Chromium Code Reviews
chromiumcodereview-hr@appspot.gserviceaccount.com (chromiumcodereview-hr) | Please choose your nickname with Settings | Help | Chromium Project | Gerrit Changes | Sign out
(255)

Issues Search
    My Issues | Starred     Open | Closed | All

Unified Diff: tracing/tracing/base/statistics_test.html

Issue 2771723003: [tracing] Move math utilities from base into their own subdirectory (attempt 2) (Closed)
Patch Set: rebase Created 3 years, 9 months ago
Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.
Jump to:
View side-by-side diff with in-line comments
Download patch
« no previous file with comments | « tracing/tracing/base/statistics.html ('k') | tracing/tracing/base/unittest/test_case.html » ('j') | no next file with comments »
Expand Comments ('e') | Collapse Comments ('c') | Show Comments Hide Comments ('s')
Index: tracing/tracing/base/statistics_test.html
diff --git a/tracing/tracing/base/statistics_test.html b/tracing/tracing/base/statistics_test.html
deleted file mode 100644
index ab4d7d0da1ea5ff7b81f0ff618d987f6bc4a2679..0000000000000000000000000000000000000000
--- a/tracing/tracing/base/statistics_test.html
+++ /dev/null
@@ -1,577 +0,0 @@
-<!DOCTYPE html>
-<!--
-Copyright (c) 2014 The Chromium Authors. All rights reserved.
-Use of this source code is governed by a BSD-style license that can be
-found in the LICENSE file.
--->
-<link rel="import" href="/tracing/base/statistics.html">
-<script>
-'use strict';
-
-// TODO(charliea): Remove:
-/* eslint-disable catapult-camelcase */
-
-tr.b.unittest.testSuite(function() {
- var Statistics = tr.b.Statistics;
-
- /**
- * Lloyd relaxation in 1D.
- *
- * Keeps the position of the first and last sample.
- **/
- function relax(samples, opt_iterations) {
- opt_iterations = opt_iterations || 10;
- for (var i = 0; i < opt_iterations; i++) {
- var voronoiBoundaries = [];
- for (var j = 1; j < samples.length; j++)
- voronoiBoundaries.push((samples[j] + samples[j - 1]) * 0.5);
-
- var relaxedSamples = [];
- relaxedSamples.push(samples[0]);
- for (var j = 1; j < samples.length - 1; j++) {
- relaxedSamples.push(
- (voronoiBoundaries[j - 1] + voronoiBoundaries[j]) * 0.5);
- }
- relaxedSamples.push(samples[samples.length - 1]);
- samples = relaxedSamples;
- }
- return samples;
- }
-
- function createRandomSamples(numSamples) {
- var samples = [];
- var position = 0.0;
- samples.push(position);
- for (var i = 1; i < numSamples; i++) {
- position += Math.random();
- samples.push(position);
- }
- return samples;
- }
-
- test('normalDistribution', function() {
- for (var mean = -100; mean <= 100; mean += 25) {
- for (var stddev = 0.1; stddev < 2; stddev += 0.2) {
- var dist = new Statistics.NormalDistribution(mean, stddev * stddev);
- assert.closeTo(mean, dist.mean, 1e-6);
- assert.closeTo(stddev, dist.standardDeviation, 1e-6);
- assert.closeTo(0, dist.standardDeviation * dist.computeDensity(
- -1e10), 1e-5);
- assert.closeTo(0.05399, dist.standardDeviation * dist.computeDensity(
- dist.mean - 2 * dist.standardDeviation), 1e-5);
- assert.closeTo(0.24197, dist.standardDeviation * dist.computeDensity(
- dist.mean - dist.standardDeviation), 1e-5);
- assert.closeTo(0.39894, dist.standardDeviation * dist.computeDensity(
- dist.mean), 1e-5);
- assert.closeTo(0.24197, dist.standardDeviation * dist.computeDensity(
- dist.mean + dist.standardDeviation), 1e-5);
- assert.closeTo(0.054, dist.standardDeviation * dist.computeDensity(
- dist.mean + 2 * dist.standardDeviation), 1e-5);
- assert.closeTo(0, dist.standardDeviation * dist.computeDensity(
- 1e10), 1e-5);
-
- assert.closeTo(0, dist.computePercentile(-1e10), 1e-5);
- assert.closeTo(0.02275, dist.computePercentile(
- dist.mean - 2 * dist.standardDeviation), 1e-5);
- assert.closeTo(0.15866, dist.computePercentile(
- dist.mean - dist.standardDeviation), 1e-5);
- assert.closeTo(0.5, dist.computePercentile(dist.mean), 1e-5);
- assert.closeTo(0.841344, dist.computePercentile(
- dist.mean + dist.standardDeviation), 1e-5);
- assert.closeTo(0.97725, dist.computePercentile(
- dist.mean + 2 * dist.standardDeviation), 1e-5);
- assert.closeTo(1, dist.computePercentile(1e10), 1e-5);
- }
- }
- });
-
- test('logNormalDistribution', function() {
- // Unlike the Normal distribution, the LogNormal distribution can look very
- // different depending on its parameters, and it's defined in terms of the
- // Normal distribution anyway, so only test the standard LogNormal
- // distribution.
