gistfile1.py

from google.appengine.ext import webapp
from google.appengine.ext.webapp.util import run_wsgi_app
import logging
from google.appengine.api import urlfetch

from BeautifulSoup import *

class SoapLinks(webapp.RequestHandler):

    def get(self, url):

        url = url.replace('%3A', ':', 1)
        params = url.split('/')
        del params[-1]
        baseUrl = "/".join(params)
        baseUrl = baseUrl + "/"

        resultStr = ""

        response = urlfetch.fetch(url, method=urlfetch.GET)
        if response.status_code == 200:
            content = response.content

            resultList = list()
            #logging.debug(r"SoapLinks get %s" % content)
            self.response.headers['Content-Type'] = 'text/html;charset=utf-8'

            soaplinkstmp = BeautifulSoup(content, parseOnlyThese=SoupStrainer('dd'))
            for dda in soaplinkstmp:
                if dda.contents[0].find("_img") >= 0:
                    self.response.out.write("<a href='%s'>%s</a>" % (dda.contents[0], dda.contents[0]))
                    self.response.out.write("<br />")

                logging.debug(r"SoapLinks Tmp get %s", dda.contents[0])


            self.response.out.write("<br />")
            self.response.out.write("<br />")
            self.response.out.write("<br />")
            self.response.out.write("<br />")

            for dda in soaplinkstmp:
                if dda.contents[0].find("_img") >= 0:
                    thumb_url = dda.contents[0].replace('http://', '', 1)
                    self.response.out.write("<img src='http://cdn-cloud.appspot.com/a/%s?url=%s' />" % (thumb_url, dda.contents[0]))
                    self.response.out.write("<br />")



def main():
    application = webapp.WSGIApplication(
        [
        (r'/soaplinks/(.*)', SoapLinks)],
        debug=True)
    run_wsgi_app(application)

if __name__ == '__main__':
    main()