Gathering detailed insights and metrics for cellx-lite
Gathering detailed insights and metrics for cellx-lite
npm install cellx-lite
Typescript
Module System
Node Version
NPM Version
JavaScript (69.29%)
TypeScript (18.32%)
CSS (6.91%)
HTML (5.48%)
Love this project? Help keep it running — sponsor us today! 🚀
Total Downloads
10,375
Last Day
3
Last Week
31
Last Month
83
Last Year
482
2 Stars
25 Commits
2 Watching
1 Branches
1 Contributors
Minified
Minified + Gzipped
Latest Version
1.8.1
Package Id
cellx-lite@1.8.1
Unpacked Size
1.09 MB
Size
244.06 kB
File Count
48
NPM Version
6.4.1
Node Version
10.10.0
Cumulative downloads
Total Downloads
Last day
0%
3
Compared to previous day
Last week
-3.1%
31
Compared to previous week
Last month
84.4%
83
Compared to previous month
Last year
-45.5%
482
Compared to previous year
Ultra-fast implementation of reactivity for javascript.
The following command installs cellx as a npm package:
npm install cellx --save
1let user = { 2 firstName: cellx('Matroskin'), 3 lastName: cellx('Cat'), 4 5 fullName: cellx(function() { 6 return (this.firstName() + ' ' + this.lastName()).trim(); 7 }) 8}; 9 10user.fullName('subscribe', function() { 11 console.log('fullName: ' + this.fullName()); 12}); 13 14console.log(user.fullName()); 15// => 'Matroskin Cat' 16 17user.firstName('Sharik'); 18user.lastName('Dog'); 19// => 'fullName: Sharik Dog'
Despite the fact that the two dependencies of the cell fullName
has been changed, event handler worked only once.
Important feature of cellx is that it tries to get rid of unnecessary calls
of the event handlers as well as of unnecessary calls of the dependent cells calculation formulas.
In combination with some special optimizations, this leads to an ideal speed of calculation of
the complex dependencies networks.
One test, which is used for measuring the performance, generates grid with multiply "layers" each of which is composed of 4 cells. Cells are calculated from the previous layer of cells (except the first one, which contains initial values) by the formula A2=B1, B2=A1-C1, C2=B1+D1, D2=C1. After that start time is stored, values of all first layer cells are changed and time needed to update all last layer cells is measured. Test results (in milliseconds) for different number of layers (for Google Chrome 53.0.2785.116 (64-bit)):
Library ↓ \ Number of computed layers → | 10 | 20 | 30 | 50 | 100 | 1000 | 5000 | 25000 |
---|---|---|---|---|---|---|---|---|
cellx | <~1 | <~1 | <~1 | <~1 | <~1 | 4 | 20 | 100 |
VanillaJS (naive) | <~1 | 15 | 1750 | >300000 | >300000 | >300000 | >300000 | >300000 |
Knockout | 10 | 750, increases in subsequent runs | 67250, increases in subsequent runs | >300000 | >300000 | >300000 | >300000 | >300000 |
$jin.atom | 2 | 3 | 3 | 4 | 6 | 40 | 230 | 1100 |
$mol_atom | <~1 | <~1 | <~1 | 1 | 2 | 20 | RangeError: Maximum call stack size exceeded | RangeError: Maximum call stack size exceeded |
Warp9 | 2 | 3 | 4 | 6 | 10 | 140 | 900, increases in subsequent runs | 4200, increases in subsequent runs |
Reactor.js | <~1 | <~1 | 2 | 3 | 5 | 50 | 230 | >300000 |
Reactive.js | <~1 | <~1 | 2 | 3 | 5 | 140 | RangeError: Maximum call stack size exceeded | RangeError: Maximum call stack size exceeded |
Kefir.js | 25 | 2500 | >300000 | >300000 | >300000 | >300000 | >300000 | >300000 |
MobX | <~1 | <~1 | <~1 | 2 | 3 | 40 | RangeError: Maximum call stack size exceeded | RangeError: Maximum call stack size exceeded |
Matreshka.js | 11 | 1150 | 143000 | >300000 | >300000 | >300000 | >300000 | >300000 |
Test sources can be found in the folder perf. Density of connections in real applications is usually lower than in the present test, that is, if a certain delay in the test is visible in 100 calculated cells (25 layers), in a real application, this delay will either be visible in the greater number of cells, or cells formulas will include some complex calculations (e.g., computation of one array from other).
