]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/animeondemand.py
prepare NMU
[youtubedl] / youtube_dl / extractor / animeondemand.py
index a7d8daf7b4788bbaff020ccfca1ed7ba46e5d6f8..00ce684d1c9811ef3ba885194c23fd012d935b6b 100644 (file)
@@ -3,13 +3,14 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_urlparse
+from ..compat import compat_str
 from ..utils import (
     determine_ext,
-    encode_dict,
+    extract_attributes,
     ExtractorError,
-    sanitized_Request,
+    url_or_none,
     urlencode_postdata,
+    urljoin,
 )
 
 
@@ -18,7 +19,10 @@ class AnimeOnDemandIE(InfoExtractor):
     _LOGIN_URL = 'https://www.anime-on-demand.de/users/sign_in'
     _APPLY_HTML5_URL = 'https://www.anime-on-demand.de/html5apply'
     _NETRC_MACHINE = 'animeondemand'
-    _TEST = {
+    # German-speaking countries of Europe
+    _GEO_COUNTRIES = ['AT', 'CH', 'DE', 'LI', 'LU']
+    _TESTS = [{
+        # jap, OmU
         'url': 'https://www.anime-on-demand.de/anime/161',
         'info_dict': {
             'id': '161',
@@ -26,16 +30,40 @@ class AnimeOnDemandIE(InfoExtractor):
             'description': 'md5:6681ce3c07c7189d255ac6ab23812d31',
         },
         'playlist_mincount': 4,
-    }
+    }, {
+        # Film wording is used instead of Episode, ger/jap, Dub/OmU
+        'url': 'https://www.anime-on-demand.de/anime/39',
+        'only_matching': True,
+    }, {
+        # Episodes without titles, jap, OmU
+        'url': 'https://www.anime-on-demand.de/anime/162',
+        'only_matching': True,
+    }, {
+        # ger/jap, Dub/OmU, account required
+        'url': 'https://www.anime-on-demand.de/anime/169',
+        'only_matching': True,
+    }, {
+        # Full length film, non-series, ger/jap, Dub/OmU, account required
+        'url': 'https://www.anime-on-demand.de/anime/185',
+        'only_matching': True,
+    }, {
+        # Flash videos
+        'url': 'https://www.anime-on-demand.de/anime/12',
+        'only_matching': True,
+    }]
 
     def _login(self):
-        (username, password) = self._get_login_info()
+        username, password = self._get_login_info()
         if username is None:
             return
 
         login_page = self._download_webpage(
             self._LOGIN_URL, None, 'Downloading login page')
 
+        if '>Our licensing terms allow the distribution of animes only to German-speaking countries of Europe' in login_page:
+            self.raise_geo_restricted(
+                '%s is only available in German-speaking countries of Europe' % self.IE_NAME)
+
         login_form = self._form_hidden_inputs('new_user', login_page)
 
         login_form.update({
@@ -48,19 +76,18 @@ class AnimeOnDemandIE(InfoExtractor):
             'post url', default=self._LOGIN_URL, group='url')
 
         if not post_url.startswith('http'):
-            post_url = compat_urlparse.urljoin(self._LOGIN_URL, post_url)
-
-        request = sanitized_Request(
-            post_url, urlencode_postdata(encode_dict(login_form)))
-        request.add_header('Referer', self._LOGIN_URL)
+            post_url = urljoin(self._LOGIN_URL, post_url)
 
         response = self._download_webpage(
-            request, None, 'Logging in as %s' % username)
+            post_url, None, 'Logging in',
+            data=urlencode_postdata(login_form), headers={
+                'Referer': self._LOGIN_URL,
+            })
 
         if all(p not in response for p in ('>Logout<', 'href="/users/sign_out"')):
             error = self._search_regex(
-                r'<p class="alert alert-danger">(.+?)</p>',
-                response, 'error', default=None)
+                r'<p[^>]+\bclass=(["\'])(?:(?!\1).)*\balert\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</p>',
+                response, 'error', default=None, group='error')
             if error:
                 raise ExtractorError('Unable to login: %s' % error, expected=True)
             raise ExtractorError('Unable to log in')
@@ -91,70 +118,176 @@ class AnimeOnDemandIE(InfoExtractor):
 
         entries = []
 
