debug vs winston vs pino vs loglevel vs log4js vs bunyan
Node.js 日志库
debugwinstonpinologlevellog4jsbunyan类似的npm包:
Node.js 日志库

这些日志库用于 Node.js 应用程序中,帮助开发者记录和管理应用的日志信息。它们提供了不同的功能和特性,适用于各种日志记录需求,从简单的调试信息到复杂的日志管理。选择合适的日志库可以提高应用程序的可维护性和调试效率。

npm下载趋势
3 年
GitHub Stars 排名
统计详情
npm包名称
下载量
Stars
大小
Issues
发布时间
License
debug415,191,20111,40342.8 kB843 个月前MIT
winston17,255,29024,244275 kB51712 天前MIT
pino16,521,66116,998637 kB1402 个月前MIT
loglevel11,810,8002,73486.2 kB181 年前MIT
log4js6,784,3105,844160 kB973 年前Apache-2.0
bunyan2,460,6537,216-2945 年前MIT
功能对比: debug vs winston vs pino vs loglevel vs log4js vs bunyan

性能

  • debug:

    Debug 是一个轻量级的库,性能开销极小,适合在开发环境中使用,能够快速输出调试信息。

  • winston:

    Winston 性能良好,适合多种场景,但在复杂配置下可能会影响性能。

  • pino:

    Pino 是一个专注于性能的日志库,能够以极低的开销记录日志,适合对性能要求极高的应用。

  • loglevel:

    Loglevel 设计简单,性能开销小,适合快速开发和小型项目,能够满足基本的日志需求。

  • log4js:

    Log4js 性能较为一般,适合中小型应用。其灵活性和功能丰富性可能导致性能开销增加。

  • bunyan:

    Bunyan 提供高效的 JSON 日志记录,适合需要快速写入日志的应用。其流式 API 使得日志记录的性能非常高。

结构化日志

  • debug:

    Debug 输出的日志通常是字符串,不支持结构化日志。

  • winston:

    Winston 支持结构化日志,可以通过不同的格式化器进行自定义。

  • pino:

    Pino 支持结构化日志,能够方便地与日志分析工具集成。

  • loglevel:

    Loglevel 不支持结构化日志,主要以文本形式输出。

  • log4js:

    Log4js 支持结构化日志,但需要额外配置。

  • bunyan:

    Bunyan 默认输出结构化的 JSON 日志,便于后续分析和处理。

易用性

  • debug:

    Debug 的 API 非常简单,适合快速使用,特别是在开发过程中。

  • winston:

    Winston 的 API 灵活,功能强大,但可能需要一些时间来掌握其复杂的配置选项。

  • pino:

    Pino 的 API 直观,易于使用,适合需要高性能的应用。

  • loglevel:

    Loglevel 提供简单的 API,易于使用,适合快速开发。

  • log4js:

    Log4js 的功能丰富,但配置相对复杂,学习曲线较陡。

  • bunyan:

    Bunyan 的 API 设计合理,易于上手,但需要理解 JSON 格式的日志。

输出方式

  • debug:

    Debug 主要在控制台输出,适合开发过程中的调试。

  • winston:

    Winston 支持多种传输选项,可以灵活配置输出目标。

  • pino:

    Pino 支持多种输出方式,能够与流式处理结合使用,适合高性能应用。

  • loglevel:

    Loglevel 主要在控制台输出,适合简单的日志需求。

  • log4js:

    Log4js 支持多种输出方式,包括文件、控制台、网络等,灵活性高。

  • bunyan:

    Bunyan 支持多种输出方式,包括文件、控制台和流式处理,适合多种使用场景。

日志级别控制

  • debug:

    Debug 通过命名空间控制日志级别,方便管理不同模块的日志。

  • winston:

    Winston 提供丰富的日志级别控制选项,适合复杂的日志管理需求。

  • pino:

    Pino 支持动态日志级别控制,适合高性能应用。

  • loglevel:

    Loglevel 提供简单的日志级别控制,适合快速开发。

  • log4js:

    Log4js 提供灵活的日志级别配置,适合复杂应用。

  • bunyan:

    Bunyan 支持多种日志级别,能够灵活控制输出的日志信息。

如何选择: debug vs winston vs pino vs loglevel vs log4js vs bunyan
  • debug:

    选择 Debug 如果你需要一个轻量级的调试工具,适合在开发过程中快速输出调试信息。它支持命名空间功能,方便管理不同模块的日志。

  • winston:

    选择 Winston 如果你需要一个通用的日志库,支持多种传输和格式化选项。它非常灵活,适合需要复杂日志管理的应用。

  • pino:

    选择 Pino 如果你需要一个高性能的日志库,能够以极低的开销记录日志。它支持结构化日志和异步日志记录,适合对性能要求极高的应用。

  • loglevel:

