llog
v0.3.3
Published
simple multi-level application logging module
Downloads
966
Readme
llog
llog is a simple logging module. llog is intended for use in applications, not libraries. It provides level-based debugging (trace|debug|info|warn|error|fatal). note: For adding detailed debug statements in libraries, use TJ's
debug module.
As an added bonus, llog provides a simple upgrade path from plain text to json logging via either pino or bunyan. llog
will look for pino
or bunyan
at load time and automatically use them as a provider if present. This means you can write your apps with non-json logging when starting, and auto-magically all your logs to json logging with a simple npm i --save pino
.
level-based debugging
To use llog:
var log = require('llog');
log.debug('some info'); // prints 'debug some info' to stdout
log.info('some info'); // prints 'info some info' to stdout
log.warn('some info'); // prints 'warn some info' to stdout
log.error('some error'); // prints 'error some error' to stdout
Specifying log levels
llog is based on debug, and as such extends the same mechanism debug uses to specify which items to log - using process.env.DEBUG
.
To log only info, warn, and error in llog, for instance, you would run your application similar to the following:
DEBUG=info,warn,error node app.js
Adding trace is simple:
DEBUG=trace,info,warn,error node app.js
Magic upgrade to json logging
llog will automatically detect if bunyan
or pino
are installed as a peer dependency and, if so, automatically upgrade to json logging. Because bunyan
and pino
use process.env.LOG_LEVEL
as a level indicator (as opposed to debug
's DEBUG
variable, the steps for moving to json logs are:
npm install --save buynan
ornpm install --save pino
- execute your application using
LOG_LEVEL=10 node app
instead of usingDEBUG
. Higher levels are always included when specifying a level.
Bunyan log levels can be found at https://github.com/trentm/node-bunyan#levels. Pino log levels can be found at https://github.com/pinojs/pino
This is particularly useful when moving from early stages of application development to having unified logging via Logstash or Splunk.