Margaret Lanterman

A file logging library.

Name: Margaret Lanterman - the Log Lady of Twin Peaks

Installation

npm i margaret-lanterman

Basic Usage

import lanterman from 'margaret-lanterman';

lanterman.buffer.addStream(); // @todo - finish off docs

// This will write a section
lanterman.section('Start doing thing X', async () => {
	lanterman.write('Did a thing...', 'note'); // prefixes [note] to the line
	lanterman.write('Did another thing...', 'note');

	// a sub-section
	lanterman.section('A sub-section!', async () => {

		// JSON encodes an object, prefixes [response] to the line
		lanterman.write({ url: 'request.com' }, 'response');
		lanterman.write({ response: 'response from request' }, 'response');
	});
});

Feedback

You can set up a feedback handler to respond to messages - for example to provide some visual feedback.

lanterman.feedback.withFeedback(
	(message, verbosity) => verbosity === 'verbose' ? stderr.write(`[info] ${message}`),
	() => {
		lanterman.feedback('Sending API Request', 'verbose');
		// ... send a request
		lanterman.feedback('Response was succesful', 'superVerbose');
	},
);

Uses the gloucester verbosity library for verbosity levels.

Integration with Commander JS

If you are using CommanderJS, there is a helper function to set up commander.

This will set up two log streams - in the directory ~/.logs/{program-name}. One "latest" file, which will always have the latest log stream in it, and one "per-diem" file which will have a file for every execution, and will prune any files from previous calendar days.

import { Command } from 'commander';
import lanterman from 'lanterman';
import setupCommanderAndLanterman from 'margaret-lanterman/lib/integrations/commander';

const program = new Command();

program.name('my-cli-app');

setupCommanderAndLanterman(program, gloucester);

Integration with Listr2

If you are using Listr2, there are two helper functions to wrap tasks.

Wrap Tasks In Sections

This will take a list of tasks, and will wrap each of the tasks inside a lanterman.section() call, using the task's title as the section title.

import {Listr, delay} from 'listr2';
import {wrapListrTasksInSections} from 'margaret-lanterman/lib/integrations/listr';

new Listr(wrapListrTasksInSections([
	{
		title: 'Task One',
		async task() {
			lanterman.write('Some details about task one');
		}
	},
	{
		title: 'Task Two',
		async task() {
			lanterman.write('Some details about task two');
		}
	},
]));

Wrap Tasks In Feedback

This will take a list of tasks, and will wrap each of the tasks inside a lanterman.feedback.withFeedback() call, adding a few features:

• Sets future tasks to gray
• Sets current tasks to cyan
• Sets completed tasks to green
• Appends any calls to feedback lanterman.feedback.send() to the title (overwriting any previous ones) - or if using a verbosity level higher than normal, puts it on a line below (using task.output)

import {Listr, delay} from 'listr2';
import {wrapListrTasksInFeedback} from 'margaret-lanterman/lib/integrations/listr';

new Listr(wrapListrTasksInFeedback([
	{
		title: 'Task One',
		async task() {
			await delay(1000);
			lanterman.feedback.send('Doing a thing', 'verbose');
			await delay(1000);
			lanterman.feedback.send('Trying somethign else', 'verbose');
			await delay(1000);
		}
	},
]));

All

If you want to use both of these, there is a wrapListrTasks function which will apply both of these.

import {Listr, delay} from 'listr2';
import {wrapListrTasks} from 'margaret-lanterman/lib/integrations/listr';

new Listr(wrapListrTasks([]));