Annotate changelog with (some) bugs closed.

[youtubedl] / youtube_dl / extractor / crunchyroll.py

diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py
index fe1324fe3cc047b87e605d2724e3bab015b3ed28..d1b6d7366e847015af4581160c918d4a1ee6e11f 100644 (file)
--- a/youtube_dl/extractor/crunchyroll.py
+++ b/youtube_dl/extractor/crunchyroll.py
@@ -9,11 +9,14 @@ import xml.etree.ElementTree
 
 from hashlib import sha1
 from math import pow, sqrt, floor
 
 from hashlib import sha1
 from math import pow, sqrt, floor

-from .subtitles import SubtitlesInfoExtractor
-from ..utils import (
-    ExtractorError,
+from .common import InfoExtractor
+from ..compat import (

     compat_urllib_parse,
     compat_urllib_parse,

+    compat_urllib_parse_unquote,

     compat_urllib_request,
     compat_urllib_request,

+)
+from ..utils import (
+    ExtractorError,

     bytes_to_intlist,
     intlist_to_bytes,
     unified_strdate,
     bytes_to_intlist,
     intlist_to_bytes,
     unified_strdate,


@@ -21,16 +24,14 @@ from ..utils import (
 )
 from ..aes import (
     aes_cbc_decrypt,
 )
 from ..aes import (
     aes_cbc_decrypt,

-    inc,

 )
 )

-from .common import InfoExtractor

 
 
 
 

-class CrunchyrollIE(SubtitlesInfoExtractor):
-    _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?:[^/]*/[^/?&]*?|media/\?id=)(?P<video_id>[0-9]+))(?:[/?&]|$)'
-    _TEST = {
+class CrunchyrollIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)'
+    _NETRC_MACHINE = 'crunchyroll'
+    _TESTS = [{

         'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
         'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',

-        #'md5': 'b1639fd6ddfaa43788c85f6d1dddd412',

         'info_dict': {
             'id': '645513',
             'ext': 'flv',
         'info_dict': {
             'id': '645513',
             'ext': 'flv',


@@ -45,7 +46,26 @@ class CrunchyrollIE(SubtitlesInfoExtractor):
             # rtmp
             'skip_download': True,
         },
             # rtmp
             'skip_download': True,
         },

-    }
+    }, {
+        'url': 'http://www.crunchyroll.com/media-589804/culture-japan-1',
+        'info_dict': {
+            'id': '589804',
+            'ext': 'flv',
+            'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11',
+            'description': 'md5:fe2743efedb49d279552926d0bd0cd9e',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'Danny Choo Network',
+            'upload_date': '20120213',
+        },
+        'params': {
+            # rtmp
+            'skip_download': True,
+        },
+
+    }, {
+        'url': 'http://www.crunchyroll.fr/girl-friend-beta/episode-11-goodbye-la-mode-661697',
+        'only_matching': True,
+    }]

 
     _FORMAT_IDS = {
         '360': ('60', '106'),
 
     _FORMAT_IDS = {
         '360': ('60', '106'),


@@ -69,14 +89,12 @@ class CrunchyrollIE(SubtitlesInfoExtractor):
         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         self._download_webpage(login_request, None, False, 'Wrong login info')
 
         login_request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         self._download_webpage(login_request, None, False, 'Wrong login info')
 

-

     def _real_initialize(self):
         self._login()
 
     def _real_initialize(self):
         self._login()
 

-

     def _decrypt_subtitles(self, data, iv, id):
     def _decrypt_subtitles(self, data, iv, id):

-        data = bytes_to_intlist(data)
-        iv = bytes_to_intlist(iv)
+        data = bytes_to_intlist(base64.b64decode(data.encode('utf-8')))
+        iv = bytes_to_intlist(base64.b64decode(iv.encode('utf-8')))

         id = int(id)
 
         def obfuscate_key_aux(count, modulo, start):
         id = int(id)
 
         def obfuscate_key_aux(count, modulo, start):


@@ -99,12 +117,7 @@ class CrunchyrollIE(SubtitlesInfoExtractor):
             return shaHash + [0] * 12
 
         key = obfuscate_key(id)
             return shaHash + [0] * 12
 
         key = obfuscate_key(id)

-        class Counter:
-            __value = iv
-            def next_value(self):
-                temp = self.__value
-                self.__value = inc(self.__value)
-                return temp
+

         decrypted_data = intlist_to_bytes(aes_cbc_decrypt(data, key, iv))
         return zlib.decompress(decrypted_data)
 
         decrypted_data = intlist_to_bytes(aes_cbc_decrypt(data, key, iv))
         return zlib.decompress(decrypted_data)
 


@@ -183,7 +196,35 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 
         return output
 
 
         return output
 

-    def _real_extract(self,url):
+    def _extract_subtitles(self, subtitle):
+        sub_root = xml.etree.ElementTree.fromstring(subtitle)
+        return [{
+            'ext': 'srt',
+            'data': self._convert_subtitles_to_srt(sub_root),
+        }, {
+            'ext': 'ass',
+            'data': self._convert_subtitles_to_ass(sub_root),
+        }]
+
+    def _get_subtitles(self, video_id, webpage):
+        subtitles = {}
+        for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage):
+            sub_page = self._download_webpage(
+                'http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id=' + sub_id,
+                video_id, note='Downloading subtitles for ' + sub_name)
+            id = self._search_regex(r'id=\'([0-9]+)', sub_page, 'subtitle_id', fatal=False)
+            iv = self._search_regex(r'<iv>([^<]+)', sub_page, 'subtitle_iv', fatal=False)
+            data = self._search_regex(r'<data>([^<]+)', sub_page, 'subtitle_data', fatal=False)
+            if not id or not iv or not data:
+                continue
+            subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8')
+            lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False)
+            if not lang_code:
+                continue
+            subtitles[lang_code] = self._extract_subtitles(subtitle)
+        return subtitles
+
+    def _real_extract(self, url):