    选择 Loglevel 如果你需要一个简单易用的日志库,适合前端和后端应用。它提供了简单的 API 和灵活的日志级别控制,适合快速开发和小型项目。

  • log4js:

    选择 Log4js 如果你需要一个功能丰富的日志库,支持多种输出方式(如文件、控制台、网络等),并且需要灵活的日志级别配置。适合大型应用程序。

  • bunyan:

    选择 Bunyan 如果你需要一个结构化的 JSON 日志输出,适合进行日志分析和监控。它支持流式日志记录和多种传输方式,适合需要高性能和高可用性的应用。

debug的README

debug

OpenCollective OpenCollective

A tiny JavaScript debugging utility modelled after Node.js core's debugging technique. Works in Node.js and web browsers.

Installation

$ npm install debug

Usage

debug exposes a function; simply pass this function the name of your module, and it will return a decorated version of console.error for you to pass debug statements to. This will allow you to toggle the debug output for different parts of your module as well as the module as a whole.

Example app.js:

var debug = require('debug')('http')
  , http = require('http')
  , name = 'My App';

// fake app

debug('booting %o', name);

http.createServer(function(req, res){
  debug(req.method + ' ' + req.url);
  res.end('hello\n');
}).listen(3000, function(){
  debug('listening');
});

// fake worker of some kind

require('./worker');

Example worker.js:

var a = require('debug')('worker:a')
  , b = require('debug')('worker:b');

function work() {
  a('doing lots of uninteresting work');
  setTimeout(work, Math.random() * 1000);
}

work();

function workb() {
  b('doing some work');
  setTimeout(workb, Math.random() * 2000);
}

workb();

The DEBUG environment variable is then used to enable these based on space or comma-delimited names.

Here are some examples:

screen shot 2017-08-08 at 12 53 04 pm screen shot 2017-08-08 at 12 53 38 pm screen shot 2017-08-08 at 12 53 25 pm

Windows command prompt notes

CMD

On Windows the environment variable is set using the set command.

set DEBUG=*,-not_this

Example:

set DEBUG=* & node app.js
PowerShell (VS Code default)

PowerShell uses different syntax to set environment variables.

$env:DEBUG = "*,-not_this"

Example:

$env:DEBUG='app';node app.js

Then, run the program to be debugged as usual.

npm script example:

  "windowsDebug": "@powershell -Command $env:DEBUG='*';node app.js",

Namespace Colors

Every debug instance has a color generated for it based on its namespace name. This helps when visually parsing the debug output to identify which debug instance a debug line belongs to.

Node.js

In Node.js, colors are enabled when stderr is a TTY. You also should install the supports-color module alongside debug, otherwise debug will only use a small handful of basic colors.

Web Browser

Colors are also enabled on "Web Inspectors" that understand the %c formatting option. These are WebKit web inspectors, Firefox (since version 31) and the Firebug plugin for Firefox (any version).

Millisecond diff

When actively developing an application it can be useful to see when the time spent between one debug() call and the next. Suppose for example you invoke debug() before requesting a resource, and after as well, the "+NNNms" will show you how much time was spent between calls.

When stdout is not a TTY, Date#toISOString() is used, making it more useful for logging the debug information as shown below:

Conventions

If you're using this in one or more of your libraries, you should use the name of your library so that developers may toggle debugging as desired without guessing names. If you have more than one debuggers you should prefix them with your library name and use ":" to separate features. For example "bodyParser" from Connect would then be "connect:bodyParser". If you append a "*" to the end of your name, it will always be enabled regardless of the setting of the DEBUG environment variable. You can then use it for normal output as well as debug output.

Wildcards

The * character may be used as a wildcard. Suppose for example your library has debuggers named "connect:bodyParser", "connect:compress", "connect:session", instead of listing all three with DEBUG=connect:bodyParser,connect:compress,connect:session, you may simply do DEBUG=connect:*, or to run everything using this module simply use DEBUG=*.

You can also exclude specific debuggers by prefixing them with a "-" character. For example, DEBUG=*,-connect:* would include all debuggers except those starting with "connect:".

Environment Variables

When running through Node.js, you can set a few environment variables that will change the behavior of the debug logging:

NamePurpose
DEBUGEnables/disables specific debugging namespaces.
DEBUG_HIDE_DATEHide date from debug output (non-TTY).
DEBUG_COLORSWhether or not to use colors in the debug output.
DEBUG_DEPTHObject inspection depth.
DEBUG_SHOW_HIDDENShows hidden properties on inspected objects.

Note: The environment variables beginning with DEBUG_ end up being converted into an Options object that gets used with %o/%O formatters. See the Node.js documentation for util.inspect() for the complete list.

