commit 24e0cd709fae13d9192361dae27ecd292c6b1fd6
parent 4779420ce8e13d0df33189907e477146a2a0cefe
Author: Sergey M․ <dstftw@gmail.com>
Date:   Mon, 13 Aug 2018 00:15:59 +0700

[raywenderlich] Adapt to site redesign (closes #17225)

Diffstat:
Myoutube_dl/extractor/extractors.py | 5++++-
Myoutube_dl/extractor/raywenderlich.py | 189+++++++++++++++++++++++++++++++++++++++++++++++++++++++------------------------
2 files changed, 137 insertions(+), 57 deletions(-)

diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py @@ -899,7 +899,10 @@ from .rai import ( RaiPlayPlaylistIE, RaiIE, ) -from .raywenderlich import RayWenderlichIE +from .raywenderlich import ( + RayWenderlichIE, + RayWenderlichCourseIE, +) from .rbmaradio import RBMARadioIE from .rds import RDSIE from .redbulltv import RedBullTVIE diff --git a/youtube_dl/extractor/raywenderlich.py b/youtube_dl/extractor/raywenderlich.py @@ -4,24 +4,37 @@ import re from .common import InfoExtractor from .vimeo import VimeoIE +from ..compat import compat_str from ..utils import ( - extract_attributes, ExtractorError, - smuggle_url, - unsmuggle_url, + int_or_none, + merge_dicts, + try_get, + unescapeHTML, + unified_timestamp, urljoin, ) class RayWenderlichIE(InfoExtractor): - _VALID_URL = r'https?://videos\.raywenderlich\.com/courses/(?P<course_id>[^/]+)/lessons/(?P<id>\d+)' + _VALID_URL = r'''(?x) + https?:// + (?: + videos\.raywenderlich\.com/courses| + (?:www\.)?raywenderlich\.com + )/ + (?P<course_id>[^/]+)/lessons/(?P<id>\d+) + ''' _TESTS = [{ - 'url': 'https://videos.raywenderlich.com/courses/105-testing-in-ios/lessons/1', + 'url': 'https://www.raywenderlich.com/3530-testing-in-ios/lessons/1', 'info_dict': { 'id': '248377018', 'ext': 'mp4', - 'title': 'Testing In iOS Episode 1: Introduction', + 'title': 'Introduction', + 'description': 'md5:804d031b3efa9fcb49777d512d74f722', + 'timestamp': 1513906277, + 'upload_date': '20171222', 'duration': 133, 'uploader': 'Ray Wenderlich', 'uploader_id': 'user3304672', @@ -34,69 +47,133 @@ class RayWenderlichIE(InfoExtractor): 'expected_warnings': ['HTTP Error 403: Forbidden'], }, { 'url': 'https://videos.raywenderlich.com/courses/105-testing-in-ios/lessons/1', + 'only_matching': True, + }] + + @staticmethod + def _extract_video_id(data, lesson_id): + if not data: + return + groups = try_get(data, lambda x: x['groups'], list) or [] + if not groups: + return + for group in groups: + if not isinstance(group, dict): + continue + contents = try_get(data, lambda x: x['contents'], list) or [] + for content in contents: + if not isinstance(content, dict): + continue + ordinal = int_or_none(content.get('ordinal')) + if ordinal != lesson_id: + continue + video_id = content.get('identifier') + if video_id: + return compat_str(video_id) + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + course_id, lesson_id = mobj.group('course_id', 'id') + display_id = '%s/%s' % (course_id, lesson_id) + + webpage = self._download_webpage(url, display_id) + + thumbnail = self._og_search_thumbnail( + webpage, default=None) or self._html_search_meta( + 'twitter:image', webpage, 'thumbnail') + + if '>Subscribe to unlock' in webpage: + raise ExtractorError( + 'This content is only available for subscribers', + expected=True) + + info = { + 'thumbnail': thumbnail, + } + + vimeo_id = self._search_regex( + r'data-vimeo-id=["\'](\d+)', webpage, 'vimeo id', default=None) + + if not vimeo_id: + data = self._parse_json( + self._search_regex( + r'data-collection=(["\'])(?P<data>{.+?