-
-from slughifi import slughifi
-
-META_REGEX = re.compile(r'\s*<!--\s(.*?)-->', re.DOTALL | re.MULTILINE)
-STAGE_TAGS_RE = re.compile(r'^#stage-finished: (.*)$', re.MULTILINE)
-AUTHOR_RE = re.compile(r'\s*(.*?)\s*<(.*)>\s*')
-
-
-def urlunquote(url):
- """Unqotes URL
-
- # >>> urlunquote('Za%C5%BC%C3%B3%C5%82%C4%87_g%C4%99%C5%9Bl%C4%85_ja%C5%BA%C5%84')
- # u'Za\u017c\xf3\u0142\u0107_g\u0119\u015bl\u0105 ja\u017a\u0144'
- """
- return unicode(urllib.unquote(url), 'utf-8', 'ignore')
-
-
-def split_name(name):
- parts = name.split('__')
- return parts
-
-
-def file_to_title(fname):
- """ Returns a title-like version of a filename. """
- parts = (p.replace('_', ' ').title() for p in fname.split('__'))
- return ' / '.join(parts)
-
-
-def make_patch(src, dst):
- if isinstance(src, unicode):
- src = src.encode('utf-8')
- if isinstance(dst, unicode):
- dst = dst.encode('utf-8')
- return cPickle.dumps(mdiff.textdiff(src, dst))
-
-
-def plain_text(text):
- return re.sub(META_REGEX, '', text, 1)
-
-
-def gallery(slug, text):
- result = {}
-
- m = re.match(META_REGEX, text)
- if m:
- for line in m.group(1).split('\n'):
- try:
- k, v = line.split(':', 1)
- result[k.strip()] = v.strip()
- except ValueError:
- continue
-
- gallery = result.get('gallery', slughifi(slug))
-
- if gallery.startswith('/'):
- gallery = os.path.basename(gallery)
-
- return gallery
-
-
-def migrate_file_from_hg(orm, fname, entry):
- fname = urlunquote(fname)
- print fname
- if fname.endswith('.xml'):
- fname = fname[:-4]
- title = file_to_title(fname)
- fname = slughifi(fname)
- # create all the needed objects
- # what if it already exists?
- book = orm.Book.objects.create(
- title=title,
- slug=fname)
- chunk = orm.Chunk.objects.create(
- book=book,
- number=1,
- slug='1')
- head = orm.ChunkChange.objects.create(
- tree=chunk,
- revision=-1,
- patch=make_patch('', ''),
- created_at=datetime.datetime.fromtimestamp(entry.filectx(0).date()[0]),
- description=''
- )
- chunk.head = head
- try:
- chunk.stage = orm.ChunkTag.objects.order_by('ordering')[0]
- except IndexError:
- chunk.stage = None
- old_data = ''
-
- maxrev = entry.filerev()
- gallery_link = None
-
- for rev in xrange(maxrev + 1):
- fctx = entry.filectx(rev)
- data = fctx.data()
- gallery_link = gallery(fname, data)
- data = plain_text(data)
-
- # get tags from description
- description = fctx.description().decode("utf-8", 'replace')
- tags = STAGE_TAGS_RE.findall(description)
- tags = [orm.ChunkTag.objects.get(slug=slug.strip()) for slug in tags]
-
- if tags:
- max_ordering = max(tags, key=lambda x: x.ordering).ordering
- try:
- chunk.stage = orm.ChunkTag.objects.filter(ordering__gt=max_ordering).order_by('ordering')[0]
- except IndexError:
- chunk.stage = None
-
- description = STAGE_TAGS_RE.sub('', description)
-
- author = author_name = author_email = None
- author_desc = fctx.user().decode("utf-8", 'replace')
- m = AUTHOR_RE.match(author_desc)
- if m:
- try:
- author = orm['auth.User'].objects.get(username=m.group(1), email=m.group(2))
- except orm['auth.User'].DoesNotExist:
- author_name = m.group(1)
- author_email = m.group(2)
- else:
- author_name = author_desc
-
- head = orm.ChunkChange.objects.create(
- tree=chunk,
- revision=rev + 1,
- patch=make_patch(old_data, data),
- created_at=datetime.datetime.fromtimestamp(fctx.date()[0]),
- description=description,
- author=author,
- author_name=author_name,
- author_email=author_email,
- parent=chunk.head
- )
- head.tags = tags
- chunk.head = head
- old_data = data
-
- chunk.save()
- if gallery_link:
- book.gallery = gallery_link
- book.save()
-
-
-def migrate_from_hg(orm):
- try:
- hg_path = settings.WIKI_REPOSITORY_PATH
- except:
- pass
-
- print 'migrate from', hg_path
- repo = hg.repository(ui.ui(), hg_path)
- tip = repo['tip']
- for fname in tip:
- if fname.startswith('.'):
- continue
- migrate_file_from_hg(orm, fname, tip[fname])