npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2024 – Pkg Stats / Ryan Hefner

power-merge

v1.6.1

Published

Merge multiple objects and/or arrays in a configurable way by defining rules that govern whether an attribute should be cloned, referenced, recursed, mutated or ignored

Downloads

25

Readme

⚡ Power Merge ⚡

NPM version NPM downloads Node.js CI Code Climate Test Coverage Dependency Status devDependencies Status

power-merge is a library for custom merging of two or more documents. If your merge requirments are simple and you are happy with how Ramda's mergeDeepLeft or Lodash's merge works with regard to:

  • arrays
  • null values
  • undefined values
  • inherited properties
  • non enumerable properties
  • functions
  • regular expressions
  • dates
  • objects with a clone function, e.g. moment.clone
  • circular references
  • mutability

then you're probably better off using one of those libraries. They will be faster, use fewer system resources and are heavily battle tested. However it your merge requirements are somewhat bespoke, then you've come to the right place.

TL;DR

const pm = require('power-merge');
const { ignoreNull, deepClone } = pm.ruleSets;

// Compile the rules
const merge = pm.compile({ rules: [ ignoreNull, deepClone ] });

// Define the documents
const a = {
  poll: {
    delay: null,
    frequency: '5s',
  },
};

const b = {
  poll: {
    delay: '1m',
    frequency: '10s',
  },
};

// Merge the documents
const result = merge(a, b);

Result:

{
  poll: {
    frequency: '5s',
  }
};

But wait, there's more...

A rules driven merge libary wouldn't be much use if you couldn't compose your own rules. Here's how...

const pm = require('power-merge');
const { ignoreNull, deepClone } = pm.ruleSets;
const { and, eq, unionWith } = pm.commands;
const R = require('ramda');

// Compose a new rule
const unionHostsByIp = {
  when: eq('node.name', 'hosts'),
  then: unionWith(R.eqBy(R.prop('ip'))),
};

// Compile the rules
const merge = pm.compile({ rules: [ unionHostsByIp, ignoreNull, deepClone ] });

// Define the documents
const a = {
  poll: {
    delay: null,
    frequency: '5s',
  },
  hosts: [
    { ip: '192.168.1.100', port: 80 },
    { ip: '192.168.1.200', port: 80 },
  ],
};

const b = {
  poll: {
    delay: '1m',
    frequency: '10s',
  },
  hosts: [
    { ip: '192.168.1.100', port: 8080 },
    { ip: '192.168.1.101', port: 8080 },
  ],
};

// Merge the documents
const result = merge(a, b);

Result:

{
  poll: {
    frequency: '5s',
  },
  hosts: [
    { ip: '192.168.1.100', port: 80 },
    { ip: '192.168.1.101', port: 8080 },
    { ip: '192.168.1.200', port: 80 },
  ],
};

How power-merge works

API

power-merge is intended to be configurable. In addition to the merge rules you can specify

  • a synchronous(default) or asynchronous interface.
  • whether the merge should be left-to-right(default) or right-to-left
  • whether the arguments should be varardic(default) or an array.
const merge = pm.compile({
  api: {
    synchronous: false,
    direction: 'right-to-left',
    varardic: false
  },
  rules: [ ... ],
});

merge([d, c, b, a], (err, result) => {
  // profit
});

The merge function can be promisified if you prefer.

Rules

power-merge operates on an array of rules. A rule is comprised of zero or one when conditions and exactly one then condition.

const { eq, clone } = pm.commands;
const rules = [
  {
    when: eq('a.value', undefined),
    then: clone('b.value'),
  },
  {
    then: clone('a.value'),
  },
];

The when conditions are tested in order until one passes, after which the associated then condition is invoked. The result of the then condition will normally be the merge result, but could be the pm.noop token to instruct the merge function to skip over the current node instead of merging it.

To support re-use you can also provide nested arrays of rules which power-merge will automatically flatten, e.g.

const baseRules = [ rule1, rule2, rule3 ];
const customRules = [ rule4, rule5 ];
const rules = [ customRules, baseRules ];

RuleSets

power-merge ships with the following preconfigured rulesets

alwaysIgnore

Will always pass, and do nothing. Useful to place at the end of your rules array if you don't like the default behaviour of throwing an error when no rules pass.

const { alwaysIgnore } = pm.ruleSets;
const rules = [ customRules, alwaysIgnore ];

deepClone

Uses a combination of recurseKeys, iterate and clone commands to recursively clone two documents own properties.

const { deepClone } = pm.ruleSets;
const rules = [ customRules, deepClone ];

errorOnCircularReference

Does what it says on the tin

const { errorOnCircularReference } = pm.ruleSets;
const rules = [ customRules, errorOnCircularReference, deepClone ];

errorOnNoMatchingRules

Also does what it says on the tin. This rule is automatically added to the end of every ruleset. If you want different behaviour finish your ruleset with something like alwaysIgnore

Facts

Facts is a document are passed to each when and then condition. The facts are...

