Index

lognestmonster / 0a557d5

A general-purpose single-header C logging library and parser for event-based logs. (Incomplete)

Latest Commit

{#}TimeHashSubjectAuthor#(+)(-)GPG?
5928 Dec 2018 00:060a557d5Update logger init for queue compactionJoshua161N

Tree

ObjectLatest Commit SubjectTimeHash
foldersrcUpdate logger init for queue compaction28 Dec 2018 00:060a557d5
folderstaticLogo25 Dec 2018 19:585a96206
foldertestsUpdate test.js28 Dec 2018 00:0542e63e1
file.gitignoreUpdate .gitignore27 Dec 2018 21:409f3fa60
file.travis.ymlCreate .travis.yml27 Dec 2018 21:41cd4024b
fileLICENSECreate LICENSE26 Dec 2018 17:019046b13
fileREADME.mdUpdate README.md27 Dec 2018 21:400bd1bbf
fileindex.jsUpdate index.js26 Dec 2018 13:2526d3797
filepackage-lock.jsonCreate package-lock.json25 Dec 2018 21:19478c0cd
filepackage.jsonUpdate package.json27 Dec 2018 21:402754d5b

README

Advanced node.js logging for advanced programs.

  1. lognestmonster
  2. Installation
  3. Classes
    1. Logger.Logger
    2. Logger.Queue
    3. Logger.Statement
    4. Logger.Event
  4. Verbosity Levels
  5. Sample Usage
    1. Notes
  6. Log Format
  7. Contributors
    1. License

lognestmonster

Most loggers available only use a linear method of logging; there are verbosity levels and tags to narrow searches, but everything is still on the same level or plane nonetheless. The package lognestmonster is a similar type of logger, but it allows you to create multiple layers (or "nests") of log statements. This is useful because, although you may have to put in some extra organizational work on the code side, the log results are much more clean and thorough, boosting workflow efficiency. The user has absolute control over the way log statements are pushed to their log files.

Why should I use this? There are many projects that create insane amounts of data, almost impossible to sift through without a helper program. The purpose of this logging system is to be a time-saver. Although it takes more time to put it in place, it's almost immediately made up with the performance gain through using the new log format. One thing that's unique about this logger compared to others is that it allows multiple queues to be made, in turn allowing you to split up your data. For example, a Node web server could keep one log file that records everything that the backend does while it uses another to record user or traffic information for analytics. Parsing software could be used to read either one.

Installation

In your npm initiated package, type the following:

npm i lognestmonster

That's it! You can now require("lognestmonster") and begin using it.

Classes

There are 4 classes offered by the package: Logger, the driving device that organizes everything; Queue, the class that actually pushes the log statements to their respective file; Statement, the actual log data (timestamp, verbosity, tag/invoker, message); and Event, a nest layer for Statements or other Events.

The following subsections assume that lognestmonster has been required by node.js with the following code:

const Logger = require("lognestmonster");

Logger

Creates and organizes your queues. Logger is the exported package. Logger.Logger is the Logger class.

Logger.Logger()

var MyLogger = new Logger.Logger(Object config);

Where Object config defaults to:

{"name": "Logger", "locations": {"node": "./log/node"}}

name serves no purpose other than identification. locations is used to create new queues, taking each key as the queue name and each value as the queue output location.

Logger.Logger.queue()

MyLogger.queue(string name);

This returns the appropriate Queue object for the provided name, assuming it was created with the Logger object.

Queue

Manages log Statements and Events and how they're written to the final log file. Note that these are implicitly created with Logger.Logger when the locations object is properly provided in the config parameter of the Logger.Logger constructor.

Logger.Queue()

let MyQueue = new Logger.Queue(string name, string location);
// note that parameters are the same format as key-value
// pairs in `config.locations` of `Logger.Logger(config)`

This creates the queue, taking name to be used as its ID and location as the path to where the log file should be created.

Logger.Queue.push()

MyQueue.push(Statement statement);
MyQueue.push(Event event);
MyQueue.push(string verbosity, string tag, string message); // implicitly creates a `Statement` object

This adds log items or nests to the to-write queue. This returns the Queue object.

Logger.Queue.write()

MyQueue.write();

This appends every queue value to the log file, emptying the queue. This returns the Queue object.

Statement

This is the base log item where written log data is actually held.

Logger.Statement()

let MyStatement = Logger.Statement(string verbosity, string tag, string message);

The timestamp value is created automatically by the constructor.

Event

This is the proper name for a nest. Essentially, it's just an array that can hold other Event objects and Statement objects, creating a tree.

Logger.Event()

let MyEvent = Logger.Event();
let MyEvent = Logger.Event(Event event);
let MyEvent = Logger.Event(Statement statement);
let MyEvent = Logger.Event(string verbosity, string tag, string message);

Any arguments given are passed to this.push().

Logger.Event.push()

MyEvent.push(Event event);
MyEvent.push(Statement statement);
MyEvent.push(string verbosity, string tag, string message);

This message pushes an Event or Statement as items in the nest. In the case that 3 strings are given as arguments, a Statement is implicitly created. This returns the Event object.