- var dist = new Statistics.LogNormalDistribution(0, 1);
- assert.closeTo(0.3678, dist.mode, 1e-4);
- assert.closeTo(1, dist.median, 1e-6);
- assert.closeTo(1.6487, dist.mean, 1e-4);
- assert.closeTo(0.65774, dist.computeDensity(dist.mode), 1e-5);
- assert.closeTo(0.39894, dist.computeDensity(dist.median), 1e-5);
- assert.closeTo(0.21354, dist.computeDensity(dist.mean), 1e-5);
- assert.closeTo(0, dist.computePercentile(1e-10), 1e-6);
- assert.closeTo(0.15865, dist.computePercentile(dist.mode), 1e-5);
- assert.closeTo(0.5, dist.computePercentile(dist.median), 1e-6);
- assert.closeTo(0.69146, dist.computePercentile(dist.mean), 1e-5);
- assert.closeTo(1, dist.computePercentile(1e100), 1e-5);
- });
-
- test('divideIfPossibleOrZero', function() {
- assert.equal(Statistics.divideIfPossibleOrZero(1, 2), 0.5);
- assert.equal(Statistics.divideIfPossibleOrZero(0, 2), 0);
- assert.equal(Statistics.divideIfPossibleOrZero(1, 0), 0);
- assert.equal(Statistics.divideIfPossibleOrZero(0, 0), 0);
- });
-
- test('sumBasic', function() {
- assert.equal(Statistics.sum([1, 2, 3]), 6);
- });
-
- test('sumWithFunctor', function() {
- var ctx = {};
- var ary = [1, 2, 3];
- assert.equal(12, Statistics.sum(ary, function(x, i) {
- assert.equal(this, ctx);
- assert.equal(ary[i], x);
- return x * 2;
- }, ctx));
- });
-
- test('minMaxWithFunctor', function() {
- var ctx = {};
- var ary = [1, 2, 3];
- function func(x, i) {
- assert.equal(this, ctx);
- assert.equal(ary[i], x);
- return x;
- }
- assert.equal(Statistics.max(ary, func, ctx), 3);
- assert.equal(Statistics.min(ary, func, ctx), 1);
-
- var range = Statistics.range(ary, func, ctx);
- assert.isFalse(range.isEmpty);
- assert.equal(range.min, 1);
- assert.equal(range.max, 3);
- });
-
- test('maxExtrema', function() {
- assert.equal(Statistics.max([]), -Infinity);
- assert.equal(Statistics.min([]), Infinity);
- });
-
- test('meanBasic', function() {
- assert.closeTo(Statistics.mean([1, 2, 3]), 2, 1e-6);
- assert.closeTo(Statistics.mean(new Set([1, 2, 3])), 2, 1e-6);
- });
-
- test('geometricMean', function() {
- assert.strictEqual(1, Statistics.geometricMean([]));
- assert.strictEqual(1, Statistics.geometricMean([1]));
- assert.strictEqual(0, Statistics.geometricMean([-1]));
- assert.strictEqual(0, Statistics.geometricMean([0]));
- assert.strictEqual(0, Statistics.geometricMean([1, 2, 3, 0]));
- assert.strictEqual(0, Statistics.geometricMean([1, 2, 3, -1]));
- assert.strictEqual(1, Statistics.geometricMean([1, 1, 1]));
- assert.strictEqual(2, Statistics.geometricMean([2]));
- assert.closeTo(Math.sqrt(6), Statistics.geometricMean([2, 3]), 1e-6);
- assert.closeTo(6, Statistics.geometricMean(new Set([4, 9])), 1e-6);
-
- var samples = [];
- for (var i = 0; i < 1e3; ++i)
- samples.push(Number.MAX_SAFE_INTEGER);
- assert.closeTo(Number.MAX_SAFE_INTEGER, Statistics.geometricMean(samples),
- Number.MAX_SAFE_INTEGER * 1e-13);
-
- samples = [];
- for (var i = 0; i < 1e3; ++i)
- samples.push(Number.MAX_VALUE / 1e3);
- assert.closeTo(Number.MAX_VALUE / 1e3, Statistics.geometricMean(samples),
- Number.MAX_VALUE * 1e-13);
- });
-
- test('weightedMean', function() {
