mirror of
https://github.com/yt-dlp/yt-dlp.git
synced 2024-12-30 10:43:03 +01:00
73 lines
2.4 KiB
Python
73 lines
2.4 KiB
Python
from __future__ import unicode_literals
|
|
|
|
import re
|
|
import json
|
|
|
|
from .common import InfoExtractor
|
|
from ..utils import (
|
|
int_or_none,
|
|
js_to_json,
|
|
qualities,
|
|
)
|
|
|
|
|
|
class PornHdIE(InfoExtractor):
|
|
_VALID_URL = r'http://(?:www\.)?pornhd\.com/(?:[a-z]{2,4}/)?videos/(?P<id>\d+)(?:/(?P<display_id>.+))?'
|
|
_TEST = {
|
|
'url': 'http://www.pornhd.com/videos/1962/sierra-day-gets-his-cum-all-over-herself-hd-porn-video',
|
|
'md5': '956b8ca569f7f4d8ec563e2c41598441',
|
|
'info_dict': {
|
|
'id': '1962',
|
|
'display_id': 'sierra-day-gets-his-cum-all-over-herself-hd-porn-video',
|
|
'ext': 'mp4',
|
|
'title': 'Sierra loves doing laundry',
|
|
'description': 'md5:8ff0523848ac2b8f9b065ba781ccf294',
|
|
'thumbnail': 're:^https?://.*\.jpg',
|
|
'view_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 or video_id)
|
|
|
|
title = self._html_search_regex(
|
|
[r'<span[^>]+class=["\']video-name["\'][^>]*>([^<]+)',
|
|
r'<title>(.+?) - .*?[Pp]ornHD.*?</title>'], webpage, 'title')
|
|
description = self._html_search_regex(
|
|
r'<div class="description">([^<]+)</div>', webpage, 'description', fatal=False)
|
|
view_count = int_or_none(self._html_search_regex(
|
|
r'(\d+) views\s*</span>', webpage, 'view count', fatal=False))
|
|
thumbnail = self._search_regex(
|
|
r"'poster'\s*:\s*'([^']+)'", webpage, 'thumbnail', fatal=False)
|
|
|
|
quality = qualities(['sd', 'hd'])
|
|
sources = json.loads(js_to_json(self._search_regex(
|
|
r"(?s)'sources'\s*:\s*(\{.+?\})\s*\}[;,)]",
|
|
webpage, 'sources')))
|
|
formats = []
|
|
for qname, video_url in sources.items():
|
|
if not video_url:
|
|
continue
|
|
formats.append({
|
|
'url': video_url,
|
|
'format_id': qname,
|
|
'quality': quality(qname),
|
|
})
|
|
self._sort_formats(formats)
|
|
|
|
return {
|
|
'id': video_id,
|
|
'display_id': display_id,
|
|
'title': title,
|
|
'description': description,
|
|
'thumbnail': thumbnail,
|
|
'view_count': view_count,
|
|
'formats': formats,
|
|
'age_limit': 18,
|
|
}
|