[pbs] fix embed data extraction(fixes #16474)

This commit is contained in:
Remita Amine 2018-05-16 18:34:25 +01:00
parent 6843ac5b13
commit eea2fafcf5

View File

@ -505,7 +505,7 @@ def extract_redirect_urls(info):
if player:
video_info = self._parse_json(
self._search_regex(
r'(?s)PBS\.videoData\s*=\s*({.+?});\n',
[r'(?s)PBS\.videoData\s*=\s*({.+?});\n', r'window\.videoBridge\s*=\s*({.+?});'],
player, '%s video data' % page, default='{}'),
display_id, transform_source=js_to_json, fatal=False)
if video_info:
@ -513,10 +513,14 @@ def extract_redirect_urls(info):
if not info:
info = video_info
if not chapters:
raw_chapters = video_info.get('chapters') or []
if not raw_chapters:
for chapter_data in re.findall(r'(?s)chapters\.push\(({.*?})\)', player):
chapter = self._parse_json(chapter_data, video_id, js_to_json, fatal=False)
if not chapter:
continue
raw_chapters.append(chapter)
for chapter in raw_chapters:
start_time = float_or_none(chapter.get('start_time'), 1000)
duration = float_or_none(chapter.get('duration'), 1000)
if start_time is None or duration is None: