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     sys.stderr.write('Initializing branch [%s] to parent [%s]\n' %
  94         (branch,src))
  95     link=src # avoid making a merge commit for incremental import
  96   elif not heads.has_key(branch) and revision>0:
  97     # newly created branch and not the first one: connect to parent
  98     tmp=get_parent_mark(parents[0],marks)
  99     wr('from %s' % tmp)
 100     sys.stderr.write('Link new branch [%s] to parent [%s]\n' %
 101         (branch,tmp))
 102     link=tmp # avoid making a merge commit for branch fork
 103
 104   if parents:
 105     l=last.get(branch,revision)
 106     for p in parents:
 107       # 1) as this commit implicitely is the child of the most recent
 108       #    commit of this branch, ignore this parent
 109       # 2) ignore nonexistent parents
 110       # 3) merge otherwise
 111       if p==l or p==revision or p<0:
 112         continue
 113       tmp=get_parent_mark(p,marks)
 114       # if we fork off a branch, don't merge via 'merge' as we have
 115       # 'from' already above
 116       if tmp==link:
 117         continue
 118       sys.stderr.write('Merging branch [%s] with parent [%s] from [r%d]\n' %
 119           (branch,tmp,p))
 120       wr('merge %s' % tmp)
 121
 122   last[branch]=revision
 123   heads[branch]=''
 124
 125   # just wipe the branch clean, all full manifest contents
 126   wr('deleteall')
 127
 128   ctx=repo.changectx(str(revision))
 129   man=ctx.manifest()
 130
 131   #for f in man.keys():
 132   #  fctx=ctx.filectx(f)
 133   #  d=fctx.data()
 134   #  wr('M %s inline %s' % (gitmode(man.execf(f)),f))
 135   #  wr('data %d' % len(d)) # had some trouble with size()
 136   #  wr(d)
 137
 138   for fctx in ctx.filectxs():
 139     f=fctx.path()
 140     d=fctx.data()
 141     wr('M %s inline %s' % (gitmode(man.execf(f)),f))
 142     wr('data %d' % len(d)) # had some trouble with size()
 143     wr(d)
 144
 145   wr()
 146   return checkpoint(count)
 147
 148 def export_tags(ui,repo,cache,count):
 149   l=repo.tagslist()
 150   for tag,node in l:
 151     if tag=='tip':
 152       continue
 153     rev=repo.changelog.rev(node)
 154     ref=cache.get(str(rev),None)
 155     if ref==None:
 156       sys.stderr.write('Failed to find reference for creating tag'
 157           ' %s at r%d\n' % (tag,rev))
 158       continue
 159     (_,user,(time,timezone),_,desc,branch,_)=get_changeset(ui,repo,rev)
 160     sys.stderr.write('Exporting tag [%s] at [hg r%d] [git %s]\n' % (tag,rev,ref))
 161     wr('tag %s' % tag)
 162     wr('from %s' % ref)
 163     wr('tagger %s %d %s' % (user,time,timezone))
 164     msg='hg2git created tag %s for hg revision %d on branch %s on (summary):\n\t%s' % (tag,
 165         rev,branch,desc.split('\n')[0])
 166     wr('data %d' % (len(msg)+1))
 167     wr(msg)
 168     wr()
 169     count=checkpoint(count)
 170   return count
 171
 172 def load_cache(filename):
 173   cache={}
 174   if not os.path.exists(filename):
 175     return cache
 176   f=open(filename,'r')
 177   l=0
 178   for line in f.readlines():
 179     l+=1
 180     fields=line.split(' ')
 181     if fields==None or not len(fields)==2 or fields[0][0]!=':':
 182       sys.stderr.write('Invalid file format in [%s], line %d\n' % (filename,l))
 183       continue
 184     # put key:value in cache, key without ^:
 185     cache[fields[0][1:]]=fields[1].split('\n')[0]
 186   f.close()
 187   return cache
 188
 189 def save_cache(filename,cache):
 190   f=open(filename,'w+')
 191   map(lambda x: f.write(':%s %s\n' % (str(x),str(cache.get(x)))),cache.keys())
 192   f.close()
 193
 194 def verify_heads(ui,repo,cache):
 195   def getsha1(branch):
 196     f=open(os.getenv('GIT_DIR','/dev/null')+'/refs/heads/'+branch)
 197     sha1=f.readlines()[0].split('\n')[0]
 198     f.close()
 199     return sha1
 200
 201   for b in cache.keys():
 202     sys.stderr.write('Verifying branch [%s]\n' % b)
 203     sha1=getsha1(b)
 204     c=cache.get(b)
 205     if sha1!=c:
 206       sys.stderr.write('Warning: Branch [%s] modified outside hg2git:'
 207         '\n%s (repo) != %s (cache)\n' % (b,sha1,c))
 208   return True
 209
 210 if __name__=='__main__':
 211   if len(sys.argv)!=6: sys.exit(usage(1))
 212   repourl,m,marksfile,headsfile,tipfile=sys.argv[1:]
 213   _max=int(m)
 214
 215   marks_cache=load_cache(marksfile)
 216   heads_cache=load_cache(headsfile)
 217   state_cache=load_cache(tipfile)
 218
 219   ui,repo=setup_repo(repourl)
 220
 221   if not verify_heads(ui,repo,heads_cache):
 222     sys.exit(1)
 223
 224   tip=repo.changelog.count()
 225
 226   min=int(state_cache.get('tip',0))
 227   max=_max
 228   if _max<0:
 229     max=tip
 230
 231   c=int(state_cache.get('count',0))
 232   last={}
 233   for rev in range(min,max):
 234     c=export_commit(ui,repo,rev,marks_cache,heads_cache,last,tip,c)
 235
 236   c=export_tags(ui,repo,marks_cache,c)
 237
 238   state_cache['tip']=max
 239   state_cache['count']=c
 240   state_cache['repo']=repourl
 241   save_cache(tipfile,state_cache)