WebSVN - LCARS - Path Comparison - /trunk/tools/eazytrans/vuh.py Rev 299 and /trunk/tools/eazytrans/vuh.py Rev 300

Compare Revisions

This comparison shows the changes necessary to convert path

→ /trunk/tools/eazytrans/vuh.py @ 299

→ /trunk/tools/eazytrans/vuh.py @ 300

↔ Reverse comparison
Compare Path: Rev

With Path: Rev

Last modification

Rev 300 2015-02-01 05:33:48
Author: PointedEars

Log message:

eazytrans:

* Dictionary.py
  - dmsg(): Fixed argument handling
  - Dictionary.load(): Fixed/updated default values

* vuh-gol-en.dict.zdb.txt
  - Added Vim modeline
  - Clarified format description
  - Simplified expression of imperatives in translation
  - Updated IPA transcriptions
  - Updated word order
  - New words based on Surak's teachings in "Spock's World"

* vuh.py, VulcanDictionary.py
  - Organized imports
  - Moved VulcanDictionary class to own file
  - Properly split into paragraphs, sentences, and clauses
  - Find and translate phrases by looking up subsets of clauses
    (TODO: compound handling)

Ignore whitespace Rev 299 → Rev 300

 /trunk/tools/eazytrans/vuh.py
 ,21 → 6,14
 @author: Thomas 'PointedEars' Lahn <mail@PointedEars.de>
 '''
 from sys import argv, stderr
-from re import findall, DOTALL, match, sub, compile, \
-    escape, search
+from re import findall, compile
 from os.path import basename
 from functools import cmp_to_key
-from Dictionary import Dictionary, dmsg, \
-    sort_dict_alnum_english_key
+from Dictionary import dmsg, sort_dict_alnum_english_key
+from VulcanDictionary import VulcanDictionary, Text
 dictionary = {}
-prepositions = {
-    "fi'": 'on',
-    "na'": 'at|to',
-    "t'": 'of'
-}
 def cli_help():
     print('Usage: {0} TEXT...'.format(basename(argv[0])))
 ,63 → 47,7
     return cmp_to_key(sort_dict_alnum_vulcan)
-class VulcanDictionary (Dictionary):
-    """
-    """
-    def translate (self, phrase, search_prefix=True, search_plural=True):
-        """
-        :param phrase:
-        :type phrase:
-        :param search_prefix:
-        :type search_prefix:
-        :param search_plural:
-        :type search_plural:
-        """
-        translation = super().translate(phrase)
-        if translation is not None:
-            return translation
-        else:
-            expr_translation = self.translate_expression(phrase)
-            if expr_translation is not None:
-                return expr_translation
-            if search_prefix:
-                # find prefix
-                for preposition in prepositions:
-                    prefix = match(escape(preposition), phrase)
-                    if prefix is not None:
-                        prefix_translation = self.translate(prefix.group(0))
-                        if prefix_translation is not None:
-                            tail = sub(preposition, '', phrase)
-                            tail_translation = self.translate(tail, search_prefix=False)
-                            if tail_translation is not None:
-                                return [prefix_translation, tail_translation]
-            elif search_plural:
-                # find plural
-                suffix = search(r'lar$', phrase)
-                if suffix is not None:
-                    head = sub(r'lar$', '', phrase)
-                    head_translation = self.translate(head, search_prefix=False, search_plural=False)
-                    if head_translation is not None:
-                        head_translation = dict(head_translation)
-                        head_translation['en'] += ' (pl.)'
-                        return head_translation
-        return None
-    def clean_entry(self, phrase):
-        """
-        Replace GV Media Script parens with FSE parens
-        :param phrase:
-        :type phrase:
-        """
-        return sub(
-            r'(\([^)]*\))|\|([^|)]+)\|',
-            lambda m: '({0})'.format(m.group(2)) if m.group(2) else m.group(1),
-            phrase)
 if __name__ == '__main__':
     if len(argv) < 2:
         print('Nothing to translate.', end='\n\n', file=stderr)
 ,60 → 60,30
     dictionary.load('vuh-gol-en.dict.zdb.txt', 'vuh')
     dictionary.clean()
-#     try:
-#         for phrase, data in OrderedDict(sorted(
-#             dictionary.items(),
-#             key=get_sort_dict_alnum_vulcan_key()
-#         )).items():
-#             print(phrase, "=", data)
-#     except BrokenPipeError:
-#         pass
+#       try:
+#           for phrase, data in OrderedDict(sorted(
+#               dictionary.items(),
+#               key=get_sort_dict_alnum_vulcan_key()
+#           )).items():
+#               print(phrase, "=", data)
+#       except BrokenPipeError:
+#           pass
+    text = Text(text)
     dmsg("text:", text, min_level=2)
-    sentences = findall(r'(?!\s+)(?:.+?\.{1,3}|.+$)', text, DOTALL)
-    dmsg("sentences:", sentences, min_level=2)
-    for sentence in sentences:
-        dmsg("sentence:", sentence, min_level=2)
+    dmsg("text:", text.__repr__(), min_level=2)
-        clauses = findall(r'(?!\s+)(?:.+?(?:\s+-\s*|\s*[–—]\s*|\.{1,3}|.+$))', sentence, DOTALL)
-        dmsg("clauses:", clauses, min_level=2)
-        for clause in clauses:
-            dmsg("clause:", clause, min_level=2)
+    text.translate(dictionary)
-            words = findall(r'[^\s.]+', clause)
-            dmsg("words:", words, min_level=2)
-            offset = 0
-            while offset < len(words):
-                translation = None
-                for i in range(len(words), offset, -1):
-                    dmsg("words[{0}:{1}] = {2}".format(offset, i, words[offset:i]), min_level=2)
-                    phrase = ' '.join(words[offset:i])
-                    dmsg("phrase:", phrase, min_level=2)
-                    translation = dictionary.translate(phrase)
-                    if translation is not None:
-                        dmsg("phrase-translation:", translation, min_level=2)
-                        dmsg("words[{0}:{1}] = [\"{2}\"]".format(offset, i, translation), min_level=2)
-                        words[offset:i] = [translation]
-                        offset += i - offset
-                        break
-                if translation is None:
-                    dmsg("phrase-translation:", translation, min_level=2)
-                    offset += 1
-            dmsg("words-translation:", words, min_level=2)
-            dmsg("words-translation-reduced:",
-                list(map(
-                    lambda word:
-                        word['en']
-                        if (hasattr(word, "get") and word.get('en', None) is not None)
-                        else word,
-                    words)),
-                min_level=2)
-#             for key, value in dictionary._expressions.items():
-#                 dmsg(key, value, min_level=3)
+#             dmsg("words-translation:", words, min_level=2)
+#             dmsg("words-translation-reduced:",
+#                 list(map(
+#                     lambda word:
+#                         word['en']
+#                         if (hasattr(word, "get") and word.get('en', None) is not None)
+#                         else word,
+#                     words)),
+#                 min_level=2)
+#              for key, value in dictionary._expressions.items():
+#                  dmsg(key, value, min_level=3)

Subversion Repositories LCARS

Compare Revisions

Last modification

Ignore whitespace Rev 299 → Rev 300