]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/dramafever.py
Start new release.
[youtubedl] / youtube_dl / extractor / dramafever.py
index d836c1a6c27ccd757e4e1f6574e8851c90775193..db1de699fafc5885da2c48f72f5e102825449703 100644 (file)
@@ -1,26 +1,26 @@
-# encoding: utf-8
+# coding: utf-8
 from __future__ import unicode_literals
 
 import itertools
+import json
 
 from .common import InfoExtractor
 from ..compat import (
     compat_HTTPError,
-    compat_urllib_parse,
     compat_urlparse,
 )
 from ..utils import (
-    ExtractorError,
     clean_html,
-    determine_ext,
+    ExtractorError,
     int_or_none,
-    parse_iso8601,
-    sanitized_Request,
+    parse_age_limit,
+    parse_duration,
+    unified_timestamp,
+    url_or_none,
 )
 
 
 class DramaFeverBaseIE(InfoExtractor):
-    _LOGIN_URL = 'https://www.dramafever.com/accounts/login/'
     _NETRC_MACHINE = 'dramafever'
 
     _CONSUMER_SECRET = 'DA59dtVXYLxajktV'
@@ -38,11 +38,11 @@ class DramaFeverBaseIE(InfoExtractor):
             'consumer secret', default=self._CONSUMER_SECRET)
 
     def _real_initialize(self):
-        self._login()
         self._consumer_secret = self._get_consumer_secret()
+        self._login()
 
     def _login(self):
-        (username, password) = self._get_login_info()
+        username, password = self._get_login_info()
         if username is None:
             return
 
@@ -51,113 +51,163 @@ class DramaFeverBaseIE(InfoExtractor):
             'password': password,
         }
 
-        request = sanitized_Request(
-            self._LOGIN_URL, compat_urllib_parse.urlencode(login_form).encode('utf-8'))
-        response = self._download_webpage(
-            request, None, 'Logging in as %s' % username)
+        try:
+            response = self._download_json(
+                'https://www.dramafever.com/api/users/login', None, 'Logging in',
+                data=json.dumps(login_form).encode('utf-8'), headers={
+                    'x-consumer-key': self._consumer_secret,
+                })
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code in (403, 404):
+                response = self._parse_json(
+                    e.cause.read().decode('utf-8'), None)
+            else:
+                raise
 
-        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')
+        # Successful login
+        if response.get('result') or response.get('guid') or response.get('user_guid'):
+            return
+
+        errors = response.get('errors')
+        if errors and isinstance(errors, list):
+            error = errors[0]
+            message = error.get('message') or error['reason']
+            raise ExtractorError('Unable to login: %s' % message, expected=True)
+        raise ExtractorError('Unable to log in')
 
 
 class DramaFeverIE(DramaFeverBaseIE):
     IE_NAME = 'dramafever'
-    _VALID_URL = r'https?://(?:www\.)?dramafever\.com/drama/(?P<id>[0-9]+/[0-9]+)(?:/|$)'
-    _TEST = {
-        'url': 'http://www.dramafever.com/drama/4512/1/Cooking_with_Shin/',
+    _VALID_URL = r'https?://(?:www\.)?dramafever\.com/(?:[^/]+/)?drama/(?P<id>[0-9]+/[0-9]+)(?:/|$)'
+    _TESTS = [{
+        'url': 'https://www.dramafever.com/drama/4274/1/Heirs/',
         'info_dict': {
-            'id': '4512.1',
+            'id': '4274.1',
+            'ext': 'wvm',
+            'title': 'Heirs - Episode 1',
+            'description': 'md5:362a24ba18209f6276e032a651c50bc2',
+            'thumbnail': r're:^https?://.*\.jpg',
+            'duration': 3783,
+            'timestamp': 1381354993,
+            'upload_date': '20131009',
+            'series': 'Heirs',
+            'season_number': 1,
+            'episode': 'Episode 1',
+            'episode_number': 1,
+        },
+        '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': 'flv',
-            'title': 'Cooking with Shin 4512.1',
-            'description': 'md5:a8eec7942e1664a6896fcd5e1287bfd0',
-            'thumbnail': 're:^https?://.*\.jpg',
-            'timestamp': 1404336058,
-            'upload_date': '20140702',
-            'duration': 343,
-        }
-    }
+            'title': 'Mnet Asian Music Awards 2015',
+            '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': 5359,
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.dramafever.com/zh-cn/drama/4972/15/Doctor_Romantic/',
+        'only_matching': True,
+    }]
+
+    def _call_api(self, path, video_id, note, fatal=False):
+        return self._download_json(
+            'https://www.dramafever.com/api/5/' + path,
+            video_id, note=note, headers={
+                'x-consumer-key': self._consumer_secret,
+            }, fatal=fatal)
+
+    def _get_subtitles(self, video_id):
+        subtitles = {}
+        subs = self._call_api(
+            'video/%s/subtitles/webvtt/' % video_id, video_id,
+            'Downloading subtitles JSON', fatal=False)
+        if not subs or not isinstance(subs, list):
+            return subtitles
+        for sub in subs:
+            if not isinstance(sub, dict):
+                continue
+            sub_url = url_or_none(sub.get('url'))
+            if not sub_url:
+                continue
+            subtitles.setdefault(
+                sub.get('code') or sub.get('language') or 'en', []).append({
+                    'url': sub_url
+                })
+        return subtitles
 
     def _real_extract(self, url):
         video_id = self._match_id(url).replace('/', '.')
 
