From 577281b0c6b2fc774e03abf3465d47d909ba31e5 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Kacper=20Michaj=C5=82ow?= Date: Sun, 16 Oct 2016 03:04:17 +0200 Subject: [PATCH] [cda] Fix and improve extraction Fixes #10929 --- youtube_dl/extractor/cda.py | 65 ++++++++++++++++++++++++++++--------- 1 file changed, 49 insertions(+), 16 deletions(-) diff --git a/youtube_dl/extractor/cda.py b/youtube_dl/extractor/cda.py index 8af318703..e00bdaf66 100755 --- a/youtube_dl/extractor/cda.py +++ b/youtube_dl/extractor/cda.py @@ -5,14 +5,16 @@ from .common import InfoExtractor from ..utils import ( - decode_packed_codes, ExtractorError, - parse_duration + float_or_none, + int_or_none, + parse_duration, ) class CDAIE(InfoExtractor): _VALID_URL = r'https?://(?:(?:www\.)?cda\.pl/video|ebd\.cda\.pl/[0-9]+x[0-9]+)/(?P[0-9a-z]+)' + _BASE_URL = 'http://www.cda.pl/' _TESTS = [{ 'url': 'http://www.cda.pl/video/5749950c', 'md5': '6f844bf51b15f31fae165365707ae970', @@ -21,6 +23,9 @@ class CDAIE(InfoExtractor): 'ext': 'mp4', 'height': 720, 'title': 'Oto dlaczego przed zakrętem należy zwolnić.', + 'description': 'md5:269ccd135d550da90d1662651fcb9772', + 'thumbnail': 're:^https?://.*\.jpg$', + 'average_rating': float, 'duration': 39 } }, { @@ -30,6 +35,11 @@ class CDAIE(InfoExtractor): 'id': '57413289', 'ext': 'mp4', 'title': 'Lądowanie na lotnisku na Maderze', + 'description': 'md5:60d76b71186dcce4e0ba6d4bbdb13e1a', + 'thumbnail': 're:^https?://.*\.jpg$', + 'uploader': 'crash404', + 'view_count': int, + 'average_rating': float, 'duration': 137 } }, { @@ -39,31 +49,55 @@ class CDAIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - webpage = self._download_webpage('http://ebd.cda.pl/0x0/' + video_id, video_id) + self._set_cookie('cda.pl', 'cda.player', 'html5') + webpage = self._download_webpage( + self._BASE_URL + '/video/' + video_id, video_id) if 'Ten film jest dostępny dla użytkowników premium' in webpage: raise ExtractorError('This video is only available for premium users.', expected=True) - title = self._html_search_regex(r'(.+?)', webpage, 'title') - formats = [] + uploader = self._search_regex(r'''(?x) + <(span|meta)[^>]+itemprop=(["\'])author\2[^>]*> + (?:<\1[^>]*>[^<]*|(?!)(?:.|\n))*? + <(span|meta)[^>]+itemprop=(["\'])name\4[^>]*>(?P[^<]+) + ''', webpage, 'uploader', default=None, group='uploader') + view_count = self._search_regex( + r'Odsłony:(?:\s| )*([0-9]+)', webpage, + 'view_count', default=None) + average_rating = self._search_regex( + r'<(?:span|meta)[^>]+itemprop=(["\'])ratingValue\1[^>]*>(?P[0-9.]+)', + webpage, 'rating', fatal=False, group='rating_value') + info_dict = { 'id': video_id, - 'title': title, + 'title': self._og_search_title(webpage), + 'description': self._og_search_description(webpage), + 'uploader': uploader, + 'view_count': int_or_none(view_count), + 'average_rating': float_or_none(average_rating), + 'thumbnail': self._og_search_thumbnail(webpage), 'formats': formats, 'duration': None, } def extract_format(page, version): - unpacked = decode_packed_codes(page) - format_url = self._search_regex( - r"(?:file|url)\s*:\s*(\\?[\"'])(?Phttp.+?)\1", unpacked, - '%s url' % version, fatal=False, group='url') - if not format_url: + json_str = self._search_regex( + r'player_data=(\\?["\'])(?P.+?)\1', page, + '%s player_json' % version, fatal=False, group='player_data') + if not json_str: + return + player_data = self._parse_json( + json_str, '%s player_data' % version, fatal=False) + if not player_data: + return + video = player_data.get('video') + if not video or 'file' not in video: + self.report_warning('Unable to extract %s version information' % version) return f = { - 'url': format_url, + 'url': video['file'], } m = re.search( r']+data-quality="(?P[^"]+)"[^>]+href="[^"]+"[^>]+class="[^"]*quality-btn-active[^"]*">(?P[0-9]+)p', @@ -75,9 +109,7 @@ def extract_format(page, version): }) info_dict['formats'].append(f) if not info_dict['duration']: - info_dict['duration'] = parse_duration(self._search_regex( - r"duration\s*:\s*(\\?[\"'])(?P.+?)\1", - unpacked, 'duration', fatal=False, group='duration')) + info_dict['duration'] = parse_duration(video.get('duration')) extract_format(webpage, 'default') @@ -85,7 +117,8 @@ def extract_format(page, version): r']+data-quality="[^"]+"[^>]+href="([^"]+)"[^>]+class="quality-btn"[^>]*>([0-9]+p)', webpage): webpage = self._download_webpage( - href, video_id, 'Downloading %s version information' % resolution, fatal=False) + self._BASE_URL + href, video_id, + 'Downloading %s version information' % resolution, fatal=False) if not webpage: # Manually report warning because empty page is returned when # invalid version is requested.