X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fcrunchyroll.py;h=588c3c71b776ccb9ad52d721d389f1fd3833abab;hb=067aa17edf5a46a8cbc4d6b90864eddf051fa2bc;hp=af786d09699c10b60f38846bf43310ee1fec03a4;hpb=1084563eaab3a039ea8ccb65f2507e919d8e11f0;p=youtube-dl

diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py
index af786d096..588c3c71b 100644
--- a/youtube_dl/extractor/crunchyroll.py
+++ b/youtube_dl/extractor/crunchyroll.py
@@ -11,6 +11,7 @@ from .common import InfoExtractor
 from .vrv import VRVIE
 from ..compat import (
     compat_b64decode,
+    compat_etree_Element,
     compat_etree_fromstring,
     compat_urllib_parse_urlencode,
     compat_urllib_request,
@@ -45,7 +46,7 @@ class CrunchyrollBaseIE(InfoExtractor):
         data['req'] = 'RpcApi' + method
         data = compat_urllib_parse_urlencode(data).encode('utf-8')
         return self._download_xml(
-            'http://www.crunchyroll.com/xml/',
+            'https://www.crunchyroll.com/xml/',
             video_id, note, fatal=False, data=data, headers={
                 'Content-Type': 'application/x-www-form-urlencoded',
             })
@@ -55,22 +56,11 @@ class CrunchyrollBaseIE(InfoExtractor):
         if username is None:
             return
 
-        self._download_webpage(
-            'https://www.crunchyroll.com/?a=formhandler',
-            None, 'Logging in', 'Wrong login info',
-            data=urlencode_postdata({
-                'formname': 'RpcApiUser_Login',
-                'next_url': 'https://www.crunchyroll.com/acct/membership',
-                'name': username,
-                'password': password,
-            }))
-
-        '''
         login_page = self._download_webpage(
             self._LOGIN_URL, None, 'Downloading login page')
 
         def is_logged(webpage):
-            return '<title>Redirecting' in webpage
+            return 'href="/logout"' in webpage
 
         # Already logged in
         if is_logged(login_page):
@@ -109,7 +99,6 @@ class CrunchyrollBaseIE(InfoExtractor):
             raise ExtractorError('Unable to login: %s' % error, expected=True)
 
         raise ExtractorError('Unable to log in')
-        '''
 
     def _real_initialize(self):
         self._login()
@@ -118,7 +107,7 @@ class CrunchyrollBaseIE(InfoExtractor):
         request = (url_or_request if isinstance(url_or_request, compat_urllib_request.Request)
                    else sanitized_Request(url_or_request))
         # Accept-Language must be set explicitly to accept any language to avoid issues
-        # similar to https://github.com/rg3/youtube-dl/issues/6797.
+        # similar to https://github.com/ytdl-org/youtube-dl/issues/6797.
         # Along with IP address Crunchyroll uses Accept-Language to guess whether georestriction
         # should be imposed or not (from what I can see it just takes the first language
         # ignoring the priority and requires it to correspond the IP). By the way this causes
@@ -135,7 +124,7 @@ class CrunchyrollBaseIE(InfoExtractor):
         # > This content may be inappropriate for some people.
         # > Are you sure you want to continue?
         # since it's not disabled by default in crunchyroll account's settings.
-        # See https://github.com/rg3/youtube-dl/issues/7202.
+        # See https://github.com/ytdl-org/youtube-dl/issues/7202.
         qs['skip_wall'] = ['1']
         return compat_urlparse.urlunparse(
             parsed_url._replace(query=compat_urllib_parse_urlencode(qs, True)))
@@ -143,7 +132,7 @@ class CrunchyrollBaseIE(InfoExtractor):
 
 class CrunchyrollIE(CrunchyrollBaseIE, VRVIE):
     IE_NAME = 'crunchyroll'
-    _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)'
+    _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|(?:[^/]*/){1,2}[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)'
     _TESTS = [{
         'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
         'info_dict': {
@@ -268,6 +257,9 @@ class CrunchyrollIE(CrunchyrollBaseIE, VRVIE):
     }, {
         'url': 'http://www.crunchyroll.com/media-723735',
         'only_matching': True,
+    }, {
+        'url': 'https://www.crunchyroll.com/en-gb/mob-psycho-100/episode-2-urban-legends-encountering-rumors-780921',
+        'only_matching': True,
     }]
 
     _FORMAT_IDS = {
@@ -398,7 +390,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
                 'Downloading subtitles for ' + sub_name, data={
                     'subtitle_script_id': sub_id,
                 })
-            if sub_doc is None:
+            if not isinstance(sub_doc, compat_etree_Element):
                 continue
             sid = sub_doc.get('id')
             iv = xpath_text(sub_doc, 'iv', 'subtitle iv')
@@ -515,7 +507,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
                         'video_quality': stream_quality,
                         'current_page': url,
                     })
-                if streamdata is not None:
+                if isinstance(streamdata, compat_etree_Element):
                     stream_info = streamdata.find('./{default}preload/stream_info')
                     if stream_info is not None:
                         stream_infos.append(stream_info)
@@ -526,7 +518,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
                         'video_format': stream_format,
                         'video_encode_quality': stream_quality,
                     })
-                if stream_info is not None:
+                if isinstance(stream_info, compat_etree_Element):
                     stream_infos.append(stream_info)
                 for stream_info in stream_infos:
                     video_encode_id = xpath_text(stream_info, './video_encode_id')
@@ -598,10 +590,22 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         series = self._html_search_regex(
             r'(?s)<h\d[^>]+\bid=["\']showmedia_about_episode_num[^>]+>(.+?)</h\d',
             webpage, 'series', fatal=False)
-        season = xpath_text(metadata, 'series_title')
 
-        episode = xpath_text(metadata, 'episode_title') or media_metadata.get('title')
-        episode_number = int_or_none(xpath_text(metadata, 'episode_number') or media_metadata.get('episode_number'))
+        season = episode = episode_number = duration = thumbnail = None
+
+        if isinstance(metadata, compat_etree_Element):
+            season = xpath_text(metadata, 'series_title')
+            episode = xpath_text(metadata, 'episode_title')
+            episode_number = int_or_none(xpath_text(metadata, 'episode_number'))
+            duration = float_or_none(media_metadata.get('duration'), 1000)
+            thumbnail = xpath_text(metadata, 'episode_image_url')
+
+        if not episode:
+            episode = media_metadata.get('title')
+        if not episode_number:
+            episode_number = int_or_none(media_metadata.get('episode_number'))
+        if not thumbnail:
+            thumbnail = media_metadata.get('thumbnail', {}).get('url')
 
         season_number = int_or_none(self._search_regex(
             r'(?s)<h\d[^>]+id=["\']showmedia_about_episode_num[^>]+>.+?</h\d>\s*<h4>\s*Season (\d+)',
@@ -611,8 +615,8 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             'id': video_id,
             'title': video_title,
             'description': video_description,
-            'duration': float_or_none(media_metadata.get('duration'), 1000),
-            'thumbnail': xpath_text(metadata, 'episode_image_url') or media_metadata.get('thumbnail', {}).get('url'),
+            'duration': duration,
+            'thumbnail': thumbnail,
             'uploader': video_uploader,
             'upload_date': video_upload_date,
             'series': series,