import re
from .common import InfoExtractor
-from .subtitles import SubtitlesInfoExtractor
+
+from ..compat import compat_urlparse
from ..utils import (
- compat_urllib_request,
- unescapeHTML,
- parse_iso8601,
- compat_urlparse,
clean_html,
- compat_str,
+ int_or_none,
+ parse_iso8601,
+ sanitized_Request,
+ unescapeHTML,
+ xpath_text,
+ xpath_with_ns,
)
-class BlipTVIE(SubtitlesInfoExtractor):
- _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/(?:(?:.+-|rss/flash/)(?P<id>\d+)|((?:play/|api\.swf#)(?P<lookup_id>[\da-zA-Z+_TESTS]+)))'
+class BlipTVIE(InfoExtractor):
+ _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/(?:(?:.+-|rss/flash/)(?P<id>\d+)|((?:play/|api\.swf#)(?P<lookup_id>[\da-zA-Z+_]+)))'
_TESTS = [
{
'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
- 'md5': 'c6934ad0b6acf2bd920720ec888eb812',
+ 'md5': '80baf1ec5c3d2019037c1c707d676b9f',
'info_dict': {
'id': '5779306',
- 'ext': 'mov',
+ 'ext': 'm4v',
'title': 'CBR EXCLUSIVE: "Gotham City Imposters" Bats VS Jokerz Short 3',
'description': 'md5:9bc31f227219cde65e47eeec8d2dc596',
'timestamp': 1323138843,
'uploader': 'redvsblue',
'uploader_id': '792887',
}
+ },
+ {
+ 'url': 'http://blip.tv/play/gbk766dkj4Yn',
+ 'md5': 'fe0a33f022d49399a241e84a8ea8b8e3',
+ 'info_dict': {
+ 'id': '1749452',
+ 'ext': 'mp4',
+ 'upload_date': '20090208',
+ 'description': 'Witness the first appearance of the Nostalgia Critic character, as Doug reviews the movie Transformers.',
+ 'title': 'Nostalgia Critic: Transformers',
+ 'timestamp': 1234068723,
+ 'uploader': 'NostalgiaCritic',
+ 'uploader_id': '246467',
+ }
+ },
+ {
+ # https://github.com/rg3/youtube-dl/pull/4404
+ 'note': 'Audio only',
+ 'url': 'http://blip.tv/hilarios-productions/weekly-manga-recap-kingdom-7119982',
+ 'md5': '76c0a56f24e769ceaab21fbb6416a351',
+ 'info_dict': {
+ 'id': '7103299',
+ 'ext': 'flv',
+ 'title': 'Weekly Manga Recap: Kingdom',
+ 'description': 'And then Shin breaks the enemy line, and he's all like HWAH! And then he slices a guy and it's all like FWASHING! And... it's really hard to describe the best parts of this series without breaking down into sound effects, okay?',
+ 'timestamp': 1417660321,
+ 'upload_date': '20141204',
+ 'uploader': 'The Rollo T',
+ 'uploader_id': '407429',
+ 'duration': 7251,
+ 'vcodec': 'none',
+ }
+ },
+ {
+ # missing duration
+ 'url': 'http://blip.tv/rss/flash/6700880',
+ 'info_dict': {
+ 'id': '6684191',
+ 'ext': 'm4v',
+ 'title': 'Cowboy Bebop: Gateway Shuffle Review',
+ 'description': 'md5:3acc480c0f9ae157f5fe88547ecaf3f8',
+ 'timestamp': 1386639757,
+ 'upload_date': '20131210',
+ 'uploader': 'sfdebris',
+ 'uploader_id': '706520',
+ }
}
]
+ @staticmethod
+ def _extract_url(webpage):
+ mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
+ if mobj:
+ return 'http://blip.tv/a/a-' + mobj.group(1)
+ mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9_]+)', webpage)
+ if mobj:
+ return mobj.group(1)
+
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
lookup_id = mobj.group('lookup_id')
- # See https://github.com/rg3/youtube-dl/issues/857
+ # See https://github.com/rg3/youtube-dl/issues/857 and
+ # https://github.com/rg3/youtube-dl/issues/4197
if lookup_id:
- info_page = self._download_webpage(
- 'http://blip.tv/play/%s.x?p=1' % lookup_id, lookup_id, 'Resolving lookup id')
- video_id = self._search_regex(r'data-episode-id="([0-9]+)', info_page, 'video_id')
- else:
- video_id = mobj.group('id')
+ urlh = self._request_webpage(
+ 'http://blip.tv/play/%s' % lookup_id, lookup_id, 'Resolving lookup id')
+ url = compat_urlparse.urlparse(urlh.geturl())
+ qs = compat_urlparse.parse_qs(url.query)
+ mobj = re.match(self._VALID_URL, qs['file'][0])
- rss = self._download_xml('http://blip.tv/rss/flash/%s' % video_id, video_id, 'Downloading video RSS')
-
- def blip(s):
- return '{http://blip.tv/dtd/blip/1.0}%s' % s
+ video_id = mobj.group('id')
- def media(s):
- return '{http://search.yahoo.com/mrss/}%s' % s
+ rss = self._download_xml('http://blip.tv/rss/flash/%s' % video_id, video_id, 'Downloading video RSS')
- def itunes(s):
- return '{http://www.itunes.com/dtds/podcast-1.0.dtd}%s' % s
+ def _x(p):
+ return xpath_with_ns(p, {
+ 'blip': 'http://blip.tv/dtd/blip/1.0',
+ 'media': 'http://search.yahoo.com/mrss/',
+ 'itunes': 'http://www.itunes.com/dtds/podcast-1.0.dtd',
+ })
item = rss.find('channel/item')
- video_id = item.find(blip('item_id')).text
- title = item.find('./title').text
- description = clean_html(compat_str(item.find(blip('puredescription')).text))
- timestamp = parse_iso8601(item.find(blip('datestamp')).text)
- uploader = item.find(blip('user')).text
- uploader_id = item.find(blip('userid')).text
- duration = int(item.find(blip('runtime')).text)
- media_thumbnail = item.find(media('thumbnail'))
