Files
git-remote-hg/git-remote-hg
2016-08-13 14:27:45 +02:00

1714 lines
51 KiB
Python
Executable File

#!/usr/bin/env python2
#
# Copyright (c) 2012 Felipe Contreras
#
# Inspired by Rocco Rutte's hg-fast-export
# Just copy to your ~/bin, or anywhere in your $PATH.
# Then you can clone with:
# git clone hg::/path/to/mercurial/repo/
#
# For remote repositories a local clone is stored in
# "$GIT_DIR/hg/origin/clone/.hg/".
from mercurial import hg, ui, bookmarks, context, encoding
from mercurial import node, error, extensions, discovery, util
from mercurial import changegroup
import re
import sys
import os
import json
import shutil
import subprocess
import urllib
import atexit
import urlparse
import hashlib
import time as ptime
#
# If you want to see Mercurial revisions as Git commit notes:
# git config core.notesRef refs/notes/hg
#
# If you are not in hg-git-compat mode and want to disable the tracking of
# named branches:
# git config --global remote-hg.track-branches false
#
# If you want the equivalent of hg's clone/pull--insecure option:
# git config --global remote-hg.insecure true
#
# If you want to switch to hg-git compatibility mode:
# git config --global remote-hg.hg-git-compat true
#
# git:
# Sensible defaults for git.
# hg bookmarks are exported as git branches, hg branches are prefixed
# with 'branches/', HEAD is a special case.
#
# hg:
# Emulate hg-git.
# Only hg bookmarks are exported as git branches.
# Commits are modified to preserve hg information and allow bidirectionality.
#
NAME_RE = re.compile('^([^<>]+)')
AUTHOR_RE = re.compile('^([^<>]+?)? ?[<>]([^<>]*)(?:$|>)')
EMAIL_RE = re.compile(r'([^ \t<>]+@[^ \t<>]+)')
AUTHOR_HG_RE = re.compile('^(.*?) ?<(.*?)(?:>(.*))?$')
RAW_AUTHOR_RE = re.compile('^(\w+) (?:(.+)? )?<(.*)> (\d+) ([+-]\d+)')
VERSION = 2
def die(msg, *args):
sys.stderr.write('ERROR: %s\n' % (msg % args))
sys.exit(1)
def warn(msg, *args):
sys.stderr.write('WARNING: %s\n' % (msg % args))
def gitmode(flags):
return 'l' in flags and '120000' or 'x' in flags and '100755' or '100644'
def gittz(tz):
return '%+03d%02d' % (-tz / 3600, -tz % 3600 / 60)
def hgmode(mode):
m = { '100755': 'x', '120000': 'l' }
return m.get(mode, '')
def hghex(n):
return node.hex(n)
def hgbin(n):
return node.bin(n)
def hgref(ref):
return ref.replace('___', ' ')
def gitref(ref):
return ref.replace(' ', '___')
def check_version(*check):
if not hg_version:
return True
return hg_version >= check
def get_config(config):
cmd = ['git', 'config', '--get', config]
process = subprocess.Popen(cmd, stdout=subprocess.PIPE)
output, _ = process.communicate()
return output
def get_config_bool(config, default=False):
value = get_config(config).rstrip('\n')
if value == "true":
return True
elif value == "false":
return False
else:
return default
def rev_parse(rev):
cmd = ['git', 'rev-parse', '--verify', '-q', rev]
process = subprocess.Popen(cmd, stdout=subprocess.PIPE)
output, _ = process.communicate()
return output
def get_rev_hg(commit):
cmd = ['git', 'notes', '--ref', 'refs/notes/hg', 'show', commit]
process = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
hgrev, _ = process.communicate()
return hgrev
class Marks:
def __init__(self, path, repo):
self.path = path
self.repo = repo
self.clear()
self.load()
if self.version < VERSION:
if self.version == 1:
self.upgrade_one()
# upgraded?
if self.version < VERSION:
self.clear()
self.version = VERSION
def clear(self):
self.tips = {}
self.marks = {}
self.rev_marks = {}
self.last_mark = 0
self.version = 0
self.last_note = 0
def load(self):
if not os.path.exists(self.path):
return
tmp = json.load(open(self.path))
self.tips = tmp['tips']
self.marks = tmp['marks']
self.last_mark = tmp['last-mark']
self.version = tmp.get('version', 1)
self.last_note = tmp.get('last-note', 0)
for rev, mark in self.marks.iteritems():
self.rev_marks[mark] = rev
def upgrade_one(self):
def get_id(rev):
return hghex(self.repo.changelog.node(int(rev)))
self.tips = dict((name, get_id(rev)) for name, rev in self.tips.iteritems())
self.marks = dict((get_id(rev), mark) for rev, mark in self.marks.iteritems())
self.rev_marks = dict((mark, get_id(rev)) for mark, rev in self.rev_marks.iteritems())
self.version = 2
def dict(self):
return { 'tips': self.tips, 'marks': self.marks,
'last-mark': self.last_mark, 'version': self.version,
'last-note': self.last_note }
def store(self):
json.dump(self.dict(), open(self.path, 'w'))
def __str__(self):
return str(self.dict())
def from_rev(self, rev):
return self.marks[rev]
def to_rev(self, mark):
return str(self.rev_marks[mark])
def next_mark(self):
self.last_mark += 1
return self.last_mark
def get_mark(self, rev):
self.last_mark += 1
self.marks[rev] = self.last_mark
return self.last_mark
def new_mark(self, rev, mark):
self.marks[rev] = mark
self.rev_marks[mark] = rev
self.last_mark = mark
def is_marked(self, rev):
return rev in self.marks
def get_tip(self, branch):
return str(self.tips[branch])
def set_tip(self, branch, tip):
self.tips[branch] = tip
class ParserContext:
def __init__(self):
# known context attributes
self.localref = None
