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

   1 #!/usr/bin/env 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, 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_COMMITTER_NAME="%s" ;' % elems.group(1)
  63         env += 'export GIT_AUTHOR_EMAIL="%s" ;' % elems.group(2)
  64         env += 'export GIT_COMMITTER_EMAIL="%s" ;' % elems.group(2)
  65     else:
  66         env += 'export GIT_AUTHOR_NAME="%s" ;' % user
  67         env += 'export GIT_COMMITTER_NAME="%s" ;' % user
  68         env += 'export GIT_AUTHOR_EMAIL= ;'
  69         env += 'export GIT_COMMITTER_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 Exception('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 sock = os.popen('hg tip --template "{rev}"')
 110 tip = sock.read()
 111 if sock.close():
 112     sys.exit(1)
 113 if verbose:
 114     print 'tip is', tip
 115
 116 # Calculate the branches
 117 if verbose:
 118     print 'analysing the branches...'
 119 hgchildren["0"] = ()
 120 hgparents["0"] = (None, None)
 121 hgbranch["0"] = "master"
 122 for cset in range(1, int(tip) + 1):
 123     hgchildren[str(cset)] = ()
 124     prnts = os.popen('hg log -r %d --template "{parents}"' % cset).read().strip().split(' ')
 125     prnts = map(lambda x: x[:x.find(':')], prnts)
 126     if prnts[0] != '':
 127         parent = prnts[0].strip()
 128     else:
 129         parent = str(cset - 1)
 130     hgchildren[parent] += ( str(cset), )
 131     if len(prnts) > 1:
 132         mparent = prnts[1].strip()
 133         hgchildren[mparent] += ( str(cset), )
 134     else:
 135         mparent = None
 136
 137     hgparents[str(cset)] = (parent, mparent)
 138
 139     if mparent:
 140         # For merge changesets, take either one, preferably the 'master' branch
 141         if hgbranch[mparent] == 'master':
 142             hgbranch[str(cset)] = 'master'
 143         else:
 144             hgbranch[str(cset)] = hgbranch[parent]
 145     else:
 146         # Normal changesets
 147         # For first children, take the parent branch, for the others create a new branch
 148         if hgchildren[parent][0] == str(cset):
 149             hgbranch[str(cset)] = hgbranch[parent]
 150         else:
 151             hgbranch[str(cset)] = "branch-" + str(cset)
 152
 153 if not hgvers.has_key("0"):
 154     print 'creating repository'
 155     os.system('git init')
 156
 157 # loop through every hg changeset
 158 for cset in range(int(tip) + 1):
 159
 160     # incremental, already seen
 161     if hgvers.has_key(str(cset)):
 162         continue
 163     hgnewcsets += 1
 164
 165     # get info
 166     log_data = os.popen('hg log -r %d --template "{tags}\n{date|date}\n{author}\n"' % cset).readlines()
 167     tag = log_data[0].strip()
 168     date = log_data[1].strip()
 169     user = log_data[2].strip()
 170     parent = hgparents[str(cset)][0]
 171     mparent = hgparents[str(cset)][1]
 172
 173     #get comment
 174     (fdcomment, filecomment) = tempfile.mkstemp()
 175     csetcomment = os.popen('hg log -r %d --template "{desc}"' % cset).read().strip()
 176     os.write(fdcomment, csetcomment)
 177     os.close(fdcomment)
 178
 179     print '-----------------------------------------'
 180     print 'cset:', cset
 181     print 'branch:', hgbranch[str(cset)]
 182     print 'user:', user
 183     print 'date:', date
 184     print 'comment:', csetcomment
 185     if parent:
 186         print 'parent:', parent
 187     if mparent:
 188         print 'mparent:', mparent
 189     if tag:
 190         print 'tag:', tag
 191     print '-----------------------------------------'
 192
 193     # checkout the parent if necessary
 194     if cset != 0:
 195         if hgbranch[str(cset)] == "branch-" + str(cset):
 196             print 'creating new branch', hgbranch[str(cset)]
 197             os.system('git checkout -b %s %s' % (hgbranch[str(cset)], hgvers[parent]))
 198         else:
 199             print 'checking out branch', hgbranch[str(cset)]
 200             os.system('git checkout %s' % hgbranch[str(cset)])
 201
 202     # merge
 203     if mparent:
 204         if hgbranch[parent] == hgbranch[str(cset)]:
 205             otherbranch = hgbranch[mparent]
 206         else:
 207             otherbranch = hgbranch[parent]
 208         print 'merging', otherbranch, 'into', hgbranch[str(cset)]
 209         os.system(getgitenv(user, date) + 'git merge --no-commit -s ours "" %s %s' % (hgbranch[str(cset)], otherbranch))
 210
 211     # remove everything except .git and .hg directories
 212     os.system('find . \( -path "./.hg" -o -path "./.git" \) -prune -o ! -name "." -print | xargs rm -rf')
 213
 214     # repopulate with checkouted files
 215     os.system('hg update -C %d' % cset)
 216
 217     # add new files
 218     os.system('git ls-files -x .hg --others | git update-index --add --stdin')
 219     # delete removed files
 220     os.system('git ls-files -x .hg --deleted | git update-index --remove --stdin')
 221
 222     # commit
 223     os.system(getgitenv(user, date) + 'git commit --allow-empty -a -F %s' % filecomment)
 224     os.unlink(filecomment)
 225
 226     # tag
 227     if tag and tag != 'tip':
 228         os.system(getgitenv(user, date) + 'git tag %s' % tag)
 229
 230     # delete branch if not used anymore...
 231     if mparent and len(hgchildren[str(cset)]):
 232         print "Deleting unused branch:", otherbranch
 233         os.system('git branch -d %s' % otherbranch)
 234
 235     # retrieve and record the version
 236     vvv = os.popen('git show --quiet --pretty=format:%H').read()
 237     print 'record', cset, '->', vvv
 238     hgvers[str(cset)] = vvv
 239
 240 if hgnewcsets >= opt_nrepack and opt_nrepack != -1:
 241     os.system('git repack -a -d')
 242
 243 # write the state for incrementals
 244 if state:
 245     if verbose:
 246         print 'Writing state'
 247     f = open(state, 'w')
 248     pickle.dump(hgvers, f)
 249
 250 # vim: et ts=8 sw=4 sts=4