]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/dramafever.py
Initiate new release.
[youtubedl] / youtube_dl / extractor / dramafever.py
index 3b6529f4b108052e3019c8e400bbc2cd0eb5a9a1..ab32ba4ff3eee337e8dfc2c7fa11c755e2b027c4 100644 (file)
@@ -1,24 +1,26 @@
-# encoding: utf-8
+# coding: utf-8
 from __future__ import unicode_literals
 
 import itertools
+import json
 
-from .amp import AMPIE
+from .common import InfoExtractor
 from ..compat import (
     compat_HTTPError,
+    compat_str,
     compat_urlparse,
 )
 from ..utils import (
-    ExtractorError,
     clean_html,
+    ExtractorError,
     int_or_none,
-    sanitized_Request,
-    urlencode_postdata
+    parse_age_limit,
+    parse_duration,
+    unified_timestamp,
 )
 
 
-class DramaFeverBaseIE(AMPIE):
-    _LOGIN_URL = 'https://www.dramafever.com/accounts/login/'
+class DramaFeverBaseIE(InfoExtractor):
     _NETRC_MACHINE = 'dramafever'
 
     _CONSUMER_SECRET = 'DA59dtVXYLxajktV'
@@ -36,11 +38,11 @@ class DramaFeverBaseIE(AMPIE):
             '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
 
@@ -49,37 +51,49 @@ class DramaFeverBaseIE(AMPIE):
             'password': password,
         }
 
-        request = sanitized_Request(
-            self._LOGIN_URL, urlencode_postdata(login_form))
-        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
+
+        # Successful login
+        if response.get('result') or response.get('guid') or response.get('user_guid'):
+            return
 
-        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')
+        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]+)(?:/|$)'
+    _VALID_URL = r'https?://(?:www\.)?dramafever\.com/(?:[^/]+/)?drama/(?P<id>[0-9]+/[0-9]+)(?:/|$)'
     _TESTS = [{
-        'url': 'http://www.dramafever.com/drama/4512/1/Cooking_with_Shin/',
+        'url': 'https://www.dramafever.com/drama/4274/1/Heirs/',
         'info_dict': {
-            'id': '4512.1',
-            'ext': 'mp4',
-            'title': 'Cooking with Shin 4512.1',
-            'description': 'md5:a8eec7942e1664a6896fcd5e1287bfd0',
+            '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,
-            'thumbnail': 're:^https?://.*\.jpg',
-            'timestamp': 1404336058,
-            'upload_date': '20140702',
-            'duration': 343,
         },
         'params': {
             # m3u8 download
@@ -89,66 +103,119 @@ class DramaFeverIE(DramaFeverBaseIE):
         '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',
+            'ext': 'flv',
+            'title': 'Mnet Asian Music Awards 2015',
             'description': 'md5:3ff2ee8fedaef86e076791c909cf2e91',
             'episode': 'Mnet Asian Music Awards 2015 - Part 3',
             'episode_number': 4,
-            'thumbnail': 're:^https?://.*\.jpg',
+            'thumbnail': r're:^https?://.*\.jpg',
             'timestamp': 1450213200,
             'upload_date': '20151215',
-            'duration': 5602,
+            '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 = sub.get('url')
+            if not sub_url or not isinstance(sub_url, compat_str):
+                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:
-            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
+
+        video = self._call_api(
+            'series/%s/episodes/%s/' % (series_id, episode_number), video_id,
+            'Downloading video JSON')
+
+        formats = []
+        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 = format_dict.get('url')
+                if not format_url or not isinstance(format_url, compat_str):
+                    continue
+                formats.append({
+                    '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 = 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'))
+
+        if series:
+            title = '%s - %s' % (series, title)
+
+        subtitles = self.extract_subtitles(video_id)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            '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,
+        }
 
 
 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': {