[xhamster] Fix extraction (closes #14884)

This commit is contained in:
Sergey M․ 2017-12-02 19:03:24 +07:00
parent 0981585bef
commit fea92aa65d
No known key found for this signature in database
GPG Key ID: 2C393E0F18A9236D

View File

@ -6,10 +6,12 @@ from .common import InfoExtractor
from ..compat import compat_str from ..compat import compat_str
from ..utils import ( from ..utils import (
clean_html, clean_html,
determine_ext,
dict_get, dict_get,
ExtractorError, ExtractorError,
int_or_none, int_or_none,
parse_duration, parse_duration,
try_get,
unified_strdate, unified_strdate,
) )
@ -32,6 +34,7 @@ class XHamsterIE(InfoExtractor):
'display_id': 'femaleagent_shy_beauty_takes_the_bait', 'display_id': 'femaleagent_shy_beauty_takes_the_bait',
'ext': 'mp4', 'ext': 'mp4',
'title': 'FemaleAgent Shy beauty takes the bait', 'title': 'FemaleAgent Shy beauty takes the bait',
'timestamp': 1350194821,
'upload_date': '20121014', 'upload_date': '20121014',
'uploader': 'Ruseful2011', 'uploader': 'Ruseful2011',
'duration': 893, 'duration': 893,
@ -45,6 +48,7 @@ class XHamsterIE(InfoExtractor):
'display_id': 'britney_spears_sexy_booty', 'display_id': 'britney_spears_sexy_booty',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Britney Spears Sexy Booty', 'title': 'Britney Spears Sexy Booty',
'timestamp': 1379123460,
'upload_date': '20130914', 'upload_date': '20130914',
'uploader': 'jojo747400', 'uploader': 'jojo747400',
'duration': 200, 'duration': 200,
@ -61,6 +65,7 @@ class XHamsterIE(InfoExtractor):
'id': '5667973', 'id': '5667973',
'ext': 'mp4', 'ext': 'mp4',
'title': '....', 'title': '....',
'timestamp': 1454948101,
'upload_date': '20160208', 'upload_date': '20160208',
'uploader': 'parejafree', 'uploader': 'parejafree',
'duration': 72, 'duration': 72,
@ -96,6 +101,83 @@ class XHamsterIE(InfoExtractor):
if error: if error:
raise ExtractorError(error, expected=True) raise ExtractorError(error, expected=True)
age_limit = self._rta_search(webpage)
def get_height(s):
return int_or_none(self._search_regex(
r'^(\d+)[pP]', s, 'height', default=None))
initials = self._parse_json(
self._search_regex(
r'window\.initials\s*=\s*({.+?})\s*;\s*\n', webpage, 'initials',
default='{}'),
video_id, fatal=False)
if initials:
video = initials['videoModel']
title = video['title']
formats = []
for format_id, formats_dict in video['sources'].items():
if not isinstance(formats_dict, dict):
continue
for quality, format_item in formats_dict.items():
if format_id == 'download':
# Download link takes some time to be generated,
# skipping for now
continue
if not isinstance(format_item, dict):
continue
format_url = format_item.get('link')
filesize = int_or_none(
format_item.get('size'), invscale=1000000)
else:
format_url = format_item
filesize = None
if not isinstance(format_url, compat_str):
continue
formats.append({
'format_id': '%s-%s' % (format_id, quality),
'url': format_url,
'ext': determine_ext(format_url, 'mp4'),
'height': get_height(quality),
'filesize': filesize,
})
self._sort_formats(formats)
categories_list = video.get('categories')
if isinstance(categories_list, list):
categories = []
for c in categories_list:
if not isinstance(c, dict):
continue
c_name = c.get('name')
if isinstance(c_name, compat_str):
categories.append(c_name)
else:
categories = None
return {
'id': video_id,
'display_id': display_id,
'title': title,
'description': video.get('description'),
'timestamp': int_or_none(video.get('created')),
'uploader': try_get(
video, lambda x: x['author']['name'], compat_str),
'thumbnail': video.get('thumbURL'),
'duration': int_or_none(video.get('duration')),
'view_count': int_or_none(video.get('views')),
'like_count': int_or_none(try_get(
video, lambda x: x['rating']['likes'], int)),
'dislike_count': int_or_none(try_get(
video, lambda x: x['rating']['dislikes'], int)),
'comment_count': int_or_none(video.get('views')),
'age_limit': age_limit,
'categories': categories,
'formats': formats,
}
# Old layout fallback
title = self._html_search_regex( title = self._html_search_regex(
[r'<h1[^>]*>([^<]+)</h1>', [r'<h1[^>]*>([^<]+)</h1>',
r'<meta[^>]+itemprop=".*?caption.*?"[^>]+content="(.+?)"', r'<meta[^>]+itemprop=".*?caption.*?"[^>]+content="(.+?)"',
@ -119,8 +201,7 @@ class XHamsterIE(InfoExtractor):
formats.append({ formats.append({
'format_id': format_id, 'format_id': format_id,
'url': format_url, 'url': format_url,
'height': int_or_none(self._search_regex( 'height': get_height(format_id),
r'^(\d+)[pP]', format_id, 'height', default=None))
}) })
video_url = self._search_regex( video_url = self._search_regex(
@ -167,8 +248,6 @@ class XHamsterIE(InfoExtractor):
mobj = re.search(r'</label>Comments \((?P<commentcount>\d+)\)</div>', webpage) mobj = re.search(r'</label>Comments \((?P<commentcount>\d+)\)</div>', webpage)
comment_count = mobj.group('commentcount') if mobj else 0 comment_count = mobj.group('commentcount') if mobj else 0
age_limit = self._rta_search(webpage)
categories_html = self._search_regex( categories_html = self._search_regex(
r'(?s)<table.+?(<span>Categories:.+?)</table>', webpage, r'(?s)<table.+?(<span>Categories:.+?)</table>', webpage,
'categories', default=None) 'categories', default=None)