The 'timm' npm package is a small, fast, and pure immutable collections library. It provides utilities for working with immutable data structures, making it easier to manage state in JavaScript applications.
Immutable Object Update
This feature allows you to update an object immutably. The `set` function creates a new object with the updated value, leaving the original object unchanged.
const timm = require('timm');
const obj = { a: 1, b: 2 };
const newObj = timm.set(obj, 'b', 3);
console.log(newObj); // { a: 1, b: 3 }Deep Merge
The `merge` function allows you to deeply merge two objects, creating a new object that combines the properties of both input objects.
const timm = require('timm');
const obj1 = { a: 1, b: { c: 2 } };
const obj2 = { b: { d: 3 } };
const mergedObj = timm.merge(obj1, obj2);
console.log(mergedObj); // { a: 1, b: { c: 2, d: 3 } }Array Push
This feature allows you to add an element to the end of an array immutably. The `addLast` function creates a new array with the added element, leaving the original array unchanged.
const timm = require('timm');
const arr = [1, 2, 3];
const newArr = timm.addLast(arr, 4);
console.log(newArr); // [1, 2, 3, 4]Array Remove
The `removeAt` function allows you to remove an element from an array immutably. It creates a new array without the specified element, leaving the original array unchanged.
const timm = require('timm');
const arr = [1, 2, 3];
const newArr = timm.removeAt(arr, 1);
console.log(newArr); // [1, 3]The 'immutable' package provides persistent immutable data structures, including List, Map, Set, and Record. It is more comprehensive and feature-rich compared to 'timm', but also larger in size and complexity.
The 'immer' package allows you to work with immutable state by using a 'draft' state that you can modify directly. It is more intuitive for developers who are used to mutable operations, but it uses a different approach compared to 'timm'.
The 'seamless-immutable' package provides immutable data structures that are backwards-compatible with normal JavaScript arrays and objects. It is similar to 'timm' in terms of simplicity and ease of use, but it offers a different API.
Immutability helpers with fast reads and acceptable writes (blog post)
$ npm install --save timm
I know, I know... the world doesn't need yet another immutability library, especially with the likes of ImmutableJS and seamless-immutable around.
And yet... I felt the urge, at least just to cover my limited needs. ImmutableJS is a solid, comprehensive and highly-performant solution, but this power comes at a price: mixing up ImmutableJS's Maps and Lists with your plain objects can cause some friction, and reading those objects (in my case, more often than writing them) isn't that convenient.
On the other side, seamless-immutable solves the "friction" problem by using plain objects and arrays, but seems to have some performance issues (at least in my benchmarks, see below).
timm's approach: use plain objects and arrays and provide simple mutation functions to handle most common operations (suggestions are welcome!). As a bonus, timm creates new objects lazily, when it confirms that the operation will mutate the input object; in other words, operations that don't modify an object always return the object itself. This alleviates pressure on the garbage collector and lets you easily check whether an object changed after an operation: merge(obj, { a: 3 }) === obj.
Important notice: timm does not freeze the objects it provides. In other words, it doesn't protect you against inadvertently modifying them in your code. I considered deeply freezing all objects with Object.freeze(), but it is really slow. Then I considered doing this only in development (like seamless-immutable), but then modifying frozen objects will silently fail in development (unless you use strict in your code), and –worse still– succeed in production. Not good. In conclusion, be careful (or send me a suggestion to work around this!).
...Oh, I almost forgot! It's tiny: just ~200 LOC and ~1.5 kB minified & compressed!
I prepared an initial benchmarking tool comparing read/write speeds in four cases:
All four solutions are first verified for consistency (the mutable solution obviously does not pass all tests) and then benchmarked. Benchmarks cover reading and writing object attributes at different nesting levels (root level, 2 levels and 5 levels deep), merging two small objects, and replacing an object in a 1000-long array.
Feel free to run them yourself (download the repo and then npm install && npm run benchmarks). These are my results on a Windows machine for 200k iterations (Node v6.2.0):
Some conclusions from these benchmarks:
Reads are on par with native objects/arrays and seamless-immutable, and faster than ImmutableJS (the deeper, the faster). In fact, you cannot go faster than native objects for reading!
Writes are much slower than in-place edits, as expected, but are much faster than seamless-immutable (even in production mode), both for objects and arrays. Compared to ImmutableJS, object writes and merges are faster (the deeper, the faster), whereas array writes are way slower (not as slow as seamless-immutable, though). For timm and seamless-immutable, write times degrade linearly with array length (and probably object size), but much more slowly for ImmutableJS (logarithmically?). This is where ImmutableJS really shines.
Hence, what I recommend (from top to bottom):
// Consume the CommonJS ES5 module (default)
import { merge } from 'timm';
const obj = timm.merge({ a: 2 }, { b: 3 });
// Consume the native ES6 module (requires bundling)
import { merge } from 'timm/lib/timm_es6';
const obj = merge({ a: 2 }, { b: 3 });
Note: you can use the ES6 module version with Webpack 2 and its tree-shaking feature to get really small bundles, including just the timm functions you use. Make sure you disable babel-plugin-transform-es2015-modules-commonjs in your babel-loader configuration. More details here.
