sitescripts: Collect unmatched filters

sitescripts/crawler/web/crawler.py

 import MySQLdb, os, re, simplejson, sys
 from sitescripts.utils import cached, get_config
 from sitescripts.web import url_handler, basic_auth
 
 @cached(600)
 def _get_db():
   database = get_config().get("crawler", "database")
   dbuser = get_config().get("crawler", "dbuser")
   dbpasswd = get_config().get("crawler", "dbpassword")
   if os.name == "nt":
@@ skipping 19 matching lines @@
   urls = _fetch_crawlable_sites()
   start_response("200 OK", [("Content-Type", "text/plain")])
   return "\n".join(urls)
 
 def _find_site_id(site_url):
   cursor = _get_cursor()
   cursor.execute("SELECT id FROM crawler_sites WHERE url = %s", site_url)
   result = cursor.fetchone()
   return result["id"] if result else None
 
+def _find_filter_id(filter_str):
+  if filter_str is None:
+    return None
+
+  cursor = _get_cursor()
+  query = "SELECT id FROM crawler_filters WHERE filter_hash = sha1(%s)"
+  cursor.execute(query, filter_str)
+  result = cursor.fetchone()
+  return result["id"] if result else None
+
 def _read_multipart_lines(environ, line_callback):
   data_file = environ["wsgi.input"]
   content_type = environ.get("CONTENT_TYPE")
   if not content_type:
     raise ValueError("Content-Type missing from header")
 
   match = re.search(r"boundary=(.*)", content_type)
   if not match:
     raise ValueError("Multipart form data or boundary declaration missing")
 
@@ skipping 18 matching lines @@
       break
 
     if line:
       line_callback(line)
 
 def _create_run():
   cursor = _get_cursor()
   cursor.execute("INSERT INTO crawler_runs () VALUES ()")
   return cursor.lastrowid
 
-def _insert_data(run_id, site, url, filtered):
+def _insert_data(run_id, site, url, filtered, filter_str):
   site_id = _find_site_id(site)
   if site_id is None:
     print >>sys.stderr, "Unable to find site '%s' in the database" % site
     return
 
   cursor = _get_cursor()
-  cursor.execute("""
-INSERT INTO crawler_data (run, site, url, filtered)
-VALUES (%s, %s, %s, %s)""",
-                 (run_id, site_id, url, filtered))
+  insert = """
+INSERT INTO crawler_requests (run, site, url, filtered)
+VALUES (%s, %s, %s, %s)"""
+  cursor.execute(insert, (run_id, site_id, url, filtered))
 
-@url_handler("/crawlerData")
+  filter_id = _find_filter_id(filter_str)
+  if filter_id is not None:
+    update = "UPDATE crawler_requests SET filter = %s WHERE id = %s";
+    cursor.execute(update, (filter_id, cursor.lastrowid))
+
+@url_handler("/crawlerRequests")
 @basic_auth("crawler")
-def crawler_data(environ, start_response):
+def crawler_requests(environ, start_response):
   def line_callback(line):
     try:
-      url, site, filtered = simplejson.loads(line)
-      _insert_data(run_id, site, url, filtered)
+      data = simplejson.loads(line)
+      if len(data) < 3:
+        print >>sys.stderr, "Not enough elements in line '%s'" % line
+        return
+      url = data[0]
+      site = data[1]
+      filtered = data[2]
+      filter_str = data[3] if len(data) >= 4 else None
+      _insert_data(run_id, site, url, filtered, filter_str)
     except simplejson.JSONDecodeError:
       print >>sys.stderr, "Unable to parse JSON from '%s'" % line
 
   run_id = _create_run()
   try:
     _read_multipart_lines(environ, line_callback)
     start_response("200 OK", [("Content-Type", "text/plain")])
     return ""
   except ValueError as e:
     start_response("401 Bad Request", [("Content-Type", "text/plain")])
+    print >>sys.stderr, "Unable to read multipart data: %s" % e
     return e