X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9f2b33881274af98a9145c533a1d295fad71521a..e96056e11025aad125c364bd96b2a2df071dd2ae:/youtube_dl/extractor/xuite.py diff --git a/youtube_dl/extractor/xuite.py b/youtube_dl/extractor/xuite.py index 2466410..0276c0d 100644 --- a/youtube_dl/extractor/xuite.py +++ b/youtube_dl/extractor/xuite.py @@ -1,14 +1,13 @@ -# -*- coding: utf-8 -*- +# 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, + float_or_none, + get_element_by_attribute, parse_iso8601, - parse_duration, + remove_end, ) @@ -24,7 +23,8 @@ class XuiteIE(InfoExtractor): 'id': '3860914', 'ext': 'mp3', 'title': '孤單南半球-歐德陽', - 'thumbnail': 're:^https?://.*\.jpg$', + 'description': '孤單南半球-歐德陽', + 'thumbnail': r're:^https?://.*\.jpg$', 'duration': 247.246, 'timestamp': 1314932940, 'upload_date': '20110902', @@ -40,11 +40,11 @@ class XuiteIE(InfoExtractor): 'id': '25925099', 'ext': 'mp4', 'title': 'BigBuckBunny_320x180', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'duration': 596.458, 'timestamp': 1454242500, 'upload_date': '20160131', - 'uploader': 'yan12125', + 'uploader': '屁姥', 'uploader_id': '12158353', 'categories': ['個人短片'], 'description': 'http://download.blender.org/peach/bigbuckbunny_movies/BigBuckBunny_320x180.mp4', @@ -58,7 +58,7 @@ class XuiteIE(InfoExtractor): 'ext': 'mp4', 'title': '暗殺教室 02', 'description': '字幕:【極影字幕社】', - 'thumbnail': 're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.*\.jpg$', 'duration': 1384.907, 'timestamp': 1421481240, 'upload_date': '20150117', @@ -66,35 +66,29 @@ class XuiteIE(InfoExtractor): 'uploader_id': '242127761', 'categories': ['電玩動漫'], }, + '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, }] - @staticmethod - def base64_decode_utf8(data): - return base64.b64decode(data.encode('utf-8')).decode('utf-8') - - @staticmethod - def base64_encode_utf8(data): - return base64.b64encode(data.encode('utf-8')).decode('utf-8') - - def _extract_flv_config(self, media_id): - base64_media_id = self.base64_encode_utf8(media_id) - 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 = self.base64_decode_utf8(prop.attrib['id']) - # CDATA may be empty in flv config - if not prop.text: - continue - encoded_content = self.base64_decode_utf8(prop.text) - prop_dict[prop_id] = compat_urllib_parse_unquote(encoded_content) - return prop_dict - 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) webpage = self._download_webpage(url, video_id) @@ -107,46 +101,53 @@ class XuiteIE(InfoExtractor): '%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) + media_info = self._parse_json(self._search_regex( + r'var\s+mediaInfo\s*=\s*({.*});', webpage, 'media info'), video_id) - FORMATS = { - 'audio': 'mp3', - 'video': 'mp4', - } + video_id = media_info['MEDIA_ID'] formats = [] - for format_tag in ('src', 'hq_src'): - video_url = flv_config.get(format_tag) + 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=format_tag) + r'\bq=(.+?)\b', video_url, 'format id', default=None) formats.append({ 'url': video_url, - 'ext': FORMATS.get(flv_config['type'], 'mp4'), + '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 = flv_config.get('publish_datetime') + timestamp = media_info.get('PUBLISH_DATETIME') if timestamp: timestamp = parse_iso8601(timestamp + ' +0800', ' ') - category = flv_config.get('category') + 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'], - 'description': flv_config.get('description'), - 'thumbnail': flv_config.get('thumb'), + 'title': media_info['TITLE'], + 'description': remove_end(media_info.get('metaDesc'), ' (Xuite 影音)'), + 'thumbnail': media_info.get('ogImageUrl'), 'timestamp': timestamp, - 'uploader': flv_config.get('author_name'), - 'uploader_id': flv_config.get('author_id'), - 'duration': parse_duration(flv_config.get('duration')), + '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, }