commit 8fbf5d2f87fbfe0441bc20cf69d506109b2810bc
parent f81dd65ba2c1e7be549e5c8cfe6cbf0f0829edfe
Author: Remita Amine <remitamine@gmail.com>
Date:   Sat,  9 Nov 2019 13:14:23 +0100

[seeker] remove Revision3 extractors and fix extraction

Diffstat:
Myoutube_dl/extractor/extractors.py | 4----
Dyoutube_dl/extractor/revision3.py | 170-------------------------------------------------------------------------------
Myoutube_dl/extractor/seeker.py | 45+++++++++++++++++++++++----------------------
3 files changed, 23 insertions(+), 196 deletions(-)

diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py @@ -932,10 +932,6 @@ from .rentv import ( from .restudy import RestudyIE from .reuters import ReutersIE from .reverbnation import ReverbNationIE -from .revision3 import ( - Revision3EmbedIE, - Revision3IE, -) from .rice import RICEIE from .rmcdecouverte import RMCDecouverteIE from .ro220 import Ro220IE diff --git a/youtube_dl/extractor/revision3.py b/youtube_dl/extractor/revision3.py @@ -1,170 +0,0 @@ -# coding: utf-8 -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor -from ..compat import compat_str -from ..utils import ( - int_or_none, - parse_iso8601, - unescapeHTML, - qualities, -) - - -class Revision3EmbedIE(InfoExtractor): - IE_NAME = 'revision3:embed' - _VALID_URL = r'(?:revision3:(?:(?P<playlist_type>[^:]+):)?|https?://(?:(?:(?:www|embed)\.)?(?:revision3|animalist)|(?:(?:api|embed)\.)?seekernetwork)\.com/player/embed\?videoId=)(?P<playlist_id>\d+)' - _TEST = { - 'url': 'http://api.seekernetwork.com/player/embed?videoId=67558', - 'md5': '83bcd157cab89ad7318dd7b8c9cf1306', - 'info_dict': { - 'id': '67558', - 'ext': 'mp4', - 'title': 'The Pros & Cons Of Zoos', - 'description': 'Zoos are often depicted as a terrible place for animals to live, but is there any truth to this?', - 'uploader_id': 'dnews', - 'uploader': 'DNews', - } - } - _API_KEY = 'ba9c741bce1b9d8e3defcc22193f3651b8867e62' - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - playlist_id = mobj.group('playlist_id') - playlist_type = mobj.group('playlist_type') or 'video_id' - video_data = self._download_json( - 'http://revision3.com/api/getPlaylist.json', playlist_id, query={ - 'api_key': self._API_KEY, - 'codecs': 'h264,vp8,theora', - playlist_type: playlist_id, - })['items'][0] - - formats = [] - for vcodec, media in video_data['media'].items(): - for quality_id, quality in media.items(): - if quality_id == 'hls': - formats.extend(self._extract_m3u8_formats( - quality['url'], playlist_id, 'mp4', - 'm3u8_native', m3u8_id='hls', fatal=False)) - else: - formats.append({ - 'url': quality['url'], - 'format_id': '%s-%s' % (vcodec, quality_id), - 'tbr': int_or_none(quality.get('bitrate')), - 'vcodec': vcodec, - }) - self._sort_formats(formats) - - return { - 'id': playlist_id, - 'title': unescapeHTML(video_data['title']), - 'description': unescapeHTML(video_data.get('summary')), - 'uploader': video_data.get('show', {}).get('name'), - 'uploader_id': video_data.get('show', {}).get('slug'), - 'duration': int_or_none(video_data.get('duration')), - 'formats': formats, - } - - -class Revision3IE(InfoExtractor): - IE_NAME = 'revision' - _VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:revision3|animalist)\.com)/(?P<id>[^/]+(?:/[^/?#]+)?)' - _TESTS = [{ - 'url': 'http://www.revision3.com/technobuffalo/5-google-predictions-for-2016', - 'md5': 'd94a72d85d0a829766de4deb8daaf7df', - 'info_dict': { - 'id': '71089', - 'display_id': 'technobuffalo/5-google-predictions-for-2016', - 'ext': 'webm', - 'title': '5 Google Predictions for 2016', - 'description': 'Google had a great 2015, but it\'s already time to look ahead. Here are our five predictions for 2016.', - 'upload_date': '20151228', - 'timestamp': 1451325600, - 'duration': 187, - 'uploader': 'TechnoBuffalo', - 'uploader_id': 'technobuffalo', - } - }, { - # Show - 'url': 'http://revision3.com/variant', - 'only_matching': True, - }, { - # Tag - 'url': 'http://revision3.com/vr', - 'only_matching': True, - }] - _PAGE_DATA_TEMPLATE = 'http://www.%s/apiProxy/ddn/%s?domain=%s' - - def _real_extract(self, url): - domain, display_id = re.match(self._VALID_URL, url).groups() - site = domain.split('.')