contrib/hg-to-git/hg-to-git.py

   1 #! /usr/bin/python
   2
   3 """ hg-to-git.py - A Mercurial to GIT converter
   4
   5     Copyright (C)2007 Stelian Pop <stelian@popies.net>
   6
   7     This program is free software; you can redistribute it and/or modify
   8     it under the terms of the GNU General Public License as published by
   9     the Free Software Foundation; either version 2, or (at your option)
  10     any later version.
  11
  12     This program is distributed in the hope that it will be useful,
  13     but WITHOUT ANY WARRANTY; without even the implied warranty of
  14     MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  15     GNU General Public License for more details.
  16
  17     You should have received a copy of the GNU General Public License
  18     along with this program; if not, write to the Free Software
  19     Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
  20 """
  21
  22 import os, os.path, sys
  23 import tempfile, popen2, pickle, getopt
  24 import re
  25
  26 # Maps hg version -> git version
  27 hgvers = {}
  28 # List of children for each hg revision
  29 hgchildren = {}
  30 # List of parents for each hg revision
  31 hgparents = {}
  32 # Current branch for each hg revision
  33 hgbranch = {}
  34 # Number of new changesets converted from hg
  35 hgnewcsets = 0
  36
  37 #------------------------------------------------------------------------------
  38
  39 def usage():
  40
  41         print """\
  42 %s: [OPTIONS] <hgprj>
  43
  44 options:
  45     -s, --gitstate=FILE: name of the state to be saved/read
  46                          for incrementals
  47     -n, --nrepack=INT:   number of changesets that will trigger
  48                          a repack (default=0, -1 to deactivate)
  49     -v, --verbose:       be verbose
  50
  51 required:
  52     hgprj:  name of the HG project to import (directory)
  53 """ % sys.argv[0]
  54
  55 #------------------------------------------------------------------------------
  56
  57 def getgitenv(user, date):
  58     env = ''
  59     elems = re.compile('(.*?)\s+<(.*)>').match(user)
  60     if elems:
  61         env += 'export GIT_AUTHOR_NAME="%s" ;' % elems.group(1)
  62         env += 'export GIT_COMMITER_NAME="%s" ;' % elems.group(1)
  63         env += 'export GIT_AUTHOR_EMAIL="%s" ;' % elems.group(2)
  64         env += 'export GIT_COMMITER_EMAIL="%s" ;' % elems.group(2)
  65     else:
  66         env += 'export GIT_AUTHOR_NAME="%s" ;' % user
  67         env += 'export GIT_COMMITER_NAME="%s" ;' % user
  68         env += 'export GIT_AUTHOR_EMAIL= ;'
  69         env += 'export GIT_COMMITER_EMAIL= ;'
  70
  71     env += 'export GIT_AUTHOR_DATE="%s" ;' % date
  72     env += 'export GIT_COMMITTER_DATE="%s" ;' % date
  73     return env
  74
  75 #------------------------------------------------------------------------------
  76
  77 state = ''
  78 opt_nrepack = 0
  79 verbose = False
  80
  81 try:
  82     opts, args = getopt.getopt(sys.argv[1:], 's:t:n:v', ['gitstate=', 'tempdir=', 'nrepack=', 'verbose'])
  83     for o, a in opts:
  84         if o in ('-s', '--gitstate'):
  85             state = a
  86             state = os.path.abspath(state)
  87         if o in ('-n', '--nrepack'):
  88             opt_nrepack = int(a)
  89         if o in ('-v', '--verbose'):
  90             verbose = True
  91     if len(args) != 1:
  92         raise('params')
  93 except:
  94     usage()
  95     sys.exit(1)
  96
  97 hgprj = args[0]
  98 os.chdir(hgprj)
  99
 100 if state:
 101     if os.path.exists(state):
 102         if verbose:
 103             print 'State does exist, reading'
 104         f = open(state, 'r')
 105         hgvers = pickle.load(f)
 106     else:
 107         print 'State does not exist, first run'
 108
 109 tip = os.popen('hg tip --template "{rev}"').read()
 110 if verbose:
 111     print 'tip is', tip
 112
 113 # Calculate the branches
 114 if verbose:
 115     print 'analysing the branches...'
