python-oxweb/oxweb/youtube.py

113 lines
4.3 KiB
Python
Raw Normal View History

# -*- coding: utf-8 -*-
2008-06-19 09:47:02 +00:00
# vi:si:et:sw=4:sts=4:ts=4
2009-07-24 10:46:59 +00:00
from urllib import quote, unquote
2008-10-07 20:22:17 +00:00
import httplib
import xml.etree.ElementTree as ET
2008-09-30 13:58:21 +00:00
import re
2008-04-30 14:22:01 +00:00
import feedparser
2008-09-30 13:58:21 +00:00
from oxlib.cache import getUrl, getUrlUnicode
from oxlib import findString, findRe
2008-06-19 09:47:02 +00:00
2008-10-07 20:22:17 +00:00
def getVideoKey(youtubeId):
2008-12-09 18:25:35 +00:00
data = getUrl("http://www.youtube.com/watch?v=%s" % youtubeId)
match = re.compile(".*[?&]t=([^&]+).*").findall(data)
if match:
2009-07-24 10:46:59 +00:00
return unquote(match[0])
2008-10-07 20:22:17 +00:00
try:
conn = httplib.HTTPConnection ("www.youtube.com")
conn.request ("HEAD", '/v/' + youtubeId)
response = conn.getresponse ()
conn.close ()
except:
return False
if response.status >= 300 and response.status < 400:
location = response.getheader("location")
2008-12-07 14:39:39 +00:00
match = re.match(".*[?&]t=([^&]+)", location)
if match:
2009-07-24 10:46:59 +00:00
return unquote(match.groups()[0])
2008-12-07 14:39:39 +00:00
return False
2008-10-07 20:22:17 +00:00
def getVideoUrl(youtubeId, format='mp4'):
youtubeKey = getVideoKey(youtubeId)
2008-12-09 18:09:12 +00:00
if format == '720p':
fmt=22
url = "http://youtube.com/get_video.php?video_id=%s&t=%s&fmt=%s" % (youtubeId, youtubeKey, fmt)
elif format == 'mp4':
2008-06-19 09:47:02 +00:00
fmt=18
2008-10-07 20:22:17 +00:00
url = "http://youtube.com/get_video.php?video_id=%s&t=%s&fmt=%s" % (youtubeId, youtubeKey, fmt)
2008-06-19 09:47:02 +00:00
else:
2008-10-07 20:22:17 +00:00
url = "http://youtube.com/get_video.php?video_id=%s&t=%s" % (youtubeId, youtubeKey)
2008-06-19 09:47:02 +00:00
return url
2008-10-07 21:07:49 +00:00
def getMovieInfo(youtubeId, video_url_base=None):
url = "http://gdata.youtube.com/feeds/api/videos/%s" % youtubeId
2008-06-19 09:47:02 +00:00
data = getUrl(url)
fd = feedparser.parse(data)
2008-10-07 21:07:49 +00:00
return getInfoFromAtom(fd.entries[0], video_url_base)
2008-10-07 21:07:49 +00:00
def getInfoFromAtom(entry, video_url_base=None):
2008-06-19 09:47:02 +00:00
info = dict()
info['title'] = entry['title']
info['description'] = entry['description']
info['author'] = entry['author']
2008-10-07 21:07:49 +00:00
#info['published'] = entry['published_parsed']
if 'media_keywords' in entry:
info['keywords'] = entry['media_keywords'].split(', ')
2008-06-19 09:47:02 +00:00
info['url'] = entry['links'][0]['href']
info['id'] = findString(info['url'], "/watch?v=")
info['thumbnail'] = "http://img.youtube.com/vi/%s/0.jpg" % info['id']
2008-10-07 21:07:49 +00:00
if video_url_base:
info['flv'] = "%s/%s.%s" % (video_url_base, info['id'], 'flv')
info['mp4'] = "%s/%s.%s" % (video_url_base, info['id'], 'mp4')
else:
info['flv'] = getVideoUrl(info['id'], 'flv')
info['mp4'] = getVideoUrl(info['id'], 'mp4')
2008-12-09 18:09:12 +00:00
info['720p'] = getVideoUrl(info['id'], '720p')
2008-09-30 13:58:21 +00:00
info['embed'] = '<object width="425" height="355"><param name="movie" value="http://www.youtube.com/v/%s&hl=en"></param><param name="wmode" value="transparent"></param><embed src="http://www.youtube.com/v/%s&hl=en" type="application/x-shockwave-flash" wmode="transparent" width="425" height="355"></embed></object>' % (info['id'], info['id'])
2008-06-19 09:47:02 +00:00
return info
2008-10-07 21:07:49 +00:00
def find(query, max_results=10, offset=1, orderBy='relevance', video_url_base=None):
2008-06-19 09:47:02 +00:00
query = quote(query)
2008-10-07 21:07:49 +00:00
url = "http://gdata.youtube.com/feeds/api/videos?vq=%s&orderby=%s&start-index=%s&max-results=%s" % (query, orderBy, offset, max_results)
data = getUrlUnicode(url)
2008-06-19 09:47:02 +00:00
fd = feedparser.parse(data)
videos = []
for entry in fd.entries:
2008-10-07 21:07:49 +00:00
v = getInfoFromAtom(entry, video_url_base)
2008-06-19 09:47:02 +00:00
videos.append(v)
if len(videos) >= max_results:
return videos
return videos
2008-09-30 13:58:21 +00:00
2008-10-07 21:07:49 +00:00
'''
2008-09-30 13:58:21 +00:00
def find(query, max_results=10, offset=1, orderBy='relevance', video_url_base=None):
url = "http://youtube.com/results?search_query=%s&search=Search" % quote(query)
data = getUrlUnicode(url)
2008-10-07 21:07:49 +00:00
regx = re.compile(' <a href="/watch.v=(.*?)" title="(.*?)" ')
regx = re.compile('<a href="/watch\?v=(\w*?)" ><img src="(.*?)" class="vimg120" title="(.*?)" alt="video">')
2008-09-30 13:58:21 +00:00
id_title = regx.findall(data)
data_flat = data.replace('\n', ' ')
videos = {}
for video in id_title:
vid = video[0]
if vid not in videos:
v = dict()
v['id'] = vid
v['link'] = "http//youtube.com/watch.v=%s" % v['id']
v['title'] = video[2].strip()
if video_url_base:
v['video_link'] = "%s/%s" % (video_url_base, v['id'])
else:
2008-10-07 20:22:17 +00:00
v['video_url'] = getVideoUrl(v['id'])
2008-09-30 13:58:21 +00:00
v['description'] = findRe(data, 'BeginvidDesc%s">(.*?)</span>' % v['id']).strip().replace('<b>', ' ').replace('</b>', '')
v['thumbnail'] = video[1]
videos[vid] = v
if len(videos) >= max_results:
return videos.values()
return videos.values()
2008-10-07 21:07:49 +00:00
'''
2008-09-30 13:58:21 +00:00