@spreaker/logger

Spreaker Node Logger

Pino-based Logger that we use in all of Spreaker's nodeJS projects. It's possible to create 2 different instances of the logger:

• Application Logger: this is an instance of Pino Logger, initialized with Spreaker's base options, with a custom serializer that take care of stringify objects and array and transforming other types into string before log. The Application Logger add also a loglevel field to each log to have a string version of the log level.
• Access Logger: this is an instance of Pino Logger, initialized with Spreaker's base options.

Both the loggers provide a common functions:

• Errors serializer that take care of set the correct error fields in the logs

Install package

npm install @spreaker/logger

Usage

const { createLogger } = require("@spreaker/logger");

// Create logger passing the initial properties and the options: minLoglevel and destination. 
// Mandatory properties are: 
// - `type` (name of application is using the logger)
// - `context` (type of logger to create: "app" or "access")
// Options object is not mandatory:
// - `minLoglevel` (Minimum level of log enabled; if not passed is "info")
// - `destination` (Path of logs destination; useful for tests. If not passed STDOUT is the default one)
const logger = createLogger({ type: "test", context: "app"}, { minLoglevel: "info" });

// Use the logger as a simple `Pino` child instance.
logger.info({a: "b", c: 123}, "Message to log");

const loggerWithDestination = createLogger({ type: "test", context: "app"}, { destination: "/path/to/file" });

// Logs will be saved into this file: /path/to/file
loggerWithDestination.info({a: "b", c: 123}, "Message to log");

Logger types

Application logger

Create logger passing the property context: "app" if you want to create an Application Logger.

const { createLogger } = require("@spreaker/logger");
const logger = createLogger({ type: "test", context: "app"});

This instance has a custom serializer that that take care of stringify objects and array and transforming other types into string before log.

logger.info(
    {"object":{"b":123},"string":"c","number":321,"array":["a",1,["subarray"]]}, 
    "Stringify log"
);

will produce a log like:

{ 
   "level":30,
   "time":1566908680683,
   "type":"test",
   "context":"app",
   "message":"Stringify log",
   "loglevel":"INFO",
   "object":"{\"b\":123}",
   "string":"c",
   "number":"321",
   "array":"[\"a\",1,[\"subarray\"]]",
   "v":1
}

Access logger

Create logger passing the property context: "access" if you want to create an Access Logger.

const { createLogger } = require("@spreaker/logger");
const logger = createLogger({ type: "test", context: "access"});

This instance is a normal Pino child instance, initialized with Spreaker's base options.

logger.info(
    {"object":{"b":123},"string":"c","number":321,"array":["a",1,["subarray"]]}, 
    "Don't stringify log"
);

will produce a log like:

{
   "level":30,
   "time":1566908953080,
   "type":"test",
   "context":"access",
   "object":{
      "b":123
   },
   "string":"c",
   "number":321,
   "array":[
      "a",
      1,
      [
         "subarray"
      ]
   ],
   "message":"Don't stringify log",
   "v":1
}

Errors serializer

When you pass an error to the logger the possible scenarios are:

• the Error is passed as mergingObject (first param) and the log has already a message (second param). In this case the logger add field "error_message": Error.message and common error fields to the mergingObject and the log message is printed in the message field

logger.error(new Error("This is an error"), "Error with log message");

will produce a log like:

{
   "level":50,
   "time":1567688722065,
   "type":"test",
   "context":"app",
   "loglevel":"ERROR",
   "message":"Error with log message",
   "error_message":"This is an error",
   "error_stack":"Error: This is an error...",
   "error_file":"...",
   "error_line":"...",
   "v":1
}

• the Error is passed as mergingObject (first param) and the log has not a message (second param). In this case the logger add common error fields to the mergingObject and use the Error.message as log message

logger.error(new Error("Error without log message");

will produce a log like:

{
   "level":50,
   "time":1567688853208,
   "type":"test",
   "context":"app",
   "loglevel":"ERROR",
   "message":"Error without log message",
   "error_stack":"Error: Error without log message...",
   "error_file":"...",
   "error_line":"...",
   "v":1
}

• the Error is passed as log message (second param). In this case the logger add common error fields to the mergingObject and use the Error.message as log message

logger.error({ "a": "b" }, new Error("Error as log message"));

will produce a log like:

{
   "level":50,
   "time":1567689070052,
   "type":"test",
   "context":"app",
   "loglevel":"ERROR",
   "message":"Error as log message",
   "a":"b",
   "error_stack":"Error: Error as log message...",
   "error_file":"...",
   "error_line":"...",
   "v":1
}

Common error fields

In application logs

{
    error_stack: "error stack trace",
    error_code: "error code property",
    error_file: "path of the file",
    error_line: "number of the line"
}

or in access logs

{
    error_stack: "error stack trace",
    error_code: "error code property",
    error_file: "path of the file",
    error_line: "number of the line"
}