self.remoteref = None
self.gitmarks = None
self.hghelper = None
self.revs = []
class Parser:
def __init__(self, repo, cmdstream=sys.stdin, ctx=ParserContext()):
self.repo = repo
self.cmdstream = cmdstream
self.line = self.get_line()
self.context = ctx
def get_line(self):
return self.cmdstream.readline().strip()
def __getitem__(self, i):
return self.line.split()[i]
def check(self, word):
return self.line.startswith(word)
def each_block(self, separator):
while self.line != separator:
yield self.line
self.line = self.get_line()
def __iter__(self):
return self.each_block('')
def next(self):
self.line = self.get_line()
if self.line == 'done':
self.line = None
def get_mark(self):
i = self.line.index(':') + 1
return int(self.line[i:])
def get_data(self):
if not self.check('data'):
return None
i = self.line.index(' ') + 1
size = int(self.line[i:])
return self.cmdstream.read(size)
def get_author(self):
ex = None
m = RAW_AUTHOR_RE.match(self.line)
if not m:
return None
_, name, email, date, tz = m.groups()
if name and 'ext:' in name:
m = re.match('^(.+?) ext:\((.+)\)$', name)
if m:
name = m.group(1)
ex = urllib.unquote(m.group(2))
if email != bad_mail:
if name:
user = '%s <%s>' % (name, email)
else:
user = '<%s>' % (email)
else:
user = name
if ex:
user += ex
tz = int(tz)
tz = ((tz / 100) * 3600) + ((tz % 100) * 60)
return (user, int(date), -tz)
def fix_file_path(path):
path = os.path.normpath(path)
if not os.path.isabs(path):
return path
return os.path.relpath(path, '/')
def export_files(files):
final = []
for f in files:
fid = node.hex(f.filenode())
if fid in filenodes:
mark = filenodes[fid]
else:
mark = marks.next_mark()
filenodes[fid] = mark
d = f.data()
print "blob"
print "mark :%u" % mark
print "data %d" % len(d)
print d
path = fix_file_path(f.path())
final.append((gitmode(f.flags()), mark, path))
return final
def get_filechanges(repo, ctx, parent):
modified = set()
added = set()
removed = set()
# load earliest manifest first for caching reasons
prev = parent.manifest().copy()
cur = ctx.manifest()
for fn in cur:
if fn in prev:
if (cur.flags(fn) != prev.flags(fn) or cur[fn] != prev[fn]):
modified.add(fn)
del prev[fn]
else:
added.add(fn)
removed |= set(prev.keys())
return added | modified, removed
def fixup_user_git(user):
name = mail = None
user = user.replace('"', '')
m = AUTHOR_RE.match(user)
if m:
name = m.group(1)
mail = m.group(2).strip()
else:
m = EMAIL_RE.match(user)
if m:
mail = m.group(1)
else:
m = NAME_RE.match(user)
if m:
name = m.group(1).strip()
return (name, mail)
def fixup_user_hg(user):
def sanitize(name):
# stole this from hg-git
return re.sub('[<>\n]', '?', name.lstrip('< ').rstrip('> '))
m = AUTHOR_HG_RE.match(user)
if m:
name = sanitize(m.group(1))
mail = sanitize(m.group(2))
ex = m.group(3)
if ex:
name += ' ext:(' + urllib.quote(ex) + ')'
else:
name = sanitize(user)
if '@' in user:
mail = name
else:
mail = None
return (name, mail)
def fixup_user(user):
if mode == 'git':
name, mail = fixup_user_git(user)
else:
name, mail = fixup_user_hg(user)
if not name:
name = bad_name
if not mail:
mail = bad_mail
return '%s <%s>' % (name, mail)
def updatebookmarks(repo, peer):
remotemarks = peer.listkeys('bookmarks')
# delete bookmarks locally that disappeared on remote
localmarks = bookmarks.listbookmarks(repo)
remote = set(remotemarks.keys())
local = set(localmarks.keys())
for bmark in local - remote:
bookmarks.pushbookmark(repo, bmark, localmarks[bmark], '')
# also delete private ref
pbookmark = '%s/bookmarks/%s' % (prefix, bmark)
subprocess.call(['git', 'update-ref', '-d', pbookmark])
# now add or update remote bookmarks to local, if any
localmarks = repo._bookmarks
if not remotemarks:
return
for k, v in remotemarks.iteritems():
localmarks[k] = hgbin(v)
if check_version(3, 6):
lock = tr = None
try:
lock = repo.lock()
tr = repo.transaction('bookmark')
localmarks.recordchange(tr)
tr.close()
finally:
if tr is not None:
tr.release()
if lock is not None:
lock.release()
else:
if hasattr(localmarks, 'write'):
localmarks.write()
else:
bookmarks.write(repo)
def get_repo(url, alias):
global peer
myui = ui.ui()
myui.setconfig('ui', 'interactive', 'off')
myui.fout = sys.stderr
if get_config_bool('remote-hg.insecure'):
myui.setconfig('web', 'cacerts', '')
extensions.loadall(myui)
if hg.islocal(url) and not os.environ.get('GIT_REMOTE_HG_TEST_REMOTE'):
repo = hg.repository(myui, url)
if not os.path.exists(dirname):
os.makedirs(dirname)
else:
shared_path = os.path.join(gitdir, 'hg')
# check and upgrade old organization
hg_path = os.path.join(shared_path, '.hg')
if os.path.exists(shared_path) and not os.path.exists(hg_path):
repos = os.listdir(shared_path)
for x in repos:
local_hg = os.path.join(shared_path, x, 'clone', '.hg')
if not os.path.exists(local_hg):
continue
if not os.path.exists(hg_path):
shutil.move(local_hg, hg_path)
shutil.rmtree(os.path.join(shared_path, x, 'clone'))
# setup shared repo (if not there)
try:
hg.peer(myui, {}, shared_path, create=True)
except error.RepoError:
pass
if not os.path.exists(dirname):
