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, see <http://www.gnu.org/licenses/>.
  19 """
  20
  21 import os, os.path, sys
  22 import tempfile, pickle, getopt
  23 import re
  24
  25 if sys.hexversion < 0x02030000:
  26    # The behavior of the pickle module changed significantly in 2.3
  27    sys.stderr.write("hg-to-git.py: requires Python 2.3 or later.\n")
  28    sys.exit(1)
  29
  30 # Maps hg version -> git version
  31 hgvers = {}
  32 # List of children for each hg revision
  33 hgchildren = {}
  34 # List of parents for each hg revision
  35 hgparents = {}
  36 # Current branch for each hg revision
  37 hgbranch = {}
  38 # Number of new changesets converted from hg
  39 hgnewcsets = 0
  40
  41 #------------------------------------------------------------------------------
  42
  43 def usage():
  44
  45         print """\
  46 %s: [OPTIONS] <hgprj>
  47
  48 options:
  49     -s, --gitstate=FILE: name of the state to be saved/read
  50                          for incrementals
  51     -n, --nrepack=INT:   number of changesets that will trigger
  52                          a repack (default=0, -1 to deactivate)
  53     -v, --verbose:       be verbose
  54
  55 required:
  56     hgprj:  name of the HG project to import (directory)
  57 """ % sys.argv[0]
  58
  59 #------------------------------------------------------------------------------
  60
  61 def getgitenv(user, date):
  62     env = ''
  63     elems = re.compile('(.*?)\s+<(.*)>').match(user)
  64     if elems:
  65         env += 'export GIT_AUTHOR_NAME="%s" ;' % elems.group(1)
  66         env += 'export GIT_COMMITTER_NAME="%s" ;' % elems.group(1)
  67         env += 'export GIT_AUTHOR_EMAIL="%s" ;' % elems.group(2)
  68         env += 'export GIT_COMMITTER_EMAIL="%s" ;' % elems.group(2)
  69     else:
  70         env += 'export GIT_AUTHOR_NAME="%s" ;' % user
  71         env += 'export GIT_COMMITTER_NAME="%s" ;' % user
  72         env += 'export GIT_AUTHOR_EMAIL= ;'
  73         env += 'export GIT_COMMITTER_EMAIL= ;'
  74
  75     env += 'export GIT_AUTHOR_DATE="%s" ;' % date
  76     env += 'export GIT_COMMITTER_DATE="%s" ;' % date
  77     return env
  78
  79 #------------------------------------------------------------------------------
  80
  81 state = ''
  82 opt_nrepack = 0
  83 verbose = False
  84
  85 try:
  86     opts, args = getopt.getopt(sys.argv[1:], 's:t:n:v', ['gitstate=', 'tempdir=', 'nrepack=', 'verbose'])
  87     for o, a in opts:
  88         if o in ('-s', '--gitstate'):
  89             state = a
  90             state = os.path.abspath(state)
  91         if o in ('-n', '--nrepack'):
  92             opt_nrepack = int(a)
  93         if o in ('-v', '--verbose'):
  94             verbose = True
  95     if len(args) != 1:
  96         raise Exception('params')
  97 except:
  98     usage()
  99     sys.exit(1)
 100
 101 hgprj = args[0]
 102 os.chdir(hgprj)
 103
 104 if state:
 105     if os.path.exists(state):
 106         if verbose:
 107             print 'State does exist, reading'
 108         f = open(state, 'r')
 109         hgvers = pickle.load(f)
 110     else:
 111         print 'State does not exist, first run'
 112
 113 sock = os.popen('hg tip --template "{rev}"')
 114 tip = sock.read()
 115 if sock.close():
 116     sys.exit(1)
 117 if verbose:
 118     print 'tip is', tip
 119
 120 # Calculate the branches
 121 if verbose:
 122     print 'analysing the branches...'