-        try:
-            feed = self._download_json(
-                'http://www.dramafever.com/amp/episode/feed.json?guid=%s' % video_id,
-                video_id, 'Downloading episode JSON')['channel']['item']
-        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('.')
 
-        media_group = feed.get('media-group', {})
+        video = self._call_api(
+            'series/%s/episodes/%s/' % (series_id, episode_number), video_id,
+            'Downloading video JSON')
 
         formats = []
-        for media_content in media_group['media-content']:
-            src = media_content.get('@attributes', {}).get('url')
-            if not src:
-                continue
-            ext = determine_ext(src)
-            if ext == 'f4m':
-                formats.extend(self._extract_f4m_formats(
-                    src, video_id, f4m_id='hds'))
-            elif ext == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(
-                    src, video_id, 'mp4', m3u8_id='hls'))
-            else:
+        download_assets = video.get('download_assets')
+        if download_assets and isinstance(download_assets, dict):
+            for format_id, format_dict in download_assets.items():
+                if not isinstance(format_dict, dict):
+                    continue
+                format_url = url_or_none(format_dict.get('url'))
+                if not format_url:
+                    continue
                 formats.append({
-                    'url': src,
+                    'url': format_url,
+                    'format_id': format_id,
+                    'filesize': int_or_none(video.get('filesize')),
                 })
+
+        stream = self._call_api(
+            'video/%s/stream/' % video_id, video_id, 'Downloading stream JSON',
+            fatal=False)
+        if stream:
+            stream_url = stream.get('stream_url')
+            if stream_url:
+                formats.extend(self._extract_m3u8_formats(
+                    stream_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
         self._sort_formats(formats)
 
-        title = media_group.get('media-title')
-        description = media_group.get('media-description')
-        duration = int_or_none(media_group['media-content'][0].get('@attributes', {}).get('duration'))
-        thumbnail = self._proto_relative_url(
-            media_group.get('media-thumbnail', {}).get('@attributes', {}).get('url'))
-        timestamp = parse_iso8601(feed.get('pubDate'), ' ')
+        title = video.get('title') or 'Episode %s' % episode_number
+        description = video.get('description')
+        thumbnail = video.get('thumbnail')
+        timestamp = unified_timestamp(video.get('release_date'))
+        duration = parse_duration(video.get('duration'))
+        age_limit = parse_age_limit(video.get('tv_rating'))
+        series = video.get('series_title')
+        season_number = int_or_none(video.get('season'))
 
-        subtitles = {}
-        for media_subtitle in media_group.get('media-subTitle', []):
-            lang = media_subtitle.get('@attributes', {}).get('lang')
-            href = media_subtitle.get('@attributes', {}).get('href')
-            if not lang or not href:
-                continue
-            subtitles[lang] = [{
-                'ext': 'ttml',
-                'url': href,
-            }]
+        if series:
+            title = '%s - %s' % (series, title)
 
-        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 value:
-                subfile = value[0].get('subfile') or value[0].get('new_subfile')
-                if subfile and subfile != 'http://www.dramafever.com/st/':
-                    subtitles.setdefault('English', []).append({
-                        'ext': 'srt',
-                        'url': subfile,
-                    })
+        subtitles = self.extract_subtitles(video_id)
 
         return {
             'id': video_id,
             'title': title,
             'description': description,
             'thumbnail': thumbnail,
-            'timestamp': timestamp,
             'duration': duration,
+            'timestamp': timestamp,
+            'age_limit': age_limit,
+            'series': series,
+            'season_number': season_number,
+            'episode_number': int_or_none(episode_number),
             'formats': formats,
             'subtitles': subtitles,
         }
@@ -165,7 +215,7 @@ class DramaFeverIE(DramaFeverBaseIE):
 
 class DramaFeverSeriesIE(DramaFeverBaseIE):
     IE_NAME = 'dramafever:series'
-    _VALID_URL = r'https?://(?:www\.)?dramafever\.com/drama/(?P<id>[0-9]+)(?:/(?:(?!\d+(?:/|$)).+)?)?$'
+    _VALID_URL = r'https?://(?:www\.)?dramafever\.com/(?:[^/]+/)?drama/(?P<id>[0-9]+)(?:/(?:(?!\d+(?:/|$)).+)?)?$'
     _TESTS = [{
         'url': 'http://www.dramafever.com/drama/4512/Cooking_with_Shin/',
         'info_dict': {