Feature dump

.gitignore

@@ -157,4 +157,7 @@ cython_debug/
 #  be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore
 #  and can be added to the global gitignore or merged into this file.  For a more nuclear
 #  option (not recommended) you can uncomment the following to ignore the entire idea folder.
-#.idea/
+#.idea/
+test.ipynb
+.gitignore
+tests/samples/text.2.eng.srt

Dockerfile

@@ -1,7 +1,6 @@
 FROM ubuntu:focal
 
-RUN apt update && apt install software-properties-common -y
-RUN add-apt-repository ppa:alex-p/tesseract-ocr5 
+RUN apt update && apt install software-properties-common -y && add-apt-repository ppa:alex-p/tesseract-ocr5 
 RUN apt update && apt install mkvtoolnix tesseract-ocr python3-pip ffmpeg libsm6 libxext6 -y
 
 COPY ./requirements.txt requirements.txt

README.md

@@ -31,28 +31,35 @@ services:
 ## Usage:
 
 ```sh
-usage: main.py [-h] [--path PATH] [--formats {ass,srt,vtt} [{ass,srt,vtt} ...]] [--languages LANGUAGES [LANGUAGES ...]] [--overwrite] [--disable_bitmap_extraction]
-               [--postprocess_only] [--postprocessing POSTPROCESSING] [--scan_interval SCAN_INTERVAL] [--log_level LOG_LEVEL] [--log_file LOG_FILE] [--progress_bar {on,off}]
+positional arguments:
+  path                  path to media file/folder
 
 options:
   -h, --help            show this help message and exit
-  --path PATH           Path to media file/folder
   --formats {ass,srt,vtt} [{ass,srt,vtt} ...]
-                        Subtitles formats to extract to
+                        output subtitles formats
   --languages LANGUAGES [LANGUAGES ...]
-                        Select subtitle languages stream to extract from, use 'all' to extract all languages
-  --overwrite           Overwrite existing subtitle file
+                        extract subtitle for selected languages, use 'all' to extract all languages
+  --unknown_language_as UNKNOWN_LANGUAGE_AS
+                        treat unknown language as
+  --overwrite           overwrite existing subtitle file
   --disable_bitmap_extraction
-                        Disable bitmap subtitle extraction via OCR
-  --postprocess_only    Only do conduct post processing
+                        disable extraction for bitmap based subtitle extraction via OCR
+  --postprocess_only    only do conduct post processing
   --postprocessing POSTPROCESSING
-                        Path to postprocessing config file
+                        path to postprocessing config file
   --scan_interval SCAN_INTERVAL
-                        Interval to scan folder in mins (set 0 to disable and exit upon completion)
+                        interval to monitor and scan folder in mins (set 0 to disable and exit upon completion)
   --log_level LOG_LEVEL
-                        Setting logging level
-  --log_file LOG_FILE   Path to log file
-  --progress_bar {on,off}
-                        Enable progress bar
-
+                        setting logging level
+  --log_file LOG_FILE   path to log file
+  --disable_progress_bar
+                        enable progress bar
+  --exclude_videos EXCLUDE_VIDEOS
+                        path to a newline separated file with paths to video files to exclude
+  --exclude_subtitles EXCLUDE_SUBTITLES
+                        path to a newline separated file with paths to subtitles files to exclude
+  --exclude_mode {e,e+a}
+                        set file exclusion behavior, e = exclude only, e+a = exclude and append new extracted file
+  --threads THREADS     set number of extraction thread
 ```

docker-compose.yml

@@ -1,5 +1,9 @@
 services:
+
     subtitle-extract:
+        build: .
+
+    subtitle-extract-image:
         image: ghcr.io/klementng/subtitle-extract:main
         container_name: subtitle-extract
         volumes:

extract.py

@@ -109,7 +109,7 @@ def _run_subprocess(args):
         return completed_process
 
     @staticmethod
-    @cachetools.func.fifo_cache(maxsize=1)
+    @cachetools.func.fifo_cache(maxsize=10)
     def _run_ffprobe(path):
         """
         The function `_run_ffprobe` is used to probe a media file and retrieve information about its
@@ -337,7 +337,7 @@ def extract(self, media_path: str, stream_indexes: list):
           a list of file paths.
         """
 
