diff -r 078085becbf8 -r 5c757e167687 script/utils/search_topsy_scrap.py
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/script/utils/search_topsy_scrap.py	Fri Sep 12 13:03:29 2014 +0200
@@ -0,0 +1,211 @@
+import argparse
+import logging
+import math
+import re
+import time
+import urllib
+
+from blessings import Terminal
+import requests
+import twitter
+
+from iri_tweet import models, utils
+from iri_tweet.processor import TwitterProcessorStatus
+
+from selenium import webdriver
+from selenium.webdriver.common.desired_capabilities import DesiredCapabilities
+from selenium.webdriver.common.by import By
+from selenium.webdriver.support.ui import WebDriverWait
+from selenium.webdriver.support import expected_conditions as EC
+
+from lxml import html
+import json
+
+logger = logging.getLogger(__name__)
+
+APPLICATION_NAME = "Tweet recorder user"
+
+dcap = dict(DesiredCapabilities.PHANTOMJS)
+dcap["phantomjs.page.settings.userAgent"] = "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_9_3) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/36.0.1985.103 Safari/537.36"
+
+class TopsyResource(object):
+
+    def __init__(self, query, **kwargs):
+
+        self.options = {}
+        self.options['q'] = query
+        self.options.update(kwargs)
+        self.base_url = "http://topsy.com/s"
+        self.driver = webdriver.PhantomJS(desired_capabilities=dcap)
+        self.driver.set_window_size(1024, 768)
+        self.page = -1
+        self.tree = None
+
+
+    def __do_request(self, params):
+      url = "%s?%s" % (self.base_url, urllib.urlencode(params).replace('+','%20')) #calculate url with urllib
+      print('Requesting %s' % url)
+      self.driver.get(url)
+      try:
+          element = WebDriverWait(self.driver, 60).until(
+              EC.presence_of_element_located((By.CLASS_NAME, "result-tweet"))
+          )
+      except Exception as e:
+        print('Exception requesting %s : %s' % (url, e))
+        self.tree = None
+      else:
+        self.tree = html.fromstring(self.driver.page_source)
+
+    def __check_last(self):
+      if self.page < 0:
+          return False
+      if self.tree is None or len(self.tree.xpath("//*[@id=\"module-pager\"]/div/ul/li[@data-page=\"next\"and @class=\"disabled\"]")):
+          return True
+      else:
+          return False
+
+
+    def __next_page(self):
+        if self.__check_last():
+          return False
+        self.page += 1
+        params = {}
+        params.update(self.options)
+        if self.page:
+          params['offset'] = self.page*self.options.get('perpage',10)
+        self.__do_request(params)
+        return self.tree is not None
+
+    def __iter__(self):
+        result_xpath = "//*[@id=\"results\"]/div"
+        while self.__next_page():
+            for res_node in self.tree.xpath(result_xpath):
+                res_obj = {
+                  'user': "".join(res_node.xpath("./div/div/h5/a/text()")),
+                  'content': "".join(res_node.xpath("./div/div/div/text()")),
+                  'url': "".join(res_node.xpath("./div/div/ul/li[1]/small/a/@href"))
+                }
+                if res_obj['url']:
+                  yield res_obj
+
+
+def get_options():
+
+    usage = "usage: %(prog)s [options] <connection_str_or_filepath>"
+
+    parser = argparse.ArgumentParser(usage=usage)
+
+    parser.add_argument(dest="conn_str",
+                        help="write tweet to DATABASE. This is a connection string", metavar="CONNECTION_STR")
+    parser.add_argument("-Q", dest="query",
+                      help="query", metavar="QUERY")
+    parser.add_argument("-k", "--key", dest="consumer_key",
+                        help="Twitter consumer key", metavar="CONSUMER_KEY")
+    parser.add_argument("-s", "--secret", dest="consumer_secret",
