import cgi

import md5

import time

import datetime

import math

from xml.dom import minidom

import urllib

import sys



from google.appengine.api import users

from google.appengine.ext import webapp

from google.appengine.ext.webapp.util import run_wsgi_app

from google.appengine.ext import db

from google.appengine.api import urlfetch

from django.utils import simplejson as json





# Database Classes

class TumblrCache(db.Model):

    cache_id = db.StringProperty(required=True)

    content = db.StringProperty(multiline=True)

    date = db.DateTimeProperty(auto_now_add=True)

    

class Flush(webapp.RequestHandler):

  def get(self):

      self.response.headers['Content-Type'] = 'text/plain'

      w = self.response.out

      

      query = TumblrCache.all()

      for q in query:

          q.delete()

      

      w.write('FLUSHED!')

      



class MainPage(webapp.RequestHandler):

  def get(self):

      self.response.headers['Content-Type'] = 'text/plain'



class TagCloud(webapp.RequestHandler):

  def get(self):

      self.response.headers['Content-Type'] = 'text/plain'

      w = self.response.out

      # Thu thap cac bien can thiet

      tumblr_api = cgi.escape(self.request.get('url').encode('utf-8')) + '/api/read'

      tumblr_api = 'http://im.doquangtu.net/api/read'

      cache_id = md5.new(tumblr_api).hexdigest()

      # Neu da co cache, thi tra lai cache

      try:

          cache = TumblrCache.gql("WHERE cache_id = :1 LIMIT 1", cache_id)[0]

      except:

          cache = TumblrCache(cache_id=cache_id)

          

      content = cache.content

      if content == None:

          content = ''

      

      d = (((cache.date.toordinal()-719163)*24+cache.date.hour)*60+cache.date.minute)*60+cache.date.second

      d = time.time() - d

           

      if d > 24*60*60 or len(content) < 5:

          # Da qua 24h cache trong he thong

            form_fields = {

              "filter": "text"

            }

            

            loop = True

            start = 0

            taglist = {}

            



        

            try:

                while loop == True:

                    form_fields['start'] = int(start) * 20

                    form_data = urllib.urlencode(form_fields)

                    result = urlfetch.fetch(tumblr_api + '?filter=text&start=' + str(form_fields['start']))

                    result = result.content #.encode('utf-8')



                    xdom = minidom.parseString(result)

                    try:

                        posts = xdom.firstChild.getElementsByTagName("posts")[0].getElementsByTagName("post")

                        for post in posts:

                            post_id = post.getAttribute("id")

                            tags = post.getElementsByTagName("tag")

                            try:

                                for tag in tags:

                                    # Duyet qua tags

                                    t = "" . join(t.nodeValue for t in tag.childNodes if t.nodeType == t.TEXT_NODE)

                                    t = t.strip()

                                    #w.write(t)

                                    if t not in taglist:

                                        taglist[t] = 1

                                    else:

                                        taglist[t] = taglist[t] + 1

                            except:

                                # Loi & khong tim thay tags nao

                                i = 0

                    except:

                        # Loi & khong tim thay posts nao

                        i = 0

                    

                    # kiem tra xem co thoat vong lap hay chua

                    try:

                        posts = xdom.firstChild.getElementsByTagName("posts")[0].getAttribute("total")

                        posts = int(math.ceil(float(posts) / float(20)) - 1)

                        if start + 1 > posts:

                            loop = False

                        else:

                            start = start + 1

                            loop = True

                    except:

                        loop = False

            except:

                loop = False

                

            # Viet tags

            txt = json.JSONEncoder().encode( taglist )

            

            cache.content = txt

            cache.date = datetime.datetime.today()

            cache.put()            

            

            w.write(taglist)

      else:

          # Chua qua 24h cache trong he thong

          # Tra ve cache data 

          content = cache.content

          w.write( "var myJSONObject = " + content + ";" )



application = webapp.WSGIApplication(

                                     [('/', MainPage),

                                      ('/tagcloud', TagCloud),

                                      ('/flush', Flush)],

                                     debug=True)







def main():

  run_wsgi_app(application)



if __name__ == "__main__":

  main()

http://www.doquangtu.net

Tumblr taglist generator returns your Tumblog tags under JSON format. Written in Python and utilized Google App Engine.

Tumblr taglist generator

python

Python