w/youtube-dl
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

[youtube:user] Extract in terms of load_more_widget_html

Browse source
This commit is contained in:
Sergey M․ 2015-04-21 22:36:41 +06:00
parent ed553379df
commit eb0f3e7ec0
1 changed files with 4 additions and 45 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

49
youtube_dl/extractor/youtube.py
View file

@ -1358,6 +1358,7 @@ class YoutubePlaylistIE(YoutubeBaseInfoExtractor):
class YoutubeChannelIE(InfoExtractor): class YoutubeChannelIE(InfoExtractor):
IE_DESC = 'YouTube.com channels' IE_DESC = 'YouTube.com channels'
_VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)' _VALID_URL = r'https?://(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/(?P<id>[0-9A-Za-z_-]+)'
_TEMPLATE_URL = 'https://www.youtube.com/channel/%s/videos'
IE_NAME = 'youtube:channel' IE_NAME = 'youtube:channel'
_TESTS = [{ _TESTS = [{
'note': 'paginated channel', 'note': 'paginated channel',
@ -1386,7 +1387,7 @@ class YoutubeChannelIE(InfoExtractor):
def _real_extract(self, url): def _real_extract(self, url):
channel_id = self._match_id(url) channel_id = self._match_id(url)
url = 'https://www.youtube.com/channel/%s/videos' % channel_id url = self._TEMPLATE_URL % channel_id
channel_page = self._download_webpage(url, channel_id) channel_page = self._download_webpage(url, channel_id)
autogenerated = re.search(r'''(?x) autogenerated = re.search(r'''(?x)
class="[^"]*?(?: class="[^"]*?(?:
@ -1429,12 +1430,10 @@ class YoutubeChannelIE(InfoExtractor):
return self.playlist_result(_entries(), channel_id) return self.playlist_result(_entries(), channel_id)
class YoutubeUserIE(InfoExtractor): class YoutubeUserIE(YoutubeChannelIE):
IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)' IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
_VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)' _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:user/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
_TEMPLATE_URL = 'https://gdata.youtube.com/feeds/api/users/%s' _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
_GDATA_PAGE_SIZE = 50
_GDATA_URL = 'https://gdata.youtube.com/feeds/api/users/%s/uploads?max-results=%d&start-index=%d&alt=json'
IE_NAME = 'youtube:user' IE_NAME = 'youtube:user'
_TESTS = [{ _TESTS = [{
@ -1458,46 +1457,6 @@ class YoutubeUserIE(InfoExtractor):
else: else:
return super(YoutubeUserIE, cls).suitable(url) return super(YoutubeUserIE, cls).suitable(url)
def _real_extract(self, url):
username = self._match_id(url)
# Download video ids using YouTube Data API. Result size per
# query is limited (currently to 50 videos) so we need to query
# page by page until there are no video ids - it means we got
# all of them.
def download_page(pagenum):
start_index = pagenum * self._GDATA_PAGE_SIZE + 1
gdata_url = self._GDATA_URL % (username, self._GDATA_PAGE_SIZE, start_index)
page = self._download_webpage(
gdata_url, username,
'Downloading video ids from %d to %d' % (
start_index, start_index + self._GDATA_PAGE_SIZE))
try:
response = json.loads(page)
except ValueError as err:
raise ExtractorError('Invalid JSON in API response: ' + compat_str(err))
if 'entry' not in response['feed']:
return
# Extract video identifiers
entries = response['feed']['entry']
for entry in entries:
title = entry['title']['$t']
video_id = entry['id']['$t'].split('/')[-1]
yield {
'_type': 'url',
'url': video_id,
'ie_key': 'Youtube',
'id': video_id,
'title': title,
}
url_results = OnDemandPagedList(download_page, self._GDATA_PAGE_SIZE)
return self.playlist_result(url_results, playlist_title=username)
class YoutubeSearchIE(SearchInfoExtractor): class YoutubeSearchIE(SearchInfoExtractor):
IE_DESC = 'YouTube.com searches' IE_DESC = 'YouTube.com searches'