mirror of
https://github.com/ytdl-org/youtube-dl.git
synced 2024-11-10 19:37:25 +00:00
Merge branch 'peugeot-xxxymovies'
This commit is contained in:
commit
3e2bcf530b
@ -515,6 +515,7 @@ from .xminus import XMinusIE
|
||||
from .xnxx import XNXXIE
|
||||
from .xvideos import XVideosIE
|
||||
from .xtube import XTubeUserIE, XTubeIE
|
||||
from .xxxymovies import XXXYMoviesIE
|
||||
from .yahoo import (
|
||||
YahooIE,
|
||||
YahooSearchIE,
|
||||
|
81
youtube_dl/extractor/xxxymovies.py
Normal file
81
youtube_dl/extractor/xxxymovies.py
Normal file
@ -0,0 +1,81 @@
|
||||
from __future__ import unicode_literals
|
||||
|
||||
import re
|
||||
|
||||
from .common import InfoExtractor
|
||||
from ..utils import (
|
||||
parse_duration,
|
||||
int_or_none,
|
||||
)
|
||||
|
||||
|
||||
class XXXYMoviesIE(InfoExtractor):
|
||||
_VALID_URL = r'https?://(?:www\.)?xxxymovies\.com/videos/(?P<id>\d+)/(?P<display_id>[^/]+)'
|
||||
_TEST = {
|
||||
'url': 'http://xxxymovies.com/videos/138669/ecstatic-orgasm-sofcore/',
|
||||
'md5': '810b1bdbbffff89dd13bdb369fe7be4b',
|
||||
'info_dict': {
|
||||
'id': '138669',
|
||||
'display_id': 'ecstatic-orgasm-sofcore',
|
||||
'ext': 'mp4',
|
||||
'title': 'Ecstatic Orgasm Sofcore',
|
||||
'duration': 931,
|
||||
'categories': list,
|
||||
'view_count': int,
|
||||
'like_count': int,
|
||||
'dislike_count': int,
|
||||
'age_limit': 18,
|
||||
}
|
||||
}
|
||||
|
||||
def _real_extract(self, url):
|
||||
mobj = re.match(self._VALID_URL, url)
|
||||
video_id = mobj.group('id')
|
||||
display_id = mobj.group('display_id')
|
||||
|
||||
webpage = self._download_webpage(url, display_id)
|
||||
|
||||
video_url = self._search_regex(
|
||||
r"video_url\s*:\s*'([^']+)'", webpage, 'video URL')
|
||||
|
||||
title = self._html_search_regex(
|
||||
[r'<div class="block_header">\s*<h1>([^<]+)</h1>',
|
||||
r'<title>(.*?)\s*-\s*XXXYMovies\.com</title>'],
|
||||
webpage, 'title')
|
||||
|
||||
thumbnail = self._search_regex(
|
||||
r"preview_url\s*:\s*'([^']+)'",
|
||||
webpage, 'thumbnail', fatal=False)
|
||||
|
||||
categories = self._html_search_meta(
|
||||
'keywords', webpage, 'categories', default='').split(',')
|
||||
|
||||
duration = parse_duration(self._search_regex(
|
||||
r'<span>Duration:</span>\s*(\d+:\d+)',
|
||||
webpage, 'duration', fatal=False))
|
||||
|
||||
view_count = int_or_none(self._html_search_regex(
|
||||
r'<div class="video_views">\s*(\d+)',
|
||||
webpage, 'view count', fatal=False))
|
||||
like_count = int_or_none(self._search_regex(
|
||||
r'>\s*Likes? <b>\((\d+)\)',
|
||||
webpage, 'like count', fatal=False))
|
||||
dislike_count = int_or_none(self._search_regex(
|
||||
r'>\s*Dislike <b>\((\d+)\)</b>',
|
||||
webpage, 'dislike count', fatal=False))
|
||||
|
||||
age_limit = self._rta_search(webpage)
|
||||
|
||||
return {
|
||||
'id': video_id,
|
||||
'display_id': display_id,
|
||||
'url': video_url,
|
||||
'title': title,
|
||||
'thumbnail': thumbnail,
|
||||
'categories': categories,
|
||||
'duration': duration,
|
||||
'view_count': view_count,
|
||||
'like_count': like_count,
|
||||
'dislike_count': dislike_count,
|
||||
'age_limit': age_limit,
|
||||
}
|
Loading…
Reference in New Issue
Block a user