Verbosity Levels

When creating a Statement, you can pass anything you'd like for the first verbosity string, although there are some ones preset by the package:

Logger.INFO Logger.DEBUG Logger.VERBOSE Logger.VERYVERBOSE Logger.WARNING Logger.ERROR

These verbosity levels can be used to narrow down your search results when parsing the log files.

Sample Usage

Here's an example of how somebody would initiate the Logger, create and push items to the Queue, and write them to the log file. PLEASE SEE THE NOTES ABOUT THIS EXAMPLE DOWN BELOW.

// Require the package
const Logger = require("lognestmonster");

// Creates the logger object. Placing the new Logger inside the package allows cross-file usage, so you only have to initiate once.
Logger.Overseer = new Logger.Logger({
    name: "Overseer",
    locations: {
        "node": "./log/node" // Creates a queue named `node` that uses the path `./log/node`
    },
});

// Pushes a statement directly to the `node` queue
Logger.Overseer.queue("node").push(Logger.INFO, "PROCESS", "Process started");

// Creates a new event and pushes a Statement to it
let LoadEvent = new Logger.Event();
LoadEvent.push(Logger.INFO, "INIT", "Acquiring needed top-level packages.");

// Creates a new event and pushes multiple Statements to it
let LowerNestedEvent = new Logger.Event();

LowerNestedEvent.push(Logger.INFO, "INIT", "Loading fs...");
LowerNestedEvent.push(Logger.DEBUG, "INIT", "fs loaded.");

LowerNestedEvent.push(Logger.INFO, "INIT", "Loading http...");
LowerNestedEvent.push(Logger.DEBUG, "INIT", "http loaded.");

LowerNestedEvent.push(Logger.INFO, "INIT", "Loading jsonwebtoken...");
LowerNestedEvent.push(Logger.DEBUG, "INIT", "jsonwebtoken loaded.");

LowerNestedEvent.push(Logger.INFO, "INIT", "Loading lognestmonster...");
LowerNestedEvent.push(Logger.DEBUG, "INIT", "lognestmonster loaded.");

// Pushes the Event LowerNestedEvent to the Event LoadEvent
LoadEvent.push(LowerNestedEvent);

// Pushes another statement to LoadEvent
LoadEvent.push(Logger.INFO, "INIT", "Finished.");

// Pushes LoadEvent (a nest that consists of [Statement, Event, Statement] now) to the write queue
Logger.Overseer.queue("node").push(LoadEvent);

// Queue should now look like this: [Statement, Event]

// Writes the queue, effectively emptying it into the log file
Logger.Overseer.queue("node").write();

The above code creates the following JSON-like log output in the designated file:

{"timestamp":"2018-12-26T18:08:37.654Z","verbosity":"INFO","tag":"PROCESS","message":"Process started"}
[{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"INFO","tag":"INIT","message":"Acquiring needed top-level packages."},[{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"INFO","tag":"INIT","message":"Loading fs..."},{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"DEBUG","tag":"INIT","message":"fs loaded."},{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"INFO","tag":"INIT","message":"Loading http..."},{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"DEBUG","tag":"INIT","message":"http loaded."},{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"INFO","tag":"INIT","message":"Loading jsonwebtoken..."},{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"DEBUG","tag":"INIT","message":"jsonwebtoken loaded."},{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"INFO","tag":"INIT","message":"Loading lognestmonster..."},{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"DEBUG","tag":"INIT","message":"lognestmonster loaded."}],{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"INFO","tag":"INIT","message":"Finished."}]

To see how to parse this data (putting it into proper JSON), see the Log Format section.

Notes about this example

Large projects If your project spans multiple files, you could easily place your Logger object into the package itself, allowing the same Logger object to be accessed by other parts of your project. This is seen in the sample code with Logger.Overseer = new Logger.Logger(...).

Repetition There is some repetitive code; specifically, Logger.Overseer.queue("node"). Do note that this actually results in a Queue object, so you could easily make this its own variable like this:

let NodeQueue = Logger.Overseer.queue("node");
NodeQueue.push(...).write();

Queue pushing and writing When you're pushing multiple objects to a queue, be wary that they will stay there until the queue is written. Because everything is its own class, what you're really pushing is a reference to the real object, so you can make changes to a pushed object after it has been pushed. As such, it is entirely possible to prematurely write a queue before an already-pushed Event or Statement is finished. It's good practice to immediately write the queue immediately after it has been pushed. Perhaps in the future there will be some functionality where a Statement or Event can be directly written instead of placed in the queue.

Push order The developer has control over everything here. The order of your log file is the order that you push to your Events and Queue. If I were to push a new Statement to LoadEvent in the middle of the pushes to LowerNestedEvent, it would show up on the log first because LowerNestedEvent isn't itself pushed to LoadEvent until later in the code. The order of the log file is 100% logical and in the developer's control, so it would be wise to write your code neatly with this in mind. This logging package doesn't do anything you don't tell it to.

