def findunit(self, source):
# find using hash instead of index
source_hash = md5_f(source.encode("utf-8")).hexdigest()
try:
return self.units.get(source_hash=source_hash)
except Unit.DoesNotExist:
return None
def convertunit(self, unit, prefix):
if self.ignorefunc:
if self.ignorefunc(unit):
return unit
if prefix.find("@hash_placeholder@") != -1:
if unit.getlocations():
hashable = unit.getlocations()[0]
else:
hashable = unit.source
prefix = prefix.replace(
"@hash_placeholder@",
hash.md5_f(hashable).hexdigest()[:self.hash_len])
rich_source = unit.rich_source
if not isinstance(rich_source, StringElem):
rich_source = [
rich_parse(string, podebug_parsers) for string in rich_source
]
if self.rewritefunc:
rewritten = [self.rewritefunc(string) for string in rich_source]
if rewritten:
unit.rich_target = rewritten
elif not unit.istranslated():
unit.rich_target = unit.rich_source
unit.rich_target = add_prefix(prefix, unit.rich_target)
return unit
def phprep(self, text):
"""Replaces all instances of PHP with placeholder tags, and returns
the new text and a dictionary of tags. The current implementation
replaces <?foo?> with <?md5(foo)?>. The hash => code conversions
are stored in self.phpdict for later use in restoring the real PHP.
The purpose of this is to remove all potential "tag-like" code from
inside PHP. The hash looks nothing like an HTML tag, but the following
PHP::
$a < $b ? $c : ($d > $e ? $f : $g)
looks like it contains an HTML tag::
< $b ? $c : ($d >
to nearly any regex. Hence, we replace all contents of PHP with simple
strings to help our regexes out.
"""
from translate.misc import hash
self.phpdict = {}
result = re.findall('(?s)<\?(.*?)\?>', text)
for cmd in result:
h = hash.md5_f(cmd).hexdigest()
self.phpdict[h] = cmd
text = text.replace(cmd, h)
return text
def phprep(self, text):
"""Replaces all instances of PHP with placeholder tags, and returns
the new text and a dictionary of tags. The current implementation
replaces <?foo?> with <?md5(foo)?>. The hash => code conversions
are stored in self.phpdict for later use in restoring the real PHP.
The purpose of this is to remove all potential "tag-like" code from
inside PHP. The hash looks nothing like an HTML tag, but the following
PHP::
$a < $b ? $c : ($d > $e ? $f : $g)
looks like it contains an HTML tag::
< $b ? $c : ($d >
to nearly any regex. Hence, we replace all contents of PHP with simple
strings to help our regexes out.
"""
from translate.misc import hash
self.phpdict = {}
result = re.findall('(?s)<\?(.*?)\?>', text)
for cmd in result:
h = hash.md5_f(cmd).hexdigest()
self.phpdict[h] = cmd
text = text.replace(cmd, h)
return text
def findid(self, id):
if hasattr(self, "id_index"):
return self.id_index.get(id, None)
unitid_hash = md5_f(id.encode("utf-8")).hexdigest()
try:
return self.units.get(unitid_hash=unitid_hash)
except Unit.DoesNotExist:
return None
def findunits(self, source):
if hasattr(self, "sourceindex"):
return super(Store, self).findunit(source)
# find using hash instead of index
source_hash = md5_f(source.encode("utf-8")).hexdigest()
units = self.units.filter(source_hash=source_hash)
if units.count():
return units
def update(self, unit):
"""update indb translation from file"""
changed = False
if (
self.source != unit.source
or len(self.source.strings) != stringcount(unit.source)
or self.hasplural() != unit.hasplural()
):
if unit.hasplural() and len(unit.source.strings) == 1:
self.source = [unit.source, PLURAL_PLACEHOLDER]
else:
self.source = unit.source
changed = True
if self.target != unit.target or len(self.target.strings) != stringcount(unit.target):
notempty = filter(None, self.target_f.strings)
self.target = unit.target
if filter(None, self.target_f.strings) or notempty:
# FIXME: we need to do this cause we discard nplurals
# for empty plurals
changed = True
notes = unit.getnotes(origin="developer")
if self.developer_comment != notes and (self.developer_comment or notes):
self.developer_comment = notes or None
changed = True
notes = unit.getnotes(origin="translator")
if self.translator_comment != notes and (self.translator_comment or notes):
self.translator_comment = notes or None
changed = True
locations = "\n".join(unit.getlocations())
if self.locations != locations and (self.locations or locations):
self.locations = locations or None
changed = True
context = unit.getcontext()
if self.context != unit.getcontext() and (self.context or context):
self.context = context or None
changed = True
if self.isfuzzy() != unit.isfuzzy():
self.markfuzzy(unit.isfuzzy())
changed = True
if self.isobsolete() != unit.isobsolete():
if unit.isobsolete():
self.makeobsolete()
else:
self.resurrect()
changed = True
if self.unitid != unit.getid():
self.unitid = unicode(unit.getid()) or unicode(unit.source)
