diff options
Diffstat (limited to 'youtube_dl/extractor')
| -rw-r--r-- | youtube_dl/extractor/__init__.py | 8 | ||||
| -rw-r--r-- | youtube_dl/extractor/ctsnews.py | 93 | ||||
| -rw-r--r-- | youtube_dl/extractor/generic.py | 24 | ||||
| -rw-r--r-- | youtube_dl/extractor/ivi.py | 27 | ||||
| -rw-r--r-- | youtube_dl/extractor/nextmedia.py | 163 | ||||
| -rw-r--r-- | youtube_dl/extractor/srmediathek.py | 2 | ||||
| -rw-r--r-- | youtube_dl/extractor/viddler.py | 63 | ||||
| -rw-r--r-- | youtube_dl/extractor/xuite.py | 142 |
8 files changed, 493 insertions, 29 deletions
diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py index 873ae69d3..d3f51d8c4 100644 --- a/youtube_dl/extractor/__init__.py +++ b/youtube_dl/extractor/__init__.py @@ -82,6 +82,7 @@ from .crunchyroll import ( CrunchyrollShowPlaylistIE ) from .cspan import CSpanIE +from .ctsnews import CtsNewsIE from .dailymotion import ( DailymotionIE, DailymotionPlaylistIE, @@ -285,6 +286,12 @@ from .netzkino import NetzkinoIE from .nerdcubed import NerdCubedFeedIE from .newgrounds import NewgroundsIE from .newstube import NewstubeIE +from .nextmedia import ( + NextMediaIE, + NextMediaActionNewsIE, + AppleDailyRealtimeNewsIE, + AppleDailyAnimationNewsIE +) from .nfb import NFBIE from .nfl import NFLIE from .nhl import NHLIE, NHLVideocenterIE @@ -547,6 +554,7 @@ from .xminus import XMinusIE from .xnxx import XNXXIE from .xvideos import XVideosIE from .xtube import XTubeUserIE, XTubeIE +from .xuite import XuiteIE from .xxxymovies import XXXYMoviesIE from .yahoo import ( YahooIE, diff --git a/youtube_dl/extractor/ctsnews.py b/youtube_dl/extractor/ctsnews.py new file mode 100644 index 000000000..0226f8036 --- /dev/null +++ b/youtube_dl/extractor/ctsnews.py @@ -0,0 +1,93 @@ +# -*- coding: utf-8 -*- +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import parse_iso8601, ExtractorError + + +class CtsNewsIE(InfoExtractor): + # https connection failed (Connection reset) + _VALID_URL = r'http://news\.cts\.com\.tw/[a-z]+/[a-z]+/\d+/(?P<id>\d+)\.html' + _TESTS = [{ + 'url': 'http://news.cts.com.tw/cts/international/201501/201501291578109.html', + 'md5': 'a9875cb790252b08431186d741beaabe', + 'info_dict': { + 'id': '201501291578109', + 'ext': 'mp4', + 'title': '以色列.真主黨交火 3人死亡', + 'description': 'md5:95e9b295c898b7ff294f09d450178d7d', + 'timestamp': 1422528540, + 'upload_date': '20150129', + } + }, { + # News count not appear on page but still available in database + 'url': 'http://news.cts.com.tw/cts/international/201309/201309031304098.html', + 'md5': '3aee7e0df7cdff94e43581f54c22619e', + 'info_dict': { + 'id': '201309031304098', + 'ext': 'mp4', + 'title': '韓國31歲童顏男 貌如十多歲小孩', + 'description': 'md5:f183feeba3752b683827aab71adad584', + 'thumbnail': 're:^https?://.*\.jpg$', + 'timestamp': 1378205880, + 'upload_date': '20130903', + } + }, { + # With Youtube embedded video + 'url': 'http://news.cts.com.tw/cts/money/201501/201501291578003.html', + 'md5': '1d842c771dc94c8c3bca5af2cc1db9c5', + 'add_ie': ['Youtube'], + 'info_dict': { + 'id': 'OVbfO7d0_hQ', + 'ext': 'mp4', + 'title': 'iPhone6熱銷 蘋果財報亮眼', + 'description': 'md5:f395d4f485487bb0f992ed2c4b07aa7d', + 'thumbnail': 're:^https?://.