[camwithher] Improve extraction (Closes #8989)

This commit is contained in:
Sergey M․ 2016-04-01 23:47:27 +06:00
parent 04819db58e
commit 9aaaf8e8e8

View file

@ -1,55 +1,88 @@
from __future__ import unicode_literals from __future__ import unicode_literals
import re
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import (
int_or_none,
parse_duration,
unified_strdate,
)
class CamWithHerIE(InfoExtractor): class CamWithHerIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?camwithher\.tv/view_video\.php\?.*viewkey=(?P<id>\w+)' _VALID_URL = r'https?://(?:www\.)?camwithher\.tv/view_video\.php\?.*\bviewkey=(?P<id>\w+)'
_TESTS = [ _TESTS = [{
{
'url': 'http://camwithher.tv/view_video.php?viewkey=6e9a24e2c0e842e1f177&page=&viewtype=&category=', 'url': 'http://camwithher.tv/view_video.php?viewkey=6e9a24e2c0e842e1f177&page=&viewtype=&category=',
'info_dict': { 'info_dict': {
'id': '5644', 'id': '5644',
'ext': 'flv', 'ext': 'flv',
'title': 'Periscope Tease', 'title': 'Periscope Tease',
'description': 'In the clouds teasing on periscope to my favorite song',
'duration': 240,
'view_count': int,
'comment_count': int,
'uploader': 'MileenaK',
'upload_date': '20160322',
}, },
'params': { 'params': {
'skip_download': True, 'skip_download': True,
} }
}, }, {
{
'url': 'http://camwithher.tv/view_video.php?viewkey=6dfd8b7c97531a459937', 'url': 'http://camwithher.tv/view_video.php?viewkey=6dfd8b7c97531a459937',
'only_matching': True, 'only_matching': True,
}, }, {
{
'url': 'http://camwithher.tv/view_video.php?page=&viewkey=6e9a24e2c0e842e1f177&viewtype=&category=', 'url': 'http://camwithher.tv/view_video.php?page=&viewkey=6e9a24e2c0e842e1f177&viewtype=&category=',
'only_matching': True, 'only_matching': True,
}, }, {
{
'url': 'http://camwithher.tv/view_video.php?viewkey=b6c3b5bea9515d1a1fc4&page=&viewtype=&category=mv', 'url': 'http://camwithher.tv/view_video.php?viewkey=b6c3b5bea9515d1a1fc4&page=&viewtype=&category=mv',
'only_matching': True, 'only_matching': True,
} }]
]
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
flv_id = self._html_search_regex(r'<a href="/download/\?v=(\d+)', webpage, 'id') flv_id = self._html_search_regex(
r'<a[^>]+href=["\']/download/\?v=(\d+)', webpage, 'video id')
# The number "2010" was reverse-engineered from cwhplayer.swf. # Video URL construction algorithm is reverse-engineered from cwhplayer.swf
# It appears that they changed their video codec, and hence the RTMP URL rtmp_url = 'rtmp://camwithher.tv/clipshare/%s' % (
# scheme at that video's ID. ('mp4:%s.mp4' % flv_id) if int(flv_id) > 2010 else flv_id)
rtmp_url = 'rtmp://camwithher.tv/clipshare/%s' % (('mp4:%s.mp4' % flv_id) if int(flv_id) > 2010 else flv_id)
title = self._html_search_regex(
r'<div[^>]+style="float:left"[^>]*>\s*<h2>(.+?)</h2>', webpage, 'title')
description = self._html_search_regex(
r'>Description:</span>(.+?)</div>', webpage, 'description', default=None)
runtime = self._search_regex(
r'Runtime\s*:\s*(.+?) \|', webpage, 'duration', default=None)
if runtime:
runtime = re.sub(r'[\s-]', '', runtime)
duration = parse_duration(runtime)
view_count = int_or_none(self._search_regex(
r'Views\s*:\s*(\d+)', webpage, 'view count', default=None))
comment_count = int_or_none(self._search_regex(
r'Comments\s*:\s*(\d+)', webpage, 'comment count', default=None))
uploader = self._search_regex(
r'Added by\s*:\s*<a[^>]+>([^<]+)</a>', webpage, 'uploader', default=None)
upload_date = unified_strdate(self._search_regex(
r'Added on\s*:\s*([\d-]+)', webpage, 'upload date', default=None))
title = self._html_search_regex(r'<div style="float:left">\s+<h2>(.+?)</h2>', webpage, 'title')
return { return {
'id': flv_id, 'id': flv_id,
'url': rtmp_url, 'url': rtmp_url,
'no_resume': True,
'ext': 'flv', 'ext': 'flv',
'no_resume': True,
'title': title, 'title': title,
'description': description,
'duration': duration,
'view_count': view_count,
'comment_count': comment_count,
'uploader': uploader,
'upload_date': upload_date,
} }