- function getWeight(element) {
- return element.weight;
- }
- function getValue(element) {
- return element.value;
- }
-
- var data = [
- {value: 10, weight: 3},
- {value: 20, weight: 1},
- {value: 30, weight: 6}
- ];
- assert.equal(23, Statistics.weightedMean(data, getWeight, getValue));
-
- data = [
- {value: 10, weight: 0},
- {value: 20, weight: 0},
- {value: 30, weight: 0}
- ];
- assert.equal(undefined, Statistics.weightedMean(data, getWeight, getValue));
-
- data = [
- {value: 10, weight: -10},
- {value: 20, weight: 5},
- {value: 30, weight: 5}
- ];
- assert.equal(undefined, Statistics.weightedMean(data, getWeight, getValue));
- });
-
- test('weightedMean_positionDependent', function() {
- function getWeight(element, idx) {
- return idx;
- }
- // 3 has weight of 0, 6 has weight of 1, 9 has weight of 2
- assert.equal(8, Statistics.weightedMean([3, 6, 9], getWeight));
- });
-
- test('max_positionDependent', function() {
- function getValue(element, idx) {
- return element * idx;
- }
- assert.equal(6, Statistics.max([1, 2, 3], getValue));
- });
-
- test('min_positionDependent', function() {
- function getValue(element, idx) {
- return element * idx;
- }
- assert.equal(-6, Statistics.min([1, 2, -3], getValue));
- });
-
- test('varianceBasic', function() {
- // In [2, 4, 4, 2], all items have a deviation of 1.0 from the mean so the
- // population variance is 4.0 / 4 = 1.0, but the sample variance is 4.0 / 3.
- assert.equal(Statistics.variance([2, 4, 4, 2]), 4.0 / 3);
-
- // In [1, 2, 3], the squared deviations are 1.0, 0.0 and 1.0 respectively;
- // population variance 2.0 / 3 but sample variance is 2.0 / 2 = 1.0.
- assert.equal(Statistics.variance([1, 2, 3]), 1.0);
- });
-
- test('varianceWithFunctor', function() {
- var ctx = {};
- var ary = [{x: 2},
- {x: 4},
- {x: 4},
- {x: 2}];
- assert.equal(4.0 / 3, Statistics.variance(ary, function(d) {
- assert.equal(ctx, this);
- return d.x;
- }, ctx));
- });
-
- test('stddevBasic', function() {
- assert.equal(Statistics.stddev([2, 4, 4, 2]), Math.sqrt(4.0 / 3));
- });
-
- test('stddevWithFunctor', function() {
- var ctx = {};
- var ary = [{x: 2},
- {x: 4},
- {x: 4},
- {x: 2}];
- assert.equal(Math.sqrt(4.0 / 3), Statistics.stddev(ary, function(d) {
- assert.equal(ctx, this);
- return d.x;
- }, ctx));
- });
-
- test('percentile', function() {
- var ctx = {};
- var ary = [{x: 0},
- {x: 1},
- {x: 2},
- {x: 3},
- {x: 4},
- {x: 5},
- {x: 6},
- {x: 7},
- {x: 8},
- {x: 9}];
- function func(d, i) {
- assert.equal(ctx, this);
- return d.x;
- }
- assert.equal(Statistics.percentile(ary, 0, func, ctx), 0);
- assert.equal(Statistics.percentile(ary, .5, func, ctx), 4);
- assert.equal(Statistics.percentile(ary, .75, func, ctx), 6);
- assert.equal(Statistics.percentile(ary, 1, func, ctx), 9);
- });
-
- test('percentile_positionDependent', function() {
- var ctx = {};
- var ary = [{x: 0},
- {x: 1},
- {x: 2},
- {x: 3},
- {x: 4},
- {x: 5},
- {x: 6},
- {x: 7},
- {x: 8},
- {x: 9}];
- function func(d, i) {
- assert.equal(ctx, this);
- assert.equal(d.x, i);
- return d.x * i;
- }
- assert.equal(Statistics.percentile(ary, 0, func, ctx), 0);