{
  a: {
    value: '30s',
    type: 'String',
    circular: false,
  },
  b: {
    value: '1m',
    type: 'String',
    circular: false,
  },
  node: {
    depth: 3,
    name: 'delay',
    path: 'poll.delay',
  },
};

when conditions are used to check the facts. If they return true, the then condition will be executed. then conditions typically reference, clone or descend into the fact's a.value and/or b.value, but could also be written to perform operations upon any of the facts.

Context

The context contains information about the current merge. It records the depth, current node name and path. It also contains a reference to the merge function that is used to recursively descend into objects or iterate over arrays. Unless you're writing your own commands, you won't need to know about the context.

Commands

Commands are the functions which operate on facts. You specify them in the when and then conditions. e.g.

const { eq, clone } = pm.commands;
const rules = [
  {
    when: eq('a.value', 'foo'),
    then: clone('a.value'),
  },
];

references two commands, eq and clone. The eq command takes two parameters, path and value. It uses the path to extract data from the facts and compares it to the value, returning true if they are equal, and false otherwise.

The clone takes one parameter, path. It clones the data located at the specified path and returns it to the merge operation. Several commands are included with power-merge. It is also easy to write your own custom commands.

always

Always execute the then command.

const { always, clone } = pm.commands;
const rules = [
  {
    when: always(),
    then: clone('a.value'),
  },
];

Since when will default to always, you can achieve the same result by omitting the when clause altogether.

and

Boolean AND multiple commands. e.g.

const { and, eq, clone } = pm.commands;
const rules = [
  {
    when: and([
      eq('a.type', 'String'),
      eq('b.type', 'String')
    ]),
    then: clone('a.value'),
  },
];

clone

Clones the value specified by the path parameter using Ramda's clone function.

const { clone } = pm.commands;
const rules = [
  {
    then: clone('a.value')
  },
];

compose

Composes a chain of commands so the output from one will be passed to the next. This is useful post processing tasks such as sorting arrays, e.g.

const { and, eq, compose, union, invoke } = pm.commands;
const rules = [
  {
    when: and([
      eq('a.type', 'Array'),
      eq('b.type', 'Array')
    ]),
    then: compose([
      union(),
      invoke(R.sort((a, b) => a.localeCompare(b.ip)))
    ]),
  }
];

debug

Useful for debuging output to the consule while developing your merge rules, however use with care since this command will also cause the current node to be omitted from the merged document. The first parameter is a hogan.js template, the second (optional) parameter is a logger in case you want to direct output somewhere other than the console.

const { debug } = pm.commands;
const rules = [
  {
    then: debug('A: {{value.a}}, B: {{value.b}}'),
  },
];

eq

Compares the value located at the given path with the second parameter, returning true if they are equal and false otherwise

const { eq, clone } = pm.commands;
const rules = [
  {
    when: eq('a.type', 'Number'),
    then: clone('a.value'),
  },
];

error

Throws an error constructed from the given hogan.js template

const { error } = pm.commands;
const rules = [
  {
    then: error('Boom! A: {{value.a}}, B: {{value.b}}'),
  },
];

gt

Compares the value located at the given path with the second parameter, returning true if it is greater and false otherwise

const { gt, clone } = pm.commands;
const rules = [
  {
    when: gt('a.value', 10),
    then: clone('a.value'),
  },
];

gte

Compares the value located at the given path with the second parameter, returning true if it is greater or equal and false otherwise

const { gte, clone } = pm.commands;
const rules = [
  {
    when: gte('a.value', 10),
    then: clone('a.value'),
  },
];

ignore

Ignores a part of the document, e.g.

const { eq, ignore } = pm.commands;
const rules = [
  {
    when: eq('a.value', 'do-not-want'),
    then: ignore(),
  },
];

invoke (inline)

Invokes an inline function.

const { invoke } = pm.commands;
const rules = [
  {
    when: invoke(facts => facts.value.a === 'yes'),
    then: invoke(facts => true),
  },
];

invoke (named)

Invokes a named function.

const { invoke } = pm.commands;
const rules = [
  {
    when: invoke('isYes'),
    then: invoke('truism'),
  },
];
pm.compile({ rules }, {
  isYes: facts => facts.value.a === 'yes',
  truism: facts => true,
});

iterate

Iterates over two arrays, merging each item. If the arrays are different lengths, overflowing items will be merged against undefined.

const { and, eq, iterate } = pm.commands;
const rules = [
  {
    when: and([
      eq('a.type', 'Array'),
      eq('b.type', 'Array'),
    ]),
    then: iterate(),
  },
];

lt

Compares the value located at the given path with the second parameter, returning true if it is less and false otherwise

const { lt, clone } = pm.commands;
const rules = [
  {
    when: lt('a.value', 10),
    then: clone('a.value'),
  },
];

lte

Compares the value located at the given path with the second parameter, returning true if it is less or equal and false otherwise

const { lte, clone } = pm.commands;
const rules = [
  {
    when: lte('a.value', 10),
    then: clone('a.value'),
  },
];

matches

Tests the value located at the given path against a regex.

