X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fxuite.py;h=0276c0dbb26f584c0fc49cf35614f399cc203508;hb=HEAD;hp=8466d4bc58d557acf9f14513657cfdd36006e3d6;hpb=c84130e86521a7474dd2cf7f7c67e4c8b3dd40bd;p=youtube-dl diff --git a/youtube_dl/extractor/xuite.py b/youtube_dl/extractor/xuite.py index 8466d4bc5..0276c0dbb 100644 --- a/youtube_dl/extractor/xuite.py +++ b/youtube_dl/extractor/xuite.py @@ -1,159 +1,153 @@ -# -*- coding: utf-8 -*- +# coding: utf-8 from __future__ import unicode_literals -import base64 from .common import InfoExtractor -from ..compat import ( - compat_urlparse, - compat_urllib_parse_unquote, - compat_parse_qs -) from ..utils import ( ExtractorError, + float_or_none, + get_element_by_attribute, parse_iso8601, - parse_duration + remove_end, ) -# ref: http://stackoverflow.com/questions/475074/regex-to-parse-or-validate-base64-data -REGEX_BASE64 = r'(?:[A-Za-z0-9+/]{4})*(?:[A-Za-z0-9+/]{2}==|[A-Za-z0-9+/]{3}=)?' - -CST_ZONE = +8 # China Standard Time - class XuiteIE(InfoExtractor): - _VALID_URL = r'http://vlog.xuite.net/play/(?P%s)(/.*)?' % REGEX_BASE64 + IE_DESC = '隨意窩Xuite影音' + _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%s)' % _REGEX_BASE64 _TESTS = [{ # Audio 'url': 'http://vlog.xuite.net/play/RGkzc1ZULTM4NjA5MTQuZmx2', - 'md5': '63a42c705772aa53fd4c1a0027f86adf', + 'md5': 'e79284c87b371424885448d11f6398c8', 'info_dict': { - 'id': 'RGkzc1ZULTM4NjA5MTQuZmx2', + 'id': '3860914', 'ext': 'mp3', + 'title': '孤單南半球-歐德陽', + 'description': '孤單南半球-歐德陽', + 'thumbnail': r're:^https?://.*\.jpg$', + 'duration': 247.246, + 'timestamp': 1314932940, 'upload_date': '20110902', - 'uploader_id': '15973816', 'uploader': '阿能', - 'timestamp': 1314932940, - 'title': '孤單南半球-歐德陽', - 'thumbnail': 're:^https?://.*\.jpg$', - 'categories': ['個人短片'], - 'duration': 247.246 - } - }, { - # Audio with alternative form of url - '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', - 'md5': 'c91645f324de53d82ebb80930e1a73d2', - 'info_dict': { - 'id': 'S1dDUjdyLTMyOTc3NjcuZmx2', - 'ext': 'mp3', - 'upload_date': '20101226', - 'uploader_id': '10102699', - 'uploader': '蠍', - 'timestamp': 1293367080, - 'title': '孫燕姿-眼淚成詩', - 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader_id': '15973816', 'categories': ['個人短片'], - 'duration': 223.19 - } + }, }, { # Video with only one format - 'url': 'http://vlog.xuite.net/play/TkRZNjhULTM0NDE2MjkuZmx2', - 'md5': 'c45737fc8ac5dc8ac2f92ecbcecf505e', + 'url': 'http://vlog.xuite.net/play/WUxxR2xCLTI1OTI1MDk5LmZsdg==', + 'md5': '21f7b39c009b5a4615b4463df6eb7a46', 'info_dict': { - 'id': 'TkRZNjhULTM0NDE2MjkuZmx2', + 'id': '25925099', 'ext': 'mp4', - 'upload_date': '20110306', - 'uploader_id': '10400126', - 'uploader': 'Valen', - 'timestamp': 1299383640, - 'title': '孫燕姿 - 眼淚成詩', - 'thumbnail': 're:^https?://.*\.jpg$', - 'categories': ['影視娛樂'], - 'duration': 217.399 - } + 'title': 'BigBuckBunny_320x180', + 'thumbnail': r're:^https?://.*\.jpg$', + 'duration': 596.458, + 'timestamp': 1454242500, + 'upload_date': '20160131', + 'uploader': '屁姥', + 'uploader_id': '12158353', + 'categories': ['個人短片'], + 'description': 'http://download.blender.org/peach/bigbuckbunny_movies/BigBuckBunny_320x180.mp4', + }, }, { # Video with two formats 'url': 'http://vlog.xuite.net/play/bWo1N1pLLTIxMzAxMTcwLmZsdg==', 'md5': '1166e0f461efe55b62e26a2d2a68e6de', 'info_dict': { - 'id': 'bWo1N1pLLTIxMzAxMTcwLmZsdg==', + 'id': '21301170', 'ext': 'mp4', + 'title': '暗殺教室 02', + 'description': '字幕:【極影字幕社】', + 'thumbnail': r're:^https?://.*\.jpg$', + 'duration': 1384.907, + 'timestamp': 1421481240, 'upload_date': '20150117', - 'uploader_id': '242127761', 'uploader': '我只是想認真點', - 'timestamp': 1421481240, - 'title': '暗殺教室 02', - 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader_id': '242127761', 'categories': ['電玩動漫'], - 'duration': 1384.907 - } + }, + 'skip': 'Video removed', + }, { + # Video with encoded media id + # from http://forgetfulbc.blogspot.com/2016/06/date.html + 'url': 'http://vlog.xuite.net/embed/cE1xbENoLTI3NDQ3MzM2LmZsdg==?ar=0&as=0', + 'info_dict': { + 'id': '27447336', + 'ext': 'mp4', + 'title': '男女平權只是口號？