Merge pull request #233 from NSoiffer/zh

Zh

PythonScripts/translate-rules.py

@@ -47,6 +47,8 @@ def collect_phrases_to_translate(file_to_translate: str) -> (list[str], list[str
             word = WordToTranslate.search(line)
             if word:
                 words.append(word.group(1))
+
+        print(f"#phrases={len(phrases)}, #words={len(words)}")
         return (phrases, words)
 
 # break up the words into chunks to make google translate happy (and to run faster) and return a dictionary of word: translation
@@ -68,9 +70,11 @@ def do_translation_chunk(phrases: list[str]):
         # print("***Phrases to translate: {}\n".format(phrases))
         translated_phrases_str = GoogleTranslate.translate(phrases_string, src='en', dest=lang).text.lower()
         translated_phrases_str = translated_phrases_str.replace('。', '.')   # happens for Chinese
+
         translated_phrases_str = translated_phrases_str.replace('"', "'").replace("“", "'").replace("”", "'")    # google occasionally changes quotes
         translated_phrases_str = translated_phrases_str.replace("«", "'").replace("»", "'")    # google occasionally changes quotes to this form
         translated_phrases_str = translated_phrases_str.replace("、", ",")   # Chinese comma
+
         translated_phrases_list = translated_phrases_str.split('.\n')
         if len(translated_phrases_list) != len(phrases):
             print("\n!!!Problem in translation: size of translations ({}) differs from phrases to translate ({})\n".format(len(translated_phrases_list), len(phrases)))
@@ -101,7 +105,7 @@ def do_translation_chunk(phrases: list[str]):
             time.sleep(TIMEOUT)       # try to avoid google banning us
     return translations + do_translation_chunk(phrases_chunks_to_translate)
 
-TargetWord = re.compile(r"'([^']+)'")
+argetWord = re.compile(r"'([^']+)'")
 TextString = re.compile(r'([ \[{][oc]?t: )"([^"]+)"')
 def substitute_in_translated_phrase(line, translated_phrase, translated_word) -> str:
     has_phrase = PhraseToTranslate.search(line)
@@ -120,6 +124,7 @@ def substitute_in_translated_phrase(line, translated_phrase, translated_word) ->
     elif text_words:
         print(f"Failed to find quoted part in translation \"{translated_phrase}\", \
                using '{translated_word}\n   original line: {line}")
+
         replacement = text_words.group(1) + '"' + translated_word + '"'    # add the surrounding context back
         new_line = TextString.sub(replacement, line)
     return new_line
@@ -148,6 +153,7 @@ def build_new_translation(path_to_mathcat: str, lang: str, rule_file_name: str)
     (phrases_to_translate, words_to_translate) = collect_phrases_to_translate(file_to_translate)
     phrase_translations = translate_phrases(phrases_to_translate, lang)
     word_translations = translate_phrases(words_to_translate, lang)
+
     print(f"file:{rule_file_name}: #phrases={len(phrase_translations)}, #words={len(word_translations)}")
     create_new_file(file_to_translate, os.path.join(lang, rule_file_name), phrase_translations, word_translations)
     print("done\n")

PythonScripts/translate-unicode.py

@@ -93,6 +93,8 @@ def do_translate_char(match_obj):
             result['original'] = match_obj.group(1)
             result['translation'] = translation
             result['alternatives'] = alternatives
+            if line.find('divided by') != -1:
+                print(f"  divided by translation: {translation}")
             return 't: "{}"'.format(translation)
         else:
             return line
@@ -426,4 +428,4 @@ def translate_definition(start: int, lines: list[str], translated_lines: list[st
 build_new_translation("..", language, "unicode-full")
 
 # see translate_definitions comments -- you need to manually copy the file to google translate. 
-translate_definitions("..", language)
+# translate_definitions("..", language)