feat: support loose language selection

README.md

@@ -56,9 +56,9 @@ The built-in help provides handles and default values for implemented functions
 ```bash
 (venv) ~/ffmpeg2obj$ ffmpeg2obj --help
 usage: ffmpeg2obj [-h] [-v] [--noop] [--force-cleanup] [-s SRC_DIR] [-d DST_DIR] [-i IGNORED_SUBDIR] [-o OBJ_PREFIX]
-                  [--source-file-extension SOURCE_FILE_EXTENSION] [-e FILE_EXTENSION] [-vc VIDEO_CODEC] [--pix-fmt PIX_FMT]
-                  [-l LANGS] [--width TARGET_WIDTH] [--resize] [--concat] [--height TARGET_HEIGHT]
-                  (-b BUCKET_NAME | --disable-upload) [-qp TARGET_QP | -crf TARGET_CRF]
+                  [--source-file-extension SOURCE_FILE_EXTENSION] [-e FILE_EXTENSION] [-vc VIDEO_CODEC] [--pix-fmt PIX_FMT] [-l LANGS]
+                  [-ll] [--width TARGET_WIDTH] [--resize] [--concat] [--height TARGET_HEIGHT] (-b BUCKET_NAME | --disable-upload)
+                  [-qp TARGET_QP | -crf TARGET_CRF]
 
 Simple tool to compress blu ray movie library and store it in obj
 
@@ -84,6 +84,8 @@ options:
   --pix-fmt PIX_FMT     pix fmt for transcoding of the media files
   -l LANGS, --languages LANGS
                         selected languages transcoding of the media files, all keeps every track
+  -ll, --loose-languages
+                        prevents ffmpeg failures if media to be transcoded lack requested languages
   --width TARGET_WIDTH  target width for the media files to be transcoded
   --resize              scale input files to height x width
   --concat              concatenates files within same directory

src/ffmpeg2obj/helper.py

@@ -9,7 +9,7 @@
 import tempfile
 import time
 from datetime import timedelta
-from typing import Any
+from typing import Any, Optional
 
 import boto3
 import botocore
@@ -35,13 +35,15 @@ def __init__(
         video_codec: str,
         pix_fmt: str,
         langs: list[str],
+        loose_langs: bool,
         target_qp: int,
         target_crf: int,
     ) -> None:
         self.resize = resize
         self.video_codec = video_codec
         self.pix_fmt = pix_fmt
         self.langs = langs
+        self.loose_langs = loose_langs
         self.target_qp = target_qp
         self.target_crf = target_crf
         self.target_res: list[int] = [target_width, target_height]
@@ -77,6 +79,7 @@ def __init__(
         self.dst_hashed_path: str = (
             self.dst_dir + self.hashed_name + "." + self.file_extension
         )
+        self.probe_result: Optional[dict] = None
 
     def __str__(self) -> str:
         out = []
@@ -102,9 +105,10 @@ def update(self, obj_config: dict, bucket_name: str) -> None:
 
     def get_coded_res(self) -> list[int]:
         """Returns height and width for the file from real_path"""
-        probe_result = ffmpeg.probe(self.real_paths[0])
+        if self.probe_result is None:
+            self.probe_result = ffmpeg.probe(self.real_paths[0])
         video_stream = list(
-            filter(lambda x: x["codec_type"] == "video", probe_result["streams"])
+            filter(lambda x: x["codec_type"] == "video", self.probe_result["streams"])
         )[0]
         coded_res = [video_stream["coded_width"], video_stream["coded_height"]]
         return coded_res
@@ -131,8 +135,22 @@ def convert(self, verbose: bool = False) -> tuple[str, str, bool, timedelta]:
         elif self.processing_params.target_qp is not None:
             opts_dict.update({"qp": str(self.processing_params.target_qp)})
         if self.processing_params.langs != ["all"]:
+            requested_langs = set(self.processing_params.langs)
+            if self.processing_params.loose_langs:
+                langs = set()
+                if self.probe_result is None:
+                    self.probe_result = ffmpeg.probe(self.real_paths[0])
+                for stream in self.probe_result["streams"]:
+                    try:
+                        lang = stream["tags"]["language"]
+                        if lang in requested_langs:
+                            langs.add(lang)
+                    except KeyError:
+                        pass
+            else:
+                langs = requested_langs
             lang_map = []
-            for lang in self.processing_params.langs:
+            for lang in langs:
                 lang_map.append("0:m:language:" + lang)
             lang_dict = {"map": tuple(lang_map)}
             opts_dict.update(lang_dict)

src/ffmpeg2obj/script.py

@@ -139,6 +139,15 @@ def parse_args() -> argparse.Namespace:
         help="selected languages transcoding of the media files, all keeps every track",
     )
 
+    parser.add_argument(
+        "-ll",
+        "--loose-languages",
+        dest="loose_langs",
+        action="store_true",
+        default=False,
+        help="prevents ffmpeg failures if media to be transcoded lack requested languages",
+    )
+
     parser.add_argument(
         "--width",
         dest="target_width",
@@ -471,6 +480,7 @@ def main():
         args.video_codec,
         args.pix_fmt,
         args.langs,
+        args.loose_langs,
         args.target_qp,
         args.target_crf,
     )