Allowing libaries to participate in logging without dictating anything about that logging system. Because logging is a process-level decision. More information about thehelp
.
- Libraries can be logging-agnostic, just call
verbose
,info
,warn
anderror
on the returnedlogger
objects. - By default, these four logging libraries will be tried, in this order:
bunyan
-verbose
is mapped to itsdebug
level.log4js
-verbose
mapped to itsdebug
level.winston
(Note: Added second-to-last because it is installed byprompt
, now at the top level because of npm@3)debug
- no differentiation between log levels. (Note: Added last because it is likely already installed in your project already thanks tomocha
.)
- Library consumers can customize a library's logging system by overriding:
logger
- to provided a custom static logger objectloadLogger(moduleName)
- for dynamic, per-module logger generation
- If a library isn't using the same version of
thehelp-log-shim
that you have installed, you can usegetAllVersions()
to get access to all versions of this package loaded in your process
First install the project as a dependency:
npm install thehelp-log-shim --save
Then start using it:
var logShim = require('thehelp-log-shim');
var logger = logShim('my-module-name');
logger.info('blah');
If no logging system is installed, that logger.info()
call will do nothing. Install one of the four logging systems supported by default, and that call will automatically use it.
The first level of configuration is by providing your own logger
object:
var logShim = require('thehelp-log-shim');
logShim.logger = {
error: function(item) {
console.error('error:', item);
},
warn: function(item) {
console.warn('warn:', item);
}
};
var logger = logShim('my-module-name');
logger.info('info message');
logger.warn('something is flaky!');
The logger.info()
call will do nothing, and the warn()
call will be piped through your function to console.warn
.
Note: Notice that the provided logger
only specified two levels, but the code calls info()
. thehelp-log-shim
will ensure that all levels are present on any logger
it provides.
Now let's get a little more advanced. You have a number of modules, and you want to do different things for each of them:
var logShim = require('thehelp-log-shim');
logShim.loadLogger = function(moduleName) {
if (moduleName === 'left')) {
return {
verbose: console.log,
info: console.log,
warn: console.log,
error: console.error
}
}
};
var left = logShim('left');
left.warn('warning! will be shown');
var right = logShim('right');
right.warn('warning! will NOT be shown');
Since we're looking at moduleName
in our loadLogger()
override, the two logShim()
calls result in different behavior. It's okay to return null or a half-constructed object - the rest of the levels will be filled in.
Okay, now you're using a few different libraries, all of which use thehelp-log-shim
. Things are easy as long as you're okay with them all jumping in on the first logging system they find.
Things get interesting when you want to hook in for deeper behavior. npm
is a very flexible system, so you might have multiple versions of thehelp-log-shim
installed in the tree of modules under your node_modules
directory.
Time to write some code:
var logShim = require('thehelp-log-shim');
// how many are installed?
console.log(logShim.countAll());
// this is an array of string versions
var versions = logShim.getAllVersions();
// set all versions to one logger object
versions.forEach(function(version) {
logShim.applyToVersion(version, function() {
logShim.logger = myCustomLogger;
});
});
Of course, this is a bit dangerous as it doesn't take into account the differences of all the various versions. But it will get you started.
Always include some sort of post-require()
configuration option. Without it, you'll remove your users' ability to do more complex customization of thehelp-log-shim
. They can't override loadLogger()
if you never call it again after loading your library.
Because the module names you pass to logShim(moduleName)
will be global to the process, consider using something like 'library-name'. If you have a reason to subdivide your project into a few sections: 'library-name:module-name'.
Because each module will be using a named logger from the default container (via winston.loggers.get()
), you'll need to do a little more work to set the default transports.
For example, this is how you might share the console transport between default container loggers as well as the default top-level logger (direct winston.info()
calls):
var winston = require('winston');
var options = {
colorize: true,
timestamp: function() {
var date = new Date();
return date.toJSON();
},
level: 'verbose'
};
var transport = new winston.transports.Console(options);
// configure default transports for all loggers in default container
winston.loggers.options.transports = [transport];
// configure the top level default logger
var instantiatedAlready = true;
winston.remove(winston.transports.Console);
winston.add(transport, null, instantiatedAlready);
You can also set custom settings for a specific named logger from the default container:
winston.loggers.add('module-name', {
transports: [
// perhaps a console transport with a different label, or a different level?
]
})
Out of the box, bunyan
doesn't support global configuration of specific loggers, so all thehelp-log-shim
users will be logging to the console at the default 'info' level.
But, via a monkey-patch of bunyan.createLogger()
you can start centralizing. Here's an example of setting your current log level in one place for all loggers:
var bunyan = require('bunyan');
var create = bunyan.createLogger;
bunyan.createLogger = function(options) {
options.level = 'debug';
// modify streams key here to customize where each log entry goes...
return create(options);
};
More discussion of central bunyan
configuration: trentm/node-bunyan#116
Just a few things to remember here:
- all
debug
output is to stderr - you enable a given module name by putting it in the DEBUG environment variable, which is a comma- or space-delimited list
- enabled/disabled for a given module name is only activated on process load
log4js
is like winston
except for one key difference - levels are on the logger, not on the transport. So the way to customize the levels for the various modules in your appplication is modify the logger objects themselves:
var custom = log4js.getLogger('module-name');
custom.setLevel('WARN');
Detailed docs be found at this project's GitHub Pages, thanks to groc
: http://thehelp.github.io/log-shim/src/server/index.html
When you have some changes ready, please submit a pull request with:
- Justification - why is this change worthwhile? Link to issues, use code samples, etc.
- Documentation changes for your code updates. Be sure to check the groc-generated HTML with
grunt doc
- A description of how you tested the change. Don't forget about the very-useful
npm link
command :0)
I may ask you to use a git rebase
to ensure that your commits are not interleaved with commits already in the history. And of course, make sure grunt
completes successfully (take a look at the requirements for thehelp-project
). :0)
(The MIT License)
Copyright (c) 2014 Scott Nonnenberg <[email protected]>
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.