python-oxweb/oxweb/youtube.py

# -*- coding: utf-8 -*-
# vi:si:et:sw=4:sts=4:ts=4
from urllib import quote
import xml.etree.ElementTree as ET
import re

import feedparser
from oxlib.cache import getUrl, getUrlUnicode
from oxlib import findString, findRe


def getVideoUrl(youtubeId, format='mp4'):
    url = 'http://www.youtube.com/api2_rest?method=youtube.videos.get_video_token&video_id=' + youtubeId
    data = getUrl(url)
    xml = ET.fromstring(data)
    youtubeKey = xml.find('t').text
    if format == 'mp4':
        fmt=18
        url = "http://youtube.com/get_video.php?video_id=%s&t=%s&fmt=%s"%(youtubeId, youtubeKey, fmt)
    else:
        url = "http://youtube.com/get_video.php?video_id=%s&t=%s"%(youtubeId, youtubeKey)
    return url

def getMovieInfo(youtubeId):
    url = "http://gdata.youtube.com/feeds/api/videos/%s " % youtubeId
    data = getUrl(url)
    fd = feedparser.parse(data)
    return getInfoFromAtom(fd.entries[0])

'''
def getInfoFromAtom(entry):
    info = dict()
    info['title'] = entry['title']
    info['description'] = entry['description']
    info['author'] = entry['author']
    info['published'] = entry['published_parsed']
    info['keywords'] = entry['media_keywords'].split(', ')
    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']
    info['flv'] = getVideoUrl(info['id'], 'flv')
    info['mp4'] = getVideoUrl(info['id'], 'mp4')
    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'])
    return info

def find(query, max_results=10, offset=1, orderBy='relevance'):
    query = quote(query)
    url = "http://gdata.youtube.com/feeds/api/videos?vq=%s&orderby=%s&start-index=%s&max-results=%s"%(query, orderBy, offset, max_results)
    data = getUrl(url)
    fd = feedparser.parse(data)
    videos = []
    for entry in fd.entries:
        v = getInfoFromAtom(entry)
        videos.append(v)
        if len(videos) >= max_results:
            return videos
    return videos
'''

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)
  regx = re.compile(''' <a href="/watch.v=(.*?)" title="(.*?)" ''')
  regx = re.compile('''<a href="/watch\?v=(\w*?)" ><img src="(.*?)"  class="vimg120" title="(.*?)" alt="video">''')
  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:
        v['video_url'] = get_video_url(v['id'])
      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()