
+    return tagmerge.merge(repo, fcd, fco, fca)
+
 @internaltool('dump', True)
 def _idump(repo, mynode, orig, fcd, fco, fca, toolconf, files, labels=None):
     """
     Creates three versions of the files to merge, containing the
     contents of local, other and base. These files can then be used to
     perform a merge manually. If the file to be merged is named
     ``a.txt``, these files will accordingly be named ``a.txt.local``,
     ``a.txt.other`` and ``a.txt.base`` and they will be placed in the
     same directory as ``a.txt``."""
     r = _premerge(repo, toolconf, files, labels=labels)
     if r:
         a, b, c, back = files
 
         fd = fcd.path()
 
         util.copyfile(a, a + ".local")
         repo.wwrite(fd + ".other", fco.data(), fco.flags())
         repo.wwrite(fd + ".base", fca.data(), fca.flags())
     return False, r
 
 def _xmerge(repo, mynode, orig, fcd, fco, fca, toolconf, files, labels=None):
     r = _premerge(repo, toolconf, files, labels=labels)
     if r:
         tool, toolpath, binary, symlink = toolconf
         a, b, c, back = files
         out = ""
         env = {'HG_FILE': fcd.path(),
                'HG_MY_NODE': short(mynode),
                'HG_OTHER_NODE': str(fco.changectx()),
                'HG_BASE_NODE': str(fca.changectx()),
                'HG_MY_ISLINK': 'l' in fcd.flags(),
                'HG_OTHER_ISLINK': 'l' in fco.flags(),
                'HG_BASE_ISLINK': 'l' in fca.flags(),
                }
 
         ui = repo.ui
 
         args = _toolstr(ui, tool, "args", '$local $base $other')
         if "$output" in args:
             out, a = a, back # read input from backup, write to original
         replace = {'local': a, 'base': b, 'other': c, 'output': out}
         args = util.interpolate(r'\$', replace, args,
                                 lambda s: util.shellquote(util.localpath(s)))
         r = util.system(toolpath + ' ' + args, cwd=repo.root, environ=env,
                         out=ui.fout)
         return True, r
     return False, 0
 
 def _formatconflictmarker(repo, ctx, template, label, pad):
     """Applies the given template to the ctx, prefixed by the label.
 
     Pad is the minimum width of the label prefix, so that multiple markers
     can have aligned templated parts.
     """
     if ctx.node() is None:
         ctx = ctx.p1()
 
     props = templatekw.keywords.copy()
     props['templ'] = template
     props['ctx'] = ctx
     props['repo'] = repo
     templateresult = template('conflictmarker', **props)
 
     label = ('%s:' % label).ljust(pad + 1)
     mark = '%s %s' % (label, templater.stringify(templateresult))
 
     if mark:
         mark = mark.splitlines()[0] # split for safety
 
     # 8 for the prefix of conflict marker lines (e.g. '<<<<<<< ')
     return util.ellipsis(mark, 80 - 8)
 
 _defaultconflictmarker = ('{node|short} ' +
     '{ifeq(tags, "tip", "", "{tags} ")}' +
     '{if(bookmarks, "{bookmarks} ")}' +
     '{ifeq(branch, "default", "", "{branch} ")}' +
     '- {author|user}: {desc|firstline}')
 
 _defaultconflictlabels = ['local', 'other']
 
 def _formatlabels(repo, fcd, fco, labels):
     """Formats the given labels using the conflict marker template.
 
     Returns a list of formatted labels.
     """
     cd = fcd.changectx()
     co = fco.changectx()
 
     ui = repo.ui
     template = ui.config('ui', 'mergemarkertemplate', _defaultconflictmarker)
     template = templater.parsestring(template, quoted=False)
     tmpl = templater.templater(None, cache={ 'conflictmarker' : template })
 
     pad = max(len(labels[0]), len(labels[1]))
 
     return [_formatconflictmarker(repo, cd, tmpl, labels[0], pad),
             _formatconflictmarker(repo, co, tmpl, labels[1], pad)]
 
 def filemerge(repo, mynode, orig, fcd, fco, fca, labels=None):
     """perform a 3-way merge in the working directory
 
     mynode = parent node before merge
     orig = original local filename before merge
     fco = other file context
     fca = ancestor file context
     fcd = local file context for current/destination file
     """
 
