Update of /cvsroot/fasttran/fasttran/fasttran
In directory sc8-pr-cvs5.sourceforge.net:/tmp/cvs-serv10123
Modified Files:
tm.py
Log Message:
use pickle protocol version 2;
added TM meta information
Index: tm.py
===================================================================
RCS file: /cvsroot/fasttran/fasttran/fasttran/tm.py,v
retrieving revision 1.1
retrieving revision 1.2
diff -C2 -d -r1.1 -r1.2
*** tm.py 13 Feb 2007 05:46:55 -0000 1.1
--- tm.py 20 Feb 2007 06:48:07 -0000 1.2
***************
*** 26,29 ****
--- 26,31 ----
"""
"""History (most recent first):
+ 20-feb-2007 [als] use pickle protocol version 2;
+ added TM meta information
03-feb-2007 [als] created
"""
***************
*** 63,66 ****
--- 65,74 ----
@ivar tgt: target language tag.
+ @type meta: C{dict}
+ @ivar meta: TM meta-information.
+
+ At present, this dictionary has only one entry:
+ C{version} with value of C{1}.
+
@type delta: C{int}
@ivar delta: maximum allowed length difference.
***************
*** 156,159 ****
--- 164,169 ----
- we'll need this later.
+ The C{words} table also contains TM meta-information dictionary.
+
Operations
----------
***************
*** 321,324 ****
--- 331,349 ----
self._db_translations = anydbm.open(_db_prefix + ".translations", flag)
self.load_wex("default.wex", src + ".wex")
+ _save_meta = False
+ if "" in self._db_words:
+ self.meta = loads(self._db_words[""])
+ else:
+ self.meta = {}
+ _save_meta = True
+ if "version" not in self.meta:
+ self.meta["version"] = 1
+ _save_meta = True
+ if _save_meta and (flag in "cwn"):
+ self.save_meta()
+
+ def save_meta(self):
+ """Update TM meta information in the database"""
+ self._db_words[""] = dumps(self.meta, 2)
def store(self, sentence, translation):
***************
*** 354,358 ****
return
for _word in _words:
! _key = dumps((_word, len(_words)))
if _key in self._db_words:
self._db_words[_key] += "," + _id
--- 379,383 ----
return
for _word in _words:
! _key = dumps((_word, len(_words)), 2)
if _key in self._db_words:
self._db_words[_key] += "," + _id
***************
*** 369,373 ****
else:
_translations[translation] = 1
! self._db_translations[_id] = dumps(_translations)
def lookup(self, sentence):
--- 394,398 ----
else:
_translations[translation] = 1
! self._db_translations[_id] = dumps(_translations, 2)
def lookup(self, sentence):
***************
*** 455,459 ****
_slen = _cnt + delta
for _word in words:
! _ids = self._db_words.get(dumps((_word, _slen)), None)
if _ids is None:
_missing += 1
--- 480,484 ----
_slen = _cnt + delta
for _word in words:
! _ids = self._db_words.get(dumps((_word, _slen), 2), None)
if _ids is None:
_missing += 1
|