Returns a new array with an appended item or items.
Usage: addLast<T>(array: Array<T>, val: Array<T>|T): Array<T>
arr = ['a', 'b']
arr2 = addLast(arr, 'c')
// ['a', 'b', 'c']
arr2 === arr
// false
arr3 = addLast(arr, ['c', 'd'])
// ['a', 'b', 'c', 'd']
Returns a new array with a prepended item or items.
Usage: addFirst<T>(array: Array<T>, val: Array<T>|T): Array<T>
arr = ['a', 'b']
arr2 = addFirst(arr, 'c')
// ['c', 'a', 'b']
arr2 === arr
// false
arr3 = addFirst(arr, ['c', 'd'])
// ['c', 'd', 'a', 'b']
Returns a new array removing the last item.
Usage: removeLast<T>(array: Array<T>): Array<T>
arr = ['a', 'b']
arr2 = removeLast(arr)
// ['a']
arr2 === arr
// false
// The same array is returned if there are no changes:
arr3 = []
removeLast(arr3) === arr3
// true
Returns a new array removing the first item.
Usage: removeFirst<T>(array: Array<T>): Array<T>
arr = ['a', 'b']
arr2 = removeFirst(arr)
// ['b']
arr2 === arr
// false
// The same array is returned if there are no changes:
arr3 = []
removeFirst(arr3) === arr3
// true
Returns a new array obtained by inserting an item or items at a specified index.
Usage: insert<T>(array: Array<T>, idx: number, val: Array<T>|T): Array<T>
arr = ['a', 'b', 'c']
arr2 = insert(arr, 1, 'd')
// ['a', 'd', 'b', 'c']
arr2 === arr
// false
insert(arr, 1, ['d', 'e'])
// ['a', 'd', 'e', 'b', 'c']
Returns a new array obtained by removing an item at a specified index.
Usage: removeAt<T>(array: Array<T>, idx: number): Array<T>
arr = ['a', 'b', 'c']
arr2 = removeAt(arr, 1)
// ['a', 'c']
arr2 === arr
// false
// The same array is returned if there are no changes:
removeAt(arr, 4) === arr
// true
Returns a new array obtained by replacing an item at a specified index. If the provided item is the same as (referentially equal to) the previous item at that position, the original array is returned.
Usage: replaceAt<T>(array: Array<T>, idx: number, newItem: T): Array<T>
arr = ['a', 'b', 'c']
arr2 = replaceAt(arr, 1, 'd')
// ['a', 'd', 'c']
arr2 === arr
// false
// The same object is returned if there are no changes:
replaceAt(arr, 1, 'b') === arr
// true
The following types are used throughout this section
type ArrayOrObject = Array<any>|Object;
type Key = number|string;
Returns a value from an object at a given path. Works with
nested arrays and objects. If the path does not exist, it returns
undefined.
Usage: getIn(obj: ?ArrayOrObject, path: Array<Key>): any
obj = { a: 1, b: 2, d: { d1: 3, d2: 4 }, e: ['a', 'b', 'c'] }
getIn(obj, ['d', 'd1'])
// 3
getIn(obj, ['e', 1])
// 'b'
Returns a new object with a modified attribute. If the provided value is the same as (referentially equal to) the previous value, the original object is returned.
Usage: set<T>(obj: ?T, key: Key, val: any): T
obj = { a: 1, b: 2, c: 3 }
obj2 = set(obj, 'b', 5)
// { a: 1, b: 5, c: 3 }
obj2 === obj
// false
// The same object is returned if there are no changes:
set(obj, 'b', 2) === obj
// true
Returns a new object with a modified nested attribute.
Notes:
Usage: setIn<T: ArrayOrObject>(obj: T, path: Array<Key>, val: any): T
obj = { a: 1, b: 2, d: { d1: 3, d2: 4 }, e: { e1: 'foo', e2: 'bar' } }
obj2 = setIn(obj, ['d', 'd1'], 4)
// { a: 1, b: 2, d: { d1: 4, d2: 4 }, e: { e1: 'foo', e2: 'bar' } }
obj2 === obj
// false
obj2.d === obj.d
// false
obj2.e === obj.e
// true
// The same object is returned if there are no changes:
obj3 = setIn(obj, ['d', 'd1'], 3)
// { a: 1, b: 2, d: { d1: 3, d2: 4 }, e: { e1: 'foo', e2: 'bar' } }
obj3 === obj
// true
obj3.d === obj.d
// true
obj3.e === obj.e
// true
// ... unknown paths create intermediate keys. Numeric segments are treated as array indices:
setIn({ a: 3 }, ['unknown', 0, 'path'], 4)
// { a: 3, unknown: [{ path: 4 }] }
Returns a new object with a modified attribute, calculated via a user-provided callback based on the current value. If the calculated value is the same as (referentially equal to) the previous value, the original object is returned.
