hg2git.py

   1 #!/usr/bin/env python
   2
   3 # Copyright (c) 2007 Rocco Rutte <pdmef@gmx.net>
   4 # License: GPLv2
   5
   6 """hg2git.py - A mercurial-to-git filter for git-fast-import(1)
   7 Usage: hg2git.py <hg repo url> <marks file> <heads file> <tip file>
   8 """
   9
  10 from mercurial import repo,hg,cmdutil,util,ui,revlog
  11 from tempfile import mkstemp
  12 import re
  13 import sys
  14 import os
  15
  16 # silly regex to see if user field has email address
  17 user_re=re.compile('[^<]+ <[^>]+>$')
  18 # git branch for hg's default 'HEAD' branch
  19 cfg_master='master'
  20 # insert 'checkpoint' command after this many commits
  21 cfg_checkpoint_count=1000
  22
  23 def usage(ret):
  24   sys.stderr.write(__doc__)
  25   return ret
  26
  27 def setup_repo(url):
  28   myui=ui.ui()
  29   return myui,hg.repository(myui,url)
  30
  31 def get_changeset(ui,repo,revision):
  32   def get_branch(name):
  33     if name=='HEAD':
  34       name=cfg_master
  35     return name
  36   def fixup_user(user):
  37     if user_re.match(user)==None:
  38       if '@' not in user:
  39         return user+' <none@none>'
  40       return user+' <'+user+'>'
  41     return user
  42   node=repo.lookup(revision)
  43   (manifest,user,(time,timezone),files,desc,extra)=repo.changelog.read(node)
  44   tz="%+03d%02d" % (-timezone / 3600, ((-timezone % 3600) / 60))
  45   branch=get_branch(extra.get('branch','master'))
  46   return (manifest,fixup_user(user),(time,tz),files,desc,branch,extra)
  47
  48 def gitmode(x):
  49   return x and '100755' or '100644'
  50
  51 def wr(msg=''):
  52   print msg
  53   #map(lambda x: sys.stderr.write('\t[%s]\n' % x),msg.split('\n'))
  54
  55 def checkpoint(count):
  56   count=count+1
  57   if count%cfg_checkpoint_count==0:
  58     sys.stderr.write("Checkpoint after %d commits\n" % count)
  59     wr('checkpoint')
  60     wr()
  61   return count
  62
  63 def get_parent_mark(parent,marks):
  64   p=marks.get(str(parent),None)
  65   if p==None:
  66     # if we didn't see parent previously, assume we saw it in this run
  67     p=':%d' % (parent+1)
  68   return p
  69
  70 def export_commit(ui,repo,revision,marks,heads,last,max,count):
  71   sys.stderr.write('Exporting revision %d (tip %d) as [:%d]\n' % (revision,max,revision+1))
  72
  73   (_,user,(time,timezone),files,desc,branch,_)=get_changeset(ui,repo,revision)
  74   parents=repo.changelog.parentrevs(revision)
  75
  76   # we need this later to write out tags
  77   marks[str(revision)]=':%d'%(revision+1)
  78
  79   wr('commit refs/heads/%s' % branch)
  80   wr('mark :%d' % (revision+1))
  81   wr('committer %s %d %s' % (user,time,timezone))
  82   wr('data %d' % (len(desc)+1)) # wtf?
