From: Sergey M․ Date: Fri, 19 Dec 2014 15:56:44 +0000 (+0600) Subject: [xtube] Fix and modernize (Closes #4489) X-Git-Url: http://git.oshgnacknak.de/?a=commitdiff_plain;h=16ea8179681263d570e632d3ce523ed23ddbf204;p=youtube-dl [xtube] Fix and modernize (Closes #4489) --- diff --git a/youtube_dl/extractor/xtube.py b/youtube_dl/extractor/xtube.py index f9d98b83f..cf74d4fd5 100644 --- a/youtube_dl/extractor/xtube.py +++ b/youtube_dl/extractor/xtube.py @@ -1,11 +1,11 @@ from __future__ import unicode_literals import re -import json from .common import InfoExtractor from ..compat import ( compat_urllib_request, + compat_urllib_parse, ) from ..utils import ( parse_duration, @@ -14,7 +14,7 @@ from ..utils import ( class XTubeIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?(?Pxtube\.com/watch\.php\?v=(?P[^/?&]+))' + _VALID_URL = r'https?://(?:www\.)?(?Pxtube\.com/watch\.php\?v=(?P[^/?&#]+))' _TEST = { 'url': 'http://www.xtube.com/watch.php?v=kVTUy_G222_', 'md5': '092fbdd3cbe292c920ef6fc6a8a9cdab', @@ -30,41 +30,49 @@ class XTubeIE(InfoExtractor): } def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - url = 'http://www.' + mobj.group('url') + video_id = self._match_id(url) req = compat_urllib_request.Request(url) req.add_header('Cookie', 'age_verified=1') webpage = self._download_webpage(req, video_id) - video_title = self._html_search_regex(r'

([^<]+)', webpage, 'title') + video_title = self._html_search_regex( + r'

([^<]+)', webpage, 'title') video_uploader = self._html_search_regex( - r'so_s\.addVariable\("owner_u", "([^"]+)', webpage, 'uploader', fatal=False) + [r"var\s+contentOwnerId\s*=\s*'([^']+)", + r'By:\s*([^<]+)', webpage, 'description', fatal=False) + r'

([^<]+)', + webpage, 'description', fatal=False) duration = parse_duration(self._html_search_regex( - r'Runtime: ([^<]+)

', webpage, 'duration', fatal=False)) - view_count = self._html_search_regex( - r'Views: ([\d,\.]+)

', webpage, 'view count', fatal=False) - if view_count: - view_count = str_to_int(view_count) - comment_count = self._html_search_regex( - r'
([\d,\.]+) Comments
', webpage, 'comment count', fatal=False) - if comment_count: - comment_count = str_to_int(comment_count) - - player_quality_option = json.loads(self._html_search_regex( - r'playerQualityOption = ({.+?});', webpage, 'player quality option')) - - QUALITIES = ['3gp', 'mp4_normal', 'mp4_high', 'flv', 'mp4_ultra', 'mp4_720', 'mp4_1080'] - formats = [ - { - 'url': furl, + r'Runtime: ([^<]+)

', + webpage, 'duration', fatal=False)) + view_count = str_to_int(self._html_search_regex( + r'Views: ([\d,\.]+)

', + webpage, 'view count', fatal=False)) + comment_count = str_to_int(self._html_search_regex( + r'
([\d,\.]+) Comments
', + webpage, 'comment count', fatal=False)) + + formats = [] + for format_id, video_url in re.findall( + r'flashvars\.quality_(.+?)\s*=\s*"([^"]+)"', webpage): + fmt = { + 'url': compat_urllib_parse.unquote(video_url), 'format_id': format_id, - 'preference': QUALITIES.index(format_id) if format_id in QUALITIES else -1, - } for format_id, furl in player_quality_option.items() - ] + } + m = re.search(r'^(?P\d+)[pP]', format_id) + if m: + fmt['height'] = int(m.group('height')) + formats.append(fmt) + + if not formats: + video_url = compat_urllib_parse.unquote(self._search_regex( + r'flashvars\.video_url\s*=\s*"([^"]+)"', + webpage, 'video URL')) + formats.append({'url': video_url}) + self._sort_formats(formats) return {