})\1', webpage, + 'data collection', default='{}', group='data'), + display_id, transform_source=unescapeHTML, fatal=False) + video_id = self._extract_video_id( + data, lesson_id) or self._search_regex( + r'/videos/(\d+)/', thumbnail, 'video id') + headers = { + 'Referer': url, + 'X-Requested-With': 'XMLHttpRequest', + } + csrf_token = self._html_search_meta( + 'csrf-token', webpage, 'csrf token', default=None) + if csrf_token: + headers['X-CSRF-Token'] = csrf_token + video = self._download_json( + 'https://videos.raywenderlich.com/api/v1/videos/%s.json' + % video_id, display_id, headers=headers)['video'] + vimeo_id = video['clips'][0]['provider_id'] + info.update({ + '_type': 'url_transparent', + 'title': video.get('name'), + 'description': video.get('description') or video.get( + 'meta_description'), + 'duration': int_or_none(video.get('duration')), + 'timestamp': unified_timestamp(video.get('created_at')), + }) + + return merge_dicts(info, self.url_result( + VimeoIE._smuggle_referrer( + 'https://player.vimeo.com/video/%s' % vimeo_id, url), + ie=VimeoIE.ie_key(), video_id=vimeo_id)) + + +class RayWenderlichCourseIE(InfoExtractor): + _VALID_URL = r'''(?x) + https?:// + (?: + videos\.raywenderlich\.com/courses| + (?:www\.)?raywenderlich\.com + )/ + (?P<id>[^/]+) + ''' + + _TEST = { + 'url': 'https://www.raywenderlich.com/3530-testing-in-ios', 'info_dict': { 'title': 'Testing in iOS', - 'id': '105-testing-in-ios', + 'id': '3530-testing-in-ios', }, 'params': { 'noplaylist': False, }, 'playlist_count': 29, - }] + } + + @classmethod + def suitable(cls, url): + return False if RayWenderlichIE.suitable(url) else super( + RayWenderlichCourseIE, cls).suitable(url) def _real_extract(self, url): - url, smuggled_data = unsmuggle_url(url, {}) + course_id = self._match_id(url) - mobj = re.match(self._VALID_URL, url) - course_id, lesson_id = mobj.group('course_id', 'id') - video_id = '%s/%s' % (course_id, lesson_id) - - webpage = self._download_webpage(url, video_id) - - no_playlist = self._downloader.params.get('noplaylist') - if no_playlist or smuggled_data.get('force_video', False): - if no_playlist: - self.to_screen( - 'Downloading just video %s because of --no-playlist' - % video_id) - if '>Subscribe to unlock' in webpage: - raise ExtractorError( - 'This content is only available for subscribers', - expected=True) - vimeo_id = self._search_regex( - r'data-vimeo-id=["\'](\d+)', webpage, 'video id') - return self.url_result( - VimeoIE._smuggle_referrer( - 'https://player.vimeo.com/video/%s' % vimeo_id, url), - ie=VimeoIE.ie_key(), video_id=vimeo_id) - - self.to_screen( - 'Downloading playlist %s - add --no-playlist to just download video' - % course_id) - - lesson_ids = set((lesson_id, )) - for lesson in re.findall( - r'(<a[^>]+\bclass=["\']lesson-link[^>]+>)', webpage): - attrs = extract_attributes(lesson) - if not attrs: - continue - lesson_url = attrs.get('href') - if not lesson_url: - continue - lesson_id = self._search_regex( - r'/lessons/(\d+)', lesson_url, 'lesson id', default=None) - if not lesson_id: - continue - lesson_ids.add(lesson_id) + webpage = self._download_webpage(url, course_id) entries = [] - for lesson_id in sorted(lesson_ids): + lesson_urls = set() + for lesson_url in re.findall( + r'<a[^>]+\bhref=["\'](/%s/lessons/\d+)' % course_id, webpage): + if lesson_url in lesson_urls: + continue + lesson_urls.add(lesson_url) entries.append(self.url_result( - smuggle_url(urljoin(url, lesson_id), {'force_video': True}), - ie=RayWenderlichIE.ie_key())) + urljoin(url, lesson_url), ie=RayWenderlichIE.ie_key())) - title = self._search_regex( - r'class=["\']course-title[^>]+>([^<]+)', webpage, 'course title', - default=None) + title = self._og_search_title( + webpage, default=None) or self._html_search_meta( + 'twitter:title', webpage, 'title', default=None) return self.playlist_result(entries, course_id, title)