forked-synapse/synapse/metrics/metric.py

161 lines
4.7 KiB
Python

# -*- coding: utf-8 -*-
# Copyright 2015 OpenMarket Ltd
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from itertools import chain
# TODO(paul): I can't believe Python doesn't have one of these
def map_concat(func, items):
# flatten a list-of-lists
return list(chain.from_iterable(map(func, items)))
class BaseMetric(object):
def __init__(self, name, keys=[]):
self.name = name
self.keys = keys # OK not to clone as we never write it
def dimension(self):
return len(self.keys)
def is_scalar(self):
return not len(self.keys)
def _render_key(self, values):
if self.is_scalar():
return ""
# TODO: some kind of value escape
return "{%s}" % (
",".join(["%s=%s" % kv for kv in zip(self.keys, values)])
)
def render(self):
return map_concat(self.render_item, sorted(self.counts.keys()))
class CounterMetric(BaseMetric):
"""The simplest kind of metric; one that stores a monotonically-increasing
integer that counts events."""
def __init__(self, *args, **kwargs):
super(CounterMetric, self).__init__(*args, **kwargs)
self.counts = {}
# Scalar metrics are never empty
if self.is_scalar():
self.counts[()] = 0
def inc(self, *values):
if len(values) != self.dimension():
raise ValueError("Expected as many values to inc() as keys (%d)" %
(self.dimension())
)
# TODO: should assert that the tag values are all strings
if values not in self.counts:
self.counts[values] = 1
else:
self.counts[values] += 1
def fetch(self):
return dict(self.counts)
def render_item(self, k):
return ["%s%s %d" % (self.name, self._render_key(k), self.counts[k])]
class CallbackMetric(BaseMetric):
"""A metric that returns the numeric value returned by a callback whenever
it is rendered. Typically this is used to implement gauges that yield the
size or other state of some in-memory object by actively querying it."""
def __init__(self, name, callback, keys=[]):
super(CallbackMetric, self).__init__(name, keys=keys)
self.callback = callback
def render(self):
value = self.callback()
if self.is_scalar():
return ["%s %d" % (self.name, value)]
return ["%s%s %d" % (self.name, self._render_key(k), value[k])
for k in sorted(value.keys())]
class TimerMetric(CounterMetric):
"""A combination of an event counter and a time accumulator, which counts
both the number of events and how long each one takes.
TODO(paul): Try to export some heatmap-style stats?
"""
def __init__(self, *args, **kwargs):
super(TimerMetric, self).__init__(*args, **kwargs)
self.times = {}
# Scalar metrics are never empty
if self.is_scalar():
self.times[()] = 0
def inc_time(self, msec, *values):
self.inc(*values)
if values not in self.times:
self.times[values] = msec
else:
self.times[values] += msec
def render_item(self, k):
keystr = self._render_key(k)
return ["%s%s:count %d" % (self.name, keystr, self.counts[k]),
"%s%s:msec %d" % (self.name, keystr, self.times[k])]
class CacheMetric(object):
"""A combination of two CounterMetrics, one to count cache hits and one to
count misses, and a callback metric to yield the current size.
This metric generates standard metric name pairs, so that monitoring rules
can easily be applied to measure hit ratio."""
def __init__(self, name, size_callback, keys=[]):
self.name = name
self.hits = CounterMetric(name + ":hits", keys=keys)
self.misses = CounterMetric(name + ":misses", keys=keys)
self.size = CallbackMetric(name + ":size",
callback=size_callback,
keys=keys,
)
def inc_hits(self, *values):
self.hits.inc(*values)
def inc_misses(self, *values):
self.misses.inc(*values)
def render(self):
return self.hits.render() + self.misses.render() + self.size.render()