     def temp(prefix, ctx):
         pre = "%s~%s." % (os.path.basename(ctx.path()), prefix)
         (fd, name) = tempfile.mkstemp(prefix=pre)
         data = repo.wwritedata(ctx.path(), ctx.data())
         f = os.fdopen(fd, "wb")
         f.write(data)
         f.close()
         return name
 
     if not fco.cmp(fcd): # files identical?
         return None
 
     ui = repo.ui
     fd = fcd.path()
     binary = fcd.isbinary() or fco.isbinary() or fca.isbinary()
     symlink = 'l' in fcd.flags() + fco.flags()
     tool, toolpath = _picktool(repo, ui, fd, binary, symlink)
     ui.debug("picked tool '%s' for %s (binary %s symlink %s)\n" %
                (tool, fd, binary, symlink))
 
     if tool in internals:
         func = internals[tool]
         trymerge = func.trymerge
         onfailure = func.onfailure
     else:
         func = _xmerge
         trymerge = True
         onfailure = _("merging %s failed!\n")
 
     toolconf = tool, toolpath, binary, symlink
 
     if not trymerge:
         return func(repo, mynode, orig, fcd, fco, fca, toolconf)
 
     a = repo.wjoin(fd)
     b = temp("base", fca)
     c = temp("other", fco)
     back = a + ".orig"
     util.copyfile(a, back)
 
     if orig != fco.path():
         ui.status(_("merging %s and %s to %s\n") % (orig, fco.path(), fd))
     else:
         ui.status(_("merging %s\n") % fd)
 
     ui.debug("my %s other %s ancestor %s\n" % (fcd, fco, fca))
 
     markerstyle = ui.config('ui', 'mergemarkers', 'basic')
     if markerstyle == 'basic':
         formattedlabels = _defaultconflictlabels
     else:
         if not labels:
             labels = _defaultconflictlabels
 
         formattedlabels = _formatlabels(repo, fcd, fco, labels)
 
     needcheck, r = func(repo, mynode, orig, fcd, fco, fca, toolconf,
                         (a, b, c, back), labels=formattedlabels)
     if not needcheck:
         if r:
             if onfailure:
                 ui.warn(onfailure % fd)
         else:
             util.unlink(back)
 
         util.unlink(b)
         util.unlink(c)
         return r
 
     if not r and (_toolbool(ui, tool, "checkconflicts") or
                   'conflicts' in _toollist(ui, tool, "check")):
         if re.search("^(<<<<<<< .*|=======|>>>>>>> .*)$", fcd.data(),
                      re.MULTILINE):
             r = 1
 
     checked = False
     if 'prompt' in _toollist(ui, tool, "check"):
         checked = True
         if ui.promptchoice(_("was merge of '%s' successful (yn)?"
                              "$$ &Yes $$ &No") % fd, 1):
             r = 1
 
     if not r and not checked and (_toolbool(ui, tool, "checkchanged") or
                                   'changed' in _toollist(ui, tool, "check")):
         if filecmp.cmp(a, back):
             if ui.promptchoice(_(" output file %s appears unchanged\n"
                                  "was merge successful (yn)?"
                                  "$$ &Yes $$ &No") % fd, 1):
                 r = 1
 
     if _toolbool(ui, tool, "fixeol"):
         _matcheol(a, back)
 
     if r:
         if onfailure:
             ui.warn(onfailure % fd)
     else:
         util.unlink(back)
 
     util.unlink(b)
     util.unlink(c)
     return r
 
 # tell hggettext to extract docstrings from these functions:
 i18nfunctions = internals.values()

tests/test-tag.t

@@ -1,405 +1,606
   $ hg init test
   $ cd test
 
   $ echo a > a
   $ hg add a
   $ hg commit -m "test"
   $ hg history
   changeset:   0:acb14030fe0a
   tag:         tip
   user:        test
   date:        Thu Jan 01 00:00:00 1970 +0000
   summary:     test
   
 
   $ hg tag ' '
   abort: tag names cannot consist entirely of whitespace
   [255]
 
 (this tests also that editor is not invoked, if '--edit' is not
 specified)
 
   $ HGEDITOR=cat hg tag "bleah"
   $ hg history
   changeset:   1:d4f0d2909abc
   tag:         tip
   user:        test
   date:        Thu Jan 01 00:00:00 1970 +0000
   summary:     Added tag bleah for changeset acb14030fe0a
   
   changeset:   0:acb14030fe0a
   tag:         bleah
   user:        test
   date:        Thu Jan 01 00:00:00 1970 +0000
   summary:     test
   
