diff options
Diffstat (limited to 'yt_dlp/extractor/pbs.py')
| -rw-r--r-- | yt_dlp/extractor/pbs.py | 31 |
1 files changed, 10 insertions, 21 deletions
diff --git a/yt_dlp/extractor/pbs.py b/yt_dlp/extractor/pbs.py index d68855d62..0eabf9bee 100644 --- a/yt_dlp/extractor/pbs.py +++ b/yt_dlp/extractor/pbs.py @@ -600,6 +600,7 @@ class PBSIE(InfoExtractor): formats = [] http_url = None + hls_subs = {} for num, redirect in enumerate(redirects): redirect_id = redirect.get('eeid') @@ -622,8 +623,9 @@ class PBSIE(InfoExtractor): continue if determine_ext(format_url) == 'm3u8': - formats.extend(self._extract_m3u8_formats( - format_url, display_id, 'mp4', m3u8_id='hls', fatal=False)) + hls_formats, hls_subs = self._extract_m3u8_formats_and_subtitles( + format_url, display_id, 'mp4', m3u8_id='hls', fatal=False) + formats.extend(hls_formats) else: formats.append({ 'url': format_url, @@ -666,25 +668,12 @@ class PBSIE(InfoExtractor): age_limit = US_RATINGS.get(rating_str) subtitles = {} - closed_captions_url = info.get('closed_captions_url') - if closed_captions_url: - subtitles['en'] = [{ - 'ext': 'ttml', - 'url': closed_captions_url, - }] - mobj = re.search(r'/(\d+)_Encoded\.dfxp', closed_captions_url) - if mobj: - ttml_caption_suffix, ttml_caption_id = mobj.group(0, 1) - ttml_caption_id = int(ttml_caption_id) - subtitles['en'].extend([{ - 'url': closed_captions_url.replace( - ttml_caption_suffix, '/%d_Encoded.srt' % (ttml_caption_id + 1)), - 'ext': 'srt', - }, { - 'url': closed_captions_url.replace( - ttml_caption_suffix, '/%d_Encoded.vtt' % (ttml_caption_id + 2)), - 'ext': 'vtt', - }]) + captions = info.get('cc') or {} + for caption_url in captions.values(): + subtitles.setdefault('en', []).append({ + 'url': caption_url + }) + subtitles = self._merge_subtitles(subtitles, hls_subs) # info['title'] is often incomplete (e.g. 'Full Episode', 'Episode 5', etc) # Try turning it to 'program - title' naming scheme if possible |