os.makedirs(dirname)
local_path = os.path.join(dirname, 'clone')
if not os.path.exists(local_path):
hg.share(myui, shared_path, local_path, update=False)
else:
# make sure the shared path is always up-to-date
util.writefile(os.path.join(local_path, '.hg', 'sharedpath'), hg_path)
repo = hg.repository(myui, local_path)
try:
peer = hg.peer(repo.ui, {}, url)
except:
die('Repository error')
if check_version(3, 0):
from mercurial import exchange
exchange.pull(repo, peer, heads=None, force=True)
else:
repo.pull(peer, heads=None, force=True)
updatebookmarks(repo, peer)
return repo
def rev_to_mark(rev):
return marks.from_rev(rev.hex())
def mark_to_rev(mark):
return marks.to_rev(mark)
# Get a range of revisions in the form of a..b (git committish)
def gitrange(repo, a, b):
positive = []
pending = set([int(b)])
negative = int(a)
for cur in xrange(b, -1, -1):
if not pending:
break
parents = [p for p in repo.changelog.parentrevs(cur) if p >= 0]
if cur in pending:
if cur > negative:
positive.append(cur)
pending.remove(cur)
for p in parents:
if p > negative:
pending.add(p)
positive.reverse()
return positive
def export_ref(repo, name, kind, head):
ename = '%s/%s' % (kind, name)
try:
tip = marks.get_tip(ename)
tip = repo[tip]
except:
tip = repo[-1]
revs = gitrange(repo, tip, head)
total = len(revs)
tip = tip.rev()
for rev in revs:
c = repo[rev]
node = c.node()
if marks.is_marked(c.hex()):
continue
(manifest, user, (time, tz), files, desc, extra) = repo.changelog.read(node)
rev_branch = extra['branch']
author = "%s %d %s" % (fixup_user(user), time, gittz(tz))
if 'committer' in extra:
try:
cuser, ctime, ctz = extra['committer'].rsplit(' ', 2)
committer = "%s %s %s" % (fixup_user(cuser), ctime, gittz(int(ctz)))
except ValueError:
cuser = extra['committer']
committer = "%s %d %s" % (fixup_user(cuser), time, gittz(tz))
else:
committer = author
parents = [repo[p] for p in repo.changelog.parentrevs(rev) if p >= 0]
if len(parents) == 0:
modified = c.manifest().keys()
removed = []
else:
modified, removed = get_filechanges(repo, c, parents[0])
desc += '\n'
if mode == 'hg':
extra_msg = ''
if rev_branch != 'default':
extra_msg += 'branch : %s\n' % rev_branch
renames = []
for f in c.files():
if f not in c.manifest():
continue
rename = c.filectx(f).renamed()
if rename:
renames.append((rename[0], f))
for e in renames:
extra_msg += "rename : %s => %s\n" % e
for key, value in extra.iteritems():
if key in ('author', 'committer', 'encoding', 'message', 'branch', 'hg-git'):
continue
else:
extra_msg += "extra : %s : %s\n" % (key, urllib.quote(value))
if extra_msg:
desc += '\n--HG--\n' + extra_msg
if len(parents) == 0 and rev:
print 'reset %s/%s' % (prefix, ename)
modified_final = export_files(c.filectx(f) for f in modified)
print "commit %s/%s" % (prefix, ename)
print "mark :%d" % (marks.get_mark(c.hex()))
print "author %s" % (author)
print "committer %s" % (committer)
print "data %d" % (len(desc))
print desc
if len(parents) > 0:
print "from :%s" % (rev_to_mark(parents[0]))
if len(parents) > 1:
print "merge :%s" % (rev_to_mark(parents[1]))
for f in removed:
print "D %s" % (fix_file_path(f))
for f in modified_final:
print "M %s :%u %s" % f
print
progress = (rev - tip)
if (progress % 100 == 0):
print "progress revision %d '%s' (%d/%d)" % (rev, name, progress, total)
# make sure the ref is updated
print "reset %s/%s" % (prefix, ename)
print "from :%u" % rev_to_mark(head)
print
pending_revs = set(revs) - notes
if pending_revs:
note_mark = marks.next_mark()
ref = "refs/notes/hg"
print "commit %s" % ref
print "mark :%d" % (note_mark)
print "committer remote-hg <> %d %s" % (ptime.time(), gittz(ptime.timezone))
desc = "Notes for %s\n" % (name)
print "data %d" % (len(desc))
print desc
# continue incrementally on current notes branch (whenever possible)
# to avoid wiping out present content upon fetch of new repo
current_note = rev_parse(ref)
if current_note and not len(notes):
print 'from %s^0' % (ref)
# but track along with the previous ref as import goes along
elif marks.last_note:
print "from :%u" % (marks.last_note)
for rev in pending_revs:
notes.add(rev)
c = repo[rev]
print "N inline :%u" % rev_to_mark(c)
msg = c.hex()
print "data %d" % (len(msg))
print msg
print
marks.last_note = note_mark
marks.set_tip(ename, head.hex())
def export_tag(repo, tag):
export_ref(repo, tag, 'tags', repo[hgref(tag)])
def export_bookmark(repo, bmark):
head = bmarks[hgref(bmark)]
export_ref(repo, bmark, 'bookmarks', head)
def export_branch(repo, branch):
tip = get_branch_tip(repo, branch)
head = repo[tip]
export_ref(repo, branch, 'branches', head)
def export_head(repo):
export_ref(repo, g_head[0], 'bookmarks', g_head[1])
def do_capabilities(parser):
print "import"
if capability_push:
print "push"
else:
print "export"
print "refspec refs/heads/branches/*:%s/branches/*" % prefix
print "refspec refs/heads/*:%s/bookmarks/*" % prefix
print "refspec refs/tags/*:%s/tags/*" % prefix
path = os.path.join(dirname, 'marks-git')
if os.path.exists(path):
print "*import-marks %s" % path
print "*export-marks %s" % path
print "option"
