mirrors/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2025-10-31 14:45:14 +00:00
Code Issues Projects Releases Wiki Activity

[youtube] Improve ytplayer.config extraction

Browse Source
This commit is contained in:
Sergey M․
2015-11-23 21:00:06 +06:00
parent 5ae17037a3
commit a72778d364
1 changed files with 14 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
youtube_dl/extractor/youtube.py
View File

@@ -891,22 +891,24 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
return {}
return sub_lang_list
def _get_ytplayer_config(self, webpage):
patterns = [
r';ytplayer\.config\s*=\s*({.*?});ytplayer',
r';ytplayer\.config\s*=\s*({.*?});',
]
config = self._search_regex(patterns, webpage, 'ytconfig.player', default=None)
if config is not None:
return json.loads(uppercase_escape(config))
def _get_ytplayer_config(self, video_id, webpage):
patterns = (
r';ytplayer\.config\s*=\s*({.+?});ytplayer',
r';ytplayer\.config\s*=\s*({.+?});',
)
config = self._search_regex(
patterns, webpage, 'ytplayer.config', default=None)
if config:
return self._parse_json(
uppercase_escape(config), video_id, fatal=False)
def _get_automatic_captions(self, video_id, webpage):
"""We need the webpage for getting the captions url, pass it as an
argument to speed up the process."""
self.to_screen('%s: Looking for automatic captions' % video_id)
player_config = self._get_ytplayer_config(webpage)
player_config = self._get_ytplayer_config(video_id, webpage)
err_msg = 'Couldn\'t find automatic captions for %s' % video_id
if player_config is None:
if not player_config:
self._downloader.report_warning(err_msg)
return {}
try:
@@ -1115,8 +1117,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
age_gate = False
video_info = None
# Try looking directly into the video webpage
ytplayer_config = self._get_ytplayer_config(video_webpage)
if ytplayer_config is not None:
ytplayer_config = self._get_ytplayer_config(video_id, video_webpage)
if ytplayer_config:
args = ytplayer_config['args']
if args.get('url_encoded_fmt_stream_map'):
# Convert to the same format returned by compat_parse_qs