Update gTTS; fix failing unit tests

jdlorimer · Mar 29, 2021 · cb3846d · cb3846d
1 parent 94d0eb7
commit cb3846d
Show file tree

Hide file tree

Showing 21 changed files with 840 additions and 134 deletions.
diff --git a/Makefile b/Makefile
@@ -15,7 +15,7 @@
 
 PROJECT_SHORT = chinese
 PROJECT_LONG = chinese-support-redux
-VERSION = 0.14.0
+VERSION = 0.14.2
 XDG_DATA_HOME ?= $(HOME)/.local/share
 ADDON_PATH = "$(XDG_DATA_HOME)/Anki2/addons21/$(PROJECT_LONG)"
 ZIP_NAME = $(PROJECT_LONG)-v$(VERSION).zip

diff --git a/chinese/_version.py b/chinese/_version.py
@@ -1 +1 @@
-__version__ = '0.14.0'
+__version__ = '0.14.2'
diff --git a/chinese/about.py b/chinese/about.py
@@ -33,6 +33,7 @@ def showAbout():
     contributors = [
         'Alex Griffin',
         'Chris Hatch',
+        'Joe Minicucci',
         'Roland Sieker',
         'Thomas TEMPÉ',
     ]

diff --git a/chinese/config.json b/chinese/config.json
@@ -1,8 +1,8 @@
 {
     "firstRun": true,
-    "version": "0.14.0",
+    "version": "0.14.2",
     "enabledModels": [],
-    "speech": "google|zh-cn",
+    "speech": "google|zh-CN",
     "target": "pinyin",
     "max_examples": -1,
     "fields": {

diff --git a/chinese/gui.py b/chinese/gui.py
@@ -41,8 +41,8 @@
 
 SPEECH_ENGINES = {
     'Baidu Translate': 'baidu|zh',
-    'Google Mandarin (PRC)': 'google|zh-cn',
-    'Google Mandarin (Taiwan)': 'google|zh-tw',
+    'Google Mandarin (PRC)': 'google|zh-CN',
+    'Google Mandarin (Taiwan)': 'google|zh-TW',
     'Amazon Polly' : 'aws|Zhiyu',
     'Disabled': None,
 }

diff --git a/chinese/lib/gtts/lang.py b/chinese/lib/gtts/lang.py
@@ -1,4 +1,6 @@
 # -*- coding: utf-8 -*-
+from gtts.langs import _main_langs
+from warnings import warn
 import logging
 
 __all__ = ['tts_langs']
@@ -14,13 +16,13 @@ def tts_langs():
     Returns:
         dict: A dictionary of the type `{ '<lang>': '<name>'}`
 
-            Where `<lang>` is an IETF language tag such as `en` or `pt-br`,
+            Where `<lang>` is an IETF language tag such as `en` or `zh-TW`,
             and `<name>` is the full English name of the language, such as
-            `English` or `Portuguese (Brazil)`.
+            `English` or `Chinese (Mandarin/Taiwan)`.
 
     The dictionary returned combines languages from two origins:
 
-    - Languages fetched from Google Translate
+    - Languages fetched from Google Translate (pre-generated in :mod:`gtts.langs`)
     - Languages that are undocumented variations that were observed to work and
       present different dialects or accents.
 
@@ -32,112 +34,65 @@ def tts_langs():
     return langs
 
 
-def _main_langs():
-    """Define the main languages.
+def _extra_langs():
+    """Define extra languages.
 
     Returns:
-        dict: A dictionnary of the main languages extracted from
-            Google Translate.
+        dict: A dictionnary of extra languages manually defined.
+
+            Variations of the ones generated in `_main_langs`,
+            observed to provide different dialects or accents or
+            just simply accepted by the Google Translate Text-to-Speech API.
 
