[spankwire] Fix extraction and modernize

This commit is contained in:
Sergey M․ 2014-09-29 20:11:51 +07:00
parent 59c03a9bfb
commit fdb4d278bf
1 changed files with 29 additions and 19 deletions

View File

@ -9,7 +9,6 @@ from ..utils import (
compat_urllib_parse, compat_urllib_parse,
unified_strdate, unified_strdate,
str_to_int, str_to_int,
int_or_none,
) )
from ..aes import aes_decrypt_text from ..aes import aes_decrypt_text
@ -40,31 +39,42 @@ class SpankwireIE(InfoExtractor):
req.add_header('Cookie', 'age_verified=1') req.add_header('Cookie', 'age_verified=1')
webpage = self._download_webpage(req, video_id) webpage = self._download_webpage(req, video_id)
title = self._html_search_regex(r'<h1>([^<]+)', webpage, 'title') title = self._html_search_regex(
r'<h1>([^<]+)', webpage, 'title')
description = self._html_search_regex( description = self._html_search_regex(
r'<div\s+id="descriptionContent">([^<]+)<', webpage, 'description', fatal=False) r'<div\s+id="descriptionContent">([^<]+)<',
webpage, 'description', fatal=False)
thumbnail = self._html_search_regex( thumbnail = self._html_search_regex(
r'flashvars\.image_url = "([^"]+)', webpage, 'thumbnail', fatal=False) r'playerData\.screenShot\s*=\s*"([^"]+)"',
webpage, 'thumbnail', fatal=False)
uploader = self._html_search_regex( uploader = self._html_search_regex(
r'by:\s*<a [^>]*>(.+?)</a>', webpage, 'uploader', fatal=False) r'by:\s*<a [^>]*>(.+?)</a>',
webpage, 'uploader', fatal=False)
uploader_id = self._html_search_regex( uploader_id = self._html_search_regex(
r'by:\s*<a href="/Profile\.aspx\?.*?UserId=(\d+).*?"', webpage, 'uploader id', fatal=False) r'by:\s*<a href="/Profile\.aspx\?.*?UserId=(\d+).*?"',
upload_date = self._html_search_regex(r'</a> on (.+?) at \d+:\d+', webpage, 'upload date', fatal=False) webpage, 'uploader id', fatal=False)
if upload_date: upload_date = unified_strdate(self._html_search_regex(
upload_date = unified_strdate(upload_date) r'</a> on (.+?) at \d+:\d+',
webpage, 'upload date', fatal=False))
view_count = self._html_search_regex( view_count = str_to_int(self._html_search_regex(
r'<div id="viewsCounter"><span>([^<]+)</span> views</div>', webpage, 'view count', fatal=False) r'<div id="viewsCounter"><span>([\d,\.]+)</span> views</div>',
if view_count: webpage, 'view count', fatal=False))
view_count = str_to_int(view_count) comment_count = str_to_int(self._html_search_regex(
comment_count = int_or_none(self._html_search_regex( r'Comments<span[^>]+>\s*\(([\d,\.]+)\)</span>',
r'<span id="spCommentCount">\s*(\d+)</span> Comments</div>', webpage, 'comment count', fatal=False)) webpage, 'comment count', fatal=False))
video_urls = list(map(compat_urllib_parse.unquote , re.findall(r'flashvars\.quality_[0-9]{3}p = "([^"]+)', webpage))) video_urls = list(map(
compat_urllib_parse.unquote,
re.findall(r'playerData\.cdnPath[0-9]{3,}\s*=\s*"([^"]+)', webpage)))
if webpage.find('flashvars\.encrypted = "true"') != -1: if webpage.find('flashvars\.encrypted = "true"') != -1:
password = self._html_search_regex(r'flashvars\.video_title = "([^"]+)', webpage, 'password').replace('+', ' ') password = self._html_search_regex(
video_urls = list(map(lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'), video_urls)) r'flashvars\.video_title = "([^"]+)',
webpage, 'password').replace('+', ' ')
video_urls = list(map(
lambda s: aes_decrypt_text(s, password, 32).decode('utf-8'),
video_urls))
formats = [] formats = []
for video_url in video_urls: for video_url in video_urls: