aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl/extractor/youtube.py
diff options
context:
space:
mode:
authorJaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>2015-07-20 21:10:28 +0200
committerJaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>2015-07-20 21:10:28 +0200
commit7c80519cbf7e3daa029239fbbd147652817877f8 (patch)
treeaa36444b1cfe823c50a6be61fec732a68c50a80b /youtube_dl/extractor/youtube.py
parent2fe1ff85829244f6498dde8c85ecdc1415f5ad7a (diff)
[youtube] Extract start_time
From the 't=*' in the url. Currently youtube-dl doesn't use the value, but it was requested for the mpv plugin.
Diffstat (limited to 'youtube_dl/extractor/youtube.py')
-rw-r--r--youtube_dl/extractor/youtube.py14
1 files changed, 13 insertions, 1 deletions
diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index 3d8b31f98..afbd34f4a 100644
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -19,6 +19,7 @@ from ..compat import (
compat_urllib_parse,
compat_urllib_parse_unquote,
compat_urllib_parse_unquote_plus,
+ compat_urllib_parse_urlparse,
compat_urllib_request,
compat_urlparse,
compat_str,
@@ -31,6 +32,7 @@ from ..utils import (
get_element_by_id,
int_or_none,
orderedSet,
+ parse_duration,
str_to_int,
unescapeHTML,
unified_strdate,
@@ -317,7 +319,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
IE_NAME = 'youtube'
_TESTS = [
{
- 'url': 'http://www.youtube.com/watch?v=BaW_jenozKc',
+ 'url': 'http://www.youtube.com/watch?v=BaW_jenozKcj&t=1s',
'info_dict': {
'id': 'BaW_jenozKc',
'ext': 'mp4',
@@ -329,6 +331,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'categories': ['Science & Technology'],
'like_count': int,
'dislike_count': int,
+ 'start_time': 1,
}
},
{
@@ -889,6 +892,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'http' if self._downloader.params.get('prefer_insecure', False)
else 'https')
+ start_time = None
+ parsed_url = compat_urllib_parse_urlparse(url)
+ for component in [parsed_url.fragment, parsed_url.query]:
+ query = compat_parse_qs(component)
+ if 't' in query:
+ start_time = parse_duration(query['t'][0])
+ break
+
# Extract original video URL from URL with redirection, like age verification, using next_url parameter
mobj = re.search(self._NEXT_URL_RE, url)
if mobj:
@@ -1255,6 +1266,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
'formats': formats,
'is_live': is_live,
+ 'start_time': start_time,
}