*\.jpg$', + 'upload_date': '20150128', + 'uploader_id': 'TBSCTS', + 'uploader': '中華電視公司', + } + }] + + def _real_extract(self, url): + news_id = self._match_id(url) + page = self._download_webpage(url, news_id) + + if self._search_regex(r'(CTSPlayer2)', page, 'CTSPlayer2 identifier', default=None): + feed_url = self._html_search_regex( + r'(http://news\.cts\.com\.tw/action/mp4feed\.php\?news_id=\d+)', + page, 'feed url') + video_url = self._download_webpage( + feed_url, news_id, note='Fetching feed') + else: + self.to_screen('Not CTSPlayer video, trying Youtube...') + youtube_url = self._search_regex( + r'src="(//www\.youtube\.com/embed/[^"]+)"', page, 'youtube url', + default=None) + if not youtube_url: + raise ExtractorError('The news includes no videos!', expected=True) + + return { + '_type': 'url', + 'url': youtube_url, + 'ie_key': 'Youtube', + } + + description = self._html_search_meta('description', page) + title = self._html_search_meta('title', page) + thumbnail = self._html_search_meta('image', page) + + datetime_str = self._html_search_regex( + r'(\d{4}/\d{2}/\d{2} \d{2}:\d{2})', page, 'date and time') + # Transform into ISO 8601 format with timezone info + datetime_str = datetime_str.replace('/', '-') + ':00+0800' + timestamp = parse_iso8601(datetime_str, delimiter=' ') + + return { + 'id': news_id, + 'url': video_url, + 'title': title, + 'description': description, + 'thumbnail': thumbnail, + 'timestamp': timestamp, + } diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index ad16b8330..41884ed7a 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -498,6 +498,19 @@ class GenericIE(InfoExtractor): 'uploader': 'www.abc.net.au', 'title': 'Game of Thrones with dice - Dungeons and Dragons fantasy role-playing game gets new life - 19/01/2015', } + }, + # embedded viddler video + { + 'url': 'http://deadspin.com/i-cant-stop-watching-john-wall-chop-the-nuggets-with-th-1681801597', + 'info_dict': { + 'id': '4d03aad9', + 'ext': 'mp4', + 'uploader': 'deadspin', + 'title': 'WALL-TO-GORTAT', + 'timestamp': 1422285291, + 'upload_date': '20150126', + }, + 'add_ie': ['Viddler'], } ] @@ -860,9 +873,16 @@ class GenericIE(InfoExtractor): if mobj is not None: return self.url_result(mobj.group('url')) + # Look for embedded Viddler player + mobj = re.search( + r'<(?:iframe[^>]+?src|param[^>]+?value)=(["\'])(?P<url>(?:https?:)?//(?:www\.)?viddler\.com/(?:embed|player)/.+?)\1', + webpage) + if mobj is not None: + return self.url_result(mobj.group('url')) + # Look for Ooyala videos - mobj = (re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or - re.search(r'OO.Player.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage)) + mobj = (re.search(r'player\.ooyala\.com/[^"?]+\?[^"]*?(?:embedCode|ec)=(?P<ec>[^"&]+)', webpage) or + re.search(r'OO\.Player\.create\([\'"].*?[\'"],\s*[\'"](?P<ec>.{32})[\'"]', webpage)) if mobj is not None: return OoyalaIE._build_url_result(mobj.group('ec')) diff --git a/youtube_dl/extractor/ivi.py b/youtube_dl/extractor/ivi.py index 7a400323d..e82594444 100644 --- a/youtube_dl/extractor/ivi.py +++ b/youtube_dl/extractor/ivi.py @@ -16,7 +16,7 @@ from ..utils import ( class IviIE(InfoExtractor): IE_DESC = 'ivi.ru' IE_NAME = 'ivi' - _VALID_URL = r'https?://(?:www\.)