Log Format

Logs are controlled by Queue objects. Placed in the folder they're told, the logs should follow the ISO datetime format and have a .log file extension. Here's the name of a sample log file: 2018-12-26T18-08-37-653Z.log

Regarding format, logs follow a JSON-like format, as follows:

{"timestamp":"2018-12-26T18:08:37.654Z","verbosity":"INFO","tag":"PROCESS","message":"Process started"}
[{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"INFO","tag":"INIT","message":"Acquiring needed top-level packages."},[{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"INFO","tag":"INIT","message":"Loading fs..."},{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"DEBUG","tag":"INIT","message":"fs loaded."},{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"INFO","tag":"INIT","message":"Loading http..."},{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"DEBUG","tag":"INIT","message":"http loaded."},{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"INFO","tag":"INIT","message":"Loading jsonwebtoken..."},{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"DEBUG","tag":"INIT","message":"jsonwebtoken loaded."},{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"INFO","tag":"INIT","message":"Loading lognestmonster..."},{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"DEBUG","tag":"INIT","message":"lognestmonster loaded."}],{"timestamp":"2018-12-26T18:08:37.655Z","verbosity":"INFO","tag":"INIT","message":"Finished."}]

Because of the way the Queue objects push it (the most efficient way regarding computing power with changing/appending to files), you will have to do a bit of tweaking to get it in proper JSON format:

  1. Add a comma before every newline (except for the last one at the end of the file)
  2. Wrap everything in table brackets ('[' and ']')

This could be easily automated. Once finished, you get proper JSON (below), where each object {} is a Statement and each wrapping table [] is an Event (excluding the outermost one).

[
  {
    "timestamp": "2018-12-26T05:55:23.360Z",
    "verbosity": "INFO",
    "tag": "PROCESS",
    "message": "Process started"
  },
  [
    {
      "timestamp": "2018-12-26T05:55:23.360Z",
      "verbosity": "INFO",
      "tag": "INIT",
      "message": "Acquiring needed top-level packages."
    },
    [
      {
        "timestamp": "2018-12-26T05:55:23.360Z",
        "verbosity": "INFO",
        "tag": "INIT",
        "message": "Loading fs..."
      },
      {
        "timestamp": "2018-12-26T05:55:23.360Z",
        "verbosity": "DEBUG",
        "tag": "INIT",
        "message": "fs loaded."
      },
      {
        "timestamp": "2018-12-26T05:55:23.360Z",
        "verbosity": "INFO",
        "tag": "INIT",
        "message": "Loading http..."
      },
      {
        "timestamp": "2018-12-26T05:55:23.360Z",
        "verbosity": "DEBUG",
        "tag": "INIT",
        "message": "http loaded."
      },
      {
        "timestamp": "2018-12-26T05:55:23.360Z",
        "verbosity": "INFO",
        "tag": "INIT",
        "message": "Loading jsonwebtoken..."
      },
      {
        "timestamp": "2018-12-26T05:55:23.360Z",
        "verbosity": "DEBUG",
        "tag": "INIT",
        "message": "jsonwebtoken loaded."
      },
      {
        "timestamp": "2018-12-26T05:55:23.360Z",
        "verbosity": "INFO",
        "tag": "INIT",
        "message": "Loading lognestmonster..."
      },
      {
        "timestamp": "2018-12-26T05:55:23.360Z",
        "verbosity": "DEBUG",
        "tag": "INIT",
        "message": "lognestmonster loaded."
      }
    ],
    {
      "timestamp": "2018-12-26T05:55:23.360Z",
      "verbosity": "INFO",
      "tag": "INIT",
      "message": "Finished."
    }
  ]
]

The above can be taken in by parsing software and create something similar to the following:

[[LOG FILE NAME]]
[[LOG DATE]]
[[LOG FILE SIZE]]
[[LOG ITEM COUNT]]

TIMESTAMP - INFO - PROCESS - Process started
v 3 ITEMS
    TIMESTAMP - INFO - INIT - Acquiring needed top-level packages.
    v 6 ITEMS
        TIMESTAMP - INFO - INIT - Loading fs...
        TIMESTAMP - DEBUG - INIT - fs loaded.
        TIMESTAMP - INFO - INIT - Loading http...
        TIMESTAMP - DEBUG - INIT - http loaded.
        TIMESTAMP - INFO - INIT - Loading jsonwebtoken...
        TIMESTAMP - DEBUG - INIT - jsonwebtoken loaded.
    TIMESTAMP - INFO - INIT - Finished.

This is exciting! You still have the ability to narrow down your results with timestamps, verbosity levels, and tags/invokers, but now you have the organization with collapsable nests to not be overwhelmed with large amounts of data at the same time.

Contributors

Developer - Joshua 'joshuas3' Stockin <joshstockin@gmail.com> (https://www.github.com/joshuas3)

Name - Patrik 'Patrola' Xop (https://github.com/PatrikXop)

License

This software is licensed under version 3 of the GNU General Public License.

lognestmonster (c) 2018 Joshua 'joshuas3' Stockin

This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.

This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
GNU General Public License for more details.

You should have received a copy of the GNU General Public License
along with this program.  If not, see <https://www.gnu.org/licenses/>.

The license can be found here.