add --skip-lang

zyddnys · May 1, 2024 · b5f91d9 · b5f91d9
1 parent ae2b3a7
commit b5f91d9
Show file tree

Hide file tree

Showing 6 changed files with 47 additions and 0 deletions.
diff --git a/README.md b/README.md
@@ -461,6 +461,8 @@ Colorizer: **mc2**
 --save-text-file SAVE_TEXT_FILE              Like --save-text but with a specified file path.
 --filter-text FILTER_TEXT                    Filter regions by their text with a regex. Example
                                              usage: --text-filter ".*badtext.*"
+--skip-lang                                  Skip translation if source image is one of the provide languages, 
+                                             use comma to separate multiple languages. Example: JPN,ENG
 --prep-manual                                Prepare for manual typesetting by outputting blank,
                                              inpainted images, plus copies of the original for
                                              reference

diff --git a/README_CN.md b/README_CN.md
@@ -204,6 +204,8 @@ FIL: Filipino (Tagalog)
 --save-text-file SAVE_TEXT_FILE              Like --save-text but with a specified file path.
 --filter-text FILTER_TEXT                    Filter regions by their text with a regex. Example
                                              usage: --text-filter ".*badtext.*"
+--skip-lang                                  Skip translation if source image is one of the provide languages, 
+                                             use comma to separate multiple languages. Example: JPN,ENG
 --prep-manual                                Prepare for manual typesetting by outputting blank,
                                              inpainted images, plus copies of the original for
                                              reference

diff --git a/manga_translator/manga_translator.py b/manga_translator/manga_translator.py
@@ -6,6 +6,7 @@
 from aiohttp.web_middlewares import middleware
 from omegaconf import OmegaConf
 import langcodes
+import langdetect
 import requests
 import os
 import re
@@ -50,6 +51,7 @@
 from .translators import (
     TRANSLATORS,
     VALID_LANGUAGES,
+    LANGDETECT_MAP,
     LanguageUnsupportedException,
     TranslatorChain,
     dispatch as dispatch_translation,
@@ -395,6 +397,16 @@ async def _translate(self, ctx: Context) -> Context:
             # If no text was found result is intermediate image product
             ctx.result = ctx.upscaled
             return ctx
+
+        if ctx.skip_lang is not None :
+            skip_langs = ctx.skip_lang.split(',')
+            detected_text = ''.join([l.text for l in ctx.textlines])
+            source_language = LANGDETECT_MAP.get(langdetect.detect(detected_text), 'UNKNOWN')
+            if source_language in skip_langs :
+                print('skip due to', source_language, 'in', skip_langs)
+                await self._report_progress('finished', True)
+                ctx.result = ctx.upscaled
+                return ctx
 
         # -- Textline merge
         await self._report_progress('textline_merge')
@@ -846,6 +858,7 @@ async def server_process_inner(main_loop, logger_task, session, websocket, task)
 
             params = {
                 'target_lang': task.target_language,
+                'skip_lang': task.skip_language,
                 'detector': task.detector,
                 'direction': task.direction,
                 'translator': task.translator,
@@ -1269,6 +1282,7 @@ class PostSchema(Schema):
         upscaler = fields.Str(required=False, validate=lambda a: a.lower() in UPSCALERS)
         translator = fields.Str(required=False, validate=lambda a: a.lower() in TRANSLATORS)
         direction = fields.Str(required=False, validate=lambda a: a.lower() in {'auto', 'h', 'v'})
+        skip_language = fields.Str(required=False)
         upscale_ratio = fields.Integer(required=False)
         translator_chain = fields.Str(required=False)
         selective_translation = fields.Str(required=False)

diff --git a/manga_translator/translators/__init__.py b/manga_translator/translators/__init__.py
@@ -124,3 +124,30 @@ async def dispatch(chain: TranslatorChain, queries: List[str], use_mtpe: bool =
         if args is not None:
             args['translations'][tgt_lang] = queries
     return queries
+
+LANGDETECT_MAP = {
+    'zh-cn': 'CHS',
+    'zh-tw': 'CHT',
+    'cs': 'CSY',
+    'nl': 'NLD',
+    'en': 'ENG',
+    'fr': 'FRA',
+    'de': 'DEU',
+    'hu': 'HUN',
+    'it': 'ITA',
+    'ja': 'JPN',
+    'ko': 'KOR',
+    'pl': 'PLK',
+    'pt': 'PTB',
+    'ro': 'ROM',
+    'ru': 'RUS',
+    'es': 'ESP',
+    'tr': 'TRK',
+    'uk': 'UKR',
+    'vi': 'VIN',
+    'ar': 'ARA',
+    'hr': 'HRV',
+    'th': 'THA',
+    'id': 'IND',
+    'tl': 'FIL'
+}
diff --git a/pyproject.toml b/pyproject.toml
@@ -14,6 +14,7 @@ build-backend = "poetry.core.masonry.api"
 [tool.poetry.dependencies]
 python = ">=3.10, <3.13"
 langcodes = "^3.3.0"
+langdetect = ">=1.0.0"
 networkx = "^3.1"
 torch = "^2.0.1"
 torchvision = "^0.15.2"

diff --git a/requirements.txt b/requirements.txt
@@ -44,5 +44,6 @@ aiofiles
 arabic-reshaper
 pyhyphen
 langcodes
+langdetect
 manga-ocr
 pydensecrf@https://github.com/lucasb-eyer/pydensecrf/archive/refs/heads/master.zip