self.unitid_hash = md5_f(self.unitid.encode("utf-8")).hexdigest()
changed = True
if hasattr(unit, "getalttrans"):
for suggestion in unit.getalttrans():
if suggestion.source == self.source:
self.add_suggestion(suggestion.target, touch=False)
changed = True
return changed
def findunits(self, source, obsolete=False):
if not obsolete and hasattr(self, "sourceindex"):
return super(Store, self).findunits(source)
# find using hash instead of index
source_hash = md5_f(source.encode("utf-8")).hexdigest()
units = self.unit_set.filter(source_hash=source_hash)
if obsolete:
units = units.filter(state=OBSOLETE)
else:
units = units.filter(state__gt=OBSOLETE)
if units.count():
return units
def update(self, unit):
"""update indb translation from file"""
changed = False
if self.hasplural() != unit.hasplural():
self.source = unit.source
self.target = unit.target
changed = True
else:
if self.source != unit.source:
self.source = unit.source
changed = True
if self.target != unit.target:
wordcount = self.target_wordcount
self.target = unit.target
if not (wordcount == count_words(self.target_f.strings) == 0):
#FIXME: we need to do this cause we discard nplurals for empty plurals
changed = True
notes = unit.getnotes(origin="developer")
if self.developer_comment != notes:
self.developer_comment = notes
changed = True
notes = unit.getnotes(origin="translator")
if self.translator_comment != notes:
self.translator_comment = notes
changed = True
locations = "\n".join(unit.getlocations())
if self.locations != locations:
self.locations = locations
changed = True
if self.context != unit.getcontext():
self.context = unit.getcontext()
changed = True
if self.isfuzzy() != unit.isfuzzy():
self.markfuzzy(unit.isfuzzy())
changed = True
if self.isobsolete() != unit.isobsolete():
if unit.isobsoltete():
self.makeobsolete()
else:
self.resurrect()
changed = True
if self.unitid != unit.getid():
self.unitid = unit.getid()
self.unitid_hash = md5_f(self.unitid.encode("utf-8")).hexdigest()
changed = True
if hasattr(unit, 'getalttrans'):
for suggestion in unit.getalttrans():
if suggestion.source == self.source:
self.add_suggestion(suggestion.target, touch=False)
changed = True
return changed
def save(self, *args, **kwargs):
if self._source_updated:
# update source related fields
self.source_hash = md5_f(self.source_f.encode("utf-8")).hexdigest()
self.source_wordcount = count_words(self.source_f.strings)
self.source_length = len(self.source_f)
if self._target_updated:
# update target related fields
self.target_wordcount = count_words(self.target_f.strings)
self.target_length = len(self.target_f)
if filter(None, self.target_f.strings):
if self.state == UNTRANSLATED:
self.state = TRANSLATED
elif self.state > FUZZY:
self.state = UNTRANSLATED
super(Unit, self).save(*args, **kwargs)
if (
settings.AUTOSYNC
and self.store.file
and self.store.state >= PARSED
and (self._target_updated or self._source_updated)
):
# FIXME: last translator information is lost
self.sync(self.getorig())
self.store.update_store_header()
self.store.file.savestore()
if self.store.state >= CHECKED and (self._source_updated or self._target_updated):
# FIXME: are we sure only source and target affect quality checks?
self.update_qualitychecks()
# done processing source/target update remove flag
self._source_updated = False
self._target_updated = False
if self.store.state >= PARSED:
# updated caches
store = self.store
# translation_project = store.translation_project
# translation_project.update_index(
# translation_project.indexer, store, self.id)
deletefromcache(store, ["getquickstats", "getcompletestats", "get_mtime", "has_suggestions"])
def convertunit(self, unit, prefix):
if self.ignorefunc:
if self.ignorefunc(unit):
return unit
if prefix.find("@hash_placeholder@") != -1:
if unit.getlocations():
hashable = unit.getlocations()[0]
else:
hashable = unit.source
prefix = prefix.replace("@hash_placeholder@", hash.md5_f(hashable).hexdigest()[:self.hash_len])
rich_source = unit.rich_source
if not isinstance(rich_source, StringElem):
rich_source = [rich_parse(string, podebug_parsers) for string in rich_source]
if self.rewritefunc:
rewritten = [self.rewritefunc(string) for string in rich_source]
if rewritten:
unit.rich_target = rewritten
elif not unit.istranslated():
unit.rich_target = unit.rich_source
unit.rich_target = add_prefix(prefix, unit.rich_target)
return unit
def setid(self, value):
self.unitid = value
self.unitid_hash = md5_f(self.unitid.encode("utf-8")).hexdigest()
def _set_target(self, value):
self.target_f = value
self.target_hash = md5_f(self.target_f.encode("utf-8")).hexdigest()
def get_email_hash(self):
return md5_f(self.user.email).hexdigest()
def get_email_hash(self):
return md5_f(self.user.email).hexdigest()