-        logger.info(
+        logger.debug(
             f"[BitMapSubtitleExtractor] Processing subtitles for {media_path}, streams: {stream_indexes}")
 
         ffmpeg_args = []
@@ -351,11 +351,11 @@ def extract(self, media_path: str, stream_indexes: list):
                 ffmpeg_args.extend(['-map', f"0:{i}", "-c", "copy", sup_f])
 
         if ffmpeg_args != []:
-            logger.info(
+            logger.debug(
                 "[BitMapSubtitleExtractor] Extracting subtitles to .sup format...")
             self._run_ffmpeg(media_path, ffmpeg_args)
         else:
-            logger.info(
+            logger.debug(
                 "[BitMapSubtitleExtractor] .sup format found. Skipping ffmpeg extraction..")
 
 
@@ -367,12 +367,12 @@ def extract(self, media_path: str, stream_indexes: list):
             srt_f = self.format_subtitle_path(media_path, i, 'srt')
 
             if self.is_wanted(media_path, srt_f, i):
-                logger.info(
+                logger.debug(
                     "[BitMapSubtitleExtractor] Performing OCR. Converting .sup to .srt")
                 self._run_psgrip_ocr(sup_f, srt_f, lang)
                 filelist.append(srt_f)
             else:
-                logger.info(
+                logger.debug(
                     "[BitMapSubtitleExtractor] .srt format found. Skipping OCR...")
 
             # Convert SRT to desired subtitles formats
@@ -387,12 +387,12 @@ def extract(self, media_path: str, stream_indexes: list):
                     filelist.append(sub_f)
 
             if len(tmp_filelist) != 0:
-                logger.info(
+                logger.debug(
                     "[BitMapSubtitleExtractor] Converting .srt to desired format")
                 self._run_ffmpeg(srt_f, tmp_filelist)
 
             else:
-                logger.info(
+                logger.debug(
                     "[BitMapSubtitleExtractor] All desired format found, skipping...")
 
         return filelist
@@ -423,7 +423,7 @@ def extract(self, media_path: str, stream_indexes: list):
           a list of file paths for the extracted subtitles.
         """
 
-        logger.info(
+        logger.debug(
             f"[TextSubtitleExtractor] Processing subtitles for {media_path}, streams: {stream_indexes}")
 
         ffmpeg_args = []
@@ -439,10 +439,10 @@ def extract(self, media_path: str, stream_indexes: list):
                     filelist.append(f)
 
         if ffmpeg_args != []:
-            logger.info("[TextSubtitleExtractor] Extracting Subtitles...")
+            logger.debug("[TextSubtitleExtractor] Extracting Subtitles...")
             self._run_ffmpeg(media_path, ffmpeg_args)
         else:
-            logger.info(
+            logger.debug(
                 "[TextSubtitleExtractor] All desired format found, skipping...")
 
         return filelist
@@ -466,7 +466,7 @@ def extract(self, media_path):
         Returns:
           a list of files that were extracted.
         """
-        logger.info(f"[SubtitleExtractor] Processing {media_path}")
+        logger.debug(f"[SubtitleExtractor] Processing {media_path}")
 
         ffprobe_info = self.get_subtitle_info(media_path)
 
@@ -490,13 +490,13 @@ def extract(self, media_path):
 
         filelist = []
         if len(text_streams) != 0:
-            logger.info("[SubtitleExtractor] Extracting text based subtitles")
+            logger.debug("[SubtitleExtractor] Extracting text based subtitles")
             extractor = TextSubtitleExtractor.init(self)
             filelist1 = extractor.extract(media_path, text_streams)
             filelist.extend(filelist1)
 
         if len(bitmap_streams) != 0 and self.extract_bitmap:
-            logger.info("[SubtitleExtractor] Extracting bitmap based subtitles")
+            logger.debug("[SubtitleExtractor] Extracting bitmap based subtitles")
             extractor = BitmapSubtitleExtractor.init(self)
             filelist2 = extractor.extract(media_path, bitmap_streams)
             filelist.extend(filelist2)