?ivi\.ru/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<videoid>\d+)' + _VALID_URL = r'https?://(?:www\.)?ivi\.ru/(?:watch/(?:[^/]+/)?|video/player\?.*?videoId=)(?P<id>\d+)' _TESTS = [ # Single movie @@ -63,29 +63,34 @@ class IviIE(InfoExtractor): return int(m.group('commentcount')) if m is not None else 0 def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('videoid') + video_id = self._match_id(url) api_url = 'http://api.digitalaccess.ru/api/json/' - data = {'method': 'da.content.get', - 'params': [video_id, {'site': 's183', - 'referrer': 'http://www.ivi.ru/watch/%s' % video_id, - 'contentid': video_id - } - ] + data = { + 'method': 'da.content.get', + 'params': [ + video_id, { + 'site': 's183', + 'referrer': 'http://www.ivi.ru/watch/%s' % video_id, + 'contentid': video_id } + ] + } request = compat_urllib_request.Request(api_url, json.dumps(data)) - video_json_page = self._download_webpage(request, video_id, 'Downloading video JSON') + video_json_page = self._download_webpage( + request, video_id, 'Downloading video JSON') video_json = json.loads(video_json_page) if 'error' in video_json: error = video_json['error'] if error['origin'] == 'NoRedisValidData': raise ExtractorError('Video %s does not exist' % video_id, expected=True) - raise ExtractorError('Unable to download video %s: %s' % (video_id, error['message']), expected=True) + raise ExtractorError( + 'Unable to download video %s: %s' % (video_id, error['message']), + expected=True) result = video_json['result'] diff --git a/youtube_dl/extractor/nextmedia.py b/youtube_dl/extractor/nextmedia.py new file mode 100644 index 000000000..02dba4ef6 --- /dev/null +++ b/youtube_dl/extractor/nextmedia.py @@ -0,0 +1,163 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import parse_iso8601 + + +class NextMediaIE(InfoExtractor): + _VALID_URL = r'http://hk.apple.nextmedia.com/[^/]+/[^/]+/(?P<date>\d+)/(?P<id>\d+)' + _TESTS = [{ + 'url': 'http://hk.apple.nextmedia.com/realtime/news/20141108/53109199', + 'md5': 'dff9fad7009311c421176d1ac90bfe4f', + 'info_dict': { + 'id': '53109199', + 'ext': 'mp4', + 'title': '【佔領金鐘】50外國領事議員撐場 讚學生勇敢香港有希望', + 'thumbnail': 're:^https?://.*\.jpg$', + 'description': 'md5:28222b9912b6665a21011b034c70fcc7', + 'timestamp': 1415456273, + 'upload_date': '20141108', + } + }] + + _URL_PATTERN = r'\{ url: \'(.+)\' \}' + + def _real_extract(self, url): + news_id = self._match_id(url) + page = self._download_webpage(url, news_id) + return self._extract_from_nextmedia_page(news_id, url, page) + + def _extract_from_nextmedia_page(self, news_id, url, page): + title = self._fetch_title(page) + video_url = self._search_regex(self._URL_PATTERN, page, 'video url') + + attrs = { + 'id': news_id, + 'title': title, + 'url': video_url, # ext can be inferred from url + 'thumbnail': self._fetch_thumbnail(page), + 'description': self._fetch_description(page), + } + + timestamp = self._fetch_timestamp(page) + if timestamp: + attrs['timestamp'] = timestamp + else: + attrs['upload_date'] = self._fetch_upload_date(url) + + return