# nothing really depends on the private refs being up to date
# (export is limited anyway by the current git marks)
# and they are not always updated correctly (dry-run, bookmark delete, ...)
# (might resolve some dry-run breakage also)
print "no-private-update"
print
def branch_tip(branch):
return branches[branch][-1]
def get_branch_tip(repo, branch):
heads = branches.get(hgref(branch), None)
if not heads:
return None
# verify there's only one head
if (len(heads) > 1):
warn("Branch '%s' has more than one head, consider merging" % branch)
return branch_tip(hgref(branch))
return heads[0]
def list_head(repo, cur):
global g_head, fake_bmark
if 'default' not in branches:
# empty repo
return
node = repo[branch_tip('default')]
head = 'master' if 'master' not in bmarks else 'default'
fake_bmark = head
bmarks[head] = node
head = gitref(head)
print "@refs/heads/%s HEAD" % head
g_head = (head, node)
def do_list(parser):
repo = parser.repo
for bmark, node in bookmarks.listbookmarks(repo).iteritems():
bmarks[bmark] = repo[node]
cur = repo.dirstate.branch()
orig = peer if peer else repo
for branch, heads in orig.branchmap().iteritems():
# only open heads
heads = [h for h in heads if 'close' not in repo.changelog.read(h)[5]]
if heads:
branches[branch] = heads
list_head(repo, cur)
# for export command a ref's old_sha1 is taken from private namespace ref
# for push command a fake one is provided
# that avoids having the ref status reported as a new branch/tag
# (though it will be marked as FETCH_FIRST prior to push,
# but that's ok as we will provide proper status)
for_push = (parser.line.find('for-push') >= 0)
sha1 = 'f' * 40 if (capability_push and for_push) else '?'
if track_branches:
for branch in branches:
print "%s refs/heads/branches/%s" % (sha1, gitref(branch))
for bmark in bmarks:
if bmarks[bmark].hex() == '0' * 40:
warn("Ignoring invalid bookmark '%s'", bmark)
else:
print "%s refs/heads/%s" % (sha1, gitref(bmark))
for tag, node in repo.tagslist():
if tag == 'tip':
continue
print "%s refs/tags/%s" % (sha1, gitref(tag))
print
def do_import(parser):
repo = parser.repo
path = os.path.join(dirname, 'marks-git')
print "feature done"
if os.path.exists(path):
print "feature import-marks=%s" % path
print "feature export-marks=%s" % path
print "feature force"
sys.stdout.flush()
tmp = encoding.encoding
encoding.encoding = 'utf-8'
# lets get all the import lines
while parser.check('import'):
ref = parser[1]
if (ref == 'HEAD'):
export_head(repo)
elif ref.startswith('refs/heads/branches/'):
branch = ref[len('refs/heads/branches/'):]
export_branch(repo, branch)
elif ref.startswith('refs/heads/'):
bmark = ref[len('refs/heads/'):]
export_bookmark(repo, bmark)
elif ref.startswith('refs/tags/'):
tag = ref[len('refs/tags/'):]
export_tag(repo, tag)
parser.next()
encoding.encoding = tmp
print 'done'
def parse_blob(parser):
parser.next()
mark = parser.get_mark()
parser.next()
data = parser.get_data()
blob_marks[mark] = data
parser.next()
def get_file_metadata(repo, p1, files):
for e in files:
f = files[e]
if 'rename' in f and 'mode' not in f:
old = f['rename']
ctx = repo[p1][old]
m = ''
# .isexec(), .islink() only around in newer versions
if hasattr(ctx, 'isexec'):
m = ctx.isexec()
else:
m = ('x' in ctx.flags())
if hasattr(ctx, 'islink'):
m = ctx.islink()
else:
m = ('l' in ctx.flags())
f['mode'] = m
f['data'] = ctx.data()
def get_merge_files(repo, p1, p2, files):
for e in repo[p1].files():
if e not in files:
if e not in repo[p1].manifest():
continue
f = { 'ctx': repo[p1][e] }
files[e] = f
def split_line_pathnames(line):
if line[2] != '"':
return line.split(' ', 2)
else:
t = line[0]
p = 3
while p >= 0:
if line[p] == '"' and line[p - 1] != '\\':
return line[0], line[2:p+1], line[p+2:]
p = line.find('"', p + 1)
# hm, should not happen
die('Malformed file command: %s' % (line))
def c_style_unescape(string):
if string[0] == string[-1] == '"':
return string.decode('string-escape')[1:-1]
return string
def parse_commit(parser):
from_mark = merge_mark = None
remoteref = parser.context.remoteref
ref = parser[1] if not remoteref else remoteref
parser.next()
commit_mark = parser.get_mark()
parser.next()
author = parser.get_author()
parser.next()
committer = parser.get_author()
parser.next()
data = parser.get_data()
parser.next()
if parser.check('from'):
from_mark = parser.get_mark()
parser.next()
if parser.check('merge'):
merge_mark = parser.get_mark()
parser.next()
if parser.check('merge'):
die('octopus merges are not supported yet')
# fast-export adds an extra newline
if data[-1] == '\n':
data = data[:-1]
files = {}
for line in parser:
if parser.check('M'):
t, m, mark_ref, path = line.split(' ', 3)
mark = int(mark_ref[1:])
f = { 'mode': hgmode(m), 'data': blob_marks[mark] }
elif parser.check('D'):
t, path = line.split(' ', 1)
f = { 'deleted': True }
elif parser.check('R'):
t, old, path = split_line_pathnames(line)
old = c_style_unescape(old)
f = { 'rename': old }
# also mark old deleted
files[old] = { 'deleted': True }
elif parser.check('C'):
t, old, path = split_line_pathnames(line)
