[FEATURE] Support best codec type in audio_extract

src/ytdl_sub/plugins/audio_extract.py

@@ -8,6 +8,8 @@ from ytdl_sub.config.preset_options import OptionsDictValidator
 from ytdl_sub.downloaders.ytdl_options_builder import YTDLOptionsBuilder
 from ytdl_sub.entries.entry import Entry
 from ytdl_sub.utils.exceptions import FileNotDownloadedException
+from ytdl_sub.utils.file_handler import FileMetadata
+from ytdl_sub.validators.audo_codec_validator import AUDIO_CODEC_EXTS
 from ytdl_sub.validators.audo_codec_validator import AUDIO_CODEC_TYPES_EXTENSION_MAPPING
 from ytdl_sub.validators.audo_codec_validator import AudioTypeValidator
 from ytdl_sub.validators.validators import FloatValidator
@@ -105,12 +107,39 @@ class AudioExtractPlugin(Plugin[AudioExtractOptions]):
         FileNotDownloadedException
             If the audio file is not found
         """
-        new_ext = AUDIO_CODEC_TYPES_EXTENSION_MAPPING[self.plugin_options.codec]
-        extracted_audio_file = entry.get_download_file_path().removesuffix(entry.ext) + new_ext
+        extracted_audio_file = entry.get_download_file_path()
+
+        # yt-dlp doesn't record which ext is used, so try to find it
+        if self.plugin_options.codec == "best":
+            new_ext = "m4a"
+
+            for possible_ext in AUDIO_CODEC_EXTS:
+                extracted_audio_file = (
+                    entry.get_download_file_path().removesuffix(entry.ext) + possible_ext
+                )
+
+                if os.path.isfile(extracted_audio_file):
+                    new_ext = possible_ext
+                    break
+        else:
+            new_ext = AUDIO_CODEC_TYPES_EXTENSION_MAPPING[self.plugin_options.codec]
+            extracted_audio_file = entry.get_download_file_path().removesuffix(entry.ext) + new_ext
+
+        entry.add_kwargs({"ext": new_ext})
+
         if not self.is_dry_run:
             if not os.path.isfile(extracted_audio_file):
                 raise FileNotDownloadedException("Failed to find the extracted audio file")
 
-        entry.add_kwargs({"ext": new_ext})
-
         return entry
+
+    def post_process_entry(self, entry: Entry) -> Optional[FileMetadata]:
+        """
+        Warn the user that best cannot infer the format that will be used at run-time.
+        """
+        if self.plugin_options.codec == "best" and self.is_dry_run:
+            return FileMetadata(
+                "Caution: extracted audio with 'best' format is not known during dry-run. "
+                "Defaulting to m4a."
+            )
+        return None

src/ytdl_sub/validators/audo_codec_validator.py

@@ -21,7 +21,7 @@ VIDEO_CODEC_EXTS: Set[str] = {"avi", "flv", "mkv", "mov", "mp4", "webm"}
 
 class AudioTypeValidator(StringSelectValidator):
     _expected_value_type_name = "codec"
-    _select_values = AUDIO_CODEC_TYPES
+    _select_values = AUDIO_CODEC_TYPES.union({"best"})  # support 'best' in the audio extract plugin
 
 
 class FileTypeValidator(StringSelectValidator):

tests/e2e/plugins/test_audio_extract.py

@@ -43,6 +43,11 @@ def single_song_preset_dict(output_directory):
     }
 
 
+@pytest.fixture
+def single_song_best_format_preset_dict(single_song_preset_dict):
+    return dict(single_song_preset_dict, **{"audio_extract": {"codec": "best"}})
+
+
 @pytest.fixture
 def multiple_songs_preset_dict(output_directory):
     return {
@@ -88,7 +93,7 @@ class TestAudioExtract:
         )
 
     @pytest.mark.parametrize("dry_run", [True, False])
-    def test_audio_extract_single_song_new_format(
+    def test_audio_extract_single_song(
         self,
         music_audio_config,
         single_song_preset_dict,
@@ -113,6 +118,32 @@ class TestAudioExtract:
             expected_download_summary_file_name="plugins/test_audio_extract_single.json",
         )
 