[0] - page_info = self._download_json( - self._PAGE_DATA_TEMPLATE % (domain, display_id, domain), display_id) - - page_data = page_info['data'] - page_type = page_data['type'] - if page_type in ('episode', 'embed'): - show_data = page_data['show']['data'] - page_id = compat_str(page_data['id']) - video_id = compat_str(page_data['video']['data']['id']) - - preference = qualities(['mini', 'small', 'medium', 'large']) - thumbnails = [{ - 'url': image_url, - 'id': image_id, - 'preference': preference(image_id) - } for image_id, image_url in page_data.get('images', {}).items()] - - info = { - 'id': page_id, - 'display_id': display_id, - 'title': unescapeHTML(page_data['name']), - 'description': unescapeHTML(page_data.get('summary')), - 'timestamp': parse_iso8601(page_data.get('publishTime'), ' '), - 'author': page_data.get('author'), - 'uploader': show_data.get('name'), - 'uploader_id': show_data.get('slug'), - 'thumbnails': thumbnails, - 'extractor_key': site, - } - - if page_type == 'embed': - info.update({ - '_type': 'url_transparent', - 'url': page_data['video']['data']['embed'], - }) - return info - - info.update({ - '_type': 'url_transparent', - 'url': 'revision3:%s' % video_id, - }) - return info - else: - list_data = page_info[page_type]['data'] - episodes_data = page_info['episodes']['data'] - num_episodes = page_info['meta']['totalEpisodes'] - processed_episodes = 0 - entries = [] - page_num = 1 - while True: - entries.extend([{ - '_type': 'url', - 'url': 'http://%s%s' % (domain, episode['path']), - 'id': compat_str(episode['id']), - 'ie_key': 'Revision3', - 'extractor_key': site, - } for episode in episodes_data]) - processed_episodes += len(episodes_data) - if processed_episodes == num_episodes: - break - page_num += 1 - episodes_data = self._download_json(self._PAGE_DATA_TEMPLATE % ( - domain, display_id + '/' + compat_str(page_num), domain), - display_id)['episodes']['data'] - - return self.playlist_result( - entries, compat_str(list_data['id']), - list_data.get('name'), list_data.get('summary')) diff --git a/youtube_dl/extractor/seeker.py b/youtube_dl/extractor/seeker.py @@ -4,34 +4,37 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..utils import ( + get_element_by_class, + strip_or_none, +) class SeekerIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?seeker\.com/(?P<display_id>.*)-(?P<article_id>\d+)\.html' _TESTS = [{ - # player.loadRevision3Item 'url': 'http://www.seeker.com/should-trump-be-required-to-release-his-tax-returns-1833805621.html', - 'md5': '30c1dc4030cc715cf05b423d0947ac18', + 'md5': '897d44bbe0d8986a2ead96de565a92db', 'info_dict': { - 'id': '76243', - 'ext': 'webm', + 'id': 'Elrn3gnY', + 'ext': 'mp4', 'title': 'Should Trump Be Required To Release His Tax Returns?', - 'description': 'Donald Trump has been secretive about his "big," "beautiful" tax returns. So what can we learn if he decides to release them?', - 'uploader': 'Seeker Daily', - 'uploader_id': 'seekerdaily', + 'description': 'md5:41efa8cfa8d627841045eec7b018eb45', + 'timestamp': 1490090165, + 'upload_date': '20170321', } }, { 'url': 'http://www.seeker.com/changes-expected-at-zoos-following-recent-gorilla-lion-shootings-1834116536.html', 'playlist': [ { - 'md5': '83bcd157cab89ad7318dd7b8c9cf1306', + 'md5': '0497b9f20495174be73ae136949707d2', 'info_dict': { - 'id': '67558', + 'id': 'FihYQ8AE', 'ext': 'mp4', 'title': 'The Pros & Cons Of Zoos', - 'description': 'Zoos are often depicted as a terrible place for animals to live, but is there any truth to this?', - 'uploader': 'DNews', - 'uploader_id': 'dnews', + 'description': 'md5:d88f99a8ea8e7d25e6ff77f271b1271c', + 'timestamp': 1490039133, + 'upload_date': '20170320', }, } ], @@ -45,13 +48,11 @@ class SeekerIE(InfoExtractor): def _real_extract(self, url): display_id, article_id = re.match(self._VALID_URL, url).groups() webpage = self._download_webpage(url, display_id) - mobj = re.search(r"player\.loadRevision3Item\('([^']+)'\s*,\s*(\d+)\);", webpage) - if mobj: - playlist_type, playlist_id = mobj.groups() - return self.url_result( - 'revision3:%s:%s' % (playlist_type, playlist_id), 'Revision3Embed', playlist_id) - else: - entries = [self.url_result('revision3:video_id:%s' % video_id, 'Revision3Embed', video_id) for video_id in re.findall( - r'<iframe[^>]+src=[\'"](?:https?:)?//api\.seekernetwork\.com/player/embed\?videoId=(\d+)', webpage)] - return self.playlist_result( - entries, article_id, self._og_search_title(webpage), self._og_search_description(webpage)) + entries = [] + for jwp_id in re.findall(r'data-video-id="([a-zA-Z0-9]{8})"', webpage): + entries.append(self.url_result( + 'jwplatform:' + jwp_id, 'JWPlatform', jwp_id)) + return self.playlist_result( + entries, article_id, + self._og_search_title(webpage), + strip_or_none(get_element_by_class('subtitle__text', webpage)) or self._og_search_description(webpage))