 
   $ echo foo >> .hgtags
   $ hg tag "bleah2"
   abort: working copy of .hgtags is changed (please commit .hgtags manually)
   [255]
 
   $ hg revert .hgtags
   $ hg tag -r 0 x y z y y z
   abort: tag names must be unique
   [255]
   $ hg tag tap nada dot tip
   abort: the name 'tip' is reserved
   [255]
   $ hg tag .
   abort: the name '.' is reserved
   [255]
   $ hg tag null
   abort: the name 'null' is reserved
   [255]
   $ hg tag "bleah"
   abort: tag 'bleah' already exists (use -f to force)
   [255]
   $ hg tag "blecch" "bleah"
   abort: tag 'bleah' already exists (use -f to force)
   [255]
 
   $ hg tag --remove "blecch"
   abort: tag 'blecch' does not exist
   [255]
   $ hg tag --remove "bleah" "blecch" "blough"
   abort: tag 'blecch' does not exist
   [255]
 
   $ hg tag -r 0 "bleah0"
   $ hg tag -l -r 1 "bleah1"
   $ hg tag gack gawk gorp
   $ hg tag -f gack
   $ hg tag --remove gack gorp
 
   $ hg tag "bleah "
   abort: tag 'bleah' already exists (use -f to force)
   [255]
   $ hg tag " bleah"
   abort: tag 'bleah' already exists (use -f to force)
   [255]
   $ hg tag " bleah"
   abort: tag 'bleah' already exists (use -f to force)
   [255]
   $ hg tag -r 0 "  bleahbleah  "
   $ hg tag -r 0 " bleah bleah "
 
   $ cat .hgtags
   acb14030fe0a21b60322c440ad2d20cf7685a376 bleah
   acb14030fe0a21b60322c440ad2d20cf7685a376 bleah0
   336fccc858a4eb69609a291105009e484a6b6b8d gack
   336fccc858a4eb69609a291105009e484a6b6b8d gawk
   336fccc858a4eb69609a291105009e484a6b6b8d gorp
   336fccc858a4eb69609a291105009e484a6b6b8d gack
   799667b6f2d9b957f73fa644a918c2df22bab58f gack
   799667b6f2d9b957f73fa644a918c2df22bab58f gack
   0000000000000000000000000000000000000000 gack
   336fccc858a4eb69609a291105009e484a6b6b8d gorp
   0000000000000000000000000000000000000000 gorp
   acb14030fe0a21b60322c440ad2d20cf7685a376 bleahbleah
   acb14030fe0a21b60322c440ad2d20cf7685a376 bleah bleah
 
   $ cat .hg/localtags
   d4f0d2909abc9290e2773c08837d70c1794e3f5a bleah1
 
 tagging on a non-head revision
 
   $ hg update 0
   0 files updated, 0 files merged, 1 files removed, 0 files unresolved
   $ hg tag -l localblah
   $ hg tag "foobar"
   abort: not at a branch head (use -f to force)
   [255]
   $ hg tag -f "foobar"
   $ cat .hgtags
   acb14030fe0a21b60322c440ad2d20cf7685a376 foobar
   $ cat .hg/localtags
   d4f0d2909abc9290e2773c08837d70c1794e3f5a bleah1
   acb14030fe0a21b60322c440ad2d20cf7685a376 localblah
 
   $ hg tag -l 'xx
   > newline'
   abort: '\n' cannot be used in a name
   [255]
   $ hg tag -l 'xx:xx'
   abort: ':' cannot be used in a name
   [255]
 
 cloning local tags
 
   $ cd ..
   $ hg -R test log -r0:5
   changeset:   0:acb14030fe0a
   tag:         bleah
   tag:         bleah bleah
   tag:         bleah0
   tag:         bleahbleah
   tag:         foobar
   tag:         localblah
   user:        test
   date:        Thu Jan 01 00:00:00 1970 +0000
   summary:     test
   
   changeset:   1:d4f0d2909abc
   tag:         bleah1
   user:        test
   date:        Thu Jan 01 00:00:00 1970 +0000
   summary:     Added tag bleah for changeset acb14030fe0a
   