Formatters

Debug uses printf-style formatting. Below are the officially supported formatters:

FormatterRepresentation
%OPretty-print an Object on multiple lines.
%oPretty-print an Object all on a single line.
%sString.
%dNumber (both integer and float).
%jJSON. Replaced with the string '[Circular]' if the argument contains circular references.
%%Single percent sign ('%'). This does not consume an argument.

Custom formatters

You can add custom formatters by extending the debug.formatters object. For example, if you wanted to add support for rendering a Buffer as hex with %h, you could do something like:

const createDebug = require('debug')
createDebug.formatters.h = (v) => {
  return v.toString('hex')
}

// …elsewhere
const debug = createDebug('foo')
debug('this is hex: %h', new Buffer('hello world'))
//   foo this is hex: 68656c6c6f20776f726c6421 +0ms

Browser Support

You can build a browser-ready script using browserify, or just use the browserify-as-a-service build, if you don't want to build it yourself.

Debug's enable state is currently persisted by localStorage. Consider the situation shown below where you have worker:a and worker:b, and wish to debug both. You can enable this using localStorage.debug:

localStorage.debug = 'worker:*'

And then refresh the page.

a = debug('worker:a');
b = debug('worker:b');

setInterval(function(){
  a('doing some work');
}, 1000);

setInterval(function(){
  b('doing some work');
}, 1200);

In Chromium-based web browsers (e.g. Brave, Chrome, and Electron), the JavaScript console will—by default—only show messages logged by debug if the "Verbose" log level is enabled.

Output streams

By default debug will log to stderr, however this can be configured per-namespace by overriding the log method:

Example stdout.js:

var debug = require('debug');
var error = debug('app:error');

// by default stderr is used
error('goes to stderr!');

var log = debug('app:log');
// set this namespace to log via console.log
log.log = console.log.bind(console); // don't forget to bind to console!
log('goes to stdout');
error('still goes to stderr!');

// set all output to go via console.info
// overrides all per-namespace log settings
debug.log = console.info.bind(console);
error('now goes to stdout via console.info');
log('still goes to stdout, but via console.info now');

Extend

You can simply extend debugger

const log = require('debug')('auth');

//creates new debug instance with extended namespace
const logSign = log.extend('sign');
const logLogin = log.extend('login');

log('hello'); // auth hello
logSign('hello'); //auth:sign hello
logLogin('hello'); //auth:login hello

Set dynamically

You can also enable debug dynamically by calling the enable() method :

let debug = require('debug');

console.log(1, debug.enabled('test'));

debug.enable('test');
console.log(2, debug.enabled('test'));

debug.disable();
console.log(3, debug.enabled('test'));

print :

1 false
2 true
3 false

Usage :
enable(namespaces)
namespaces can include modes separated by a colon and wildcards.

Note that calling enable() completely overrides previously set DEBUG variable :

$ DEBUG=foo node -e 'var dbg = require("debug"); dbg.enable("bar"); console.log(dbg.enabled("foo"))'
=> false

disable()

Will disable all namespaces. The functions returns the namespaces currently enabled (and skipped). This can be useful if you want to disable debugging temporarily without knowing what was enabled to begin with.

For example:

let debug = require('debug');
debug.enable('foo:*,-foo:bar');
let namespaces = debug.disable();
debug.enable(namespaces);

Note: There is no guarantee that the string will be identical to the initial enable string, but semantically they will be identical.

Checking whether a debug target is enabled

After you've created a debug instance, you can determine whether or not it is enabled by checking the enabled property:

const debug = require('debug')('http');

if (debug.enabled) {
  // do stuff...
}

You can also manually toggle this property to force the debug instance to be enabled or disabled.

Usage in child processes

Due to the way debug detects if the output is a TTY or not, colors are not shown in child processes when stderr is piped. A solution is to pass the DEBUG_COLORS=1 environment variable to the child process.
For example:

worker = fork(WORKER_WRAP_PATH, [workerPath], {
  stdio: [
    /* stdin: */ 0,
    /* stdout: */ 'pipe',
    /* stderr: */ 'pipe',
    'ipc',
  ],
  env: Object.assign({}, process.env, {
    DEBUG_COLORS: 1 // without this settings, colors won't be shown
  }),
});

worker.stderr.pipe(process.stderr, { end: false });

Authors

  • TJ Holowaychuk
  • Nathan Rajlich
  • Andrew Rhyne
  • Josh Junon

Backers

Support us with a monthly donation and help us continue our activities. [Become a backer]

Sponsors

Become a sponsor and get your logo on our README on Github with a link to your site. [Become a sponsor]

License

(The MIT License)

Copyright (c) 2014-2017 TJ Holowaychuk <tj@vision-media.ca> Copyright (c) 2018-2021 Josh Junon

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the 'Software'), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED 'AS IS', WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.