[youtube] Fix extraction.
[youtube-dl] / youtube_dl / extractor / funimation.py
index 8de28029f277452e5ddd48a39f25c7f6bc7bc97b..8bbedca269233b2ba4bdd02febf7d8e63007feb4 100644 (file)
 # coding: utf-8
 from __future__ import unicode_literals
+
+import random
+import string
+
 from .common import InfoExtractor
 from ..compat import compat_HTTPError
 from ..utils import (
-    encode_dict,
-    sanitized_Request,
+    determine_ext,
+    int_or_none,
+    js_to_json,
     ExtractorError,
     urlencode_postdata
 )
-import re
 
 
 class FunimationIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?funimation\.com/shows/.+[^ ]/videos/official/(?P<id>[^?]+)'
+    _VALID_URL = r'https?://(?:www\.)?funimation(?:\.com|now\.uk)/shows/[^/]+/(?P<id>[^/?#&]+)'
 
-    _TEST = {
-        'url': 'http://www.funimation.com/shows/air/videos/official/breeze',
+    _NETRC_MACHINE = 'funimation'
+    _TOKEN = None
+
+    _TESTS = [{
+        'url': 'https://www.funimation.com/shows/hacksign/role-play/',
         'info_dict': {
-            'id': 'AIRENG0001',
-            'title': 'Air - 1 - Breeze ',
+            'id': '91144',
+            'display_id': 'role-play',
             'ext': 'mp4',
-            'thumbnail': 'http://www.funimation.com/admin/uploads/default/recap_thumbnails/7555590/home_spotlight/AIR0001.jpg',
-            'description': 'Travelling puppeteer Yukito arrives in a small town where he hopes to earn money through the magic of his puppets. When a young girl named Misuzu lures him to her home with the promise of food, his life changes forever. ',
-        }
-    }
+            'title': '.hack//SIGN - Role Play',
+            'description': 'md5:b602bdc15eef4c9bbb201bb6e6a4a2dd',
+            'thumbnail': r're:https?://.*\.jpg',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.funimation.com/shows/attack-on-titan-junior-high/broadcast-dub-preview/',
+        'info_dict': {
+            'id': '210051',
+            'display_id': 'broadcast-dub-preview',
+            'ext': 'mp4',
+            'title': 'Attack on Titan: Junior High - Broadcast Dub Preview',
+            'thumbnail': r're:https?://.*\.(?:jpg|png)',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }, {
+        'url': 'https://www.funimationnow.uk/shows/puzzle-dragons-x/drop-impact/simulcast/',
+        'only_matching': True,
+    }]
 
     def _login(self):
-        (username, password) = self._get_login_info()
+        username, password = self._get_login_info()
         if username is None:
             return
-        login_url = 'http://www.funimation.com/login'
-        data = urlencode_postdata(encode_dict({
-            'loginForm2': 'loginform',
-            'email_field': username,
-            'password_field': password,
-        }))
-        login_request = sanitized_Request(login_url, data)
-        login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         try:
-            login = self._download_webpage(
-                login_request, None, 'Logging in as %s' % username)
+            data = self._download_json(
+                'https://prod-api-funimationnow.dadcdigital.com/api/auth/login/',
+                None, 'Logging in', data=urlencode_postdata({
+                    'username': username,
+                    'password': password,
+                }))
+            self._TOKEN = data['token']
         except ExtractorError as e:
-            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
-                raise ExtractorError('Funimation is not available in your region.', expected=True)
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
+                error = self._parse_json(e.cause.read().decode(), None)['error']
+                raise ExtractorError(error, expected=True)
             raise
-        if re.search(r'<meta property="og:url" content="http://www.funimation.com/login"/>', login) is not None:
-            raise ExtractorError('Unable to login, wrong username or password.', expected=True)
 
     def _real_initialize(self):
         self._login()
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+
+        def _search_kane(name):
+            return self._search_regex(
+                r"KANE_customdimensions\.%s\s*=\s*'([^']+)';" % name,
+                webpage, name, default=None)
+
+        title_data = self._parse_json(self._search_regex(
+            r'TITLE_DATA\s*=\s*({[^}]+})',
+            webpage, 'title data', default=''),
+            display_id, js_to_json, fatal=False) or {}
+
+        video_id = title_data.get('id') or self._search_regex([
+            r"KANE_customdimensions.videoID\s*=\s*'(\d+)';",
+            r'<iframe[^>]+src="/player/(\d+)',
+        ], webpage, 'video_id', default=None)
+        if not video_id:
+            player_url = self._html_search_meta([
+                'al:web:url',
+                'og:video:url',
+                'og:video:secure_url',
+            ], webpage, fatal=True)
+            video_id = self._search_regex(r'/player/(\d+)', player_url, 'video id')
+
+        title = episode = title_data.get('title') or _search_kane('videoTitle') or self._og_search_title(webpage)
+        series = _search_kane('showName')
+        if series:
+            title = '%s - %s' % (series, title)
+        description = self._html_search_meta(['description', 'og:description'], webpage, fatal=True)
+
         try:
-            webpage = self._download_webpage(url, video_id)
+            headers = {}
+            if self._TOKEN:
+                headers['Authorization'] = 'Token %s' % self._TOKEN
+            sources = self._download_json(
+                'https://www.funimation.com/api/showexperience/%s/' % video_id,
+                video_id, headers=headers, query={
+                    'pinst_id': ''.join([random.choice(string.digits + string.ascii_letters) for _ in range(8)]),
+                })['items']
         except ExtractorError as e:
             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
-                raise ExtractorError('Funimation is not available in your region.', expected=True)
+                error = self._parse_json(e.cause.read(), video_id)['errors'][0]
+                raise ExtractorError('%s said: %s' % (
+                    self.IE_NAME, error.get('detail') or error.get('title')), expected=True)
             raise
-        if re.search(r'"sdUrl":"http', webpage) is None:
-                raise ExtractorError('You are not logged-in or the stream requires subscription.', expected=True)
 
-        m3u8 = self._search_regex(r'".+Url":"(.+?m3u8)"', webpage, 'm3u8') + self._search_regex(r'"authToken":"(.+?)"', webpage, 'm3u8')
-        formats = self._extract_m3u8_formats(m3u8.replace('\\', ''), video_id, ext='mp4', entry_protocol='m3u8_native')
-
-        video_show = self._search_regex(r'"artist":"(.+?)"', webpage, 'video_show')
-        video_track = self._search_regex(r'"videoNumber":"(\d+).0"', webpage, 'video_track')
-        video_title = self._search_regex(r'"title":"({0}.+?)"'.format(video_track), webpage, 'video_title')
-        video_id = self._search_regex(r'"FUNImationID":"(.+?)"', webpage, 'video_id')
+        formats = []
+        for source in sources:
+            source_url = source.get('src')
+            if not source_url:
+                continue
+            source_type = source.get('videoType') or determine_ext(source_url)
+            if source_type == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    source_url, video_id, 'mp4',
+                    m3u8_id='hls', fatal=False))
+            else:
+                formats.append({
+                    'format_id': source_type,
+                    'url': source_url,
+                })
+        self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'title': video_show + ' - ' + video_title + ' ',
-            'formats': formats,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
             'thumbnail': self._og_search_thumbnail(webpage),
-            'description': self._og_search_description(webpage)
+            'series': series,
+            'season_number': int_or_none(title_data.get('seasonNum') or _search_kane('season')),
+            'episode_number': int_or_none(title_data.get('episodeNum')),
+            'episode': episode,
+            'season_id': title_data.get('seriesId'),
+            'formats': formats,
         }