   changeset:   2:336fccc858a4
   tag:         gawk
   user:        test
   date:        Thu Jan 01 00:00:00 1970 +0000
   summary:     Added tag bleah0 for changeset acb14030fe0a
   
   changeset:   3:799667b6f2d9
   user:        test
   date:        Thu Jan 01 00:00:00 1970 +0000
   summary:     Added tag gack, gawk, gorp for changeset 336fccc858a4
   
   changeset:   4:154eeb7c0138
   user:        test
   date:        Thu Jan 01 00:00:00 1970 +0000
   summary:     Added tag gack for changeset 799667b6f2d9
   
   changeset:   5:b4bb47aaff09
   user:        test
   date:        Thu Jan 01 00:00:00 1970 +0000
   summary:     Removed tag gack, gorp
   
   $ hg clone -q -rbleah1 test test1
   $ hg -R test1 parents --style=compact
   1[tip]   d4f0d2909abc   1970-01-01 00:00 +0000   test
     Added tag bleah for changeset acb14030fe0a
   
   $ hg clone -q -r5 test#bleah1 test2
   $ hg -R test2 parents --style=compact
   5[tip]   b4bb47aaff09   1970-01-01 00:00 +0000   test
     Removed tag gack, gorp
   
   $ hg clone -q -U test#bleah1 test3
   $ hg -R test3 parents --style=compact
 
   $ cd test
 
 Issue601: hg tag doesn't do the right thing if .hgtags or localtags
 doesn't end with EOL
 
   $ python << EOF
   > f = file('.hg/localtags'); last = f.readlines()[-1][:-1]; f.close()
   > f = file('.hg/localtags', 'w'); f.write(last); f.close()
   > EOF
   $ cat .hg/localtags; echo
   acb14030fe0a21b60322c440ad2d20cf7685a376 localblah
   $ hg tag -l localnewline
   $ cat .hg/localtags; echo
   acb14030fe0a21b60322c440ad2d20cf7685a376 localblah
   c2899151f4e76890c602a2597a650a72666681bf localnewline
   
 
   $ python << EOF
   > f = file('.hgtags'); last = f.readlines()[-1][:-1]; f.close()
   > f = file('.hgtags', 'w'); f.write(last); f.close()
   > EOF
   $ hg ci -m'broken manual edit of .hgtags'
   $ cat .hgtags; echo
   acb14030fe0a21b60322c440ad2d20cf7685a376 foobar
   $ hg tag newline
   $ cat .hgtags; echo
   acb14030fe0a21b60322c440ad2d20cf7685a376 foobar
   a0eea09de1eeec777b46f2085260a373b2fbc293 newline
   
 
 tag and branch using same name
 
   $ hg branch tag-and-branch-same-name
   marked working directory as branch tag-and-branch-same-name
   (branches are permanent and global, did you want a bookmark?)
   $ hg ci -m"discouraged"
   $ hg tag tag-and-branch-same-name
   warning: tag tag-and-branch-same-name conflicts with existing branch name
 
 test custom commit messages
 
   $ cat > editor.sh << '__EOF__'
   > echo "==== before editing"
   > cat "$1"
   > echo "===="
   > echo "custom tag message" > "$1"
   > echo "second line" >> "$1"
   > __EOF__
 
 at first, test saving last-message.txt
 
 (test that editor is not invoked before transaction starting)
 
   $ cat > .hg/hgrc << '__EOF__'
   > [hooks]
   > # this failure occurs before editor invocation
   > pretag.test-saving-lastmessage = false
   > __EOF__
   $ rm -f .hg/last-message.txt
   $ HGEDITOR="\"sh\" \"`pwd`/editor.sh\"" hg tag custom-tag -e
   abort: pretag.test-saving-lastmessage hook exited with status 1
   [255]
   $ cat .hg/last-message.txt
   cat: .hg/last-message.txt: No such file or directory
   [1]
 
 (test that editor is invoked and commit message is saved into
 "last-message.txt")
 
   $ cat >> .hg/hgrc << '__EOF__'
   > [hooks]
   > pretag.test-saving-lastmessage =
   > # this failure occurs after editor invocation
   > pretxncommit.unexpectedabort = false
   > __EOF__
 
 (this tests also that editor is invoked, if '--edit' is specified,
 regardless of '--message')
 
   $ rm -f .hg/last-message.txt
   $ HGEDITOR="\"sh\" \"`pwd`/editor.sh\"" hg tag custom-tag -e -m "foo bar"
   ==== before editing
   foo bar
   