- assert.equal(Statistics.percentile(ary, .5, func, ctx), 16);
- assert.equal(Statistics.percentile(ary, .75, func, ctx), 36);
- assert.equal(Statistics.percentile(ary, 1, func, ctx), 81);
- });
-
- test('normalizeSamples', function() {
- var samples = [];
- var results = Statistics.normalizeSamples(samples);
- assert.deepEqual(results.normalized_samples, []);
- assert.deepEqual(results.scale, 1.0);
-
- samples = [0.0, 0.0];
- results = Statistics.normalizeSamples(samples);
- assert.deepEqual(results.normalized_samples, [0.5, 0.5]);
- assert.deepEqual(results.scale, 1.0);
-
- samples = [0.0, 1.0 / 3.0, 2.0 / 3.0, 1.0];
- results = Statistics.normalizeSamples(samples);
- assert.deepEqual(results.normalized_samples,
- [1.0 / 8.0, 3.0 / 8.0, 5.0 / 8.0, 7.0 / 8.0]);
- assert.deepEqual(results.scale, 0.75);
-
- samples = [1.0 / 8.0, 3.0 / 8.0, 5.0 / 8.0, 7.0 / 8.0];
- results = Statistics.normalizeSamples(samples);
- assert.deepEqual(results.normalized_samples, samples);
- assert.deepEqual(results.scale, 1.0);
- });
-
- /**
- *Tests NormalizeSamples and Discrepancy with random samples.
- *
- * Generates 10 sets of 10 random samples, computes the discrepancy,
- * relaxes the samples using Llloyd's algorithm in 1D, and computes the
- * discrepancy of the relaxed samples. Discrepancy of the relaxed samples
- * must be less than or equal to the discrepancy of the original samples.
- **/
- test('discrepancy_Random', function() {
- for (var i = 0; i < 10; i++) {
- var samples = createRandomSamples(10);
- var samples = Statistics.normalizeSamples(samples).normalized_samples;
- var d = Statistics.discrepancy(samples);
- var relaxedSamples = relax(samples);
- var dRelaxed = Statistics.discrepancy(relaxedSamples);
- assert.isBelow(dRelaxed, d);
- }
- });
-
-
- /* Computes discrepancy for sample sets with known statistics. */
- test('discrepancy_Analytic', function() {
- var samples = [];
- var d = Statistics.discrepancy(samples);
- assert.equal(d, 0.0);
-
- samples = [0.5];
- d = Statistics.discrepancy(samples);
- assert.equal(d, 0.5);
-
- samples = [0.0, 1.0];
- d = Statistics.discrepancy(samples);
- assert.equal(d, 1.0);
-
- samples = [0.5, 0.5, 0.5];
- d = Statistics.discrepancy(samples);
- assert.equal(d, 1.0);
-
- samples = [1.0 / 8.0, 3.0 / 8.0, 5.0 / 8.0, 7.0 / 8.0];
- d = Statistics.discrepancy(samples);
- assert.equal(d, 0.25);
-
- samples = [1.0 / 8.0, 5.0 / 8.0, 5.0 / 8.0, 7.0 / 8.0];
- d = Statistics.discrepancy(samples);
- assert.equal(d, 0.5);
-
- samples = [1.0 / 8.0, 3.0 / 8.0, 5.0 / 8.0, 5.0 / 8.0, 7.0 / 8.0];
- d = Statistics.discrepancy(samples);
- assert.equal(d, 0.4);
-
- samples = [0.0, 1.0 / 3.0, 2.0 / 3.0, 1.0];
- d = Statistics.discrepancy(samples);
- assert.equal(d, 0.5);
-
- samples = Statistics.normalizeSamples(samples).normalized_samples;
- d = Statistics.discrepancy(samples);
- assert.equal(d, 0.25);
- });
-
- test('timestampsDiscrepancy', function() {
- var timestamps = [];
- var dAbs = Statistics.timestampsDiscrepancy(timestamps, true);
- assert.equal(dAbs, 0.0);
-
- timestamps = [4];
- dAbs = Statistics.timestampsDiscrepancy(timestamps, true);