 116 hgchildren["0"] = ()
 117 hgparents["0"] = (None, None)
 118 hgbranch["0"] = "master"
 119 for cset in range(1, int(tip) + 1):
 120     hgchildren[str(cset)] = ()
 121     prnts = os.popen('hg log -r %d --template "{parents}"' % cset).read().strip().split(' ')
 122     prnts = map(lambda x: x[:x.find(':')], prnts)
 123     if prnts[0] != '':
 124         parent = prnts[0].strip()
 125     else:
 126         parent = str(cset - 1)
 127     hgchildren[parent] += ( str(cset), )
 128     if len(prnts) > 1:
 129         mparent = prnts[1].strip()
 130         hgchildren[mparent] += ( str(cset), )
 131     else:
 132         mparent = None
 133
 134     hgparents[str(cset)] = (parent, mparent)
 135
 136     if mparent:
 137         # For merge changesets, take either one, preferably the 'master' branch
 138         if hgbranch[mparent] == 'master':
 139             hgbranch[str(cset)] = 'master'
 140         else:
 141             hgbranch[str(cset)] = hgbranch[parent]
 142     else:
 143         # Normal changesets
 144         # For first children, take the parent branch, for the others create a new branch
 145         if hgchildren[parent][0] == str(cset):
 146             hgbranch[str(cset)] = hgbranch[parent]
 147         else:
 148             hgbranch[str(cset)] = "branch-" + str(cset)
 149
 150 if not hgvers.has_key("0"):
 151     print 'creating repository'
 152     os.system('git-init-db')
 153
 154 # loop through every hg changeset
 155 for cset in range(int(tip) + 1):
 156
 157     # incremental, already seen
 158     if hgvers.has_key(str(cset)):
 159         continue
 160     hgnewcsets += 1
 161
 162     # get info
 163     log_data = os.popen('hg log -r %d --template "{tags}\n{date|date}\n{author}\n"' % cset).readlines()
 164     tag = log_data[0].strip()
 165     date = log_data[1].strip()
 166     user = log_data[2].strip()
 167     parent = hgparents[str(cset)][0]
 168     mparent = hgparents[str(cset)][1]
 169
 170     #get comment
 171     (fdcomment, filecomment) = tempfile.mkstemp()
 172     csetcomment = os.popen('hg log -r %d --template "{desc}"' % cset).read().strip()
 173     os.write(fdcomment, csetcomment)
 174     os.close(fdcomment)
 175
 176     print '-----------------------------------------'
 177     print 'cset:', cset
 178     print 'branch:', hgbranch[str(cset)]
 179     print 'user:', user
 180     print 'date:', date
 181     print 'comment:', csetcomment
 182     if parent:
 183         print 'parent:', parent
 184     if mparent:
 185         print 'mparent:', mparent
 186     if tag:
 187         print 'tag:', tag
 188     print '-----------------------------------------'
 189
 190     # checkout the parent if necessary
 191     if cset != 0:
 192         if hgbranch[str(cset)] == "branch-" + str(cset):
 193             print 'creating new branch', hgbranch[str(cset)]
 194             os.system('git-checkout -b %s %s' % (hgbranch[str(cset)], hgvers[parent]))
 195         else:
 196             print 'checking out branch', hgbranch[str(cset)]
 197             os.system('git-checkout %s' % hgbranch[str(cset)])
 198
 199     # merge
 200     if mparent:
 201         if hgbranch[parent] == hgbranch[str(cset)]:
 202             otherbranch = hgbranch[mparent]
 203         else:
 204             otherbranch = hgbranch[parent]
 205         print 'merging', otherbranch, 'into', hgbranch[str(cset)]
 206         os.system(getgitenv(user, date) + 'git-merge --no-commit -s ours "" %s %s' % (hgbranch[str(cset)], otherbranch))
 207
 208     # remove everything except .git and .hg directories
 209     os.system('find . \( -path "./.hg" -o -path "./.git" \) -prune -o ! -name "." -print | xargs rm -rf')
 210
 211     # repopulate with checkouted files
 212     os.system('hg update -C %d' % cset)
 213
 214     # add new files
 215     os.system('git-ls-files -x .hg --others | git-update-index --add --stdin')
 216     # delete removed files
 217     os.system('git-ls-files -x .hg --deleted | git-update-index --remove --stdin')
 218
 219     # commit
 220     os.system(getgitenv(user, date) + 'git commit --allow-empty -a -F %s' % filecomment)
 221     os.unlink(filecomment)
 222
 223     # tag
 224     if tag and tag != 'tip':
 225         os.system(getgitenv(user, date) + 'git-tag %s' % tag)
 226
 227     # delete branch if not used anymore...
 228     if mparent and len(hgchildren[str(cset)]):
 229         print "Deleting unused branch:", otherbranch
 230         os.system('git-branch -d %s' % otherbranch)
 231
 232     # retrieve and record the version
 233     vvv = os.popen('git-show --quiet --pretty=format:%H').read()
 234     print 'record', cset, '->', vvv
 235     hgvers[str(cset)] = vvv
 236
 237 if hgnewcsets >= opt_nrepack and opt_nrepack != -1:
 238     os.system('git-repack -a -d')
 239
 240 # write the state for incrementals
 241 if state:
 242     if verbose:
 243         print 'Writing state'
 244     f = open(state, 'w')
 245     pickle.dump(hgvers, f)
 246
 247 # vim: et ts=8 sw=4 sts=4