  83   wr(desc)
  84   wr()
  85
  86   src=heads.get(branch,'')
  87   link=''
  88   if src!='':
  89     # if we have a cached head, this is an incremental import: initialize it
  90     # and kill reference so we won't init it again
  91     wr('from %s' % src)
  92     heads[branch]=''
  93   elif not heads.has_key(branch) and revision>0:
  94     # newly created branch and not the first one: connect to parent
  95     tmp=get_parent_mark(parents[0],marks)
  96     wr('from %s' % tmp)
  97     sys.stderr.write('Link new branch [%s] to parent [%s]\n' %
  98         (branch,tmp))
  99     link=tmp # avoid making a merge commit for branch fork
 100
 101   if parents:
 102     l=last.get(branch,revision)
 103     for p in parents:
 104       # 1) as this commit implicitely is the child of the most recent
 105       #    commit of this branch, ignore this parent
 106       # 2) ignore nonexistent parents
 107       # 3) merge otherwise
 108       if p==l or p==revision or p<0:
 109         continue
 110       tmp=get_parent_mark(p,marks)
 111       # if we fork off a branch, don't merge via 'merge' as we have
 112       # 'from' already above
 113       if tmp==link:
 114         continue
 115       sys.stderr.write('Merging branch [%s] with parent [%s] from [r%d]\n' %
 116           (branch,tmp,p))
 117       wr('merge %s' % tmp)
 118
 119   last[branch]=revision
 120   heads[branch]=''
 121
 122   ctx=repo.changectx(str(revision))
 123   man=ctx.manifest()
 124
 125   wr('deleteall')
 126
 127   for f in man.keys():
 128     fctx=ctx.filectx(f)
 129     d=fctx.data()
 130     wr('M %s inline %s' % (gitmode(man.execf(f)),f))
 131     wr('data %d' % len(d)) # had some trouble with size()
 132     wr(d)
 133
 134   wr()
 135   return checkpoint(count)
 136
 137 def export_tags(ui,repo,cache,count):
 138   l=repo.tagslist()
 139   for tag,node in l:
 140     if tag=='tip':
 141       continue
 142     rev=repo.changelog.rev(node)
 143     ref=cache.get(str(rev),None)
 144     if ref==None:
 145       sys.stderr.write('Failed to find reference for creating tag'
 146           ' %s at r%d\n' % (tag,rev))
 147       continue
 148     (_,user,(time,timezone),_,desc,branch,_)=get_changeset(ui,repo,rev)
 149     sys.stderr.write('Exporting tag [%s] at [hg r%d] [git %s]\n' % (tag,rev,ref))
 150     wr('tag %s' % tag)
 151     wr('from %s' % ref)
 152     wr('tagger %s %d %s' % (user,time,timezone))
 153     msg='hg2git created tag %s for hg revision %d on branch %s on (summary):\n\t%s' % (tag,
 154         rev,branch,desc.split('\n')[0])
 155     wr('data %d' % (len(msg)+1))
 156     wr(msg)
 157     wr()
 158     count=checkpoint(count)
 159   return count
 160
 161 def load_cache(filename):
 162   cache={}
 163   if not os.path.exists(filename):
 164     return cache
 165   f=open(filename,'r')
 166   l=0
 167   for line in f.readlines():
 168     l+=1
 169     fields=line.split(' ')
 170     if fields==None or not len(fields)==2 or fields[0][0]!=':':
 171       sys.stderr.write('Invalid file format in [%s], line %d\n' % (filename,l))
 172       continue
 173     # put key:value in cache, key without ^:
 174     cache[fields[0][1:]]=fields[1].split('\n')[0]
 175   f.close()
 176   return cache
 177
 178 def save_cache(filename,cache):
 179   f=open(filename,'w+')
 180   map(lambda x: f.write(':%s %s\n' % (str(x),str(cache.get(x)))),cache.keys())
 181   f.close()
 182
 183 def verify_heads(ui,repo,cache):
 184   def getsha1(branch):
 185     f=open(os.getenv('GIT_DIR','/dev/null')+'/refs/heads/'+branch)
 186     sha1=f.readlines()[0].split('\n')[0]
 187     f.close()
 188     return sha1
 189
 190   for b in cache.keys():
 191     sys.stderr.write('Verifying branch [%s]\n' % b)
 192     sha1=getsha1(b)
 193     c=cache.get(b)
 194     if sha1!=c:
 195       sys.stderr.write('Warning: Branch [%s] modified outside hg2git:'
 196         '\n%s (repo) != %s (cache)\n' % (b,sha1,c))
 197   return True
 198
 199 if __name__=='__main__':
 200   if len(sys.argv)!=6: sys.exit(usage(1))
 201   repourl,m,marksfile,headsfile,tipfile=sys.argv[1:]
 202   _max=int(m)
 203
 204   marks_cache=load_cache(marksfile)
 205   heads_cache=load_cache(headsfile)
 206   state_cache=load_cache(tipfile)
 207
 208   ui,repo=setup_repo(repourl)
 209
 210   if not verify_heads(ui,repo,heads_cache):
 211     sys.exit(1)
 212
 213   tip=repo.changelog.count()
 214
 215   min=int(state_cache.get('tip',0))
 216   max=_max
 217   if _max<0:
 218     max=tip
 219
 220   c=int(state_cache.get('count',0))
 221   last={}
 222   for rev in range(min,max):
 223     c=export_commit(ui,repo,rev,marks_cache,heads_cache,last,tip,c)
 224
 225   c=export_tags(ui,repo,marks_cache,c)
 226
 227   state_cache['tip']=max
 228   state_cache['count']=c
 229   state_cache['repo']=repourl
 230   save_cache(tipfile,state_cache)