attrs + + def _fetch_title(self, page): + return self._og_search_title(page) + + def _fetch_thumbnail(self, page): + return self._og_search_thumbnail(page) + + def _fetch_timestamp(self, page): + dateCreated = self._search_regex('"dateCreated":"([^"]+)"', page, 'created time') + return parse_iso8601(dateCreated) + + def _fetch_upload_date(self, url): + return self._search_regex(self._VALID_URL, url, 'upload date', group='date') + + def _fetch_description(self, page): + return self._og_search_property('description', page) + + +class NextMediaActionNewsIE(NextMediaIE): + _VALID_URL = r'http://hk.dv.nextmedia.com/actionnews/[^/]+/(?P<date>\d+)/(?P<id>\d+)/\d+' + _TESTS = [{ + 'url': 'http://hk.dv.nextmedia.com/actionnews/hit/20150121/19009428/20061460', + 'md5': '05fce8ffeed7a5e00665d4b7cf0f9201', + 'info_dict': { + 'id': '19009428', + 'ext': 'mp4', + 'title': '【壹週刊】細10年男友偷食 50歲邵美琪再失戀', + 'thumbnail': 're:^https?://.*\.jpg$', + 'description': 'md5:cd802fad1f40fd9ea178c1e2af02d659', + 'timestamp': 1421791200, + 'upload_date': '20150120', + } + }] + + def _real_extract(self, url): + news_id = self._match_id(url) + actionnews_page = self._download_webpage(url, news_id) + article_url = self._og_search_url(actionnews_page) + article_page = self._download_webpage(article_url, news_id) + return self._extract_from_nextmedia_page(news_id, url, article_page) + + +class AppleDailyRealtimeNewsIE(NextMediaIE): + _VALID_URL = r'http://(www|ent).appledaily.com.tw/(realtimenews|enews)/[^/]+/[^/]+/(?P<date>\d+)/(?P<id>\d+)(/.*)?' + _TESTS = [{ + 'url': 'http://ent.appledaily.com.tw/enews/article/entertainment/20150128/36354694', + 'md5': 'a843ab23d150977cc55ef94f1e2c1e4d', + 'info_dict': { + 'id': '36354694', + 'ext': 'mp4', + 'title': '周亭羽走過摩鐵陰霾2男陪吃 九把刀孤寒看醫生', + 'thumbnail': 're:^https?://.*\.jpg$', + 'description': 'md5:b23787119933404ce515c6356a8c355c', + 'upload_date': '20150128', + } + }, { + 'url': 'http://www.appledaily.com.tw/realtimenews/article/strange/20150128/550549/%E4%B8%8D%E6%BB%BF%E8%A2%AB%E8%B8%A9%E8%85%B3%E3%80%80%E5%B1%B1%E6%9D%B1%E5%85%A9%E5%A4%A7%E5%AA%BD%E4%B8%80%E8%B7%AF%E6%89%93%E4%B8%8B%E8%BB%8A', + 'md5': '86b4e9132d158279c7883822d94ccc49', + 'info_dict': { + 'id': '550549', + 'ext': 'mp4', + 'title': '不滿被踩腳 山東兩大媽一路打下車', + 'thumbnail': 're:^https?://.*\.jpg$', + 'description': 'md5:2648aaf6fc4f401f6de35a91d111aa1d', + 'upload_date': '20150128', + } + }] + + _URL_PATTERN = r'\{url: \'(.+)\'\}' + + def _fetch_title(self, page): + return self._html_search_regex(r'<h1 id="h1">([^<>]+)</h1>', page, 'news title') + + def _fetch_thumbnail(self, page): + return self._html_search_regex(r"setInitialImage\(\'([^']+)'\)", page, 'video thumbnail', fatal=False) + + def _fetch_timestamp(self, page): + return None + + +class AppleDailyAnimationNewsIE(AppleDailyRealtimeNewsIE): + _VALID_URL = 'http://www.appledaily.com.tw/animation/[^/]+/[^/]+/(?P<date>\d+)/(?P<id>\d+)(/.*)?' + _TESTS = [{ + 'url': 'http://www.appledaily.com.tw/animation/realtimenews/new/20150128/5003671', + 'md5': '03df296d95dedc2d5886debbb80cb43f', + 'info_dict': { + 'id': '5003671', + 'ext': 'mp4', + 'title': '20正妹熱舞 《刀龍傳說Online》火辣上市', + 'thumbnail': 're:^https?://.