   
   HG: Enter commit message.  Lines beginning with 'HG:' are removed.
   HG: Leave message empty to abort commit.
   HG: --
   HG: user: test
   HG: branch 'tag-and-branch-same-name'
   HG: changed .hgtags
   ====
   transaction abort!
   rollback completed
   note: commit message saved in .hg/last-message.txt
   abort: pretxncommit.unexpectedabort hook exited with status 1
   [255]
   $ cat .hg/last-message.txt
   custom tag message
   second line
 
   $ cat >> .hg/hgrc << '__EOF__'
   > [hooks]
   > pretxncommit.unexpectedabort =
   > __EOF__
   $ hg status .hgtags
   M .hgtags
   $ hg revert --no-backup -q .hgtags
 
 then, test custom commit message itself
 
   $ HGEDITOR="\"sh\" \"`pwd`/editor.sh\"" hg tag custom-tag -e
   ==== before editing
   Added tag custom-tag for changeset 75a534207be6
   
   
   HG: Enter commit message.  Lines beginning with 'HG:' are removed.
   HG: Leave message empty to abort commit.
   HG: --
   HG: user: test
   HG: branch 'tag-and-branch-same-name'
   HG: changed .hgtags
   ====
   $ hg log -l1 --template "{desc}\n"
   custom tag message
   second line
 
 
 local tag with .hgtags modified
 
   $ hg tag hgtags-modified
   $ hg rollback
   repository tip rolled back to revision 13 (undo commit)
   working directory now based on revision 13
   $ hg st
   M .hgtags
   ? .hgtags.orig
   ? editor.sh
   $ hg tag --local baz
   $ hg revert --no-backup .hgtags
 
 
 tagging when at named-branch-head that's not a topo-head
 
   $ hg up default
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge -t internal:local
   0 files updated, 1 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
   $ hg ci -m 'merge named branch'
   $ hg up 13
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg tag new-topo-head
 
 tagging on null rev
 
   $ hg up null
   0 files updated, 0 files merged, 2 files removed, 0 files unresolved
   $ hg tag nullrev
   abort: not at a branch head (use -f to force)
   [255]
 
   $ hg init empty
   $ hg tag -R empty nullrev
   abort: cannot tag null revision
   [255]
 
   $ hg tag -R empty -r 00000000000 -f nulltag
   abort: cannot tag null revision
   [255]
 
   $ cd ..
 
 tagging on an uncommitted merge (issue2542)
 
   $ hg init repo-tag-uncommitted-merge
   $ cd repo-tag-uncommitted-merge
   $ echo c1 > f1
   $ hg ci -Am0
   adding f1
   $ echo c2 > f2
   $ hg ci -Am1
   adding f2
   $ hg co -q 0
   $ hg branch b1
   marked working directory as branch b1
   (branches are permanent and global, did you want a bookmark?)
   $ hg ci -m2
   $ hg up default
   1 files updated, 0 files merged, 0 files removed, 0 files unresolved
   $ hg merge b1
   0 files updated, 0 files merged, 0 files removed, 0 files unresolved
   (branch merge, don't forget to commit)
 
   $ hg tag t1
   abort: uncommitted merge
   [255]
   $ hg status
   $ hg tag --rev 1 t2
   abort: uncommitted merge
   [255]
   $ hg tag --rev 1 --local t3
   $ hg tags -v
   tip                                2:2a156e8887cc
   t3                                 1:c3adabd1a5f4 local
 
   $ cd ..
 
 commit hook on tag used to be run without write lock - issue3344
 
   $ hg init repo-tag
   $ touch repo-tag/test
   $ hg -R repo-tag commit -A -m "test"
   adding test
   $ hg init repo-tag-target
   $ hg -R repo-tag --config hooks.commit="\"hg\" push \"`pwd`/repo-tag-target\"" tag tag
   pushing to $TESTTMP/repo-tag-target (glob)
   searching for changes
   adding changesets
   adding manifests
   adding file changes
   added 2 changesets with 2 changes to 2 files
 