f = { 'rename': c_style_unescape(old) }
else:
die('Unknown file command: %s' % line)
path = c_style_unescape(path)
files[path] = files.get(path, {})
files[path].update(f)
# only export the commits if we are on an internal proxy repo
if dry_run and not peer:
parsed_refs[ref] = None
return
# check if this is an hg commit we have in some other repo
gitmarks = parser.context.gitmarks
if gitmarks:
gitcommit = gitmarks.to_rev(commit_mark)
hgrev = get_rev_hg(gitcommit)
if hgrev:
hghelper = parser.context.hghelper
if not hghelper:
print "error %s rejected not pushing hg based commit %s" % (ref, gitcommit)
raise UserWarning("check-hg-commits")
# must be in some local repo
# find it and push it to the target local repo
# (rather than making a commit into it)
# probably not already in target repo, but let's make sure
if hgrev not in parser.repo:
srepo = hghelper.githgrepo.find_hg_repo(hgrev)
if not srepo:
# pretty bad, if identified as hg revision, we should have it somewhere
# but is possible if the originating repo has been removed now
# warn elaborately and fail given the current settings
description = "\n" \
"commit %s corresponds \nto hg revision %s,\n" \
"but could not find latter in any fetched hg repo.\n" \
"Please resolve the inconsistency or disable pushing hg commits" \
% (gitcommit, hgrev)
die(description)
warn('Pushing hg changeset %s for %s' % (hgrev, gitcommit))
# target is local repo so should have a root
# force push since otherwise forcibly commit anyway
# (and needed for multiple head case etc)
push(srepo, hg.peer(srepo.ui, {}, parser.repo.root), [hgbin(hgrev)], True)
else:
# could already be present, particularly in shared proxy repo
warn('Using hg changeset %s for %s' % (hgrev, gitcommit))
# track mark and are done here
parsed_refs[ref] = hgrev
marks.new_mark(hgrev, commit_mark)
return
def getfilectx(repo, memctx, f):
of = files[f]
if 'deleted' in of:
if check_version(3, 2):
return None
else:
raise IOError
if 'ctx' in of:
if mode == 'hg':
ctx = of['ctx']
is_exec = ctx.isexec()
is_link = ctx.islink()
if check_version(3, 1):
return context.memfilectx(repo, f, ctx.data(),
is_link, is_exec)
else:
return context.memfilectx(f, ctx.data(),
is_link, is_exec)
else:
return of['ctx']
is_exec = of['mode'] == 'x'
is_link = of['mode'] == 'l'
rename = of.get('rename', None)
if check_version(3, 1):
return context.memfilectx(repo, f, of['data'],
is_link, is_exec, rename)
else:
return context.memfilectx(f, of['data'],
is_link, is_exec, rename)
repo = parser.repo
user, date, tz = author
extra = {}
if committer != author:
extra['committer'] = "%s %u %u" % committer
if from_mark:
p1 = mark_to_rev(from_mark)
else:
p1 = '0' * 40
if merge_mark:
p2 = mark_to_rev(merge_mark)
else:
p2 = '0' * 40
#
# If files changed from any of the parents, hg wants to know, but in git if
# nothing changed from the first parent, nothing changed.
#
if merge_mark:
get_merge_files(repo, p1, p2, files)
# need to obtain file metadata for copied and renamed files that have
# no filemodify line; let's get that from the old file in parent revision
if from_mark:
get_file_metadata(repo, p1, files)
# Check if the ref is supposed to be a named branch
if ref.startswith('refs/heads/branches/'):
branch = ref[len('refs/heads/branches/'):]
extra['branch'] = hgref(branch)
if mode == 'hg':
i = data.find('\n--HG--\n')
if i >= 0:
tmp = data[i + len('\n--HG--\n'):].strip()
for k, v in [e.split(' : ', 1) for e in tmp.split('\n')]:
if k == 'rename':
old, new = v.split(' => ', 1)
files[new]['rename'] = old
elif k == 'branch':
extra[k] = v
elif k == 'extra':
ek, ev = v.split(' : ', 1)
extra[ek] = urllib.unquote(ev)
data = data[:i]
ctx = context.memctx(repo, (p1, p2), data,
files.keys(), getfilectx,
user, (date, tz), extra)
tmp = encoding.encoding
encoding.encoding = 'utf-8'
node = hghex(repo.commitctx(ctx))
encoding.encoding = tmp
parsed_refs[ref] = node
marks.new_mark(node, commit_mark)
parser.context.revs.append(node)
def parse_reset(parser):
remoteref = parser.context.remoteref
ref = parser[1] if not remoteref else remoteref
parser.next()
# ugh
if parser.check('commit'):
parse_commit(parser)
return
if not parser.check('from'):
return
from_mark = parser.get_mark()
parser.next()
try:
rev = mark_to_rev(from_mark)
except KeyError:
rev = None
parsed_refs[ref] = rev
def parse_tag(parser):
name = parser[1]
parser.next()
from_mark = parser.get_mark()
parser.next()
tagger = parser.get_author()
parser.next()
data = parser.get_data()
parser.next()
parsed_tags[name] = (tagger, data)
def write_tag(repo, tag, node, msg, author):
branch = repo[node].branch()
tip = branch_tip(branch)
tip = repo[tip]
def getfilectx(repo, memctx, f):
try:
fctx = tip.filectx(f)
data = fctx.data()
except error.ManifestLookupError:
data = ""
content = data + "%s %s\n" % (node, tag)
if check_version(3, 1):
return context.memfilectx(repo, f, content, False, False, None)
else:
return context.memfilectx(f, content, False, False, None)
p1 = tip.hex()
p2 = '0' * 40
if author:
user, date, tz = author
date_tz = (date, tz)
else:
