From 9f6d1f6fc2d099c0fb2a72f1bb2cefd8943bde19 Mon Sep 17 00:00:00 2001 From: Joyee Cheung <joyeec9h3@gmail.com> Date: Sun, 25 Dec 2016 03:32:26 +0800 Subject: [PATCH] util: improve readability of normalizeEncoding * Improve readability of util.normalizeEncoding and add some comments * Add a benchmark for util.normalizeEncoding PR-URL: https://github.com/nodejs/node/pull/10439 Reviewed-By: James M Snell <jasnell@gmail.com> Reviewed-By: Andreas Madsen <amwebdk@gmail.com> --- benchmark/util/normalize-encoding.js | 65 ++++++++++++++++++++++++++++ lib/internal/util.js | 14 ++++-- 2 files changed, 75 insertions(+), 4 deletions(-) create mode 100644 benchmark/util/normalize-encoding.js diff --git a/benchmark/util/normalize-encoding.js b/benchmark/util/normalize-encoding.js new file mode 100644 index 00000000000000..8c4d03478104b0 --- /dev/null +++ b/benchmark/util/normalize-encoding.js @@ -0,0 +1,65 @@ +'use strict'; + +const common = require('../common.js'); +const assert = require('assert'); + +const groupedInputs = { + group_common: ['undefined', 'utf8', 'utf-8', 'base64', 'binary', 'latin1'], + group_upper: ['UTF-8', 'UTF8', 'UCS2', 'UTF-16LE', 'UTF16LE', 'BASE64'], + group_uncommon: [ 'foo', '1', 'false', 'undefined', '[]'], + group_misc: ['', 'utf16le', 'usc2', 'hex', 'HEX', 'BINARY'] +}; + +const inputs = [ + '', 'utf8', 'utf-8', 'UTF-8', + 'UTF8', 'Utf8', 'uTf-8', 'utF-8', 'ucs2', + 'UCS2', 'utf16le', 'utf-16le', 'UTF-16LE', 'UTF16LE', + 'binary', 'BINARY', 'latin1', 'base64', 'BASE64', + 'hex', 'HEX', 'foo', '1', 'false', 'undefined', '[]']; + +const bench = common.createBenchmark(main, { + input: inputs.concat(Object.keys(groupedInputs)), + n: [1e5] +}, { + flags: '--expose-internals' +}); + +function getInput(input) { + switch (input) { + case 'group_common': + return groupedInputs.group_common; + case 'group_upper': + return groupedInputs.group_upper; + case 'group_uncommon': + return groupedInputs.group_uncommon; + case 'group_misc': + return groupedInputs.group_misc; + case '1': + return [1]; + case 'false': + return [false]; + case 'undefined': + return [undefined]; + case '[]': + return [[]]; + default: + return [input]; + } +} + +function main(conf) { + const normalizeEncoding = require('internal/util').normalizeEncoding; + + const n = conf.n | 0; + const inputs = getInput(conf.input); + var noDead = ''; + + bench.start(); + for (var i = 0; i < n; i += 1) { + for (var j = 0; j < inputs.length; ++j) { + noDead = normalizeEncoding(inputs[j]); + } + } + bench.end(n); + assert.ok(noDead === undefined || noDead.length > 0); +} diff --git a/lib/internal/util.js b/lib/internal/util.js index 0c7f391eb1fad1..b7facfbbbfda68 100644 --- a/lib/internal/util.js +++ b/lib/internal/util.js @@ -103,10 +103,16 @@ exports.assertCrypto = function(exports) { }; exports.kIsEncodingSymbol = Symbol('node.isEncoding'); + +// The loop should only run at most twice, retrying with lowercased enc +// if there is no match in the first pass. +// We use a loop instead of branching to retry with a helper +// function in order to avoid the performance hit. +// Return undefined if there is no match. exports.normalizeEncoding = function normalizeEncoding(enc) { if (!enc) return 'utf8'; - var low; - for (;;) { + var retried; + while (true) { switch (enc) { case 'utf8': case 'utf-8': @@ -124,9 +130,9 @@ exports.normalizeEncoding = function normalizeEncoding(enc) { case 'hex': return enc; default: - if (low) return; // undefined + if (retried) return; // undefined enc = ('' + enc).toLowerCase(); - low = true; + retried = true; } } };