hg2git.py

   1 #!/usr/bin/env python
   2
   3 # Copyright (c) 2007 Rocco Rutte <pdmef@gmx.net>
   4 # License: MIT <http://www.opensource.org/licenses/mit-license.php>
   5
   6 from mercurial import repo,hg,cmdutil,util,ui,revlog,node
   7 import re
   8 import os
   9 import sys
  10
  11 # git branch for hg's default 'HEAD' branch
  12 cfg_master='master'
  13 # silly regex to see if user field has email address
  14 user_re=re.compile('([^<]+) (<[^>]+>)$')
  15 # silly regex to clean out user names
  16 user_clean_re=re.compile('^["]([^"]+)["]$')
  17
  18 def setup_repo(url):
  19   myui=ui.ui(interactive=False)
  20   return myui,hg.repository(myui,url)
  21
  22 def fixup_user(user,authors):
  23   if authors!=None:
  24     # if we have an authors table, try to get mapping
  25     # by defaulting to the current value of 'user'
  26     user=authors.get(user,user)
  27   name,mail,m='','',user_re.match(user)
  28   if m==None:
  29     # if we don't have 'Name <mail>' syntax, use 'user
  30     # <devnull@localhost>' if use contains no at and
  31     # 'user <user>' otherwise
  32     name=user
  33     if '@' not in user:
  34       mail='<devnull@localhost>'
  35     else:
  36       mail='<%s>' % user
  37   else:
  38     # if we have 'Name <mail>' syntax, everything is fine :)
  39     name,mail=m.group(1),m.group(2)
  40
  41   # remove any silly quoting from username
  42   m2=user_clean_re.match(name)
  43   if m2!=None:
  44     name=m2.group(1)
  45   return '%s %s' % (name,mail)
  46
  47 def get_branch(name):
  48   # HEAD may be from CVS imports into hg
  49   if name=='HEAD' or name=='default' or name=='':
  50     name=cfg_master
  51   return name
  52
  53 def get_changeset(ui,repo,revision,authors={}):
  54   node=repo.lookup(revision)
  55   (manifest,user,(time,timezone),files,desc,extra)=repo.changelog.read(node)
  56   tz="%+03d%02d" % (-timezone / 3600, ((-timezone % 3600) / 60))
  57   branch=get_branch(extra.get('branch','master'))
  58   return (node,manifest,fixup_user(user,authors),(time,tz),files,desc,branch,extra)
  59
  60 def load_cache(filename):
  61   cache={}
  62   if not os.path.exists(filename):
  63     return cache
  64   f=open(filename,'r')
  65   l=0
  66   for line in f.readlines():
  67     l+=1
  68     fields=line.split(' ')
  69     if fields==None or not len(fields)==2 or fields[0][0]!=':':
  70       sys.stderr.write('Invalid file format in [%s], line %d\n' % (filename,l))
  71       continue
  72     # put key:value in cache, key without ^:
  73     cache[fields[0][1:]]=fields[1].split('\n')[0]
  74   f.close()
  75   return cache
  76
  77 def save_cache(filename,cache):
  78   f=open(filename,'w+')
  79   map(lambda x: f.write(':%s %s\n' % (str(x),str(cache.get(x)))),cache.keys())
  80   f.close()
  81
  82 def get_git_sha1(name,type='heads'):
  83   try:
  84     # use git-rev-parse to support packed refs
  85     cmd="GIT_DIR='%s' git-rev-parse --verify refs/%s/%s 2>/dev/null" % (os.getenv('GIT_DIR','/dev/null'),type,name)
  86     p=os.popen(cmd)
  87     l=p.readline()
  88     p.close()
  89     if l == None or len(l) == 0:
  90       return None
  91     return l[0:40]
  92   except IOError:
  93     return None