+automatically merge resolvable tag conflicts (i.e. tags that differ in rank)
+create two clones with some different tags as well as some common tags
+check that we can merge tags that differ in rank
+
+  $ hg init repo-automatic-tag-merge
+  $ cd repo-automatic-tag-merge
+  $ echo c0 > f0
+  $ hg ci -A -m0
+  adding f0
+  $ hg tag tbase
+  $ cd ..
+  $ hg clone repo-automatic-tag-merge repo-automatic-tag-merge-clone
+  updating to branch default
+  2 files updated, 0 files merged, 0 files removed, 0 files unresolved
+  $ cd repo-automatic-tag-merge-clone
+  $ echo c1 > f1
+  $ hg ci -A -m1
+  adding f1
+  $ hg tag t1 t2 t3
+  $ hg tag --remove t2
+  $ hg tag t5
+  $ echo c2 > f2
+  $ hg ci -A -m2
+  adding f2
+  $ hg tag -f t3
+
+  $ cd ../repo-automatic-tag-merge
+  $ echo c3 > f3
+  $ hg ci -A -m3
+  adding f3
+  $ hg tag -f t4 t5 t6
+  $ hg tag --remove t5
+  $ echo c4 > f4
+  $ hg ci -A -m4
+  adding f4
+  $ hg tag t2
+  $ hg tag -f t6
+
+  $ cd ../repo-automatic-tag-merge-clone
+  $ hg pull
+  pulling from $TESTTMP/repo-automatic-tag-merge (glob)
+  searching for changes
+  adding changesets
+  adding manifests
+  adding file changes
+  added 6 changesets with 6 changes to 3 files (+1 heads)
+  (run 'hg heads' to see heads, 'hg merge' to merge)
+  $ hg merge --tool internal:tagmerge
+  merging .hgtags
+  2 files updated, 1 files merged, 0 files removed, 0 files unresolved
+  (branch merge, don't forget to commit)
+  $ hg status
+  M .hgtags
+  M f3
+  M f4
+  $ hg resolve -l
+  R .hgtags
+  $ cat .hgtags
+  9aa4e1292a27a248f8d07339bed9931d54907be7 t4
+  9aa4e1292a27a248f8d07339bed9931d54907be7 t6
+  9aa4e1292a27a248f8d07339bed9931d54907be7 t6
+  09af2ce14077a94effef208b49a718f4836d4338 t6
+  6cee5c8f3e5b4ae1a3996d2f6489c3e08eb5aea7 tbase
+  4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t1
+  929bca7b18d067cbf3844c3896319a940059d748 t2
+  4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t2
+  4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t3
+  4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t2
+  0000000000000000000000000000000000000000 t2
+  875517b4806a848f942811a315a5bce30804ae85 t5
+  9aa4e1292a27a248f8d07339bed9931d54907be7 t5
+  9aa4e1292a27a248f8d07339bed9931d54907be7 t5
+  0000000000000000000000000000000000000000 t5
+  4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t3
+  79505d5360b07e3e79d1052e347e73c02b8afa5b t3
+
+check that the merge tried to minimize the diff witht he first merge parent
+
+  $ hg diff --git -r 'p1()' .hgtags
+  diff --git a/.hgtags b/.hgtags
+  --- a/.hgtags
+  +++ b/.hgtags
+  @@ -1,9 +1,17 @@
+  +9aa4e1292a27a248f8d07339bed9931d54907be7 t4
+  +9aa4e1292a27a248f8d07339bed9931d54907be7 t6
+  +9aa4e1292a27a248f8d07339bed9931d54907be7 t6
+  +09af2ce14077a94effef208b49a718f4836d4338 t6
+   6cee5c8f3e5b4ae1a3996d2f6489c3e08eb5aea7 tbase
+   4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t1
+  +929bca7b18d067cbf3844c3896319a940059d748 t2
+   4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t2
+   4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t3
+   4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t2
+   0000000000000000000000000000000000000000 t2
+   875517b4806a848f942811a315a5bce30804ae85 t5
+  +9aa4e1292a27a248f8d07339bed9931d54907be7 t5
+  +9aa4e1292a27a248f8d07339bed9931d54907be7 t5
+  +0000000000000000000000000000000000000000 t5
+   4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t3
+   79505d5360b07e3e79d1052e347e73c02b8afa5b t3
+
+detect merge tag conflicts
+
+  $ hg update -C -r tip
+  3 files updated, 0 files merged, 2 files removed, 0 files unresolved
+  $ hg tag t7
+  $ hg update -C -r 'first(sort(head()))'
+  3 files updated, 0 files merged, 2 files removed, 0 files unresolved
+  $ printf "%s %s\n" `hg log -r . --template "{node} t7"` >> .hgtags