+                        help="Twitter consumer secret", metavar="CONSUMER_SECRET")
+    parser.add_argument("-t", dest="token_filename", metavar="TOKEN_FILENAME", default=".oauth_token",
+                      help="Token file name")
+    parser.add_argument("-T", dest="topsy_apikey", metavar="TOPSY_APIKEY", default=None,
+                      help="Topsy apikey")
+
+    utils.set_logging_options(parser)
+
+    return parser.parse_args()
+
+
+
+if __name__ == "__main__":
+
+    options = get_options()
+
+    utils.set_logging(options);
+
+
+    acess_token_key, access_token_secret = utils.get_oauth_token(consumer_key=options.consumer_key, consumer_secret=options.consumer_secret, token_file_path=options.token_filename, application_name=APPLICATION_NAME)
+
+    t = twitter.Twitter(domain="api.twitter.com", auth=twitter.OAuth(acess_token_key, access_token_secret, options.consumer_key, options.consumer_secret), secure=True)
+    t.secure = True
+
+    conn_str = options.conn_str.strip()
+    if not re.match("^\w+://.+", conn_str):
+        conn_str = 'sqlite:///' + conn_str
+
+    engine, metadata, Session = models.setup_database(conn_str, echo=((options.verbose-options.quiet)>0), create_all=True)
+    session = None
+
+
+    topsy_parameters = {
+        'perpage': 10,
+        'window': 'a',
+        'type': 'tweet',
+        'hidden': 1,
+        'sort': 'date'
+    }
+
+    term = Terminal()
+
+    try:
+        session = Session()
+
+        results = None
+        page = 1
+        print options.query
+
+        tr = TopsyResource(options.query, **topsy_parameters)
+
+        move_up = 0
+
+        for i,item in enumerate(tr):
+            # get id
+            url = item.get("url")
+            tweet_id = url.split("/")[-1]
+
+            if move_up > 0:
+                print((move_up+1)*term.move_up())
+                move_up = 0
+
+            print ("%d: %s - %r" % (i+1, tweet_id, item.get("content") ) + term.clear_eol())
+            move_up += 1
+
+            count_tweet = session.query(models.Tweet).filter_by(id_str=tweet_id).count()
+
+            if count_tweet:
+                continue
+            try:
+                tweet = t.statuses.show(id=tweet_id, include_entities=True)
+            except twitter.api.TwitterHTTPError as e:
+                if e.e.code == 404 or e.e.code == 403:
+                    continue
+                else:
+                    raise
+
+            processor = TwitterProcessorStatus(tweet, None, None, session, None, options.token_filename, logger)
+            processor.process()
+            session.flush()
+            session.commit()
+
+            print("rate limit remaining %s of %s" % (str(tweet.rate_limit_remaining), str(tweet.headers.getheader('X-Rate-Limit-Limit'))) + term.clear_eol())
+            move_up += 1
+            rate_limit_limit = int(tweet.headers.getheader('X-Rate-Limit-Limit'))
+            rate_limit_remaining = int(tweet.rate_limit_remaining)
+
+            if rate_limit_remaining < rate_limit_limit:
+                time_to_sleep = 0
+            else:
+                time_to_sleep = int(math.ceil((tweet.rate_limit_reset - time.mktime(time.gmtime())) / tweet.rate_limit_remaining))
+
+            for i in xrange(time_to_sleep):
+                if i:
+                    print(2*term.move_up())
+                else:
+                    move_up += 1
+                print(("Sleeping for %d seconds, %d remaining" % (time_to_sleep, time_to_sleep-i)) + term.clear_eol())
+                time.sleep(1)
+
+    except twitter.api.TwitterHTTPError as e:
+        fmt = ("." + e.format) if e.format else ""
+        print "Twitter sent status %s for URL: %s%s using parameters: (%s)\ndetails: %s" % (repr(e.e.code), repr(e.uri), repr(fmt), repr(e.uriparts), repr(e.response_data))
+
+    finally:
+        if session:
+            session.close()