from __future__ import unicode_literals
-import datetime
-import json
import re
-import socket
from .common import InfoExtractor
+from .subtitles import SubtitlesInfoExtractor
from ..utils import (
- compat_http_client,
- compat_str,
- compat_urllib_error,
compat_urllib_request,
-
- ExtractorError,
unescapeHTML,
+ parse_iso8601,
+ compat_urlparse,
+ clean_html,
+ compat_str,
)
-class BlipTVIE(InfoExtractor):
- """Information extractor for blip.tv"""
-
- _VALID_URL = r'^(?:https?://)?(?:\w+\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(.+)$'
-
- _TEST = {
- 'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
- 'file': '5779306.mov',
- 'md5': 'c6934ad0b6acf2bd920720ec888eb812',
- 'info_dict': {
- 'upload_date': '20111205',
- 'description': 'md5:9bc31f227219cde65e47eeec8d2dc596',
- 'uploader': 'Comic Book Resources - CBR TV',
- 'title': 'CBR EXCLUSIVE: "Gotham City Imposters" Bats VS Jokerz Short 3',
+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 = [
+ {
+ 'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
+ 'md5': 'c6934ad0b6acf2bd920720ec888eb812',
+ 'info_dict': {
+ 'id': '5779306',
+ 'ext': 'mov',
+ 'title': 'CBR EXCLUSIVE: "Gotham City Imposters" Bats VS Jokerz Short 3',
+ 'description': 'md5:9bc31f227219cde65e47eeec8d2dc596',
+ 'timestamp': 1323138843,
+ 'upload_date': '20111206',
+ 'uploader': 'cbr',
+ 'uploader_id': '679425',
+ 'duration': 81,
+ }
+ },
+ {
+ # https://github.com/rg3/youtube-dl/pull/2274
+ 'note': 'Video with subtitles',
+ 'url': 'http://blip.tv/play/h6Uag5OEVgI.html',
+ 'md5': '309f9d25b820b086ca163ffac8031806',
+ 'info_dict': {
+ 'id': '6586561',
+ 'ext': 'mp4',
+ 'title': 'Red vs. Blue Season 11 Episode 1',
+ 'description': 'One-Zero-One',
+ 'timestamp': 1371261608,
+ 'upload_date': '20130615',
+ 'uploader': 'redvsblue',
+ 'uploader_id': '792887',
+ 'duration': 279,
+ }
}
- }
-
- def report_direct_download(self, title):
- """Report information extraction."""
- self.to_screen('%s: Direct download detected' % title)
+ ]
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
- if mobj is None:
- raise ExtractorError('Invalid URL: %s' % url)
+ lookup_id = mobj.group('lookup_id')
# See https://github.com/rg3/youtube-dl/issues/857
- embed_mobj = re.search(r'^(?:https?://)?(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)([a-zA-Z0-9]+)', url)
- if embed_mobj:
- info_url = 'http://blip.tv/play/%s.x?p=1' % embed_mobj.group(1)
- info_page = self._download_webpage(info_url, embed_mobj.group(1))
- video_id = self._search_regex(r'data-episode-id="(\d+)', info_page, 'video_id')
- return self.url_result('http://blip.tv/a/a-' + video_id, 'BlipTV')
-
- if '?' in url:
- cchar = '&'
+ 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:
- cchar = '?'
- json_url = url + cchar + 'skin=json&version=2&no_wrap=1'
- request = compat_urllib_request.Request(json_url)
- request.add_header('User-Agent', 'iTunes/10.6.1')
- self.report_extraction(mobj.group(1))
- urlh = self._request_webpage(request, None, False,
- 'unable to download video info webpage')
-
- try:
- json_code_bytes = urlh.read()
- json_code = json_code_bytes.decode('utf-8')
- except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
- raise ExtractorError('Unable to read video info webpage: %s' % compat_str(err))
-
- try:
- json_data = json.loads(json_code)
- if 'Post' in json_data:
- data = json_data['Post']
- else:
- data = json_data
-
- upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
- formats = []
- if 'additionalMedia' in data:
- for f in sorted(data['additionalMedia'], key=lambda f: int(f['media_height'])):
- if not int(f['media_width']): # filter m3u8
- continue
- formats.append({
- 'url': f['url'],
- 'format_id': f['role'],
- 'width': int(f['media_width']),
- 'height': int(f['media_height']),
- })
- else:
+ video_id = mobj.group('id')
+
+ 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
+
+ def media(s):
+ return '{http://search.yahoo.com/mrss/}%s' % s
+
+ def itunes(s):
+ return '{http://www.itunes.com/dtds/podcast-1.0.dtd}%s' % s
+
+ 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')]
+
+ formats = []
+ subtitles = {}
+
+ media_group = item.find(media('group'))
+ for media_content in media_group.findall(media('content')):
+ url = media_content.get('url')
+ role = media_content.get(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]
+
+ media_type = media_content.get('type')
+ if media_type == 'text/srt' or url.endswith('.srt'):
+ LANGS = {
+ 'english': 'en',
+ }
+ lang = role.rpartition('-')[-1].strip().lower()
+ langcode = LANGS.get(lang, lang)
+ subtitles[langcode] = url
+ elif media_type.startswith('video/'):
formats.append({
- 'url': data['media']['url'],
- 'width': int(data['media']['width']),
- 'height': int(data['media']['height']),
+ 'url': real_url,
+ 'format_id': role,
+ 'format_note': media_type,
+ 'vcodec': media_content.get(blip('vcodec')),
+ 'acodec': media_content.get(blip('acodec')),
+ 'filesize': media_content.get('filesize'),
+ 'width': int(media_content.get('width')),
+ 'height': int(media_content.get('height')),
})
+ 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
+
+ return {
+ 'id': video_id,
+ 'title': title,
+ 'description': description,
+ 'timestamp': timestamp,
+ 'uploader': uploader,
+ 'uploader_id': uploader_id,
+ 'duration': duration,
+ 'thumbnail': thumbnail,
+ 'categories': categories,
+ 'formats': formats,
+ 'subtitles': video_subtitles,
+ }
- self._sort_formats(formats)
-
- return {
- 'id': compat_str(data['item_id']),
- 'uploader': data['display_name'],
- 'upload_date': upload_date,
- 'title': data['title'],
- 'thumbnail': data['thumbnailUrl'],
- 'description': data['description'],
- 'user_agent': 'iTunes/10.6.1',
- 'formats': formats,
- }
- except (ValueError, KeyError) as err:
- raise ExtractorError('Unable to parse video information: %s' % repr(err))
+ 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)
class BlipTVUserIE(InfoExtractor):
- """Information Extractor for blip.tv users."""
-
_VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)([^/]+)/*$'
_PAGE_SIZE = 12
IE_NAME = 'blip.tv:user'
def _real_extract(self, url):
- # Extract username
mobj = re.match(self._VALID_URL, url)
- if mobj is None:
- raise ExtractorError('Invalid URL: %s' % url)
-
username = mobj.group(1)
page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
mobj = re.search(r'data-users-id="([^"]+)"', page)
page_base = page_base % mobj.group(1)
-
# Download video ids using BlipTV Ajax calls. Result size per
# query is limited (currently to 12 videos) so we need to query
# page by page until there are no video ids - it means we got
while True:
url = page_base + "&page=" + str(pagenum)
- page = self._download_webpage(url, username,
- 'Downloading video ids from page %d' % pagenum)
+ page = self._download_webpage(
+ url, username, 'Downloading video ids from page %d' % pagenum)
# Extract video identifiers
ids_in_page = []
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=username)]
Rapsys Git / youtubedl / blobdiff