+  $ hg commit -m "manually add conflicting t7 tag"
+  $ hg merge --tool internal:tagmerge
+  merging .hgtags
+  automatic .hgtags merge failed
+  the following 1 tags are in conflict: t7
+  automatic tag merging of .hgtags failed! (use 'hg resolve --tool internal:merge' or another merge tool of your choice)
+  2 files updated, 0 files merged, 0 files removed, 1 files unresolved
+  use 'hg resolve' to retry unresolved file merges or 'hg update -C .' to abandon
+  [1]
+  $ hg resolve -l
+  U .hgtags
+  $ cat .hgtags
+  6cee5c8f3e5b4ae1a3996d2f6489c3e08eb5aea7 tbase
+  4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t1
+  4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t2
+  4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t3
+  4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t2
+  0000000000000000000000000000000000000000 t2
+  875517b4806a848f942811a315a5bce30804ae85 t5
+  4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t3
+  79505d5360b07e3e79d1052e347e73c02b8afa5b t3
+  ea918d56be86a4afc5a95312e8b6750e1428d9d2 t7
+
+  $ cd ..
+
+handle the loss of tags
+
+  $ hg clone repo-automatic-tag-merge-clone repo-merge-lost-tags
+  updating to branch default
+  4 files updated, 0 files merged, 0 files removed, 0 files unresolved
+  $ cd repo-merge-lost-tags
+  $ echo c5 > f5
+  $ hg ci -A -m5
+  adding f5
+  $ hg tag -f t7
+  $ hg update -r 'p1(t7)'
+  1 files updated, 0 files merged, 1 files removed, 0 files unresolved
+  $ printf '' > .hgtags
+  $ hg commit -m 'delete all tags'
+  created new head
+  $ hg update -r 'max(t7::)'
+  2 files updated, 0 files merged, 0 files removed, 0 files unresolved
+  $ hg merge -r tip --tool internal:tagmerge
+  merging .hgtags
+  0 files updated, 1 files merged, 0 files removed, 0 files unresolved
+  (branch merge, don't forget to commit)
+  $ hg resolve -l
+  R .hgtags
+  $ cat .hgtags
+  6cee5c8f3e5b4ae1a3996d2f6489c3e08eb5aea7 tbase
+  0000000000000000000000000000000000000000 tbase
+  4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t1
+  0000000000000000000000000000000000000000 t1
+  4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t2
+  4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t3
+  4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t2
+  0000000000000000000000000000000000000000 t2
+  875517b4806a848f942811a315a5bce30804ae85 t5
+  0000000000000000000000000000000000000000 t5
+  4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t3
+  79505d5360b07e3e79d1052e347e73c02b8afa5b t3
+  0000000000000000000000000000000000000000 t3
+  ea918d56be86a4afc5a95312e8b6750e1428d9d2 t7
+  0000000000000000000000000000000000000000 t7
+  ea918d56be86a4afc5a95312e8b6750e1428d9d2 t7
+  fd3a9e394ce3afb354a496323bf68ac1755a30de t7
+
+also check that we minimize the diff with the 1st merge parent
+
+  $ hg diff --git -r 'p1()' .hgtags
+  diff --git a/.hgtags b/.hgtags
+  --- a/.hgtags
+  +++ b/.hgtags
+  @@ -1,12 +1,17 @@
+   6cee5c8f3e5b4ae1a3996d2f6489c3e08eb5aea7 tbase
+  +0000000000000000000000000000000000000000 tbase
+   4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t1
+  +0000000000000000000000000000000000000000 t1
+   4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t2
+   4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t3
+   4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t2
+   0000000000000000000000000000000000000000 t2
+   875517b4806a848f942811a315a5bce30804ae85 t5
+  +0000000000000000000000000000000000000000 t5
+   4f3e9b90005b68b4d8a3f4355cedc302a8364f5c t3
+   79505d5360b07e3e79d1052e347e73c02b8afa5b t3
+  +0000000000000000000000000000000000000000 t3
+   ea918d56be86a4afc5a95312e8b6750e1428d9d2 t7
+  +0000000000000000000000000000000000000000 t7
+   ea918d56be86a4afc5a95312e8b6750e1428d9d2 t7
+   fd3a9e394ce3afb354a496323bf68ac1755a30de t7
+