const { matches, clone } = pm.commands;
const rules = [
  {
    when: matches('a.value', /foo/i),
    then: clone('a.value'),
  },
];

ne

Compares the value located at the given path with the second parameter, returning false if they are equal and true otherwise

const { ne } = pm.commands;
const rules = [
  {
    when: ne('a.type', 'Number'),
    then: clone('a.value'),
  },
];

never

Never execute the then command.

const { never, clone } = pm.commands;
const rules = [
  {
    when: never(),
    then: clone('a.value'),
  },
];

Only useful for tests or to temporarily disable a rule.

or

Boolean OR multiple commands. e.g.

const { or, eq, clone } = pm.commands;
const rules = [
  {
    when: or([
      eq('a.type', 'String'),
      eq('a.type', 'Number'),
    ]),
    then: clone('a.value'),
  },
]

recurseKeys

Recursively merge union of Object.keys('a') and Object.keys('b'), i.e. all enumerable own properties. Only sensible when both 'a' and 'b' are objects (use the iterate command to recurively merge two arrays).

const { and, eq, recurseKeys } = pm.commands;
const rules = [
  {
    when: and([
      eq('a.type', 'Object'),
      eq('b.type', 'Object'),
    ]),
    then: recurseKeys(),
  },
];

recurseKeysIn

Recursively merge union of R.keysIn('a') and R.keysIn('b'), i.e. all enumerable own and inherited properties. Only sensible when both 'a' and 'b' are objects (use the iterate command to recurively merge two arrays).

const { and, eq, recurseKeysIn } = pm.commands;
const rules = [
  {
    when: and([
      eq('a.type', 'Object'),
      eq('b.type', 'Object'),
    ]),
    then: recurseKeysIn(),
  },
];

reference

References the value specified by the path parameter.

const { reference } = pm.commands;
const rules = [
  {
    then: reference('a.value'),
  },
];

union

Union the "a" and "b" values using Ramda's union function (which also dedupes). Only sensible when both "a" and "b" are arrays.

const { and, eq, union } = pm.commands;
const rules = [
  {
    when: and([
      eq('a.type', 'Array'),
      eq('b.type', 'Array'),
    ]),
    then: union(),
  },
];

unionWith

Union the "a" and "b" values using Ramda's unionWith function (which dedupes based on the result of the given function). Only sensible when both "a" and "b" are arrays.

const { and, eq, unionWith } = pm.commands;
const rules = [
  {
    when: and([
      eq('a.type', 'Array'),
      eq('b.type', 'Array'),
    ]),
    then: unionWith(function(v) {
      return v.id,
    }),
  },
];

Custom Commands

power-merge commands are easy to write, once you understand that they must be expressed as 3 levels of nested functions.

module.exports = function one(param1, param2) {
  return function two(context) {
    return function three(facts) {
      return result; // or pm.noop
    }
  }
};
  1. The outer function takes the command's configuration parameters
  2. The middle function takes the context
  3. The inner function takes the facts, e.g.
const debug = require('debug')('power-merge:commands:highlight');

module.exports = function __highlight(str) {

  return function _highlight(context) {

    return function highlight(facts) {

      debug('path: %o, facts: %o', path, facts);
      const result = str + view(facts) + str;

      debug('return: %o', result);
      return result;
    }
  }
};

Commands that should cause an attribute to be ignored, rather than merged should return the special pm.noop token. i.e.

const debug = require('debug')('power-merge:commands:log');
const noop = require('pm').noop;

module.exports = function __log(text) {

  return function _log(context) {

    return function log(facts) {
      debug('path: %o, facts: %o', path, facts);
      console.log(text);
      return noop;
    }
  }
};

Paths

Several of the bundled commands take a path parameter to locate a value within the facts. In the readme and examples this is always expressed as a dotpath, e.g. a.value, however under the hood this is converted to an array ['a', 'value'], which is passed to Ramda's lensPath function. If you can't use dots in your path for any reason, you can pass in an array, e.g.

const { eq, ignore } = pm.commands;
const rules = [
  {
    when: eq(['a', 'value'], 'nothing to see here'),
    then: ignore(),
  },
];

Circular References

If you use the clone command circular references will be handled automatically. Clone uses Ramda's clone function under the hood, which makes a copy of the original item, but uses references to the copy if they are encountered in a circular context. If you prefer some other action, then you can explicitly handle circular references as with any other fact.

Debugging

power-merge uses debug. You can enable as follows...

DEBUG='power-merge:*' node your-application.js