Usage: update<T: ArrayOrObject>(obj: T, key: Key, fnUpdate: (prevValue: any) => any): T
obj = { a: 1, b: 2, c: 3 }
obj2 = update(obj, 'b', (val) => val + 1)
// { a: 1, b: 3, c: 3 }
obj2 === obj
// false
// The same object is returned if there are no changes:
update(obj, 'b', (val) => val) === obj
// true
Returns a new object with a modified nested attribute, calculated via a user-provided callback based on the current value. If the calculated value is the same as (referentially equal to) the previous value, the original object is returned.
Usage: updateIn<T: ArrayOrObject>(obj: T, path: Array<Key>, fnUpdate: (prevValue: any) => any): T
obj = { a: 1, d: { d1: 3, d2: 4 } }
obj2 = updateIn(obj, ['d', 'd1'], (val) => val + 1)
// { a: 1, d: { d1: 4, d2: 4 } }
obj2 === obj
// false
// The same object is returned if there are no changes:
obj3 = updateIn(obj, ['d', 'd1'], (val) => val)
// { a: 1, d: { d1: 3, d2: 4 } }
obj3 === obj
// true
Returns a new object built as follows: the overlapping keys from the
second one overwrite the corresponding entries from the first one.
Similar to Object.assign(), but immutable.
Usage:
merge(obj1: Object, obj2: ?Object): Objectmerge(obj1: Object, ...objects: Array<?Object>): ObjectThe unmodified obj1 is returned if obj2 does not provide something
new to obj1, i.e. if either of the following
conditions are true:
obj2 is null or undefinedobj2 is an object, but it is emptyobj2 are referentially equal to the
corresponding attributes of objobj1 = { a: 1, b: 2, c: 3 }
obj2 = { c: 4, d: 5 }
obj3 = merge(obj1, obj2)
// { a: 1, b: 2, c: 4, d: 5 }
obj3 === obj1
// false
// The same object is returned if there are no changes:
merge(obj1, { c: 3 }) === obj1
// true
Similar to merge(), but merging the value at a given nested path.
Note that the returned type is the same as that of the first argument.
Usage:
mergeIn<T: ArrayOrObject>(obj1: T, path: Array<Key>, obj2: ?Object): TmergeIn<T: ArrayOrObject>(obj1: T, path: Array<Key>, ...objects: Array<?Object>): Tobj1 = { a: 1, d: { b: { d1: 3, d2: 4 } } }
obj2 = { d3: 5 }
obj3 = mergeIn(obj1, ['d', 'b'], obj2)
// { a: 1, d: { b: { d1: 3, d2: 4, d3: 5 } } }
obj3 === obj1
// false
// The same object is returned if there are no changes:
mergeIn(obj1, ['d', 'b'], { d2: 4 }) === obj1
// true
Returns an object excluding one or several attributes.
Usage: omit(obj: Object, attrs: Array<string>|string): Object
obj = { a: 1, b: 2, c: 3, d: 4 }
omit(obj, 'a')
// { b: 2, c: 3, d: 4 }
omit(obj, ['b', 'c'])
// { a: 1, d: 4 }
// The same object is returned if there are no changes:
omit(obj, 'z') === obj1
// true
Returns a new object built as follows: undefined keys in the first one
are filled in with the corresponding values from the second one
(even if they are null).
Usage:
addDefaults(obj: Object, defaults: Object): ObjectaddDefaults(obj: Object, ...defaultObjects: Array<?Object>): Objectobj1 = { a: 1, b: 2, c: 3 }
obj2 = { c: 4, d: 5, e: null }
obj3 = addDefaults(obj1, obj2)
// { a: 1, b: 2, c: 3, d: 5, e: null }
obj3 === obj1
// false
// The same object is returned if there are no changes:
addDefaults(obj1, { c: 4 }) === obj1
// true
Copyright (c) Guillermo Grau Panea 2016-present
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.
1.3.0 (October 3, 2017)
FAQs
Immutability helpers with fast reads and acceptable writes
The npm package timm receives a total of 0 weekly downloads. As such, timm popularity was classified as not popular.
We found that timm demonstrated a not healthy version release cadence and project activity because the last version was released a year ago. It has 1 open source maintainer collaborating on the project.
Did you know?
Socket for GitHub automatically highlights issues in each pull request and monitors the health of all your open source dependencies. Discover the contents of your packages and block harmful activity before you install or update your dependencies.
Security News
Research
A malicious npm package disguised as a WhatsApp client is exploiting authentication flows with a remote kill switch to exfiltrate data and destroy files.
Security News
PyPI now supports digital attestations, enhancing security and trust by allowing package maintainers to verify the authenticity of Python packages.
Security News
GitHub removed 27 malicious pull requests attempting to inject harmful code across multiple open source repositories, in another round of low-effort attacks.