cmd = ['git', 'var', 'GIT_COMMITTER_IDENT']
process = subprocess.Popen(cmd, stdout=subprocess.PIPE)
output, _ = process.communicate()
m = re.match('^.* <.*>', output)
if m:
user = m.group(0)
else:
user = repo.ui.username()
date_tz = None
ctx = context.memctx(repo, (p1, p2), msg,
['.hgtags'], getfilectx,
user, date_tz, {'branch': branch})
tmp = encoding.encoding
encoding.encoding = 'utf-8'
tagnode = repo.commitctx(ctx)
encoding.encoding = tmp
return (tagnode, branch)
def checkheads_bmark(repo, ref, ctx, force):
bmark = ref[len('refs/heads/'):]
if bmark not in bmarks:
# new bmark
return True
ctx_old = bmarks[bmark]
ctx_new = ctx
if not ctx.rev():
print "error %s unknown" % ref
return False
if not repo.changelog.descendant(ctx_old.rev(), ctx_new.rev()):
if force:
print "ok %s forced update" % ref
else:
print "error %s non-fast forward" % ref
return False
return True
def checkheads(repo, remote, p_revs, force):
remotemap = remote.branchmap()
if not remotemap:
# empty repo
return True
new = {}
ret = True
for node, ref in p_revs.iteritems():
ctx = repo[node]
branch = ctx.branch()
if branch not in remotemap:
# new branch
continue
if not ref.startswith('refs/heads/branches'):
if ref.startswith('refs/heads/'):
if not checkheads_bmark(repo, ref, ctx, force):
ret = False
# only check branches
continue
new.setdefault(branch, []).append(ctx.rev())
for branch, heads in new.iteritems():
old = [repo.changelog.rev(x) for x in remotemap[branch]]
for rev in heads:
if check_version(2, 3):
ancestors = repo.changelog.ancestors([rev], stoprev=min(old))
else:
ancestors = repo.changelog.ancestors(rev)
found = False
for x in old:
if x in ancestors:
found = True
break
if found:
continue
node = repo.changelog.node(rev)
ref = p_revs[node]
if force:
print "ok %s forced update" % ref
else:
print "error %s non-fast forward" % ref
ret = False
return ret
def push_unsafe(repo, remote, p_revs, force):
fci = discovery.findcommonincoming
commoninc = fci(repo, remote, force=force)
common, _, remoteheads = commoninc
fco = discovery.findcommonoutgoing
outgoing = fco(repo, remote, onlyheads=list(p_revs), commoninc=commoninc, force=force)
# nice to know about this rather than assume a bogus error
# also, some remote peertypes might otherwise be surprised further down
if not outgoing.missing:
return None
if check_version(3, 2):
cg = changegroup.getchangegroup(repo, 'push', heads=list(p_revs), common=common)
elif check_version(3, 0):
cg = changegroup.getbundle(repo, 'push', heads=list(p_revs), common=common)
else:
cg = repo.getbundle('push', heads=list(p_revs), common=common)
unbundle = remote.capable('unbundle')
if unbundle:
if force:
remoteheads = ['force']
ret = remote.unbundle(cg, remoteheads, 'push')
else:
ret = remote.addchangegroup(cg, 'push', repo.url())
phases = remote.listkeys('phases')
if phases:
for head in p_revs:
# update to public
remote.pushkey('phases', hghex(head), '1', '0')
return ret
def push(repo, remote, p_revs, force):
if hasattr(remote, 'canpush') and not remote.canpush():
print "error cannot push"
if not p_revs:
# nothing to push
return
lock = None
unbundle = remote.capable('unbundle')
if not unbundle:
lock = remote.lock()
try:
ret = push_unsafe(repo, remote, p_revs, force)
finally:
if lock is not None:
lock.release()
return ret
def check_tip(ref, kind, name, heads):
try:
ename = '%s/%s' % (kind, name)
tip = marks.get_tip(ename)
except KeyError:
return True
else:
return tip in heads
def do_export(parser):
do_push_hg(parser)
print
def do_push_hg(parser):
global parsed_refs, parsed_tags
p_bmarks = []
p_revs = {}
parsed_refs = {}
parsed_tags = {}
parser.next()
remoteref = parser.context.remoteref
if remoteref and not parser.line:
# if remoteref is in past exported
# git-fast-export might not produce anything at all
# that is ok'ish, we will determine parsed_ref another way
localref = parser.context.localref
hgrev = get_rev_hg(localref)
if not hgrev:
# maybe the notes are not updated
# happens only on fetch for now ... let's ask for that
print "error %s fetch first" % remoteref
return False
parsed_refs[remoteref] = hgrev
# now make parser happy
parser.line = 'done'
for line in parser.each_block('done'):
if parser.check('blob'):
parse_blob(parser)
elif parser.check('commit'):
parse_commit(parser)
elif parser.check('reset'):
parse_reset(parser)
elif parser.check('tag'):
parse_tag(parser)
elif parser.check('feature'):
pass
else:
die('unhandled export command: %s' % line)
need_fetch = False
for ref, node in parsed_refs.iteritems():
bnode = hgbin(node) if node else None
if ref.startswith('refs/heads/branches'):
branch = ref[len('refs/heads/branches/'):]
if branch in branches and bnode in branches[branch]:
# up to date
print "ok %s up to date" % ref
continue
if peer:
remotemap = peer.branchmap()
if remotemap and branch in remotemap:
heads = [hghex(e) for e in remotemap[branch]]
if not check_tip(ref, 'branches', branch, heads):
print "error %s fetch first" % ref
need_fetch = True
continue
p_revs[bnode] = ref
print "ok %s" % ref
elif ref.startswith('refs/heads/'):
