[pornhd] Fix formats extraction (Closes #3898)

This commit is contained in:
Sergey M․ 2014-10-08 20:08:29 +07:00
parent b3826f6c8d
commit 842cca7d56

View File

@ -4,19 +4,27 @@ import re
import json import json
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import int_or_none from ..utils import (
int_or_none,
js_to_json,
qualities,
determine_ext,
)
class PornHdIE(InfoExtractor): class PornHdIE(InfoExtractor):
_VALID_URL = r'http://(?:www\.)?pornhd\.com/(?:[a-z]{2,4}/)?videos/(?P<id>\d+)' _VALID_URL = r'http://(?:www\.)?pornhd\.com/(?:[a-z]{2,4}/)?videos/(?P<id>\d+)(?:/(?P<display_id>.+))?'
_TEST = { _TEST = {
'url': 'http://www.pornhd.com/videos/1962/sierra-day-gets-his-cum-all-over-herself-hd-porn-video', 'url': 'http://www.pornhd.com/videos/1962/sierra-day-gets-his-cum-all-over-herself-hd-porn-video',
'md5': '956b8ca569f7f4d8ec563e2c41598441', 'md5': '956b8ca569f7f4d8ec563e2c41598441',
'info_dict': { 'info_dict': {
'id': '1962', 'id': '1962',
'display_id': 'sierra-day-gets-his-cum-all-over-herself-hd-porn-video',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Sierra loves doing laundry', 'title': 'Sierra loves doing laundry',
'description': 'md5:8ff0523848ac2b8f9b065ba781ccf294', 'description': 'md5:8ff0523848ac2b8f9b065ba781ccf294',
'thumbnail': 're:^https?://.*\.jpg',
'view_count': int,
'age_limit': 18, 'age_limit': 18,
} }
} }
@ -24,8 +32,9 @@ class PornHdIE(InfoExtractor):
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id') video_id = mobj.group('id')
display_id = mobj.group('display_id')
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, display_id or video_id)
title = self._html_search_regex( title = self._html_search_regex(
r'<title>(.+) porn HD.+?</title>', webpage, 'title') r'<title>(.+) porn HD.+?</title>', webpage, 'title')
@ -33,38 +42,21 @@ class PornHdIE(InfoExtractor):
r'<div class="description">([^<]+)</div>', webpage, 'description', fatal=False) r'<div class="description">([^<]+)</div>', webpage, 'description', fatal=False)
view_count = int_or_none(self._html_search_regex( view_count = int_or_none(self._html_search_regex(
r'(\d+) views\s*</span>', webpage, 'view count', fatal=False)) r'(\d+) views\s*</span>', webpage, 'view count', fatal=False))
thumbnail = self._search_regex(
r"'poster'\s*:\s*'([^']+)'", webpage, 'thumbnail', fatal=False)
videos = re.findall( quality = qualities(['SD', 'HD'])
r'var __video([\da-zA-Z]+?)(Low|High)StreamUrl = \'(http://.+?)\?noProxy=1\'', webpage) formats = [{
'url': source['file'],
mobj = re.search(r'flashVars = (?P<flashvars>{.+?});', webpage) 'format_id': '%s-%s' % (source['label'], determine_ext(source['file'])),
if mobj: 'quality': quality(source['label']),
flashvars = json.loads(mobj.group('flashvars')) } for source in json.loads(js_to_json(self._search_regex(
for key, quality in [('hashlink', 'low'), ('hd', 'high')]: r"(?s)'sources'\s*:\s*(\[.+?\])", webpage, 'sources')))]
redirect_url = flashvars.get(key)
if redirect_url:
videos.append(('flv', quality, redirect_url))
thumbnail = flashvars['urlWallpaper']
else:
thumbnail = self._og_search_thumbnail(webpage)
formats = []
for format_, quality, redirect_url in videos:
format_id = '%s-%s' % (format_.lower(), quality.lower())
video_url = self._download_webpage(
redirect_url, video_id, 'Downloading %s video link' % format_id, fatal=False)
if not video_url:
continue
formats.append({
'url': video_url,
'ext': format_.lower(),
'format_id': format_id,
'quality': 1 if quality.lower() == 'high' else 0,
})
self._sort_formats(formats) self._sort_formats(formats)
return { return {
'id': video_id, 'id': video_id,
'display_id': display_id,
'title': title, 'title': title,
'description': description, 'description': description,
'thumbnail': thumbnail, 'thumbnail': thumbnail,