Custom logging to handle measurements and counts. We use this module as a universal logging method across all hostinng platforms. We abstract logging into this module to allow us to change the underlying reporting mechanism easily.
The function call causes a command line statistic to be produced, which can be read by CloudWatch.
Each function has a property, an assigned value, and optional tags. The tag is a key:value object, or array of such objects (see example below).
l.count(property, value, tags);To return the string value of the log, instead of sending it to the console log, the functions returnCount, returnMeasure, and returnTime can be used.
If 'process.env.AWS_LAMBDA_FUNCTION_NAME' is set, then the 'host' tag is set to "lambda" and the 'lambdaFunction' tag is set to the name of the lambda function.
The code:
const l = new require("bristlr-logger")();
l.count("accountCreated");
l.count("accountCreated", 2);
l.count("accountCreated", 2, "tag");
l.count("accountCreated", 2, {"animal":"doggo"});
l.measure("someStat", 100);
l.measure("someStat", 100, [{"foo": "baa"}, {"animals": "cat"}]);
l.time("someStat", 100);The output (0000000000 has been used in place of the timestamp):
MONITORING|0000000000|1|count|accountCreated|
MONITORING|0000000000|2|count|accountCreated|
MONITORING|0000000000|2|count|accountCreated|#tags:tag
MONITORING|0000000000|2|count|accountCreated|#animal:doggo
MONITORING|0000000000|100|measure|someStat|
MONITORING|0000000000|100|measure|someStat|#foo:baa,animals:cat
MONITORING|0000000000|100ms|measure|someStat|