Fix bug with parsing timestamps out.

Rory McCann · Rory McCann · commit e83c49bccaa0 · 2015-03-21T09:48:38.000+01:00
hacked it by presuming that %t (timestamps) are always wrapped in [ & ]. Tried to look at the httpd/apache source code, and I think it uses APR_TIME_T_FMT but couldn't see where that was defined. see #9
diff --git a/apache_log_parser/__init__.py b/apache_log_parser/__init__.py
@@ -107,7 +107,7 @@ def format_time(matched_strings):
     [make_regex('%r'), '.*?', lambda match: 'request_first_line', extra_request_from_first_line], #	First line of request
     [make_regex('%R'), '.*?', lambda match: 'handler', lambda matched_strings: matched_strings], #	The handler generating the response (if any).
     [make_regex('%s'), '[0-9]+?', lambda match: 'status', lambda matched_strings: matched_strings], #	Status. For requests that got internally redirected, this is the status of the *original* request --- %>s for the last.
-    [make_regex('%t'), '.*?', lambda match: 'time_received', format_time], #	Time the request was received (standard english format)
+    [make_regex('%t'), '\[.*?\]', lambda match: 'time_received', format_time], #	Time the request was received (standard english format)
     [make_regex('%\{[^\}]+?\}t'), '.*?', extract_inner_value("time_", "t") , lambda matched_strings: matched_strings], #	The time, in the form given by format, which should be in strftime(3) format. (potentially localized)
     [make_regex('%\{[^\}]+?\}x'), '.*?', extract_inner_value("extension_", "x") , lambda matched_strings: matched_strings], # Extension value, e.g. mod_ssl protocol and cipher
     [make_regex('%T'), '.*?', lambda match: 'time_s', lambda matched_strings: matched_strings], #	The time taken to serve the request, in seconds.
diff --git a/apache_log_parser/tests.py b/apache_log_parser/tests.py
@@ -52,5 +52,28 @@ def test_pr8(self):
             'env_unique_id': 'VHiZx6wQGCMAAEiBE8kAAAAA:VHiZx6wQGiMAAGPkBnMAAAAH:VHiZx6wQGiMAAGPkBnMAAAAH',
             'remote_ip': '192.168.1.100'})
 
+    def test_issue9(self):
+        parser = apache_log_parser.Parser("%h %v %V %l %u %t %r %>s %b %{Referer}i %{User-agent}i")
+        log = "10.1.1.1 T1 blah.foo.com - - [08/Mar/2015:18:06:58 -0400] GET /content_images/3/American-University-in-Cairo-AUC.jpeg.jpg HTTP/1.1 404 344 http://www.google.ie AppleWebKit/537.36 (KHTML, like Gecko) Chrome/41.0.2272.76 Safari/537.36"
+        data = parser.parse(log)
+        self.assertEqual(data, {
+            'status': '404',
+            'request_header_referer': 'http://www.google.ie',
+            'remote_user': '-',
+            'server_name': 'T1',
+            'request_http_ver': '1.1',
+            'request_header_user_agent': '',
+            'request_first_line': 'GET /content_images/3/American-University-in-Cairo-AUC.jpeg.jpg HTTP/1.1',
+            'remote_logname': '-',
+            'request_method': 'GET',
+            'time_received_datetimeobj': datetime.datetime(2015, 3, 8, 18, 6, 58),
+            'response_bytes_clf': '344',
+            'server_name2': 'blah.foo.com',
+            'request_url': '/content_images/3/American-University-in-Cairo-AUC.jpeg.jpg',
+            'time_received_isoformat': '2015-03-08T18:06:58',
+            'remote_host': '10.1.1.1',
+            'time_received': '[08/Mar/2015:18:06:58 -0400]'
+        })
+
 if __name__ == '__main__':
     unittest.main()