*\.jpg$', + 'description': 'md5:23c0aac567dc08c9c16a3161a2c2e3cd', + 'upload_date': '20150128', + } + }, { + # No thumbnail + 'url': 'http://www.appledaily.com.tw/animation/realtimenews/new/20150128/5003673/', + 'md5': 'b06182cd386ea7bc6115ec7ff0f72aeb', + 'info_dict': { + 'id': '5003673', + 'ext': 'mp4', + 'title': '半夜尿尿 好像會看到___', + 'description': 'md5:61d2da7fe117fede148706cdb85ac066', + 'upload_date': '20150128', + }, + 'expected_warnings': [ + 'video thumbnail', + ] + }] + + def _fetch_title(self, page): + return self._html_search_meta('description', page, 'news title') + + def _fetch_description(self, page): + return self._html_search_meta('description', page, 'news description') diff --git a/youtube_dl/extractor/srmediathek.py b/youtube_dl/extractor/srmediathek.py index 666a7dcc8..5d583c720 100644 --- a/youtube_dl/extractor/srmediathek.py +++ b/youtube_dl/extractor/srmediathek.py @@ -8,7 +8,7 @@ from ..utils import js_to_json class SRMediathekIE(InfoExtractor): - IE_DESC = 'Süddeutscher Rundfunk' + IE_DESC = 'Saarländischer Rundfunk' _VALID_URL = r'https?://sr-mediathek\.sr-online\.de/index\.php\?.*?&id=(?P<id>[0-9]+)' _TEST = { diff --git a/youtube_dl/extractor/viddler.py b/youtube_dl/extractor/viddler.py index 0faa729c6..8516a2940 100644 --- a/youtube_dl/extractor/viddler.py +++ b/youtube_dl/extractor/viddler.py @@ -5,27 +5,58 @@ from ..utils import ( float_or_none, int_or_none, ) +from ..compat import ( + compat_urllib_request +) class ViddlerIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?viddler\.com/(?:v|embed|player)/(?P<id>[a-z0-9]+)' - _TEST = { - "url": "http://www.viddler.com/v/43903784", + _TESTS = [{ + 'url': 'http://www.viddler.com/v/43903784', 'md5': 'ae43ad7cb59431ce043f0ff7fa13cbf4', 'info_dict': { 'id': '43903784', 'ext': 'mp4', - "title": "Video Made Easy", - 'description': 'You don\'t need to be a professional to make high-quality video content. Viddler provides some quick and easy tips on how to produce great video content with limited resources. ', - "uploader": "viddler", + 'title': 'Video Made Easy', + 'description': 'md5:6a697ebd844ff3093bd2e82c37b409cd', + 'uploader': 'viddler', 'timestamp': 1335371429, 'upload_date': '20120425', - "duration": 100.89, + 'duration': 100.89, 'thumbnail': 're:^https?://.*\.jpg$', 'view_count': int, + 'comment_count': int, 'categories': ['video content', 'high quality video', 'video made easy', 'how to produce video with limited resources', 'viddler'], } - } + }, { + 'url': 'http://www.viddler.com/v/4d03aad9/', + 'md5': 'faa71fbf70c0bee7ab93076fd007f4b0', + 'info_dict': { + 'id': '4d03aad9', + 'ext': 'mp4', + 'title': 'WALL-TO-GORTAT', + 'upload_date': '20150126', + 'uploader': 'deadspin', + 'timestamp': 1422285291, + 'view_count': int, + 'comment_count': int, + } + }, { + 'url': 'http://www.viddler.com/player/221ebbbd/0/', + 'md5': '0defa2bd0ea613d14a6e9bd1db6be326', + 'info_dict': { + 'id': '221ebbbd', + 'ext': 'mp4', + 'title': 'LETeens-Grammar-snack-third-conditional', + 'description': ' ', + 'upload_date': '20140929', + 'uploader': 'BCLETeens', + 'timestamp': 