Merge ca13b48 into ad76cf9

CHANGES.rst

@@ -10,6 +10,9 @@ CHANGES
 - Add `--json` output command line option.
   [valleedelisle]
 
+- Add `cmd_requests_per_hour`
+  [valleedelisle]
+
 3.0.0 (2019-06-10)
 ------------------
 

README.rst

@@ -149,7 +149,11 @@ See the ``--help`` (or the section above) to know how to run them.
 ``requests_per_minute``
   Reports on how many requests were made per minute.
   It works best when used with ``-s`` and ``-d`` command line arguments,
-  as the output can be huge.
+  as the output can be huge. We can also use ``requests_per_hour``
+  if it's too much data.
+
+``requests_per_hour``
+  Reports on how many requests were made per hour.
 
 ``print``
   Prints the raw lines.

haproxy/logfile.py

@@ -3,6 +3,7 @@
 from datetime import timedelta
 from haproxy.line import Line
 
+import datetime
 import os
 
 
@@ -336,48 +337,50 @@ def cmd_requests_per_minute(self):
 
         .. note::
           Try to combine it with time constrains (``-s`` and ``-d``) as this
-          command output can be huge otherwise.
+          command output can be huge otherwise. We can also use
+          ``requests_per_hour`` if it's too much data.
         """
-        if len(self._valid_lines) == 0:
-            return
-
-        current_minute = self._valid_lines[0].accept_date
-        current_minute_counter = 0
-        requests = []
-        one_minute = timedelta(minutes=1)
-
-        def format_and_append(append_to, date, counter):
-            seconds_and_micro = timedelta(
-                seconds=date.second, microseconds=date.microsecond
-            )
-            minute_formatted = date - seconds_and_micro
-            append_to.append((minute_formatted, counter))
-
-        # note that _valid_lines is kept sorted by date
-        for line in self._valid_lines:
-            line_date = line.accept_date
-            if (
-                line_date - current_minute < one_minute
-                and line_date.minute == current_minute.minute
-            ):
-                current_minute_counter += 1
-
-            else:
-                format_and_append(requests, current_minute, current_minute_counter)
-                current_minute_counter = 1
-                current_minute = line_date
-
-        if current_minute_counter > 0:
-            format_and_append(requests, current_minute, current_minute_counter)
+        return self._count_by_time(1)
 
-        return requests
+    def cmd_requests_per_hour(self):
+        """Generates statistics on how many requests were made per hour.
+        """
+        return self._count_by_time(60)
 
     def cmd_print(self):
         """Returns the raw lines to be printed."""
         if not self._valid_lines:
             return ''
         return '\n'.join([line.raw_line for line in self._valid_lines]) + '\n'
 
+    def _count_by_time(self, aggregate):
+        """
+        Helper function to compile stats by time frame.
+        Used by requests_per_minute and requests_per_hour
+        params:
+          aggregate: Number of minutes to aggregate on
+        """
+        if len(self._valid_lines) == 0:
+            return
+
+        requests = {}
+
+        # note that _valid_lines is kept sorted by date
+        for line in self._valid_lines:
+            tm = line.accept_date
+            tm += datetime.timedelta(minutes=aggregate / 2)
+            tm -= datetime.timedelta(
+                minutes=tm.minute % aggregate,
+                seconds=tm.second,
+                microseconds=tm.microsecond,
+            )
+            try:
+                requests[str(tm)] += 1
+            except KeyError:
+                requests[str(tm)] = 1
+
+        return sorted(requests)
+
     def _sort_lines(self):
         """Haproxy writes its logs after having gathered all information
         related to each specific connection. A simple request can be

haproxy/tests/test_log_file.py

@@ -1,5 +1,4 @@
 # -*- coding: utf-8 -*-
-from datetime import datetime
 from haproxy import filters
 from haproxy.logfile import Log
 from haproxy.main import main
@@ -269,11 +268,21 @@ def test_cmd_requests_per_minute(self):
 
         self.assertEqual(len(requests), 5)
 
-        self.assertEqual(requests[0], (datetime(2013, 12, 11, 11, 2), 8))
-        self.assertEqual(requests[1], (datetime(2013, 12, 11, 11, 3), 3))
-        self.assertEqual(requests[2], (datetime(2013, 12, 11, 11, 13), 5))
-        self.assertEqual(requests[3], (datetime(2013, 12, 11, 11, 52), 7))
-        self.assertEqual(requests[4], (datetime(2013, 12, 11, 12, 2), 9))
+        self.assertEqual(requests[0], '2013-12-11 11:02:00')
+        self.assertEqual(requests[1], '2013-12-11 11:03:00')
+        self.assertEqual(requests[2], '2013-12-11 11:13:00')
+        self.assertEqual(requests[3], '2013-12-11 11:52:00')
+        self.assertEqual(requests[4], '2013-12-11 12:02:00')
+
+    def test_cmd_requests_per_hour(self):
+        """Check that the requests per minute command reports as expected."""
+        log_file = Log(logfile='haproxy/tests/files/requests_per_minute.log')
+        requests = log_file.cmd_requests_per_hour()
+
+        self.assertEqual(len(requests), 2)
+
+        self.assertEqual(requests[0], '2013-12-11 11:00:00')
+        self.assertEqual(requests[1], '2013-12-11 12:00:00')
 
     def test_cmd_requests_per_minute_empty(self):
         """Check that the requests per minute command reports nothing if