Cells can be stored in the variables:
1let num = cellx(1); 2let plusOne = cellx(() => num() + 1); 3 4console.log(plusOne()); 5// => 2
or in the callable properties:
1function User(name) { 2 this.name = cellx(name); 3 this.nameInitial = cellx(function() { return this.name().charAt(0).toUpperCase(); }); 4} 5 6let user = new User('Matroskin'); 7 8console.log(user.nameInitial()); 9// => 'M'
including in the prototype:
1function User(name) { 2 this.name(name); 3} 4User.prototype.name = cellx(); 5User.prototype.friends = cellx(() => []); // each instance of the user will get its own instance of the array 6 7let user1 = new User('Matroskin'); 8let user2 = new User('Sharik'); 9 10console.log(user1.friends() == user2.friends()); 11// => false
or in simple properties:
1function User(name) { 2 cellx.define(this, { 3 name: name, 4 nameInitial: function() { return this.name.charAt(0).toUpperCase(); } 5 }); 6} 7 8let user = new User('Matroskin'); 9 10console.log(user.nameInitial); 11// => 'M'
Use npm module cellx-decorators.
Use npm module cellx-react.
When you create a cell, you can pass some options:
Additional processing of value during reading:
1// array that you can't mess up accidentally, the messed up thing will be a copy 2let arr = cellx([1, 2, 3], { 3 get: arr => arr.slice() 4}); 5 6console.log(arr()[0]); 7// => 1 8 9arr()[0] = 5; 10 11console.log(arr()[0]); 12// => 1
Used to create recordable calculated cells:
1function User() { 2 this.firstName = cellx(''); 3 this.lastName = cellx(''); 4 5 this.fullName = cellx(function() { 6 return (this.firstName() + ' ' + this.lastName()).trim(); 7 }, { 8 put: function(name) { 9 name = name.split(' '); 10 11 this.firstName(name[0]); 12 this.lastName(name[1]); 13 } 14 }); 15} 16 17let user = new User(); 18 19user.fullName('Matroskin Cat'); 20 21console.log(user.firstName()); 22// => 'Matroskin' 23console.log(user.lastName()); 24// => 'Cat'
Validates the value during recording and calculating.
Validation during recording into the cell:
1let num = cellx(5, { 2 validate: value => { 3 if (typeof value != 'number') { 4 throw new TypeError('Oops!'); 5 } 6 } 7}); 8 9try { 10 num('I string'); 11} catch (err) { 12 console.log(err.message); 13 // => 'Oops!' 14} 15 16console.log(num()); 17// => 5
Validation during the calculation of the cell:
1let value = cellx(5); 2 3let num = cellx(() => value(), { 4 validate: value => { 5 if (typeof value != 'number') { 6 throw new TypeError('Oops!'); 7 } 8 } 9}); 10 11num('subscribe', err => { 12 console.log(err.message); 13}); 14 15value('I string'); 16// => 'Oops!' 17 18console.log(value()); 19// => 'I string' 20 21console.log(num()); 22// => 5
Calling the cell method is somewhat unusual — the cell itself is called, the first argument passes the method name,
rest ones — the arguments. In this case, there must be at least one argument, or call of the cell will be counted as its
recording. If the method has no arguments, you need to transfer an additional undefined
with a call or to shorten it
just 0
(see dispose
).