bmark = ref[len('refs/heads/'):]
new = node
old = bmarks[bmark].hex() if bmark in bmarks else ''
if old == new:
print "ok %s up to date" % ref
continue
print "ok %s" % ref
if bmark != fake_bmark and \
not (bmark == 'master' and bmark not in parser.repo._bookmarks):
p_bmarks.append((ref, bmark, old, new))
if peer:
remote_old = peer.listkeys('bookmarks').get(bmark)
if remote_old:
if not check_tip(ref, 'bookmarks', bmark, remote_old):
print "error %s fetch first" % ref
need_fetch = True
continue
p_revs[bnode] = ref
elif ref.startswith('refs/tags/'):
if dry_run:
print "ok %s" % ref
continue
tag = ref[len('refs/tags/'):]
tag = hgref(tag)
author, msg = parsed_tags.get(tag, (None, None))
if mode == 'git':
if not msg:
msg = 'Added tag %s for changeset %s' % (tag, node[:12])
tagnode, branch = write_tag(parser.repo, tag, node, msg, author)
p_revs[tagnode] = 'refs/heads/branches/' + gitref(branch)
else:
fp = parser.repo.opener('localtags', 'a')
fp.write('%s %s\n' % (node, tag))
fp.close()
p_revs[bnode] = ref
print "ok %s" % ref
else:
# transport-helper/fast-export bugs
continue
if need_fetch:
return
if dry_run:
if peer and not force_push:
checkheads(parser.repo, peer, p_revs, force_push)
return
success = True
if peer:
if not checkheads(parser.repo, peer, p_revs, force_push):
return False
ret = push(parser.repo, peer, p_revs, force_push)
# None ok: nothing to push
if ret != None and not ret:
# do not update bookmarks
return
# update remote bookmarks
remote_bmarks = peer.listkeys('bookmarks')
for ref, bmark, old, new in p_bmarks:
if force_push:
old = remote_bmarks.get(bmark, '')
if not peer.pushkey('bookmarks', bmark, old, new):
success = False
print "error %s" % ref
else:
# update local bookmarks
for ref, bmark, old, new in p_bmarks:
if not bookmarks.pushbookmark(parser.repo, bmark, old, new):
success = False
print "error %s" % ref
return success
def delete_bookmark(parser, ref):
bmark = ref[len('refs/heads/'):]
if bmark == fake_bmark:
return False
# delete local (proxy or target)
old = bmarks[bmark].hex() if bmark in bmarks else ''
if not old:
return False
ok = False
if old:
ok = bookmarks.pushbookmark(parser.repo, bmark, old, '')
# propagate to peer if appropriate
if ok and peer:
remote_bmarks = peer.listkeys('bookmarks')
old = remote_bmarks.get(bmark, '')
ok = peer.pushkey('bookmarks', bmark, old, '')
# delete private ref
if ok:
pbookmark = '%s/bookmarks/%s' % (prefix, bmark)
subprocess.call(['git', 'update-ref', '-d', pbookmark])
return ok
def do_push_refspec(parser, refspec, revs):
global force_push
force = (refspec[0] == '+')
refs = refspec.strip('+').split(':')
# check for delete
if (not refs[0]) and refs[1].startswith('refs/heads') and \
not refs[1].startswith('refs/heads/branches'):
if not delete_bookmark(parser, refs[1]):
print "error %s could not delete"% (refs[1])
else:
print "ok %s" % (refs[1])
return
# sanity check on remote ref
if not (refs[1].startswith('refs/heads') or refs[1].startswith('refs/tags')):
print "error %s refspec not supported " % refs[1]
return
ctx = ParserContext()
if refs[0] != refs[1]:
# would work and tag as requested, but pushing to a hg permanent branch
# based on a rename rather than a git branch is probably not a good idea
if refs[1].startswith('refs/heads/branches'):
print "error %s not allowed for permanent branch" % refs[1]
return
ctx.remoteref = refs[1]
ctx.localref = refs[0]
# ok, fire up git-fast-export and process it
cmd = ['git', 'fast-export', '--use-done-feature']
fast_export_options = get_config('remote-hg.fast-export-options')
if not fast_export_options:
fast_export_options = '-M -C'
cmd.extend(fast_export_options.strip().split())
marks = os.path.join(dirname, 'marks-git')
if os.path.exists(marks):
cmd.append('--import-marks=%s' % marks)
# optionally reuse existing hg commits in local repos
check_hg_commits = get_config('remote-hg.check-hg-commits').strip()
use_hg_commits = check_hg_commits in ('fail', 'push')
# no commit of marks if dry-dry_run
# and only commit if all went ok,
# otherwise some commits may no longer be exported next time/try around
tmpmarks = ''
if use_hg_commits or not dry_run:
tmpmarks = os.path.join(dirname, 'marks-git-%d' % (os.getpid()))
cmd.append('--export-marks=%s' % tmpmarks)
cmd.append(refs[0])
# a parameter would obviously be nicer here ...
force_push = force
ok = False
tmpfastexport = None
try:
if use_hg_commits:
# we need the mapping from marks to commit
# so store the output first to a file (and marks get saved also),
# and then process that file
tmpfastexport = open(os.path.join(dirname, 'git-fast-export-%d' % (os.getpid())), 'w+b')
subprocess.check_call(cmd, stdin=None, stdout=tmpfastexport)
try:
import imp
ctx.hghelper = imp.load_source('hghelper', \
os.path.join(os.path.dirname(__file__), 'git-hg-helper'))
ctx.hghelper.init_git(gitdir)
ctx.gitmarks = ctx.hghelper.GitMarks(tmpmarks)
# let processing know it should not bother pushing if not requested
if check_hg_commits != 'push':
ctx.hghelper = None
except:
die("check-hg-commits setup failed; is git-hg-helper also installed?")
