X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fdramafever.py;h=1edd8e7bda41467c76d6b78c42199ff8e2a7b85d;hb=ec85ded83cbfa652ba94cb080aab52d8b270212a;hp=40787ffcd2118d26d7bce98076c6d3f686653001;hpb=ad49fe7c8f22c78a001eb1a53d7df9ef54424b89;p=youtube-dl diff --git a/youtube_dl/extractor/dramafever.py b/youtube_dl/extractor/dramafever.py index 40787ffcd..1edd8e7bd 100644 --- a/youtube_dl/extractor/dramafever.py +++ b/youtube_dl/extractor/dramafever.py @@ -1,104 +1,160 @@ -# encoding: utf-8 +# coding: utf-8 from __future__ import unicode_literals -import re +import itertools + +from .amp import AMPIE +from ..compat import ( + compat_HTTPError, + compat_urlparse, +) +from ..utils import ( + ExtractorError, + clean_html, + int_or_none, + sanitized_Request, + urlencode_postdata +) + + +class DramaFeverBaseIE(AMPIE): + _LOGIN_URL = 'https://www.dramafever.com/accounts/login/' + _NETRC_MACHINE = 'dramafever' + + _CONSUMER_SECRET = 'DA59dtVXYLxajktV' + + _consumer_secret = None + + def _get_consumer_secret(self): + mainjs = self._download_webpage( + 'http://www.dramafever.com/static/51afe95/df2014/scripts/main.js', + None, 'Downloading main.js', fatal=False) + if not mainjs: + return self._CONSUMER_SECRET + return self._search_regex( + r"var\s+cs\s*=\s*'([^']+)'", mainjs, + 'consumer secret', default=self._CONSUMER_SECRET) + + def _real_initialize(self): + self._login() + self._consumer_secret = self._get_consumer_secret() + + def _login(self): + (username, password) = self._get_login_info() + if username is None: + return + + login_form = { + 'username': username, + 'password': password, + } + + request = sanitized_Request( + self._LOGIN_URL, urlencode_postdata(login_form)) + response = self._download_webpage( + request, None, 'Logging in as %s' % username) -from .common import InfoExtractor + if all(logout_pattern not in response + for logout_pattern in ['href="/accounts/logout/"', '>Log out<']): + error = self._html_search_regex( + r'(?s)class="hidden-xs prompt"[^>]*>(.+?)<', + response, 'error message', default=None) + if error: + raise ExtractorError('Unable to login: %s' % error, expected=True) + raise ExtractorError('Unable to log in') -class DramaFeverIE(InfoExtractor): +class DramaFeverIE(DramaFeverBaseIE): IE_NAME = 'dramafever' - _VALID_URL = r'^https?://(?:www\.)?dramafever\.com/drama/(?P[0-9]+/[0-9]+)/' + _VALID_URL = r'https?://(?:www\.)?dramafever\.com/drama/(?P[0-9]+/[0-9]+)(?:/|$)' _TESTS = [{ 'url': 'http://www.dramafever.com/drama/4512/1/Cooking_with_Shin/', 'info_dict': { 'id': '4512.1', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Cooking with Shin 4512.1', + 'description': 'md5:a8eec7942e1664a6896fcd5e1287bfd0', + 'episode': 'Episode 1', + 'episode_number': 1, + 'thumbnail': r're:^https?://.*\.jpg', + 'timestamp': 1404336058, 'upload_date': '20140702', - 'description': 'Served at all special occasions and featured in the hit drama Heirs, Shin cooks Red Bean Rice.', - } + 'duration': 343, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + }, { + 'url': 'http://www.dramafever.com/drama/4826/4/Mnet_Asian_Music_Awards_2015/?ap=1', + 'info_dict': { + 'id': '4826.4', + 'ext': 'mp4', + 'title': 'Mnet Asian Music Awards 2015 4826.4', + 'description': 'md5:3ff2ee8fedaef86e076791c909cf2e91', + 'episode': 'Mnet Asian Music Awards 2015 - Part 3', + 'episode_number': 4, + 'thumbnail': r're:^https?://.*\.jpg', + 'timestamp': 1450213200, + 'upload_date': '20151215', + 'duration': 5602, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, }] def _real_extract(self, url): - video_id = self._match_id(url).replace("/", ".") - - consumer_secret = self._get_consumer_secret(video_id) - - ep_json = self._download_json( - "http://www.dramafever.com/amp/episode/feed.json?guid=%s" % video_id, - video_id, note='Downloading episode metadata', - errnote="Video may not be available for your location")["channel"]["item"] - - title = ep_json["media-group"]["media-title"] - description = ep_json["media-group"]["media-description"] - thumbnail = ep_json["media-group"]["media-thumbnail"]["@attributes"]["url"] - duration = int(ep_json["media-group"]["media-content"][0]["@attributes"]["duration"]) - mobj = re.match(r"([0-9]{4})-([0-9]{2})-([0-9]{2})", ep_json["pubDate"]) - upload_date = mobj.group(1) + mobj.group(2) + mobj.group(3) if mobj is not None else None - - formats = [] - for vid_format in ep_json["media-group"]["media-content"]: - src = vid_format["@attributes"]["url"] - if '.f4m' in src: - formats.extend(self._extract_f4m_formats(src, video_id)) - - self._sort_formats(formats) - video_subtitles = self.extract_subtitles(video_id, consumer_secret) - - return { - 'id': video_id, - 'title': title, - 'description': description, - 'thumbnail': thumbnail, - 'upload_date': upload_date, - 'duration': duration, - 'formats': formats, - 'subtitles': video_subtitles, - } - - def _get_consumer_secret(self, video_id): - df_js = self._download_webpage( - "http://www.dramafever.com/static/126960d/v2/js/plugins/jquery.threadedcomments.js", video_id) - return self._search_regex(r"'cs': '([0-9a-zA-Z]+)'", df_js, "cs") - - def _get_episodes(self, series_id, consumer_secret, episode_filter=None): - _PAGE_SIZE = 60 - - curr_page = 1 - max_pages = curr_page + 1 - results = [] - while max_pages >= curr_page: - page_url = "http://www.dramafever.com/api/4/episode/series/?cs=%s&series_id=%s&page_size=%d&page_number=%d" % \ - (consumer_secret, series_id, _PAGE_SIZE, curr_page) - series = self._download_json( - page_url, series_id, note="Downloading series json page #%d" % curr_page) - max_pages = series['num_pages'] - results.extend([ep for ep in series['value'] if episode_filter is None or episode_filter(ep)]) - curr_page += 1 - return results - - def _get_subtitles(self, video_id, consumer_secret): - - res = None - info = self._get_episodes( - video_id.split(".")[0], consumer_secret, - episode_filter=lambda x: x['guid'] == video_id) - - if len(info) == 1 and info[0]['subfile'] != '': - res = {'en': [{'url': info[0]['subfile'], 'ext': 'srt'}]} - return res - - -class DramaFeverSeriesIE(DramaFeverIE): + video_id = self._match_id(url).replace('/', '.') + + try: + info = self._extract_feed_info( + 'http://www.dramafever.com/amp/episode/feed.json?guid=%s' % video_id) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError): + raise ExtractorError( + 'Currently unavailable in your country.', expected=True) + raise + + series_id, episode_number = video_id.split('.') + episode_info = self._download_json( + # We only need a single episode info, so restricting page size to one episode + # and dealing with page number as with episode number + r'http://www.dramafever.com/api/4/episode/series/?cs=%s&series_id=%s&page_number=%s&page_size=1' + % (self._consumer_secret, series_id, episode_number), + video_id, 'Downloading episode info JSON', fatal=False) + if episode_info: + value = episode_info.get('value') + if isinstance(value, list): + for v in value: + if v.get('type') == 'Episode': + subfile = v.get('subfile') or v.get('new_subfile') + if subfile and subfile != 'http://www.dramafever.com/st/': + info.setdefault('subtitles', {}).setdefault('English', []).append({ + 'ext': 'srt', + 'url': subfile, + }) + episode_number = int_or_none(v.get('number')) + episode_fallback = 'Episode' + if episode_number: + episode_fallback += ' %d' % episode_number + info['episode'] = v.get('title') or episode_fallback + info['episode_number'] = episode_number + break + + return info + + +class DramaFeverSeriesIE(DramaFeverBaseIE): IE_NAME = 'dramafever:series' - _VALID_URL = r'^https?://(?:www\.)?dramafever\.com/drama/(?P[0-9]+)/\d*[a-zA-Z_][a-zA-Z0-9_]*/' + _VALID_URL = r'https?://(?:www\.)?dramafever\.com/drama/(?P[0-9]+)(?:/(?:(?!\d+(?:/|$)).+)?)?$' _TESTS = [{ 'url': 'http://www.dramafever.com/drama/4512/Cooking_with_Shin/', 'info_dict': { 'id': '4512', 'title': 'Cooking with Shin', - 'description': 'Professional chef and cooking instructor Shin Kim takes some of the delicious dishes featured in your favorite dramas and shows you how to make them right at home.', + 'description': 'md5:84a3f26e3cdc3fb7f500211b3593b5c1', }, 'playlist_count': 4, }, { @@ -106,25 +162,38 @@ class DramaFeverSeriesIE(DramaFeverIE): 'info_dict': { 'id': '124', 'title': 'IRIS', - 'description': 'Lee Byung Hun and Kim Tae Hee star in this powerhouse drama and ratings megahit of action, intrigue and romance.', + 'description': 'md5:b3a30e587cf20c59bd1c01ec0ee1b862', }, 'playlist_count': 20, }] + _PAGE_SIZE = 60 # max is 60 (see http://api.drama9.com/#get--api-4-episode-series-) + def _real_extract(self, url): series_id = self._match_id(url) - consumer_secret = self._get_consumer_secret(series_id) - series_json = self._download_json( - "http://www.dramafever.com/api/4/series/query/?cs=%s&series_id=%s" % (consumer_secret, series_id), - series_id, note='Downloading series metadata')["series"][series_id] + series = self._download_json( + 'http://www.dramafever.com/api/4/series/query/?cs=%s&series_id=%s' + % (self._consumer_secret, series_id), + series_id, 'Downloading series JSON')['series'][series_id] - title = series_json["name"] - description = series_json["description_short"] + title = clean_html(series['name']) + description = clean_html(series.get('description') or series.get('description_short')) - episodes = self._get_episodes(series_id, consumer_secret) entries = [] - for ep in episodes: - entries.append(self.url_result( - 'http://www.dramafever.com%s' % ep['episode_url'], 'DramaFever', ep['guid'])) + for page_num in itertools.count(1): + episodes = self._download_json( + 'http://www.dramafever.com/api/4/episode/series/?cs=%s&series_id=%s&page_size=%d&page_number=%d' + % (self._consumer_secret, series_id, self._PAGE_SIZE, page_num), + series_id, 'Downloading episodes JSON page #%d' % page_num) + for episode in episodes.get('value', []): + episode_url = episode.get('episode_url') + if not episode_url: + continue + entries.append(self.url_result( + compat_urlparse.urljoin(url, episode_url), + 'DramaFever', episode.get('guid'))) + if page_num == episodes['num_pages']: + break + return self.playlist_result(entries, series_id, title, description)