feat(subtitle): Add filtering for unwanted cues in WebVTT subtitles

unshackle/commands/dl.py

@@ -148,6 +148,7 @@ class dl:
         help="Language wanted for Video, you would use this if the video language doesn't match the audio.",
     )
     @click.option("-sl", "--s-lang", type=LANGUAGE_RANGE, default=["all"], help="Language wanted for Subtitles.")
+    @click.option("-fs", "--forced-subs", is_flag=True, default=False, help="Include forced subtitle tracks.")
     @click.option(
         "--proxy",
         type=str,
@@ -405,6 +406,7 @@ class dl:
         lang: list[str],
         v_lang: list[str],
         s_lang: list[str],
+        forced_subs: bool,
         sub_format: Optional[Subtitle.Codec],
         video_only: bool,
         audio_only: bool,
@@ -672,7 +674,8 @@ class dl:
                             self.log.error(f"There's no {s_lang} Subtitle Track...")
                             sys.exit(1)
 
-                    title.tracks.select_subtitles(lambda x: not x.forced or is_close_match(x.language, lang))
+                    if not forced_subs:
+                        title.tracks.select_subtitles(lambda x: not x.forced or is_close_match(x.language, lang))
 
                 # filter audio tracks
                 # might have no audio tracks if part of the video, e.g. transport stream hls

unshackle/core/tracks/subtitle.py

@@ -233,6 +233,7 @@ class Subtitle(Track):
             try:
                 caption_set = pycaption.WebVTTReader().read(text)
                 Subtitle.merge_same_cues(caption_set)
+                Subtitle.filter_unwanted_cues(caption_set)
                 subtitle_text = pycaption.WebVTTWriter().write(caption_set)
                 self.path.write_text(subtitle_text, encoding="utf8")
             except pycaption.exceptions.CaptionReadSyntaxError:
@@ -241,6 +242,7 @@ class Subtitle(Track):
                 try:
                     caption_set = pycaption.WebVTTReader().read(text)
                     Subtitle.merge_same_cues(caption_set)
+                    Subtitle.filter_unwanted_cues(caption_set)
                     subtitle_text = pycaption.WebVTTWriter().write(caption_set)
                     self.path.write_text(subtitle_text, encoding="utf8")
                 except Exception:
@@ -444,6 +446,8 @@ class Subtitle(Track):
 
         caption_set = self.parse(self.path.read_bytes(), self.codec)
         Subtitle.merge_same_cues(caption_set)
+        if codec == Subtitle.Codec.WebVTT:
+            Subtitle.filter_unwanted_cues(caption_set)
         subtitle_text = writer().write(caption_set)
 
         output_path.write_text(subtitle_text, encoding="utf8")
@@ -520,6 +524,8 @@ class Subtitle(Track):
 
             caption_set = self.parse(self.path.read_bytes(), self.codec)
             Subtitle.merge_same_cues(caption_set)
+            if codec == Subtitle.Codec.WebVTT:
+                Subtitle.filter_unwanted_cues(caption_set)
             subtitle_text = writer().write(caption_set)
 
             output_path.write_text(subtitle_text, encoding="utf8")
@@ -681,6 +687,24 @@ class Subtitle(Track):
             if merged_captions:
                 caption_set.set_captions(lang, merged_captions)
 
+    @staticmethod
+    def filter_unwanted_cues(caption_set: pycaption.CaptionSet):
+        """
+        Filter out subtitle cues containing only   or whitespace.
+        """
+        for lang in caption_set.get_languages():
+            captions = caption_set.get_captions(lang)
+            filtered_captions = pycaption.CaptionList()
+
+            for caption in captions:
+                text = caption.get_text().strip()
+                if not text or text == " " or all(c in " \t\n\r\xa0" for c in text.replace(" ", "\xa0")):
+                    continue
+
+                filtered_captions.append(caption)
+
+            caption_set.set_captions(lang, filtered_captions)
+
     @staticmethod
     def merge_segmented_wvtt(data: bytes, period_start: float = 0.0) -> tuple[CaptionList, Optional[str]]:
         """