- assert.equal(dAbs, 0.5);
-
- var timestampsA = [0, 1, 2, 3, 5, 6];
- var timestampsB = [0, 1, 2, 3, 5, 7];
- var timestampsC = [0, 2, 3, 4];
- var timestampsD = [0, 2, 3, 4, 5];
-
-
- var dAbsA = Statistics.timestampsDiscrepancy(timestampsA, true);
- var dAbsB = Statistics.timestampsDiscrepancy(timestampsB, true);
- var dAbsC = Statistics.timestampsDiscrepancy(timestampsC, true);
- var dAbsD = Statistics.timestampsDiscrepancy(timestampsD, true);
- var dRelA = Statistics.timestampsDiscrepancy(timestampsA, false);
- var dRelB = Statistics.timestampsDiscrepancy(timestampsB, false);
- var dRelC = Statistics.timestampsDiscrepancy(timestampsC, false);
- var dRelD = Statistics.timestampsDiscrepancy(timestampsD, false);
-
-
- assert.isBelow(dAbsA, dAbsB);
- assert.isBelow(dRelA, dRelB);
- assert.isBelow(dRelD, dRelC);
- assert.closeTo(dAbsD, dAbsC, 0.0001);
- });
-
- test('discrepancyMultipleRanges', function() {
- var samples = [[0.0, 1.2, 2.3, 3.3], [6.3, 7.5, 8.4], [4.2, 5.4, 5.9]];
- var d0 = Statistics.timestampsDiscrepancy(samples[0]);
- var d1 = Statistics.timestampsDiscrepancy(samples[1]);
- var d2 = Statistics.timestampsDiscrepancy(samples[2]);
- var d = Statistics.timestampsDiscrepancy(samples);
- assert.equal(d, Math.max(d0, d1, d2));
- });
-
- /**
- * Tests approimate discrepancy implementation by comparing to exact
- * solution.
- **/
- test('approximateDiscrepancy', function() {
- for (var i = 0; i < 5; i++) {
- var samples = createRandomSamples(10);
- samples = Statistics.normalizeSamples(samples).normalized_samples;
- var d = Statistics.discrepancy(samples);
- var dApprox = Statistics.discrepancy(samples, 500);
- assert.closeTo(d, dApprox, 0.01);
- }
- });
-
- test('durationsDiscrepancy', function() {
- var durations = [];
- var d = Statistics.durationsDiscrepancy(durations);
- assert.equal(d, 0.0);
-
- durations = [4];
- d = Statistics.durationsDiscrepancy(durations);
- assert.equal(d, 4.0);
-
- var durationsA = [1, 1, 1, 1, 1];
- var durationsB = [1, 1, 2, 1, 1];
- var durationsC = [1, 2, 1, 2, 1];
-
- var dA = Statistics.durationsDiscrepancy(durationsA);
- var dB = Statistics.durationsDiscrepancy(durationsB);
- var dC = Statistics.durationsDiscrepancy(durationsC);
-
- assert.isBelow(dA, dB);
- assert.isBelow(dB, dC);
- });
-
- test('uniformlySampleArray', function() {
- var samples = ['A', 'B', 'C', 'D', 'E'];
- for (var i = samples.length; i >= 0; --i) {
- Statistics.uniformlySampleArray(samples, i);
- assert.lengthOf(samples, i);
- }
- });
-
- test('uniformlySampleStream', function() {
- var samples = [];
- Statistics.uniformlySampleStream(samples, 1, 'A', 5);
- assert.deepEqual(['A'], samples);
- Statistics.uniformlySampleStream(samples, 2, 'B', 5);
- Statistics.uniformlySampleStream(samples, 3, 'C', 5);
- Statistics.uniformlySampleStream(samples, 4, 'D', 5);
- Statistics.uniformlySampleStream(samples, 5, 'E', 5);
- assert.deepEqual(['A', 'B', 'C', 'D', 'E'], samples);
-
- Statistics.uniformlySampleStream(samples, 6, 'F', 5);
- // Can't really assert anything more than the length since the elements are
- // drawn at random.
