aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl/extractor/crunchyroll.py
diff options
context:
space:
mode:
Diffstat (limited to 'youtube_dl/extractor/crunchyroll.py')
-rw-r--r--youtube_dl/extractor/crunchyroll.py156
1 files changed, 92 insertions, 64 deletions
diff --git a/youtube_dl/extractor/crunchyroll.py b/youtube_dl/extractor/crunchyroll.py
index 18ef3da10..b53f2d705 100644
--- a/youtube_dl/extractor/crunchyroll.py
+++ b/youtube_dl/extractor/crunchyroll.py
@@ -38,6 +38,16 @@ class CrunchyrollBaseIE(InfoExtractor):
_LOGIN_FORM = 'login_form'
_NETRC_MACHINE = 'crunchyroll'
+ def _call_rpc_api(self, method, video_id, note=None, data=None):
+ data = data or {}
+ data['req'] = 'RpcApi' + method
+ data = compat_urllib_parse_urlencode(data).encode('utf-8')
+ return self._download_xml(
+ 'http://www.crunchyroll.com/xml/',
+ video_id, note, fatal=False, data=data, headers={
+ 'Content-Type': 'application/x-www-form-urlencoded',
+ })
+
def _login(self):
(username, password) = self._get_login_info()
if username is None:
@@ -377,15 +387,19 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
def _get_subtitles(self, video_id, webpage):
subtitles = {}
for sub_id, sub_name in re.findall(r'\bssid=([0-9]+)"[^>]+?\btitle="([^"]+)', 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:
+ sub_doc = self._call_rpc_api(
+ 'Subtitle_GetXml', video_id,
+ 'Downloading subtitles for ' + sub_name, data={
+ 'subtitle_script_id': sub_id,
+ })
+ if not sub_doc:
continue
- subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8')
+ sid = sub_doc.get('id')
+ iv = xpath_text(sub_doc, 'iv', 'subtitle iv')
+ data = xpath_text(sub_doc, 'data', 'subtitle data')
+ if not sid or not iv or not data:
+ continue
+ subtitle = self._decrypt_subtitles(data, iv, sid).decode('utf-8')
lang_code = self._search_regex(r'lang_code=["\']([^"\']+)', subtitle, 'subtitle_lang_code', fatal=False)
if not lang_code:
continue
@@ -456,65 +470,79 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
for fmt in available_fmts:
stream_quality, stream_format = self._FORMAT_IDS[fmt]
video_format = fmt + 'p'
- streamdata_req = sanitized_Request(
- 'http://www.crunchyroll.com/xml/?req=RpcApiVideoPlayer_GetStandardConfig&media_id=%s&video_format=%s&video_quality=%s'
- % (video_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 = self._download_xml(
- streamdata_req, video_id,
- note='Downloading media info for %s' % video_format)
- stream_info = streamdata.find('./{default}preload/stream_info')
- video_encode_id = xpath_text(stream_info, './video_encode_id')
- if video_encode_id in video_encode_ids:
- continue
- video_encode_ids.append(video_encode_id)
-
- video_file = xpath_text(stream_info, './file')
- if not video_file:
- continue
- if video_file.startswith('http'):
- formats.extend(self._extract_m3u8_formats(
- video_file, video_id, 'mp4', entry_protocol='m3u8_native',
- m3u8_id='hls', fatal=False))
- continue
-
- video_url = xpath_text(stream_info, './host')
- if not video_url:
- continue
- metadata = stream_info.find('./metadata')
- format_info = {
- 'format': video_format,
- 'format_id': video_format,
- 'height': int_or_none(xpath_text(metadata, './height')),
- 'width': int_or_none(xpath_text(metadata, './width')),
- }
-
- if '.fplive.net/' in video_url:
- video_url = re.sub(r'^rtmpe?://', 'http://', video_url.strip())
- parsed_video_url = compat_urlparse.urlparse(video_url)
- direct_video_url = compat_urlparse.urlunparse(parsed_video_url._replace(
- netloc='v.lvlt.crcdn.net',
- path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_file.split(':')[-1])))
- if self._is_valid_url(direct_video_url, video_id, video_format):
- format_info.update({
- 'url': direct_video_url,
- })
- formats.append(format_info)
+ stream_infos = []
+ streamdata = self._call_rpc_api(
+ 'VideoPlayer_GetStandardConfig', video_id,
+ 'Downloading media info for %s' % video_format, data={
+ 'media_id': video_id,
+ 'video_format': stream_format,
+ 'video_quality': stream_quality,
+ 'current_page': url,
+ })
+ if streamdata:
+ stream_info = streamdata.find('./{default}preload/stream_info')
+ if stream_info:
+ stream_infos.append(stream_info)
+ stream_info = self._call_rpc_api(
+ 'VideoEncode_GetStreamInfo', video_id,
+ 'Downloading stream info for %s' % video_format, data={
+ 'media_id': video_id,
+ 'video_format': stream_format,
+ 'video_encode_quality': stream_quality,
+ })
+ if stream_info:
+ stream_infos.append(stream_info)
+ for stream_info in stream_infos:
+ video_encode_id = xpath_text(stream_info, './video_encode_id')
+ if video_encode_id in video_encode_ids:
continue
+ video_encode_ids.append(video_encode_id)
- format_info.update({
- 'url': video_url,
- 'play_path': video_file,
- 'ext': 'flv',
- })
- formats.append(format_info)
- self._sort_formats(formats)
+ video_file = xpath_text(stream_info, './file')
+ if not video_file:
+ continue
+ if video_file.startswith('http'):
+ formats.extend(self._extract_m3u8_formats(
+ video_file, video_id, 'mp4', entry_protocol='m3u8_native',
+ m3u8_id='hls', fatal=False))
+ continue
- metadata = self._download_xml(
- 'http://www.crunchyroll.com/xml', video_id,
- note='Downloading media info', query={
- 'req': 'RpcApiVideoPlayer_GetMediaMetadata',
+ video_url = xpath_text(stream_info, './host')
+ if not video_url:
+ continue
+ metadata = stream_info.find('./metadata')
+ format_info = {
+ 'format': video_format,
+ 'height': int_or_none(xpath_text(metadata, './height')),
+ 'width': int_or_none(xpath_text(metadata, './width')),
+ }
+
+ if '.fplive.net/' in video_url:
+ video_url = re.sub(r'^rtmpe?://', 'http://', video_url.strip())
+ parsed_video_url = compat_urlparse.urlparse(video_url)
+ direct_video_url = compat_urlparse.urlunparse(parsed_video_url._replace(
+ netloc='v.lvlt.crcdn.net',
+ path='%s/%s' % (remove_end(parsed_video_url.path, '/'), video_file.split(':')[-1])))
+ if self._is_valid_url(direct_video_url, video_id, video_format):
+ format_info.update({
+ 'format_id': 'http-' + video_format,
+ 'url': direct_video_url,
+ })
+ formats.append(format_info)
+ continue
+
+ format_info.update({
+ 'format_id': 'rtmp-' + video_format,
+ 'url': video_url,
+ 'play_path': video_file,
+ 'ext': 'flv',
+ })
+ formats.append(format_info)
+ self._sort_formats(formats, ('height', 'width', 'tbr', 'fps'))
+
+ metadata = self._call_rpc_api(
+ 'VideoPlayer_GetMediaMetadata', video_id,
+ note='Downloading media info', data={
'media_id': video_id,
})