 123 hgchildren["0"] = ()
 124 hgparents["0"] = (None, None)
 125 hgbranch["0"] = "master"
 126 for cset in range(1, int(tip) + 1):
 127     hgchildren[str(cset)] = ()
 128     prnts = os.popen('hg log -r %d --template "{parents}"' % cset).read().strip().split(' ')
 129     prnts = map(lambda x: x[:x.find(':')], prnts)
 130     if prnts[0] != '':
 131         parent = prnts[0].strip()
 132     else:
 133         parent = str(cset - 1)
 134     hgchildren[parent] += ( str(cset), )
 135     if len(prnts) > 1:
 136         mparent = prnts[1].strip()
 137         hgchildren[mparent] += ( str(cset), )
 138     else:
 139         mparent = None
 140
 141     hgparents[str(cset)] = (parent, mparent)
 142
 143     if mparent:
 144         # For merge changesets, take either one, preferably the 'master' branch
 145         if hgbranch[mparent] == 'master':
 146             hgbranch[str(cset)] = 'master'
 147         else:
 148             hgbranch[str(cset)] = hgbranch[parent]
 149     else:
 150         # Normal changesets
 151         # For first children, take the parent branch, for the others create a new branch
 152         if hgchildren[parent][0] == str(cset):
 153             hgbranch[str(cset)] = hgbranch[parent]
 154         else:
 155             hgbranch[str(cset)] = "branch-" + str(cset)
 156
 157 if not hgvers.has_key("0"):
 158     print 'creating repository'
 159     os.system('git init')
 160
 161 # loop through every hg changeset
 162 for cset in range(int(tip) + 1):
 163
 164     # incremental, already seen
 165     if hgvers.has_key(str(cset)):
 166         continue
 167     hgnewcsets += 1
 168
 169     # get info
 170     log_data = os.popen('hg log -r %d --template "{tags}\n{date|date}\n{author}\n"' % cset).readlines()
 171     tag = log_data[0].strip()
 172     date = log_data[1].strip()
 173     user = log_data[2].strip()
 174     parent = hgparents[str(cset)][0]
 175     mparent = hgparents[str(cset)][1]
 176
 177     #get comment
 178     (fdcomment, filecomment) = tempfile.mkstemp()
 179     csetcomment = os.popen('hg log -r %d --template "{desc}"' % cset).read().strip()
 180     os.write(fdcomment, csetcomment)
 181     os.close(fdcomment)
 182
 183     print '-----------------------------------------'
 184     print 'cset:', cset
 185     print 'branch:', hgbranch[str(cset)]
 186     print 'user:', user
 187     print 'date:', date
 188     print 'comment:', csetcomment
 189     if parent:
 190         print 'parent:', parent
 191     if mparent:
 192         print 'mparent:', mparent
 193     if tag:
 194         print 'tag:', tag
 195     print '-----------------------------------------'
 196
 197     # checkout the parent if necessary
 198     if cset != 0:
 199         if hgbranch[str(cset)] == "branch-" + str(cset):
 200             print 'creating new branch', hgbranch[str(cset)]
 201             os.system('git checkout -b %s %s' % (hgbranch[str(cset)], hgvers[parent]))
 202         else:
 203             print 'checking out branch', hgbranch[str(cset)]
 204             os.system('git checkout %s' % hgbranch[str(cset)])
 205
 206     # merge
 207     if mparent:
 208         if hgbranch[parent] == hgbranch[str(cset)]:
 209             otherbranch = hgbranch[mparent]
 210         else:
 211             otherbranch = hgbranch[parent]
 212         print 'merging', otherbranch, 'into', hgbranch[str(cset)]
 213         os.system(getgitenv(user, date) + 'git merge --no-commit -s ours "" %s %s' % (hgbranch[str(cset)], otherbranch))
 214
 215     # remove everything except .git and .hg directories
 216     os.system('find . \( -path "./.hg" -o -path "./.git" \) -prune -o ! -name "." -print | xargs rm -rf')
 217
 218     # repopulate with checkouted files
 219     os.system('hg update -C %d' % cset)
 220
 221     # add new files
 222     os.system('git ls-files -x .hg --others | git update-index --add --stdin')
 223     # delete removed files
 224     os.system('git ls-files -x .hg --deleted | git update-index --remove --stdin')
 225
 226     # commit
 227     os.system(getgitenv(user, date) + 'git commit --allow-empty --allow-empty-message -a -F %s' % filecomment)
 228     os.unlink(filecomment)
 229
 230     # tag
 231     if tag and tag != 'tip':
 232         os.system(getgitenv(user, date) + 'git tag %s' % tag)
 233
 234     # delete branch if not used anymore...
 235     if mparent and len(hgchildren[str(cset)]):
 236         print "Deleting unused branch:", otherbranch
 237         os.system('git branch -d %s' % otherbranch)
 238
 239     # retrieve and record the version
 240     vvv = os.popen('git show --quiet --pretty=format:%H').read()
 241     print 'record', cset, '->', vvv
 242     hgvers[str(cset)] = vvv
 243
 244 if hgnewcsets >= opt_nrepack and opt_nrepack != -1:
 245     os.system('git repack -a -d')
 246
 247 # write the state for incrementals
 248 if state:
 249     if verbose:
 250         print 'Writing state'
 251     f = open(state, 'w')
 252     pickle.dump(hgvers, f)
 253
 254 # vim: et ts=8 sw=4 sts=4