X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/ca4d08063804fb264eb0ae9cc57894198f66e1fb..435d023e9162287eb720b3cfb54d0ad750e3ac92:/youtube_dl/extractor/bliptv.py diff --git a/youtube_dl/extractor/bliptv.py b/youtube_dl/extractor/bliptv.py index 5e33a69..35375f7 100644 --- a/youtube_dl/extractor/bliptv.py +++ b/youtube_dl/extractor/bliptv.py @@ -1,156 +1,255 @@ -import datetime -import json -import os +from __future__ import unicode_literals + import re -import socket from .common import InfoExtractor + +from ..compat import compat_urlparse from ..utils import ( - compat_http_client, - compat_parse_qs, - compat_str, - compat_urllib_error, - compat_urllib_parse_urlparse, - compat_urllib_request, - - ExtractorError, + clean_html, + int_or_none, + parse_iso8601, + sanitized_Request, unescapeHTML, + xpath_text, + xpath_with_ns, ) class BlipTVIE(InfoExtractor): - """Information extractor for blip.tv""" - - _VALID_URL = r'^(?:https?://)?(?:\w+\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(.+)$' - _URL_EXT = r'^.*\.([a-z0-9]+)$' - IE_NAME = u'blip.tv' - _TEST = { - u'url': u'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352', - u'file': u'5779306.m4v', - u'md5': u'80baf1ec5c3d2019037c1c707d676b9f', - u'info_dict': { - u"upload_date": u"20111205", - u"description": u"md5:9bc31f227219cde65e47eeec8d2dc596", - u"uploader": u"Comic Book Resources - CBR TV", - u"title": u"CBR EXCLUSIVE: \"Gotham City Imposters\" Bats VS Jokerz Short 3" + _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/(?:(?:.+-|rss/flash/)(?P\d+)|((?:play/|api\.swf#)(?P[\da-zA-Z+_]+)))' + + _TESTS = [ + { + 'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352', + 'md5': '80baf1ec5c3d2019037c1c707d676b9f', + 'info_dict': { + 'id': '5779306', + 'ext': 'm4v', + '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, + } + }, + { + # https://bugzilla.redhat.com/show_bug.cgi?id=967465 + 'url': 'http://a.blip.tv/api.swf#h6Uag5KbVwI', + 'md5': '314e87b1ebe7a48fcbfdd51b791ce5a6', + 'info_dict': { + 'id': '6573122', + 'ext': 'mov', + 'upload_date': '20130520', + 'description': 'Two hapless space marines argue over what to do when they realize they have an astronomically huge problem on their hands.', + 'title': 'Red vs. Blue Season 11 Trailer', + 'timestamp': 1369029609, + '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', + } } - } + ] - def report_direct_download(self, title): - """Report information extraction.""" - self.to_screen(u'%s: Direct download detected' % title) + @staticmethod + def _extract_url(webpage): + mobj = re.search(r']*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) - if mobj is None: - raise ExtractorError(u'Invalid URL: %s' % url) - - # See https://github.com/rg3/youtube-dl/issues/857 - api_mobj = re.match(r'http://a\.blip\.tv/api\.swf#(?P[\d\w]+)', url) - if api_mobj is not None: - url = 'http://blip.tv/play/g_%s' % api_mobj.group('video_id') - urlp = compat_urllib_parse_urlparse(url) - if urlp.path.startswith('/play/'): - response = self._request_webpage(url, None, False) - redirecturl = response.geturl() - rurlp = compat_urllib_parse_urlparse(redirecturl) - file_id = compat_parse_qs(rurlp.fragment)['file'][0].rpartition('/')[2] - url = 'http://blip.tv/a/a-' + file_id - return self._real_extract(url) - - - if '?' in url: - cchar = '&' - 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)) - info = None - urlh = self._request_webpage(request, None, False, - u'unable to download video info webpage') - if urlh.headers.get('Content-Type', '').startswith('video/'): # Direct download - basename = url.split('/')[-1] - title,ext = os.path.splitext(basename) - title = title.decode('UTF-8') - ext = ext.replace('.', '') - self.report_direct_download(title) - info = { - 'id': title, - 'url': url, - 'uploader': None, - 'upload_date': None, - 'title': title, - 'ext': ext, - 'urlhandle': urlh - } - if info is None: # Regular URL - 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(u'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') - if 'additionalMedia' in data: - formats = sorted(data['additionalMedia'], key=lambda f: int(f['media_height'])) - best_format = formats[-1] - video_url = best_format['url'] - else: - video_url = data['media']['url'] - umobj = re.match(self._URL_EXT, video_url) - if umobj is None: - raise ValueError('Can not determine filename extension') - ext = umobj.group(1) - - info = { - 'id': compat_str(data['item_id']), - 'url': video_url, - 'uploader': data['display_name'], - 'upload_date': upload_date, - 'title': data['title'], - 'ext': ext, - 'format': data['media']['mimeType'], - 'thumbnail': data['thumbnailUrl'], - 'description': data['description'], - 'player_url': data['embedUrl'], - 'user_agent': 'iTunes/10.6.1', + lookup_id = mobj.group('lookup_id') + + # See https://github.com/rg3/youtube-dl/issues/857 and + # https://github.com/rg3/youtube-dl/issues/4197 + if lookup_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]) + + video_id = mobj.group('id') + + rss = self._download_xml('http://blip.tv/rss/flash/%s' % video_id, video_id, 'Downloading video RSS') + + 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 = 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_urls = {} + + 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(_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.strip())['message'][0] + + media_type = media_content.get('type') + if media_type == 'text/srt' or url.endswith('.srt'): + LANGS = { + 'english': 'en', } - except (ValueError,KeyError) as err: - raise ExtractorError(u'Unable to parse video information: %s' % repr(err)) + lang = role.rpartition('-')[-1].strip().lower() + langcode = LANGS.get(lang, lang) + 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(_x('blip:vcodec')) or 'none', + 'acodec': media_content.get(_x('blip:acodec')), + 'filesize': media_content.get('filesize'), + '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 = self.extract_subtitles(video_id, subtitles_urls) + + 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': subtitles, + } - return [info] + 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): - """Information Extractor for blip.tv users.""" - - _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)([^/]+)/*$' + _VALID_URL = r'(?:(?:https?://(?:\w+\.)?blip\.tv/)|bliptvuser:)(?!api\.swf)([^/]+)/*$' _PAGE_SIZE = 12 - IE_NAME = u'blip.tv:user' + 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): - # Extract username mobj = re.match(self._VALID_URL, url) - if mobj is None: - raise ExtractorError(u'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' - page = self._download_webpage(url, username, u'Downloading user page') + 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 @@ -162,8 +261,8 @@ class BlipTVUserIE(InfoExtractor): while True: url = page_base + "&page=" + str(pagenum) - page = self._download_webpage(url, username, - u'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 = [] @@ -185,6 +284,7 @@ class BlipTVUserIE(InfoExtractor): pagenum += 1 - urls = [u'http://blip.tv/%s' % video_id for video_id in video_ids] + 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)