Crawler backend

sitescripts/crawler/web/crawler.py

Index: sitescripts/crawler/web/crawler.py
===================================================================
--- a/sitescripts/crawler/web/crawler.py
+++ b/sitescripts/crawler/web/crawler.py
@@ -1,29 +1,109 @@
-import MySQLdb, os
+import MySQLdb, os, re, simplejson, sys
 from sitescripts.utils import cached, get_config
-from sitescripts.web import url_handler
-
-@url_handler('/crawlableUrls')
-def listUrls(environ, start_response):
-  urls = fetch_crawlable_urls()
-  start_response('200 OK', [('Content-Type', 'text/plain')])
-  return '\n'.join(urls)
-
-def fetch_crawlable_urls():
-  cursor = get_db().cursor(MySQLdb.cursors.DictCursor)
-  executeQuery(cursor, 'SELECT url from crawler_urls')
-  results = cursor.fetchall()
-  urls = [result['url'] for result in results]
-  return urls
+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':
-    return MySQLdb.connect(user=dbuser, passwd=dbpasswd, db=database, use_unicode=True, charset='utf8', named_pipe=True)
+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":
+    return MySQLdb.connect(user=dbuser, passwd=dbpasswd, db=database,
+                           use_unicode=True, charset="utf8", named_pipe=True)
   else:
-    return MySQLdb.connect(user=dbuser, passwd=dbpasswd, db=database, use_unicode=True, charset='utf8')
+    return MySQLdb.connect(user=dbuser, passwd=dbpasswd, db=database,
+                           use_unicode=True, charset="utf8")
 
-def executeQuery(cursor, query, args=None):
-  cursor.execute(query, args)
+def get_cursor():
+  return _get_db().cursor(MySQLdb.cursors.DictCursor)
+
+def _fetch_crawlable_sites():
+  cursor = get_cursor()
+  cursor.execute("SELECT url from crawler_sites")
+  results = cursor.fetchall()
+  sites = [result["url"] for result in results]
+  return sites
+
+@url_handler("/crawlableSites")
+@basic_auth

[Wladimir Palant, 2012/09/27 13:44:51]

Shouldn't this be @basic_auth(config_section="crawler")?

[Felix Dahlke, 2012/09/27 14:15:33]

Done.

+def crawlable_sites(environ, start_response):
+  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 _read_multipart_lines(environ, line_callback):
+  data_file = environ["wsgi.input"]
+  content_type = environ["CONTENT_TYPE"]
+  if not content_type:
+    raise ValueError("Content-Type missing from header")

[Wladimir Palant, 2012/09/27 13:44:51]

Python doesn't work like that - accessing a non-existing key will cause an
exception. So you better change that condition into "if not 'CONTENT_TYPE' in
environ:" and only access the key after the check. Alternatively you can use
environ.get("CONTENT_TYPE", "") and leave out the check here (an empty string
will trigger an error below).

And: yes, I am used to looking at the edge conditions when I write code rather
than leaving it up to testing.

[Felix Dahlke, 2012/09/27 14:15:33]

Done.
I'll work on it.

+
+  match = re.search(r"boundary=(.*)", content_type)
+  if not match:
+    raise ValueError("Multipart form data or boundary declaration missing")
+
+  boundary = match.group(1)
+  boundary_passed = False
+  header_passed = False
+
+  for line in data_file:
+    line = line.strip()
+
+    if not boundary_passed:
+      if line == "--" + boundary:
+        boundary_passed = True
+      continue
+
+    if not header_passed:
+      if not line:
+        header_passed = True
+      continue
+
+    if line == "--" + boundary + "--":
+      break
+
+    if line:
+      line_callback(line)
+
+  return True

[Wladimir Palant, 2012/09/27 13:44:51]

That return value is unused, left-over from earlier testing?

[Felix Dahlke, 2012/09/27 14:15:33]

Done.

+
+def _create_run():
+  cursor = get_cursor()
+  cursor.execute("INSERT INTO crawler_runs () VALUES ()")
+  return cursor.lastrowid
+
+def _insert_data(run_id, site, url, filtered):
+  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))
+
+@url_handler("/crawlerData")
+@basic_auth

[Wladimir Palant, 2012/09/27 13:44:51]

Shouldn't this be @basic_auth(config_section="crawler")?

[Felix Dahlke, 2012/09/27 14:15:33]

Done.

+def crawler_data(environ, start_response):
+  def line_callback(line):
+    try:
+      url, site, filtered = simplejson.loads(line)
+      _insert_data(run_id, site, url, filtered)
+    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")])
+    return e