專家解釋約會時男生是否該幫女生付錢 (中字)', + 'description': 'md5:1223810fa123b179083a3aed53574706', + 'timestamp': 1466160960, + 'upload_date': '20160617', + 'uploader': 'B.C. & Lowy', + 'uploader_id': '232279340', + }, + }, { + '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 _flv_config(self, media_id): - base64_media_id = base64.b64encode(media_id.encode('utf-8')).decode('utf-8') - flv_config_url = 'http://vlog.xuite.net/flash/player?media=' + base64_media_id - flv_config = self._download_xml(flv_config_url, 'flv config') - - prop_dict = {} - for prop in flv_config.findall('./property'): - prop_id = base64.b64decode(prop.attrib['id']).decode('utf-8') - - if not prop.text: - continue # CDATA may be empty in flv config - - encoded_content = base64.b64decode(prop.text).decode('utf-8') - prop_dict[prop_id] = compat_urllib_parse_unquote(encoded_content) - - return prop_dict - - def _type_string(self, media_url): - query_string = compat_urlparse.urlparse(media_url).query - type_string = compat_parse_qs(query_string)['q'][0] - return type_string - - def _guess_ext(self, media_url): - type_string = self._type_string(media_url) - if type_string == 'mp3': - return 'mp3' - elif type_string == '360' or type_string == '720': - return 'mp4' - else: - raise ExtractorError('Unknown type string %s' % type_string) - def _real_extract(self, url): + # /play/ URLs provide embedded video URL and more metadata + url = url.replace('/embed/', '/play/') video_id = self._match_id(url) - page = self._download_webpage(url, video_id) - media_id = self._html_search_regex(r'data-mediaid="(\d+)"', page, 'media id') - flv_config = self._flv_config(media_id) - - timestamp_local = parse_iso8601(flv_config['publish_datetime'], ' ') - timestamp_gmt = timestamp_local - CST_ZONE * 3600 - - ret_attrs = { + webpage = self._download_webpage(url, video_id) + + error_msg = self._search_regex( + r'
([^<]+)', + webpage, 'error message', default=None) + if error_msg: + raise ExtractorError( + '%s returned error: %s' % (self.IE_NAME, error_msg), + expected=True) + + media_info = self._parse_json(self._search_regex( + r'var\s+mediaInfo\s*=\s*({.*});', webpage, 'media info'), video_id) + + video_id = media_info['MEDIA_ID'] + + formats = [] + for key in ('html5Url', 'html5HQUrl'): + video_url = media_info.get(key) + if not video_url: + continue + format_id = self._search_regex( + r'\bq=(.+?)\b', video_url, 'format id', default=None) + formats.append({ + 'url': video_url, + 'ext': 'mp4' if format_id.isnumeric() else format_id, + 'format_id': format_id, + 'height': int(format_id) if format_id.isnumeric() else None, + }) + self._sort_formats(formats) + + timestamp = media_info.get('PUBLISH_DATETIME') + if timestamp: + timestamp = parse_iso8601(timestamp + ' +0800', ' ') + + category = media_info.get('catName') + categories = [category] if category else [] + + uploader = media_info.get('NICKNAME') + uploader_url = None + + author_div = get_element_by_attribute('itemprop', 'author', webpage) + if author_div: + uploader = uploader or self._html_search_meta('name', author_div) + uploader_url = self._html_search_regex( + r']+itemprop="url"[^>]+href="([^"]+)"', author_div, + 'uploader URL', fatal=False) + + return { 'id': video_id, - 'title': flv_config['title'], - 'thumbnail': flv_config['thumb'], - 'uploader': flv_config['author_name'], - 'timestamp': timestamp_gmt, - 'uploader_id': flv_config['author_id'], - 'duration': parse_duration(flv_config['duration']), - 'categories': [flv_config['category']] + 'title': media_info['TITLE'], + 'description': remove_end(media_info.get('metaDesc'), ' (Xuite 影音)'), + 'thumbnail': media_info.get('ogImageUrl'), + 'timestamp': timestamp, + 'uploader': uploader, + 'uploader_id': media_info.get('MEMBER_ID'), + 'uploader_url': uploader_url, + 'duration': float_or_none(media_info.get('MEDIA_DURATION'), 1000000), + 'categories': categories, + 'formats': formats, } - - if 'hq_src' in flv_config: - urls = [flv_config['src'], flv_config['hq_src']] - - ret_attrs['formats'] = [] - - for url in urls: - ret_attrs['formats'].append({ - 'url': url, - 'ext': self._guess_ext(url), - 'format_id': self._type_string(url), - 'height': int(self._type_string(url)) - }) - else: - ret_attrs['url'] = flv_config['src'] - ret_attrs['ext'] = self._guess_ext(flv_config['src']) - - return ret_attrs