Adds a change listener:
1let num = cellx(5); 2 3num('addChangeListener', evt => { 4 console.log(evt); 5}); 6 7num(10); 8// => { prevValue: 5, value: 10 }
Removes previously added change listener.
Adds a error listener:
1let value = cellx(1); 2 3let num = cellx(() => value(), { 4 validate: v => { 5 if (v > 1) { 6 throw new TypeError('Oops!'); 7 } 8 } 9}); 10 11num('addErrorListener', evt => { 12 console.log(evt.error.message); 13}); 14 15value(2); 16// => 'Oops!'
Removes previously added error listener.
Subscribes to the events change
and error
. First argument comes into handler is an error object, second — an event.
1user.fullName('subscribe', (err, evt) => { 2 if (err) { 3 // 4 } else { 5 // 6 } 7});
Unsubscribes from events change
and error
.
cellx.define
Subscribe to changes in the properties created with help of cellx.define
possible through EventEmitter
:
1class User extends cellx.EventEmitter { 2 constructor(name) { 3 cellx.define(this, { 4 name, 5 nameInitial: function() { return this.name.charAt(0).toUpperCase(); } 6 }); 7 } 8} 9 10let user = new User('Matroskin'); 11 12user.on('change:nameInitial', evt => { 13 console.log('nameInitial: ' + evt.value); 14}); 15 16console.log(user.nameInitial); 17// => 'M' 18 19user.name = 'Sharik'; 20// => 'nameInitial: S'
In many reactivity engines calculated cell (atom, observable-property) should be seen as a normal event handler for other cells, that is, for "killing" the cell it is not enough to simply remove all handlers from it and lose the link to it, it is also necessary to decouple it from its dependencies. Calculated cells in cellx constantly monitor the presence of handlers for themselves and all their descendants, and in cases of their (handlers) absence went to the passive updates mode, i.e. unsubscribe themselves from their dependencies and are evaluated immediately upon reading. Thus, to "kill" of the cell you just calculated remove from it all handlers added before and forget the link to it; you do not need to think about the other cells, from which it is calculated or which are calculated from it. After this, garbage collector will clean everything.
You can call the dispose
, just in case:
1user.name('dispose', 0);
This will remove all the handlers, not only from the cell itself, but also from all cells calculated from it, and in the absence of links all branch of dependencies will "die".
To minimize redraw of UI cellx may "collapse" several events into one. Link to the previous event is stored in
evt.prevEvent
:
1let num = cellx(5); 2 3num('addChangeListener', evt => { 4 console.log(evt); 5}); 6 7num(10); 8num(15); 9num(20); 10// => { 11// prevEvent: { 12// prevEvent: { 13// prevEvent: null 14// prevValue: 5, 15// value: 10 16// } 17// prevValue: 10, 18// value: 15 19// } 20// prevValue: 15, 21// value: 20 22// }
In cases when the cell comes to the initial value before generation of event, it does not generate it at all:
1let num = cellx(5); 2 3num('addChangeListener', evt => { 4 console.log(evt); 5}); 6 7num(10); 8num(15); 9num(5); // return the original value 10// but there's nothing here
Upon changing the number of the calculated cell dependencies, it is evaluated only once and creates only one event:
1let inited = false; 2let num1 = cellx(5); 3let num2 = cellx(10); 4let sum = cellx(() => { 5 if (inited) { 6 console.log('sum.formula'); 7 } 8 9 return num1() + num2(); 10}); 11 12sum('addChangeListener', evt => { 13 console.log(evt); 14}); 15 16inited = true; 17 18num1(10); 19num2(15); 20// => 'sum.formula' 21// => { 22// prevEvent: null 23// prevValue: 15, 24// value: 25 25// }
Calculated cell formula can be written so that a set of dependencies may change over time. For example:
1let user = { 2 firstName: cellx(''), 3 lastName: cellx(''), 4 5 name: cellx(function() { 6 return this.firstName() || this.lastName(); 7 }) 8};
There, while firstName
is still empty string, cell name
is signed for firstName
and lastName
,
and change in any of them will lead to the change in its value. If you assign to the firstName
some not empty
string, then during recalculation of value name
it simply will not come to reading lastName
in the formula,
i.e. the value of the cell name
from this moment will not depend on lastName
.
