svn2git.py

##############################################################################
#
# Copyright (c) 2012 Zope Foundation and Contributors.
# All Rights Reserved.
#
# This software is subject to the provisions of the Zope Public License,
# Version 2.1 (ZPL).  A copy of the ZPL should accompany this distribution.
# THIS SOFTWARE IS PROVIDED "AS IS" AND ANY AND ALL EXPRESS OR IMPLIED
# WARRANTIES ARE DISCLAIMED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
# WARRANTIES OF TITLE, MERCHANTABILITY, AGAINST INFRINGEMENT, AND FITNESS
# FOR A PARTICULAR PURPOSE.
#
##############################################################################
"""%prog [options] subversion_url

svn to git conversion helper

Based loosely on the svn2git ruby tool.

You must have git and the svn git plugin installed.

A repository directory will be created with a name derived from the
subversion URL.

The subversion URL must point to a subversion project with a standard
layout, with trunk, branches, and tags.
"""

import optparse

parser = optparse.OptionParser(__doc__)

githelp="""
Create a github repository. The argument is a string of the form
DEST/token.  where DEST is either 'users/USERNAME' or 'orgs/ORGNAME' and
token is a github OAuth token:

  https://help.github.com/articles/creating-an-oauth-token-for-command-line-use

curl must be installed if you use this.
"""
bitbuckethelp="""
Create a bitbucket repository. The argument is a string of the form

    user/password

Where user is either your username or the organisation user name.

curl must be installed if you use this.
"""
parser.add_option('--github', '-g', help=githelp)
parser.add_option('--bitbucket', '-b', help=bitbuckethelp)
parser.add_option('--private', '-p', action='store_true',
                  help="Create a private repository (bitbucket only currently)")
parser.add_option('--restart', '-r', action='store_true',
                  help="Restart after fetch times out")
parser.add_option(
    '--authors', '-a',
    help = 'Path to authors.txt file (required)')

import os
import sys
import xml.etree.ElementTree

def s(command):
    print 'Running command:', command
    if os.system(command):
        raise SystemError

def r(command):
    f = os.popen(command)
    result = f.read()
    f.close()
    return result

def main(args=None):
    if args is None:
        args = sys.argv[1:]

    options, args = parser.parse_args(args)
    url, = args
    
    if url.endswith('/'):
        url = url[:-1]
    name = url.rsplit('/', 1)[1]
    
    if options.private and options.github:
        raise NotImplementedError(
            "Currently only --bitbucket supports --private")
    
    if not options.restart:
        os.mkdir(name)
    
    os.chdir(name)

    if not options.restart:
        s('git svn init --no-metadata -s %s' % url)
    assert options.authors
    s('git config svn.authorsfile %s' % options.authors)
    s('git svn fetch')

    for tag in r('svn ls %s/tags' % url).strip().split():
        if tag[-1] == '/':
            tag = tag[:-1]
        f = os.popen('svn log --xml -l1 %s/tags/%s' % (url, tag))
        date = xml.etree.ElementTree.ElementTree(
            file=f).find('logentry').find('date').text
        f.close()
        s("GIT_COMMITTER_DATE=%r git tag %r 'tags/%s'" % (
            date.replace('T', ' ').replace('Z', ' +0000'),
            tag, tag,
            ))

    for branch in r('svn ls %s/branches' % url).strip().split():
        if branch[-1] == '/':
            branch = branch[:-1]
        s('git checkout %s' % branch)
        s('git checkout -b %s' % branch)

        # Not sure if this is necessary, or sufficient. The Ruby
        # version ran into trouble when git left files around between
        # branche checkouts.  I haven't had the problem, with this
        # script, which unlike the Ruby version, doesn't process
        # deleted branches.
        s('git reset --hard HEAD')

    s('git checkout trunk')
    s('git branch -D master')
    s('git checkout -f -b master')
    s('git branch -d -r trunk')
    s('git gc')

    if options.github:
        github_dest, github_token = options.github.rsplit('/', 1)
        if github_dest.startswith('users/'):
            github_user = github_dest[6:]
            github_dest = 'user'
        else:
            assert github_dest.startswith('orgs/')
            github_user = github_dest.split('/')[1]
        s('curl -v -XPOST -H "Authorization: token %s"'
          ' https://api.github.com/%s/repos -d '
          """'{"name": "'"%s"'"}'"""
          % (github_token, github_dest, name))
        s('git remote add origin ssh://[email protected]/%s/%s.git'
          % (github_user, name))
        s("git push -u origin '*' --tags")
    
    if options.bitbucket:
        bitbucket_user, bitbucket_password = options.bitbucket.split('/', 1)
        is_private = 0
        if options.private:
            is_private = 1
        s('curl -k -X POST --user "%s:%s" ' 
            '"https://api.bitbucket.org/1.0/repositories" -d "name=%s" ' 
            '-d "is_private=%d" -d "scm=git"'
            % (bitbucket_user, bitbucket_password, name, is_private))
        s('git remote add origin ssh://[email protected]/%s/%s.git'
          % (bitbucket_user, name))
        s("git push -u origin '*' --tags")

if __name__ == '__main__':
    main()