     """
     return {
-        'af': 'Afrikaans',
-        'ar': 'Arabic',
-        'bn': 'Bengali',
-        'bs': 'Bosnian',
-        'ca': 'Catalan',
-        'cs': 'Czech',
-        'cy': 'Welsh',
-        'da': 'Danish',
-        'de': 'German',
-        'el': 'Greek',
-        'en': 'English',
-        'eo': 'Esperanto',
-        'es': 'Spanish',
-        'et': 'Estonian',
-        'fi': 'Finnish',
-        'fr': 'French',
-        'gu': 'Gujarati',
-        'hi': 'Hindi',
-        'hr': 'Croatian',
-        'hu': 'Hungarian',
-        'hy': 'Armenian',
-        'id': 'Indonesian',
-        'is': 'Icelandic',
-        'it': 'Italian',
-        'ja': 'Japanese',
-        'jw': 'Javanese',
-        'km': 'Khmer',
-        'kn': 'Kannada',
-        'ko': 'Korean',
-        'la': 'Latin',
-        'lv': 'Latvian',
-        'mk': 'Macedonian',
-        'ml': 'Malayalam',
-        'mr': 'Marathi',
-        'my': 'Myanmar (Burmese)',
-        'ne': 'Nepali',
-        'nl': 'Dutch',
-        'no': 'Norwegian',
-        'pl': 'Polish',
-        'pt': 'Portuguese',
-        'ro': 'Romanian',
-        'ru': 'Russian',
-        'si': 'Sinhala',
-        'sk': 'Slovak',
-        'sq': 'Albanian',
-        'sr': 'Serbian',
-        'su': 'Sundanese',
-        'sv': 'Swedish',
-        'sw': 'Swahili',
-        'ta': 'Tamil',
-        'te': 'Telugu',
-        'th': 'Thai',
-        'tl': 'Filipino',
-        'tr': 'Turkish',
-        'uk': 'Ukrainian',
-        'ur': 'Urdu',
-        'vi': 'Vietnamese',
-        'zh-CN': 'Chinese'
+        # Chinese
+        'zh-TW': 'Chinese (Mandarin/Taiwan)',
+        'zh': 'Chinese (Mandarin)'
     }
 
 
-def _extra_langs():
-    """Define extra languages.
+def _fallback_deprecated_lang(lang):
+    """Languages Google Text-to-Speech used to support.
+
+    Language tags that don't work anymore, but that can
+    fallback to a more general language code to maintain
+    compatibility.
+
+    Args:
+        lang (string): The language tag.
 
     Returns:
-        dict: A dictionnary of extra languages manually defined.
+        string: The language tag, as-is if not deprecated,
+            or a fallack if it exits.
 
-            Variations of the ones fetched by `_main_langs`,
-            observed to provide different dialects or accents or
-            just simply accepted by the Google Translate Text-to-Speech API.
+    Example:
+        ``en-GB`` returns ``en``.
+        ``en-gb`` returns ``en``.
 