1411997190, + 'view_count': int, + 'comment_count': int, + } + }] def _real_extract(self, url): video_id = self._match_id(url) @@ -33,14 +64,17 @@ class ViddlerIE(InfoExtractor): json_url = ( 'http://api.viddler.com/api/v2/viddler.videos.getPlaybackDetails.json?video_id=%s&key=v0vhrt7bg2xq1vyxhkct' % video_id) - data = self._download_json(json_url, video_id)['video'] + headers = {'Referer': 'http://static.cdn-ec.viddler.com/js/arpeggio/v2/embed.html'} + request = compat_urllib_request.Request(json_url, None, headers) + data = self._download_json(request, video_id)['video'] formats = [] for filed in data['files']: if filed.get('status', 'ready') != 'ready': continue + format_id = filed.get('profile_id') or filed['profile_name'] f = { - 'format_id': filed['profile_id'], + 'format_id': format_id, 'format_note': filed['profile_name'], 'url': self._proto_relative_url(filed['url']), 'width': int_or_none(filed.get('width')), @@ -53,16 +87,15 @@ class ViddlerIE(InfoExtractor): if filed.get('cdn_url'): f = f.copy() - f['url'] = self._proto_relative_url(filed['cdn_url']) - f['format_id'] = filed['profile_id'] + '-cdn' + f['url'] = self._proto_relative_url(filed['cdn_url'], 'http:') + f['format_id'] = format_id + '-cdn' f['source_preference'] = 1 formats.append(f) if filed.get('html5_video_source'): f = f.copy() - f['url'] = self._proto_relative_url( - filed['html5_video_source']) - f['format_id'] = filed['profile_id'] + '-html5' + f['url'] = self._proto_relative_url(filed['html5_video_source']) + f['format_id'] = format_id + '-html5' f['source_preference'] = 0 formats.append(f) self._sort_formats(formats) @@ -71,7 +104,6 @@ class ViddlerIE(InfoExtractor): t.get('text') for t in data.get('tags', []) if 'text' in t] return { - '_type': 'video', 'id': video_id, 'title': data['title'], 'formats': formats, @@ -81,5 +113,6 @@ class ViddlerIE(InfoExtractor): 'uploader': data.get('author'), 'duration': float_or_none(data.get('length')), 'view_count': int_or_none(data.get('view_count')), + 'comment_count': int_or_none(data.get('comment_count')), 'categories': categories, } diff --git a/youtube_dl/extractor/xuite.py b/youtube_dl/extractor/xuite.py new file mode 100644 index 000000000..4971965f9 --- /dev/null +++ b/youtube_dl/extractor/xuite.py @@ -0,0 +1,142 @@ +# -*- coding: utf-8 -*- +from __future__ import unicode_literals + +import base64 + +from .common import InfoExtractor +from ..compat import compat_urllib_parse_unquote +from ..utils import ( + ExtractorError, + parse_iso8601, + parse_duration, +) + + +class XuiteIE(InfoExtractor): + _REGEX_BASE64 = r'(?:[A-Za-z0-9+/]{4})*(?:[A-Za-z0-9+/]{2}==|[A-Za-z0-9+/]{3}=)?' + _VALID_URL = r'https?://vlog\.xuite\.net/(?:play|embed)/(?P<id>%s)' % _REGEX_BASE64 + _TESTS = [{ + # Audio + 'url': 'http://vlog.xuite.net/play/RGkzc1ZULTM4NjA5MTQuZmx2', + 'md5': '63a42c705772aa53fd4c1a0027f86adf', + 'info_dict': { + 'id': '3860914', + 'ext': 'mp3', + 'title': '孤單南半球-歐德陽', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 247.246, + 'timestamp': 1314932940, + 'upload_date': '20110902', + 'uploader': '阿能', + 'uploader_id': '15973816', + 'categories': ['個人短片'], + }, + }, { + # Video