         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('video_id')
 
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('video_id')
 


@@ -214,7 +255,7 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             video_upload_date = unified_strdate(video_upload_date)
         video_uploader = self._html_search_regex(r'<div>\s*Publisher:(.+?)</div>', webpage, 'video_uploader', fatal=False, flags=re.DOTALL)
 
             video_upload_date = unified_strdate(video_upload_date)
         video_uploader = self._html_search_regex(r'<div>\s*Publisher:(.+?)</div>', webpage, 'video_uploader', fatal=False, flags=re.DOTALL)
 

-        playerdata_url = compat_urllib_parse.unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url'))
+        playerdata_url = compat_urllib_parse_unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url'))

         playerdata_req = compat_urllib_request.Request(playerdata_url)
         playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url})
         playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
         playerdata_req = compat_urllib_request.Request(playerdata_url)
         playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url})
         playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')


@@ -224,19 +265,20 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False)
 
         formats = []
         video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False)
 
         formats = []

-        for fmt in re.findall(r'\?p([0-9]{3,4})=1', webpage):
+        for fmt in re.findall(r'showmedia\.([0-9]{3,4})p', webpage):

             stream_quality, stream_format = self._FORMAT_IDS[fmt]
             stream_quality, stream_format = self._FORMAT_IDS[fmt]

-            video_format = fmt+'p'
-            streamdata_req = compat_urllib_request.Request('http://www.crunchyroll.com/xml/')
-            # urlencode doesn't work!
-            streamdata_req.data = 'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality='+stream_quality+'&media%5Fid='+stream_id+'&video%5Fformat='+stream_format
+            video_format = fmt + 'p'
+            streamdata_req = compat_urllib_request.Request(
+                'http://www.crunchyroll.com/xml/?req=RpcApiVideoPlayer_GetStandardConfig&media_id=%s&video_format=%s&video_quality=%s'
+                % (stream_id, stream_format, stream_quality),
+                compat_urllib_parse.urlencode({'current_page': url}).encode('utf-8'))

             streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
             streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')

-            streamdata_req.add_header('Content-Length', str(len(streamdata_req.data)))

             streamdata = self._download_xml(
                 streamdata_req, video_id,
                 note='Downloading media info for %s' % video_format)
             streamdata = self._download_xml(
                 streamdata_req, video_id,
                 note='Downloading media info for %s' % video_format)

-            video_url = streamdata.find('.//host').text
-            video_play_path = streamdata.find('.//file').text
+            stream_info = streamdata.find('./{default}preload/stream_info')
+            video_url = stream_info.find('./host').text
+            video_play_path = stream_info.find('./file').text

             formats.append({
                 'url': video_url,
                 'play_path': video_play_path,
             formats.append({
                 'url': video_url,
                 'play_path': video_play_path,


@@ -245,43 +287,17 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
                 'format_id': video_format,
             })
 
                 'format_id': video_format,
             })
 

-        subtitles = {}
-        sub_format = self._downloader.params.get('subtitlesformat', 'srt')
-        for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage):
-            sub_page = self._download_webpage('http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id='+sub_id,\
-                                              video_id, note='Downloading subtitles for '+sub_name)
-            id = self._search_regex(r'id=\'([0-9]+)', sub_page, 'subtitle_id', fatal=False)
-            iv = self._search_regex(r'<iv>([^<]+)', sub_page, 'subtitle_iv', fatal=False)
-            data = self._search_regex(r'<data>([^<]+)', sub_page, 'subtitle_data', fatal=False)
-            if not id or not iv or not data:
-                continue
-            id = int(id)
-            iv = base64.b64decode(iv)
-            data = base64.b64decode(data)
-
-            subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8')
-            lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False)
-            if not lang_code:
-                continue
-            sub_root = xml.etree.ElementTree.fromstring(subtitle)
-            if sub_format == 'ass':
-                subtitles[lang_code] = self._convert_subtitles_to_ass(sub_root)
-            else:
-                subtitles[lang_code] = self._convert_subtitles_to_srt(sub_root)
-
-        if self._downloader.params.get('listsubtitles', False):
-            self._list_available_subtitles(video_id, subtitles)
-            return
+        subtitles = self.extract_subtitles(video_id, webpage)

 
         return {
 
         return {

-            'id':          video_id,
-            'title':       video_title,
+            'id': video_id,
+            'title': video_title,

             'description': video_description,
             'description': video_description,

-            'thumbnail':   video_thumbnail,
-            'uploader':    video_uploader,
+            'thumbnail': video_thumbnail,
+            'uploader': video_uploader,

             'upload_date': video_upload_date,
             'upload_date': video_upload_date,

-            'subtitles':   subtitles,
-            'formats':     formats,
+            'subtitles': subtitles,
+            'formats': formats,

         }
 
 
         }