- assert.equal(samples.length, 5);
-
- // Try starting with a non-empty array.
- samples = [0, 0, 0];
- Statistics.uniformlySampleStream(samples, 1, 'G', 5);
- assert.deepEqual(['G', 0, 0], samples);
- });
-
- test('mergeSampledStreams', function() {
- var samples = [];
- Statistics.mergeSampledStreams(samples, 0, ['A'], 1, 5);
- assert.deepEqual(['A'], samples);
- Statistics.mergeSampledStreams(samples, 1, ['B', 'C', 'D', 'E'], 4, 5);
- assert.deepEqual(['A', 'B', 'C', 'D', 'E'], samples);
-
- Statistics.mergeSampledStreams(samples, 9, ['F', 'G', 'H', 'I', 'J'], 7, 5);
- // Can't really assert anything more than the length since the elements are
- // drawn at random.
- assert.equal(samples.length, 5);
-
- var samples = ['A', 'B'];
- Statistics.mergeSampledStreams(samples, 2, ['F', 'G', 'H', 'I', 'J'], 7, 5);
- assert.equal(samples.length, 5);
- });
-
- test('mannWhitneyUTestSmokeTest', function() {
- // x < 0.01
- var sampleA = [1, 2, 2.1, 2.2, 2, 1];
- var sampleB = [12, 13, 13.1, 13.2, 13, 12];
- var results = Statistics.mwu(sampleA, sampleB);
- assert.isBelow(results.p, 0.1);
-
- // 0.01 < x < 0.1
- sampleA = [1, 2, 2.1, 2.2, 2, 1];
- sampleB = [2, 3, 3.1, 3.2, 3, 2];
- results = Statistics.mwu(sampleA, sampleB);
- assert.isBelow(results.p, 0.1);
- assert.isAbove(results.p, 0.01);
-
- // 0.1 < x
- sampleA = [1, 2, 2.1, 2.2, 2, 1];
- sampleB = [1, 2, 2.1, 2.2, 2, 1];
- results = Statistics.mwu(sampleA, sampleB);
- assert.isAbove(results.p, 0.1);
- });
-
- test('mannWhitneyUEdgeCases', function() {
- var longRepeatingSample = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1];
- var emptySample = [];
- var singleLargeValue = [1000000];
- // mean 10, std 2
- var normallyDistributedSample = [
- 8.341540e+0, 7.216640e+0, 8.844310e+0, 9.801980e+0, 1.048760e+1,
- 6.915150e+0, 7.881740e+0, 1.131160e+1, 9.959400e+0, 9.030880e+0
- ];
- // Identical samples should not cause the null to be rejected.
- var results = Statistics.mwu(longRepeatingSample, longRepeatingSample);
- assert.isAbove(results.p, 0.05);
- results = Statistics.mwu(normallyDistributedSample,
- normallyDistributedSample);
- assert.isAbove(results.p, 0.05);
- results = Statistics.mwu(singleLargeValue, singleLargeValue);
-
- // A single value is generally not sufficient to reject the null, no matter
- // how far off it is.
- results = Statistics.mwu(normallyDistributedSample, singleLargeValue);
- assert.isAbove(results.p, 0.05);
-
- // A single value way outside the first sample may be enough to reject,
- // if the first sample is large enough.
- results = Statistics.mwu(longRepeatingSample, singleLargeValue);
- assert.isBelow(results.p, 0.005);
-
- // Empty samples should not be comparable.
- results = Statistics.mwu(emptySample, emptySample);
- assert(isNaN(results.p));
-
- // The result of comparing a sample against an empty sample should not be a
- // valid p value. NOTE: The current implementation returns 0, it is up to
- // the caller to interpret this.
- results = Statistics.mwu(normallyDistributedSample, emptySample);
- assert(!results.p);
- });
-});
-</script>
« no previous file with comments | « tracing/tracing/base/statistics.html ('k') | tracing/tracing/base/unittest/test_case.html » ('j') | no next file with comments »

Powered by Google App Engine
This is Rietveld 408576698