wow_tree.py

import json
import sys
import time
import urllib2

nodes = ["307035796257525760"]
max_branches = 2
sleep = 1

nodes_done = []

while len(nodes) > 0:
  node = nodes.pop(0)

  try:
    # First get this node
    j = json.loads(urllib2.urlopen('https://api.twitter.com/1/statuses/show.json?id=%s&include_entities=true' % node).read())

    # Add the node this one points to
    points_to = j["entities"]["urls"][0]["expanded_url"]
    points_to = points_to.split('/')[-1]
    print '%s -> %s' % (node, points_to)
    if not points_to in nodes_done:
      nodes.append(points_to)

    # Then search for this URL
    j = json.loads(urllib2.urlopen('http://search.twitter.com/search.json?q=%s&rpp=%d&include_entities=true&result_type=mixed' % (node, max_branches)).read())

    # Add everything that links to it
    for result in j["results"]:
      points_from = result["id_str"]
      print '%s -> %s' % (points_from, node)
      if not points_from in nodes_done:
        nodes.append(points_from)
  except:
    sys.stderr.write('caught error with %s\n' % node)

  nodes_done.append(node)
  sys.stderr.write('%d done, %d in queue\n' % (len(nodes_done), len(nodes)))
  if len(nodes_done) > 1000:
    break
  time.sleep(1)