-        for episode_html in re.findall(r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', webpage):
-            m = re.search(
-                r'class="episodebox-title"[^>]+title="Episode (?P<number>\d+) - (?P<title>.+?)"', episode_html)
-            if not m:
-                continue
-
-            episode_number = int(m.group('number'))
-            episode_title = m.group('title')
-            video_id = 'episode-%d' % episode_number
-
-            common_info = {
-                'id': video_id,
-                'series': anime_title,
-                'episode': episode_title,
-                'episode_number': episode_number,
-            }
-
+        def extract_info(html, video_id, num=None):
+            title, description = [None] * 2
             formats = []
 
-            playlist_url = self._search_regex(
-                r'data-playlist=(["\'])(?P<url>.+?)\1',
-                episode_html, 'data playlist', default=None, group='url')
-            if playlist_url:
-                request = sanitized_Request(
-                    compat_urlparse.urljoin(url, playlist_url),
-                    headers={
-                        'X-Requested-With': 'XMLHttpRequest',
-                        'X-CSRF-Token': csrf_token,
-                        'Referer': url,
-                        'Accept': 'application/json, text/javascript, */*; q=0.01',
-                    })
+            for input_ in re.findall(
+                    r'<input[^>]+class=["\'].*?streamstarter[^>]+>', html):
+                attributes = extract_attributes(input_)
+                title = attributes.get('data-dialog-header')
+                playlist_urls = []
+                for playlist_key in ('data-playlist', 'data-otherplaylist', 'data-stream'):
+                    playlist_url = attributes.get(playlist_key)
+                    if isinstance(playlist_url, compat_str) and re.match(
+                            r'/?[\da-zA-Z]+', playlist_url):
+                        playlist_urls.append(attributes[playlist_key])
+                if not playlist_urls:
+                    continue
+
+                lang = attributes.get('data-lang')
+                lang_note = attributes.get('value')
 
-                playlist = self._download_json(
-                    request, video_id, 'Downloading playlist JSON', fatal=False)
-                if playlist:
-                    playlist = playlist['playlist'][0]
-                    title = playlist['title']
+                for playlist_url in playlist_urls:
+                    kind = self._search_regex(
+                        r'videomaterialurl/\d+/([^/]+)/',
+                        playlist_url, 'media kind', default=None)
+                    format_id_list = []
+                    if lang:
+                        format_id_list.append(lang)
+                    if kind:
+                        format_id_list.append(kind)
+                    if not format_id_list and num is not None:
+                        format_id_list.append(compat_str(num))
+                    format_id = '-'.join(format_id_list)
+                    format_note = ', '.join(filter(None, (kind, lang_note)))
+                    item_id_list = []
+                    if format_id:
+                        item_id_list.append(format_id)
+                    item_id_list.append('videomaterial')
+                    playlist = self._download_json(
+                        urljoin(url, playlist_url), video_id,
+                        'Downloading %s JSON' % ' '.join(item_id_list),
+                        headers={
+                            'X-Requested-With': 'XMLHttpRequest',
+                            'X-CSRF-Token': csrf_token,
+                            'Referer': url,
+                            'Accept': 'application/json, text/javascript, */*; q=0.01',
+                        }, fatal=False)
+                    if not playlist:
+                        continue
+                    stream_url = url_or_none(playlist.get('streamurl'))
+                    if stream_url:
+                        rtmp = re.search(
+                            r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+/))(?P<playpath>mp[34]:.+)',
+                            stream_url)
+                        if rtmp:
+                            formats.append({
+                                'url': rtmp.group('url'),
+                                'app': rtmp.group('app'),
+                                'play_path': rtmp.group('playpath'),
+                                'page_url': url,
+                                'player_url': 'https://www.anime-on-demand.de/assets/jwplayer.flash-55abfb34080700304d49125ce9ffb4a6.swf',
+                                'rtmp_real_time': True,
+                                'format_id': 'rtmp',
+                                'ext': 'flv',
+                            })
+                            continue
+                    start_video = playlist.get('startvideo', 0)
+                    playlist = playlist.get('playlist')
+                    if not playlist or not isinstance(playlist, list):
+                        continue
+                    playlist = playlist[start_video]
+                    title = playlist.get('title')
+                    if not title:
+                        continue
                     description = playlist.get('description')
                     for source in playlist.get('sources', []):
                         file_ = source.get('file')
-                        if file_ and determine_ext(file_) == 'm3u8':
-                            formats = self._extract_m3u8_formats(
+                        if not file_:
+                            continue
+                        ext = determine_ext(file_)
+                        format_id_list = [lang, kind]
+                        if ext == 'm3u8':
+                            format_id_list.append('hls')
+                        elif source.get('type') == 'video/dash' or ext == 'mpd':
+                            format_id_list.append('dash')
+                        format_id = '-'.join(filter(None, format_id_list))
+                        if ext == 'm3u8':
+                            file_formats = self._extract_m3u8_formats(
                                 file_, video_id, 'mp4',
-                                entry_protocol='m3u8_native', m3u8_id='hls')
+                                entry_protocol='m3u8_native', m3u8_id=format_id, fatal=False)
+                        elif source.get('type') == 'video/dash' or ext == 'mpd':
+                            continue
+                            file_formats = self._extract_mpd_formats(
+                                file_, video_id, mpd_id=format_id, fatal=False)
+                        else:
+                            continue
+                        for f in file_formats:
+                            f.update({
+                                'language': lang,
+                                'format_note': format_note,
+                            })
+                        formats.extend(file_formats)
 