with only one format + 'url': 'http://vlog.xuite.net/play/TkRZNjhULTM0NDE2MjkuZmx2', + 'md5': 'c45737fc8ac5dc8ac2f92ecbcecf505e', + 'info_dict': { + 'id': '3441629', + 'ext': 'mp4', + 'title': '孫燕姿 - 眼淚成詩', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 217.399, + 'timestamp': 1299383640, + 'upload_date': '20110306', + 'uploader': 'Valen', + 'uploader_id': '10400126', + 'categories': ['影視娛樂'], + }, + }, { + # Video with two formats + 'url': 'http://vlog.xuite.net/play/bWo1N1pLLTIxMzAxMTcwLmZsdg==', + 'md5': '1166e0f461efe55b62e26a2d2a68e6de', + 'info_dict': { + 'id': '21301170', + 'ext': 'mp4', + 'title': '暗殺教室 02', + 'description': '字幕:【極影字幕社】', + 'thumbnail': 're:^https?://.*\.jpg$', + 'duration': 1384.907, + 'timestamp': 1421481240, + 'upload_date': '20150117', + 'uploader': '我只是想認真點', + 'uploader_id': '242127761', + 'categories': ['電玩動漫'], + }, + }, { + 'url': 'http://vlog.xuite.net/play/S1dDUjdyLTMyOTc3NjcuZmx2/%E5%AD%AB%E7%87%95%E5%A7%BF-%E7%9C%BC%E6%B7%9A%E6%88%90%E8%A9%A9', + 'only_matching': True, + }] + + def _extract_flv_config(self, media_id): + base64_media_id = base64.b64encode(media_id.encode('utf-8')).decode('utf-8') + flv_config = self._download_xml( + 'http://vlog.xuite.net/flash/player?media=%s' % base64_media_id, + 'flv config') + prop_dict = {} + for prop in flv_config.findall('./property'): + prop_id = base64.b64decode(prop.attrib['id']).decode('utf-8') + # CDATA may be empty in flv config + if not prop.text: + continue + encoded_content = base64.b64decode(prop.text).decode('utf-8') + prop_dict[prop_id] = compat_urllib_parse_unquote(encoded_content) + return prop_dict + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + error_msg = self._search_regex( + r'<div id="error-message-content">([^<]+)', + webpage, 'error message', default=None) + if error_msg: + raise ExtractorError( + '%s returned error: %s' % (self.IE_NAME, error_msg), + expected=True) + + video_id = self._html_search_regex( + r'data-mediaid="(\d+)"', webpage, 'media id') + flv_config = self._extract_flv_config(video_id) + + FORMATS = { + 'audio': 'mp3', + 'video': 'mp4', + } + + formats = [] + for format_tag in ('src', 'hq_src'): + video_url = flv_config.get(format_tag) + if not video_url: + continue + format_id = self._search_regex( + r'\bq=(.+?)\b', video_url, 'format id', default=format_tag) + formats.append({ + 'url': video_url, + 'ext': FORMATS.get(flv_config['type'], 'mp4'), + 'format_id': format_id, + 'height': int(format_id) if format_id.isnumeric() else None, + }) + self._sort_formats(formats) + + timestamp = flv_config.get('publish_datetime') + if timestamp: + timestamp = parse_iso8601(timestamp + ' +0800', ' ') + + category = flv_config.get('category') + categories = [category] if category else [] + + return { + 'id': video_id, + 'title': flv_config['title'], + 'description': flv_config.get('description'), + 'thumbnail': flv_config.get('thumb'), + 'timestamp': timestamp, + 'uploader': flv_config.get('author_name'), + 'uploader_id': flv_config.get('author_id'), + 'duration': parse_duration(flv_config.get('duration')), + 'categories': categories, + 'formats': formats, + } |