In such cases, cells automatically unsubscribe from dependencies insignificant for them and are not recalculated
when they change. In the future, if the firstName
again become an empty string, the cell name
will re-subscribe
to the lastName
.
1let foo = cellx(() => localStorage.foo || 'foo', { 2 put: function(value) { 3 localStorage.foo = value; 4 this.push(value); 5 } 6}); 7 8let foobar = cellx(() => foo() + 'bar'); 9 10console.log(foobar()); // => 'foobar' 11console.log(localStorage.foo); // => undefined 12foo('FOO'); 13console.log(foobar()); // => 'FOObar' 14console.log(localStorage.foo); // => 'FOO'
1let request = (() => { 2 let value = 1; 3 4 return { 5 get: url => new Promise((resolve, reject) => { 6 setTimeout(() => { 7 resolve({ 8 ok: true, 9 value 10 }); 11 }, 1000); 12 }), 13 14 put: (url, params) => new Promise((resolve, reject) => { 15 setTimeout(() => { 16 value = params.value; 17 18 resolve({ 19 ok: true 20 }); 21 }, 1000); 22 }) 23 }; 24})(); 25 26let foo = cellx(function(cell, next = 0) { 27 request.get('http://...').then((res) => { 28 if (res.ok) { 29 cell.push(res.value); 30 } else { 31 cell.fail(res.error); 32 } 33 }); 34 35 return next; 36}, { 37 put: (value, cell, next) => { 38 request.put('http://...', { value: value }).then(res => { 39 if (res.ok) { 40 cell.push(value); 41 } else { 42 cell.fail(res.error); 43 } 44 }); 45 } 46}); 47 48foo('subscribe', () => { 49 console.log('New foo value: ' + foo()); 50 foo(5); 51}); 52 53console.log(foo()); 54// => 0 55 56foo('then', () => { 57 console.log(foo()); 58}); 59// => 'New foo value: 1' 60// => 1 61// => 'New foo value: 5'
If you record to the cell an instance of class which inherits of cellx.EventEmitter
,
then the cell will subscribe to its change
event and will claim it as own:
1let value = cellx(new cellx.EventEmitter()); 2 3value('subscribe', (err, evt) => { 4 console.log(evt.target instanceof cellx.EventEmitter); 5}); 6 7value().emit('change'); 8// => true
Due to this, you can create your collections, upon updating those collections you will update the cell containing them and dependent cells will be recalculated. Two such collections already is added to the cellx:
The short syntax to create:
1let map = cellx.map({ 2 key1: 1, 3 key2: 2, 4 key3: 3 5});
cellx.ObservableMap
repeats
Map from ECMAScript 2015,
except for the following differences:
cellx.EventEmitter
and generates an event change
when changing their records;contains
, which let you know whether or not the value is contained in the map,
without going over all of its values;clone
, which creates a copy of map;Short creation syntax:
1let list = cellx.list([1, 2, 3]);
Like cellx.ObservableMap
, list generates an event change
upon any change of its records.
