Add median, median_low and median_high aggregation methods

README.md

@@ -38,7 +38,8 @@ Options:
   --aggregationMethod=AGGREGATIONMETHOD
                         The consolidation function to fetch from on input and
                         aggregationMethod to set on output. One of: average,
-                        last, max, min, avg_zero, absmax, absmin
+                        last, max, min, avg_zero, absmax, absmin, median,
+                        median_low, median_high
   --destinationPath=DESTINATIONPATH
                         Path to place created whisper file. Defaults to the
                         RRD file's source path.
@@ -66,7 +67,8 @@ Options:
   --xFilesFactor=XFILESFACTOR
   --aggregationMethod=AGGREGATIONMETHOD
                         Function to use when aggregating values (average, sum,
-                        last, max, min, avg_zero, absmax, absmin)
+                        last, max, min, avg_zero, absmax, absmin, median,
+                        median_low, median_high)
   --overwrite
   --estimate            Don't create a whisper file, estimate storage requirements based on archive definitions
 ```
@@ -170,7 +172,8 @@ Options:
                         Change the xFilesFactor
   --aggregationMethod=AGGREGATIONMETHOD
                         Change the aggregation function (average, sum, last,
-                        max, min, avg_zero, absmax, absmin)
+                        max, min, avg_zero, absmax, absmin, median,
+                        median_low, median_high)
   --force               Perform a destructive change
   --newfile=NEWFILE     Create a new database file without removing the
                         existing one
@@ -185,7 +188,7 @@ whisper-set-aggregation-method.py
 Change the aggregation method of an existing whisper file.
 
 ```
-Usage: whisper-set-aggregation-method.py path <average|sum|last|max|min|avg_zero|absmax|absmin>
+Usage: whisper-set-aggregation-method.py path <average|sum|last|max|min|avg_zero|absmax|absmin|median|median_low|median_high>
 
 Options:
   -h, --help  show this help message and exit

bin/rrd2whisper.py

@@ -28,6 +28,12 @@
 aggregationMethods.remove('absmax')
 # RRD doesn't have a 'absmin' type
 aggregationMethods.remove('absmin')
+# RRD doesn't have a 'median' type
+aggregationMethods.remove('median')
+# RRD doesn't have a 'median_low' type
+aggregationMethods.remove('median_low')
+# RRD doesn't have a 'median_high' type
+aggregationMethods.remove('median_high')
 
 option_parser = optparse.OptionParser(usage='''%prog rrd_path''')
 option_parser.add_option(

test_whisper.py

@@ -228,6 +228,14 @@ def test_aggregate(self):
         self.assertEqual(whisper.aggregate('absmin', [-3, -2, 1, 2]), 1)
         # absmin with negative min
         self.assertEqual(whisper.aggregate('absmin', [-2, -1, 2, 3]), -1)
+        # median with odd number of values
+        self.assertEqual(whisper.aggregate('median', [10, 1, 12, 42, 7]), 10)
+        # median with even number of values
+        self.assertEqual(whisper.aggregate('median', [9, 2, 5, 4]), 4.5)
+        # median_low
+        self.assertEqual(whisper.aggregate('median_low', [9, 2, 5, 4]), 4)
+        # median_high
+        self.assertEqual(whisper.aggregate('median_high', [9, 2, 5, 4]), 5)
 
         with AssertRaisesException(
                 whisper.InvalidAggregationMethod(

whisper.py

@@ -125,7 +125,10 @@ def _py_fallocate(fd, offset, len_):
   5: 'min',
   6: 'avg_zero',
   7: 'absmax',
-  8: 'absmin'
+  8: 'absmin',
+  9: 'median',
+  10: 'median_low',
+  11: 'median_high'
 })
 aggregationMethodToType = dict([[v, k] for k, v in aggregationTypeToMethod.items()])
 aggregationMethods = aggregationTypeToMethod.values()
@@ -570,6 +573,17 @@ def aggregate(aggregationMethod, knownValues, neighborValues=None):
     return max(knownValues, key=abs)
   elif aggregationMethod == 'absmin':
     return min(knownValues, key=abs)
+  elif aggregationMethod == 'median':
+    values = sorted(knownValues)
+    mid, isodd = divmod(len(values), 2)
+    return values[mid] if isodd else float(values[mid-1] + values[mid]) / 2.0
+  elif aggregationMethod == 'median_low':
+    values = sorted(knownValues)
+    mid, isodd = divmod(len(values), 2)
+    return values[mid] if isodd else values[mid-1]
+  elif aggregationMethod == 'median_high':
+    values = sorted(knownValues)
+    return values[len(values) // 2]
   else:
     raise InvalidAggregationMethod(
       "Unrecognized aggregation method %s" % aggregationMethod)