Merge branch 'lecture2go' of https://github.com/nichdu/youtube-dl into nichdu-lecture2go

[youtube-dl] / youtube_dl / extractor / crunchyroll.py

diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py
index 1c77df47ef346173fc11a58396c98768e5afc986..d1b6d7366e847015af4581160c918d4a1ee6e11f 100644 (file)
--- a/youtube_dl/extractor/crunchyroll.py
+++ b/youtube_dl/extractor/crunchyroll.py
@@ -12,6 +12,7 @@ from math import pow, sqrt, floor
 from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse,
+    compat_urllib_parse_unquote,
     compat_urllib_request,
 )
 from ..utils import (
@@ -27,7 +28,7 @@ from ..aes import (
 
 
 class CrunchyrollIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:[^/]*/[^/?&]*?|media/\?id=)(?P<video_id>[0-9]+))(?:[/?&]|$)'
+    _VALID_URL = r'https?://(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.(?:com|fr)/(?:media(?:-|/\?id=)|[^/]*/[^/?&]*?)(?P<video_id>[0-9]+))(?:[/?&]|$)'
     _NETRC_MACHINE = 'crunchyroll'
     _TESTS = [{
         'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
@@ -45,6 +46,22 @@ class CrunchyrollIE(InfoExtractor):
             # 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,
@@ -76,8 +93,8 @@ class CrunchyrollIE(InfoExtractor):
         self._login()
 
     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):
@@ -179,6 +196,16 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
 
         return output
 
+    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):
@@ -190,25 +217,11 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             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)
-            subtitles[lang_code] = [
-                {
-                    'ext': 'srt',
-                    'data': self._convert_subtitles_to_srt(sub_root),
-                },
-                {
-                    'ext': 'ass',
-                    'data': self._convert_subtitles_to_ass(sub_root),
-                },
-            ]
+            subtitles[lang_code] = self._extract_subtitles(subtitle)
         return subtitles
 
     def _real_extract(self, url):
@@ -242,7 +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)
 
-        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')
@@ -255,16 +268,17 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         for fmt in re.findall(r'showmedia\.([0-9]{3,4})p', webpage):
             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
+            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-Length', str(len(streamdata_req.data)))
             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,