     """
-    return {
-        # Chinese
-        'zh-cn': 'Chinese (Mandarin/China)',
-        'zh-tw': 'Chinese (Mandarin/Taiwan)',
-        # English
-        'en-us': 'English (US)',
-        'en-ca': 'English (Canada)',
-        'en-uk': 'English (UK)',
-        'en-gb': 'English (UK)',
-        'en-au': 'English (Australia)',
-        'en-gh': 'English (Ghana)',
-        'en-in': 'English (India)',
-        'en-ie': 'English (Ireland)',
-        'en-nz': 'English (New Zealand)',
-        'en-ng': 'English (Nigeria)',
-        'en-ph': 'English (Philippines)',
-        'en-za': 'English (South Africa)',
-        'en-tz': 'English (Tanzania)',
-        # French
-        'fr-ca': 'French (Canada)',
-        'fr-fr': 'French (France)',
-        # Portuguese
-        'pt-br': 'Portuguese (Brazil)',
-        'pt-pt': 'Portuguese (Portugal)',
-        # Spanish
-        'es-es': 'Spanish (Spain)',
-        'es-us': 'Spanish (United States)'
+
+    deprecated = {
+        # '<fallback>': [<list of deprecated langs>]
+        'en': ['en-us', 'en-ca', 'en-uk', 'en-gb', 'en-au', 'en-gh', 'en-in',
+               'en-ie', 'en-nz', 'en-ng', 'en-ph', 'en-za', 'en-tz'],
+        'fr': ['fr-ca', 'fr-fr'],
+        'pt': ['pt-br', 'pt-pt'],
+        'es': ['es-es', 'es-us'],
+        'zh-CN': ['zh-cn'],
+        'zh-TW': ['zh-tw'],
     }
+
+    for fallback_lang, deprecated_langs in deprecated.items():
+        if lang.lower() in deprecated_langs:
+            msg = (
+                "'{}' has been deprecated, falling back to '{}'. "
+                "This fallback will be removed in a future version."
+            ).format(lang, fallback_lang)
+
+            warn(msg, DeprecationWarning)
+            log.warning(msg)
+
+            return fallback_lang
+
+    return lang
diff --git a/chinese/lib/gtts/langs.py b/chinese/lib/gtts/langs.py
@@ -0,0 +1,64 @@
+# Note: this file is generated
+_langs = {
+    "af": "Afrikaans",
+    "ar": "Arabic",
+    "bn": "Bengali",
+    "bs": "Bosnian",
+    "ca": "Catalan",
+    "cs": "Czech",
+    "cy": "Welsh",
+    "da": "Danish",
+    "de": "German",
+    "el": "Greek",
+    "en": "English",
+    "eo": "Esperanto",
+    "es": "Spanish",
+    "et": "Estonian",
+    "fi": "Finnish",
+    "fr": "French",
+    "gu": "Gujarati",
+    "hi": "Hindi",
+    "hr": "Croatian",
+    "hu": "Hungarian",
+    "hy": "Armenian",
+    "id": "Indonesian",
+    "is": "Icelandic",
+    "it": "Italian",
+    "ja": "Japanese",
+    "jw": "Javanese",
+    "km": "Khmer",
+    "kn": "Kannada",
+    "ko": "Korean",
+    "la": "Latin",
+    "lv": "Latvian",
+    "mk": "Macedonian",
+    "ml": "Malayalam",
+    "mr": "Marathi",
+    "my": "Myanmar (Burmese)",
+    "ne": "Nepali",
+    "nl": "Dutch",
+    "no": "Norwegian",
+    "pl": "Polish",
+    "pt": "Portuguese",
+    "ro": "Romanian",
+    "ru": "Russian",
+    "si": "Sinhala",
+    "sk": "Slovak",
+    "sq": "Albanian",
+    "sr": "Serbian",
+    "su": "Sundanese",
+    "sv": "Swedish",
+    "sw": "Swahili",
+    "ta": "Tamil",
+    "te": "Telugu",
+    "th": "Thai",
+    "tl": "Filipino",
+    "tr": "Turkish",
+    "uk": "Ukrainian",
+    "ur": "Urdu",
+    "vi": "Vietnamese",
+    "zh-CN": "Chinese"
+}
+
+def _main_langs():
+    return _langs
diff --git a/chinese/lib/gtts/tests/__init__.py b/chinese/lib/gtts/tests/__init__.py
diff --git a/chinese/lib/gtts/tests/input_files/test_cli_test_ascii.txt b/chinese/lib/gtts/tests/input_files/test_cli_test_ascii.txt
@@ -0,0 +1,2 @@
+Can you make pink a little more pinkish can you make pink a little more pinkish, nor can you make the font bigger?
+How much will it cost the website doesn't have the theme i was going for.
diff --git a/chinese/lib/gtts/tests/input_files/test_cli_test_utf8.txt b/chinese/lib/gtts/tests/input_files/test_cli_test_utf8.txt
@@ -0,0 +1,5 @@
+这是一个三岁的小孩
+在讲述她从一系列照片里看到的东西。
+对这个世界， 她也许还有很多要学的东西，
+但在一个重要的任务上， 她已经是专家了：
+去理解她所看到的东西。
Original file line number	Diff line number	Diff line change
		@@ -0,0 +1,2 @@
		Can you make pink a little more pinkish can you make pink a little more pinkish, nor can you make the font bigger?
		How much will it cost the website doesn't have the theme i was going for.