- thumbnail = media_thumbnail.get('url') if media_thumbnail is not None else item.find(itunes('image')).text
- categories = [category.text for category in item.findall('category')]
+ video_id = xpath_text(item, _x('blip:item_id'), 'video id') or lookup_id
+ title = xpath_text(item, 'title', 'title', fatal=True)
+ description = clean_html(xpath_text(item, _x('blip:puredescription'), 'description'))
+ timestamp = parse_iso8601(xpath_text(item, _x('blip:datestamp'), 'timestamp'))
+ uploader = xpath_text(item, _x('blip:user'), 'uploader')
+ uploader_id = xpath_text(item, _x('blip:userid'), 'uploader id')
+ duration = int_or_none(xpath_text(item, _x('blip:runtime'), 'duration'))
+ media_thumbnail = item.find(_x('media:thumbnail'))
+ thumbnail = (media_thumbnail.get('url') if media_thumbnail is not None
+ else xpath_text(item, 'image', 'thumbnail'))
+ categories = [category.text for category in item.findall('category') if category is not None]
formats = []
- subtitles = {}
+ subtitles_urls = {}
- media_group = item.find(media('group'))
- for media_content in media_group.findall(media('content')):
+ media_group = item.find(_x('media:group'))
+ for media_content in media_group.findall(_x('media:content')):
url = media_content.get('url')
- role = media_content.get(blip('role'))
+ role = media_content.get(_x('blip:role'))
msg = self._download_webpage(
url + '?showplayer=20140425131715&referrer=http://blip.tv&mask=7&skin=flashvars&view=url',
video_id, 'Resolving URL for %s' % role)
- real_url = compat_urlparse.parse_qs(msg)['message'][0]
+ real_url = compat_urlparse.parse_qs(msg.strip())['message'][0]
media_type = media_content.get('type')
if media_type == 'text/srt' or url.endswith('.srt'):
}
lang = role.rpartition('-')[-1].strip().lower()
langcode = LANGS.get(lang, lang)
- subtitles[langcode] = url
+ subtitles_urls[langcode] = url
elif media_type.startswith('video/'):
formats.append({
'url': real_url,
'format_id': role,
'format_note': media_type,
- 'vcodec': media_content.get(blip('vcodec')),
- 'acodec': media_content.get(blip('acodec')),
+ 'vcodec': media_content.get(_x('blip:vcodec')) or 'none',
+ 'acodec': media_content.get(_x('blip:acodec')),
'filesize': media_content.get('filesize'),
- 'width': int(media_content.get('width')),
- 'height': int(media_content.get('height')),
+ 'width': int_or_none(media_content.get('width')),
+ 'height': int_or_none(media_content.get('height')),
})
+ self._check_formats(formats, video_id)
self._sort_formats(formats)
- # subtitles
- video_subtitles = self.extract_subtitles(video_id, subtitles)
- if self._downloader.params.get('listsubtitles', False):
- self._list_available_subtitles(video_id, subtitles)
- return
+ subtitles = self.extract_subtitles(video_id, subtitles_urls)
return {
'id': video_id,
'thumbnail': thumbnail,
'categories': categories,
'formats': formats,
- 'subtitles': video_subtitles,
+ 'subtitles': subtitles,
}
- def _download_subtitle_url(self, sub_lang, url):
- # For some weird reason, blip.tv serves a video instead of subtitles
- # when we request with a common UA
- req = compat_urllib_request.Request(url)
- req.add_header('Youtubedl-user-agent', 'youtube-dl')
- return self._download_webpage(req, None, note=False)
+ def _get_subtitles(self, video_id, subtitles_urls):
+ subtitles = {}
+ for lang, url in subtitles_urls.items():
+ # For some weird reason, blip.tv serves a video instead of subtitles
+ # when we request with a common UA
+ req = sanitized_Request(url)
+ req.add_header('User-Agent', 'youtube-dl')
+ subtitles[lang] = [{
+ # The extension is 'srt' but it's actually an 'ass' file
+ 'ext': 'ass',
+ 'data': self._download_webpage(req, None, note=False),
+ }]
+ return subtitles
class BlipTVUserIE(InfoExtractor):
- _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)(?!api\.swf)([^/]+)/*$'
+ _VALID_URL = r'(?:(?:https?://(?:\w+\.)?blip\.tv/)|bliptvuser:)(?!api\.swf)([^/]+)/*$'
_PAGE_SIZE = 12
IE_NAME = 'blip.tv:user'
+ _TEST = {
+ 'url': 'http://blip.tv/actone',
+ 'info_dict': {
+ 'id': 'actone',
+ 'title': 'Act One: The Series',
+ },
+ 'playlist_count': 5,
+ }
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
page = self._download_webpage(url, username, 'Downloading user page')
mobj = re.search(r'data-users-id="([^"]+)"', page)
page_base = page_base % mobj.group(1)
+ title = self._og_search_title(page)
# Download video ids using BlipTV Ajax calls. Result size per
# query is limited (currently to 12 videos) so we need to query
urls = ['http://blip.tv/%s' % video_id for video_id in video_ids]
url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
- return [self.playlist_result(url_entries, playlist_title=username)]
+ return self.playlist_result(
+ url_entries, playlist_title=title, playlist_id=username)
projects / youtube-dl / blobdiff