+    @pytest.mark.parametrize("dry_run", [True, False])
+    def test_audio_extract_single_song_best_format(
+        self,
+        music_audio_config,
+        single_song_best_format_preset_dict,
+        output_directory,
+        dry_run,
+    ):
+        subscription = Subscription.from_dict(
+            config=music_audio_config,
+            preset_name="single_song_best_test",
+            preset_dict=single_song_best_format_preset_dict,
+        )
+
+        transaction_log = subscription.download(dry_run=dry_run)
+        assert_transaction_log_matches(
+            output_directory=output_directory,
+            transaction_log=transaction_log,
+            transaction_log_summary_file_name=f"plugins/test_audio_extract_single_best{'_dry_run' if dry_run else ''}.txt",
+        )
+        assert_expected_downloads(
+            output_directory=output_directory,
+            dry_run=dry_run,
+            expected_download_summary_file_name=f"plugins/test_audio_extract_single_best{'_dry_run' if dry_run else ''}.json",
+        )
+
     @pytest.mark.parametrize("dry_run", [True, False])
     def test_audio_extract_multiple_songs(
         self,

tests/resources.py

@@ -1,7 +1,7 @@
 import shutil
 from pathlib import Path
 
-REGENERATE_FIXTURES: bool = False
+REGENERATE_FIXTURES: bool = True
 
 RESOURCE_PATH: Path = Path("tests") / "resources"
 _FILE_FIXTURE_PATH: Path = RESOURCE_PATH / "file_fixtures"

tests/resources/expected_downloads_summaries/plugins/test_audio_extract_single_best.json

@@ -0,0 +1,5 @@
+{
+  ".ytdl-sub-single_song_best_test-download-archive.json": "f179ccfe0a9b3a76ae62122a3ccb58fd",
+  "YouTube/[2019] YouTube Rewind 2019： For the Record ｜ #YouTubeRewind/01 - YouTube Rewind 2019： For the Record ｜ #YouTubeRewind.m4a": "4b03133e9422a9ce196d271beb15ceef",
+  "YouTube/[2019] YouTube Rewind 2019： For the Record ｜ #YouTubeRewind/folder.jpg": "50ee47c80f679029f5d3503bb91b045a"
+}

tests/resources/transaction_log_summaries/plugins/test_audio_extract_single_best.txt

@@ -0,0 +1,22 @@
+Files created:
+----------------------------------------
+{output_directory}
+  .ytdl-sub-single_song_best_test-download-archive.json
+{output_directory}/YouTube/[2019] YouTube Rewind 2019： For the Record ｜ #YouTubeRewind
+  01 - YouTube Rewind 2019： For the Record ｜ #YouTubeRewind.m4a
+    Music Tags:
+      album: YouTube Rewind 2019: For the Record | #YouTubeRewind
+      albumartist: YouTube
+      albumartists: YouTube
+      artist: YouTube
+      artists: YouTube
+      genre: Unset
+      genres:
+        - multi_tag_1
+        - multi_tag_2
+      title: YouTube Rewind 2019: For the Record | #YouTubeRewind
+      track: 1
+      tracktotal: 1
+      year: 2019
+    Embedded thumbnail
+  folder.jpg

tests/resources/transaction_log_summaries/plugins/test_audio_extract_single_best_dry_run.txt

@@ -0,0 +1,23 @@
+Files created:
+----------------------------------------
+{output_directory}
+  .ytdl-sub-single_song_best_test-download-archive.json
+{output_directory}/YouTube/[2019] YouTube Rewind 2019： For the Record ｜ #YouTubeRewind
+  01 - YouTube Rewind 2019： For the Record ｜ #YouTubeRewind.m4a
+    Caution: extracted audio with 'best' format is not known during dry-run. Defaulting to m4a.
+    Music Tags:
+      album: YouTube Rewind 2019: For the Record | #YouTubeRewind
+      albumartist: YouTube
+      albumartists: YouTube
+      artist: YouTube
+      artists: YouTube
+      genre: Unset
+      genres:
+        - multi_tag_1
+        - multi_tag_2
+      title: YouTube Rewind 2019: For the Record | #YouTubeRewind
+      track: 1
+      tracktotal: 1
+      year: 2019
+    Embedded thumbnail
+  folder.jpg