import json
import base64
import zlib
-import xml.etree.ElementTree
from hashlib import sha1
from math import pow, sqrt, floor
from .common import InfoExtractor
from ..compat import (
+ compat_etree_fromstring,
compat_urllib_parse,
compat_urllib_parse_unquote,
compat_urllib_request,
bytes_to_intlist,
intlist_to_bytes,
int_or_none,
+ lowercase_escape,
remove_end,
unified_strdate,
urlencode_postdata,
'id': '589804',
'ext': 'flv',
'title': 'Culture Japan Episode 1 – Rebuilding Japan after the 3.11',
- 'description': 'md5:fe2743efedb49d279552926d0bd0cd9e',
+ 'description': 'md5:2fbc01f90b87e8e9137296f37b461c12',
'thumbnail': 're:^https?://.*\.jpg$',
'uploader': 'Danny Choo Network',
'upload_date': '20120213',
return output
def _extract_subtitles(self, subtitle):
- sub_root = xml.etree.ElementTree.fromstring(subtitle)
+ sub_root = compat_etree_fromstring(subtitle)
return [{
'ext': 'srt',
'data': self._convert_subtitles_to_srt(sub_root),
def _get_subtitles(self, video_id, webpage):
subtitles = {}
- for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage):
+ 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)
if 'To view this, please log in to verify you are 18 or older.' in webpage:
self.raise_login_required()
- video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL)
+ video_title = self._html_search_regex(
+ r'(?s)<h1[^>]*>((?:(?!<h1).)*?<span[^>]+itemprop=["\']title["\'][^>]*>(?:(?!<h1).)+?)</h1>',
+ webpage, 'video_title')
video_title = re.sub(r' {2,}', ' ', video_title)
- video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='')
- if not video_description:
- video_description = None
+ video_description = self._html_search_regex(
+ r'<script[^>]*>\s*.+?\[media_id=%s\].+?"description"\s*:\s*"([^"]+)' % video_id,
+ webpage, 'description', default=None)
+ if video_description:
+ video_description = lowercase_escape(video_description.replace(r'\r\n', '\n'))
video_upload_date = self._html_search_regex(
[r'<div>Availability for free users:(.+?)</div>', r'<div>[^<>]+<span>\s*(.+?\d{4})\s*</span></div>'],
webpage, 'video_upload_date', fatal=False, flags=re.DOTALL)