Timut

Simple timing utils for the poor man's toolbox.

Install

yarn add --dev timut@next # npm install --save-dev timut@next

Example

// in somefile.js 
require('timut').push('just a message');

// in anotherfile.js
require('timut').push('another message');

// ...push as many messages as required

// wherever
require('timut').log();

.log() will output a table including time deltas to the console.

Time          |  Context     |  Message               |  Rel. Delta  |  Abs. Delta
----------------------------------------------------------------------------------
00:01:01:007  |  default...  |  just a message......  |         0ms  |         0ms
00:01:01:019  |  default...  |  another message.....  |       +12ms  |       +12ms
00:01:01:023  |  default...  |  yet another thing...  |        +4ms  |       +16ms

This is not fancy, but sometimes useful as a quicker and less complicated inspection tool than full-blown profilers.

The require('timut') syntax is recommended, as it easily allows simple sed commands to remove all occurrences in the complete codebase.

Contexts

// You can push messages to different contexts
require('timut').push('hello, message'); // pushes to context `default`
require('timut').push('another context', 'foo'); // pushes to context `foo`

// .log() takes an optional argument to log a context
require('timut').log('foo'); // logs all messages in context `foo`

// to remove all entries for a context
require('timut').resetContext('foo');

// or to reset all contexts
require('timut').reset();

Library Targets

• browser (dist/timut.min.js)
• commonjs (lib/index.js)
• es2015 modules (es/index.js)

API

See exports in index.ts.

Use Case

The original use case behind the library was debugging asynchronous Karma tests running in PhantomJS, transpiled by Babel and bundled by Webpack. The node debugger was not working reliably. Inspecting by logging to console has been a pragmatic workaround.

To inspect the timing and application flow timut provides a 'trash-it-after-usage' approach and simply a quick way to get an idea about where some code spends its time.