tmpfastexport.seek(0)
try:
nparser = Parser(parser.repo, tmpfastexport, ctx)
ok = do_push_hg(nparser)
except UserWarning:
ok = False
else:
# simply feed fast-export directly to processing
export = subprocess.Popen(cmd, stdin=None, stdout=subprocess.PIPE)
nparser = Parser(parser.repo, export.stdout, ctx)
ok = do_push_hg(nparser)
finally:
if tmpmarks and os.path.exists(tmpmarks):
if ok and not dry_run:
# the commits made it through, now we can commit
os.rename(tmpmarks, marks)
revs[:] = nparser.context.revs
else:
os.remove(tmpmarks)
if tmpfastexport and os.path.exists(tmpfastexport.name):
tmpfastexport.close()
os.remove(tmpfastexport.name)
def update_notes(revs, desc):
if revs:
# spin up fast-import
gitmarks = os.path.join(dirname, 'marks-git')
# marks should exist by now
# no export of marks since notes commits are not relevant
proc = subprocess.Popen(['git', 'fast-import', '--done', '--quiet',
'--import-marks=%s' % gitmarks], stdin=subprocess.PIPE, stdout=sys.stderr)
# now feed fast-import
dest = proc.stdin
note_mark = marks.next_mark()
ref = "refs/notes/hg"
dest.write("commit %s\n" % ref)
dest.write("mark :%d\n" % (note_mark))
dest.write("committer remote-hg <> %d %s\n" % (ptime.time(), gittz(ptime.timezone)))
dest.write("data %d\n" % (len(desc)))
dest.write(desc + '\n')
current_note = rev_parse(ref)
if current_note:
dest.write('from %s^0\n' % (ref))
for rev in revs:
dest.write("N inline :%u\n" % marks.from_rev(rev))
dest.write("data %d\n" % (len(rev)))
dest.write(rev + '\n')
dest.write('\n')
dest.write('done\n')
dest.flush()
proc.wait()
def do_push(parser):
if os.environ.get('GIT_REMOTE_HG_DEBUG_PUSH'):
dump = ''
for line in parser:
dump += line + '\n'
die('DEBUG push:\n%s' % (dump))
revs = []
for line in parser:
if parser.check('push'):
localrevs = []
do_push_refspec(parser, line.lstrip('push '), localrevs)
revs.extend(localrevs)
else:
die('unhandled push command: %s' % (line))
print
# at this stage, all external processes are done, marks files written
# so we can use those do update notes if so desired
if get_config_bool('remote-hg.push-updates-notes') and revs:
update_notes(revs, "Update notes on push")
def do_option(parser):
global dry_run, force_push
_, key, value = parser.line.split(' ')
if key == 'dry-run':
dry_run = (value == 'true')
print 'ok'
elif key == 'force':
force_push = (value == 'true')
print 'ok'
else:
print 'unsupported'
def fix_path(alias, repo, orig_url):
url = urlparse.urlparse(orig_url, 'file')
if url.scheme != 'file' or os.path.isabs(os.path.expanduser(url.path)):
return
abs_url = urlparse.urljoin("%s/" % os.getcwd(), orig_url)
cmd = ['git', 'config', 'remote.%s.url' % alias, "hg::%s" % abs_url]
subprocess.call(cmd)
def select_private_refs(alias):
show_private_refs = get_config_bool('remote-hg.show-private-refs', False)
# selection is easy, but let's also clean the refs of the alternative
# in any case, will be recreated along the way as and when needed
if show_private_refs:
path = "%s/refs" % (dirname)
if os.path.exists(path):
shutil.rmtree(path, True)
# in refs space
return 'refs/hg/%s' % alias
else:
refs = subprocess.Popen(['git', 'for-each-ref', \
'--format=delete %(refname)', 'refs/hg'], stdout=subprocess.PIPE)
update = subprocess.Popen(['git', 'update-ref', '--stdin'], stdin=refs.stdout)
refs.stdout.close() # helps with SIGPIPE
update.communicate()
# keep private implementation refs really private
return 'hg/%s/refs' % alias
def main(args):
global prefix, gitdir, dirname, branches, bmarks
global marks, blob_marks, parsed_refs
global peer, mode, bad_mail, bad_name
global track_branches, force_push, is_tmp
global parsed_tags
global filenodes
global fake_bmark, hg_version
global dry_run
global notes, alias
global capability_push
marks = None
is_tmp = False
gitdir = os.environ.get('GIT_DIR', None)
if len(args) < 3:
die('Not enough arguments.')
if not gitdir:
die('GIT_DIR not set')
alias = args[1]
url = args[2]
peer = None
hg_git_compat = get_config_bool('remote-hg.hg-git-compat')
track_branches = get_config_bool('remote-hg.track-branches', True)
capability_push = get_config_bool('remote-hg.capability-push', True)
force_push = False
if hg_git_compat:
mode = 'hg'
bad_mail = 'none@none'
bad_name = ''
else:
mode = 'git'
bad_mail = 'unknown'
bad_name = 'Unknown'
if alias[4:] == url:
is_tmp = True
alias = hashlib.sha1(alias).hexdigest()
dirname = os.path.join(gitdir, 'hg', alias)
branches = {}
bmarks = {}
blob_marks = {}
filenodes = {}
fake_bmark = None
try:
version, _, extra = util.version().partition('+')
version = list(int(e) for e in version.split('.'))
if extra:
version[1] += 1
hg_version = tuple(version)
except:
hg_version = None
dry_run = False
notes = set()
prefix = select_private_refs(alias)
repo = get_repo(url, alias)
if not is_tmp:
fix_path(alias, peer or repo, url)
marks_path = os.path.join(dirname, 'marks-hg')
marks = Marks(marks_path, repo)
if sys.platform == 'win32':
import msvcrt
msvcrt.setmode(sys.stdout.fileno(), os.O_BINARY)
parser = Parser(repo)
for line in parser:
if parser.check('capabilities'):
do_capabilities(parser)
elif parser.check('list'):
do_list(parser)
elif parser.check('import'):
do_import(parser)
elif parser.check('export'):
do_export(parser)
elif parser.check('push'):
do_push(parser)
elif parser.check('option'):
do_option(parser)
else:
die('unhandled command: %s' % line)
sys.stdout.flush()
marks.store()
def bye():
if is_tmp:
shutil.rmtree(dirname)
if __name__ == '__main__':
atexit.register(bye)
sys.exit(main(sys.argv))