-            if formats:
-                f = common_info.copy()
-                f.update({
-                    'title': title,
-                    'description': description,
-                    'formats': formats,
-                })
-                entries.append(f)
+            return {
+                'title': title,
+                'description': description,
+                'formats': formats,
+            }
+
+        def extract_entries(html, video_id, common_info, num=None):
+            info = extract_info(html, video_id, num)
 
-            m = re.search(
-                r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>Teaser<',
-                episode_html)
-            if m:
+            if info['formats']:
+                self._sort_formats(info['formats'])
                 f = common_info.copy()
-                f.update({
-                    'id': '%s-teaser' % f['id'],
-                    'title': m.group('title'),
-                    'url': compat_urlparse.urljoin(url, m.group('href')),
-                })
+                f.update(info)
                 entries.append(f)
 
+            # Extract teaser/trailer only when full episode is not available
+            if not info['formats']:
+                m = re.search(
+                    r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>(?P<kind>Teaser|Trailer)<',
+                    html)
+                if m:
+                    f = common_info.copy()
+                    f.update({
+                        'id': '%s-%s' % (f['id'], m.group('kind').lower()),
+                        'title': m.group('title'),
+                        'url': urljoin(url, m.group('href')),
+                    })
+                    entries.append(f)
+
+        def extract_episodes(html):
+            for num, episode_html in enumerate(re.findall(
+                    r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', html), 1):
+                episodebox_title = self._search_regex(
+                    (r'class="episodebox-title"[^>]+title=(["\'])(?P<title>.+?)\1',
+                     r'class="episodebox-title"[^>]+>(?P<title>.+?)<'),
+                    episode_html, 'episodebox title', default=None, group='title')
+                if not episodebox_title:
+                    continue
+
+                episode_number = int(self._search_regex(
+                    r'(?:Episode|Film)\s*(\d+)',
+                    episodebox_title, 'episode number', default=num))
+                episode_title = self._search_regex(
+                    r'(?:Episode|Film)\s*\d+\s*-\s*(.+)',
+                    episodebox_title, 'episode title', default=None)
+
+                video_id = 'episode-%d' % episode_number
+
+                common_info = {
+                    'id': video_id,
+                    'series': anime_title,
+                    'episode': episode_title,
+                    'episode_number': episode_number,
+                }
+
+                extract_entries(episode_html, video_id, common_info)
+
+        def extract_film(html, video_id):
+            common_info = {
+                'id': anime_id,
+                'title': anime_title,
+                'description': anime_description,
+            }
+            extract_entries(html, video_id, common_info)
+
+        extract_episodes(webpage)
+
+        if not entries:
+            extract_film(webpage, anime_id)
+
         return self.playlist_result(entries, anime_id, anime_title, anime_description)