During initialization the list may take option comparator
, which will implement the assortment of its values:
1let list = cellx.list([ 2 { x: 5 }, 3 { x: 1 }, 4 { x: 10 } 5], { 6 comparator: (a, b) => { 7 if (a.x < b.x) { return -1; } 8 if (a.x > b.x) { return 1; } 9 return 0; 10 } 11}); 12 13console.log(list.toArray()); 14// => [{ x: 1 }, { x: 5 }, { x: 10 }] 15 16list.addRange([{ x: 100 }, { x: -100 }, { x: 7 }]); 17 18console.log(list.toArray()); 19// => [{ x: -100 }, { x: 1 }, { x: 5 }, { x: 7 }, { x: 10 }, { x: 100 }]
If instead of comparator
you pass the option sorted
with the value true
, it will use the standard comparator
:
1let list = cellx.list([5, 1, 10], { sorted: true }); 2 3console.log(list.toArray()); 4// => [1, 5, 10] 5 6list.addRange([100, -100, 7]); 7 8console.log(list.toArray()); 9// => [-100, 1, 5, 7, 10, 100]
Length of the list. Read-only.
Function for comparing values in the sorted list. Read-only.
Whether or not the list is sorted. Read-only.
Important difference between list and array is that the list can't contain so-called "holes"
that is, when it will try to read or set the value of the index beyond the existing range of elements,
an exception will be generated.
Range extension (adding of items) occurs through methods add
, addRange
, insert
and insertRange
.
In such case, in the last two methods passed index
can not be longer than the length of the list.
Sorted list suggests that its values are always in sorted order. Methods
set
, setRange
, insert
and insertRange
are contrary to this statement, they either will break the correct order
of sorting or (for preservation of this order) will install/paste past the specified index, i.e.
will not work properly. Therefore, when you call the sorted list, they always generate an exception. It is possible to
add values to the sorted list through the methods add
and addRange
, or during initialization of the list.
Type signature: (value) -> boolean;
.
Checks if the value is in the list. In cases of a large amount of values in the list it may be significantly faster
than list.indexOf(value) != -1
.
Type signature: (value, fromIndex?: int) -> int;
.
Type signature: (value, fromIndex?: int) -> int;
.
Type signature: (index: int) -> *;
.
Type signature: (index: int, count?: uint) -> Array;
.
If count
is unspecified it makes copies till the end of the list.
Type signature: (index: int, value) -> cellx.ObservableList;
.
Type signature: (index: int, values: Array) -> cellx.ObservableList;
.
Type signature: (value, unique?: boolean) -> cellx.ObservableList;
.
Type signature: (values: Array, unique?: boolean) -> cellx.ObservableList;
.
Type signature: (index: int, value) -> cellx.ObservableList;
.
Type signature: (index: int, values: Array) -> cellx.ObservableList;
.
Type signature: (value, fromIndex?: int) -> boolean;
.
Removes the first occurrence of value
in the list.
Type signature: (value, fromIndex?: int) -> boolean;
.
It removes all occurrences of value
list.
Type signature: (values: Array, fromIndex?: int) -> boolean;
.
Type signature: (index: int) -> *;
.
Type signature: (index: int, count?: uint) -> Array;
.
If count
is unspecified it will remove everything till the end of the list.
Type signature: () -> cellx.ObservableList;
.
Type signature: (separator?: string) -> string;
.
Type signature: (cb: (item, index: uint, list: cellx.ObservableList), context?);
.
Type signature: (cb: (item, index: uint, list: cellx.ObservableList) -> *, context?) -> Array;
.
Type signature: (cb: (item, index: uint, list: cellx.ObservableList) -> ?boolean, context?) -> Array;
.
Type signature: (cb: (item, index: uint, list: cellx.ObservableList) -> ?boolean, context?) -> *;
.
Type signature: (cb: (item, index: uint, list: cellx.ObservableList) -> ?boolean, context?) -> int;
.
Type signature: (cb: (item, index: uint, list: cellx.ObservableList) -> ?boolean, context?) -> boolean;
.
Type signature: (cb: (item, index: uint, list: cellx.ObservableList) -> ?boolean, context?) -> boolean;
.
Type signature: (cb: (accumulator, item, index: uint, list: cellx.ObservableList) -> *, initialValue?) -> *;
.
Type signature: (cb: (accumulator, item, index: uint, list: cellx.ObservableList) -> *, initialValue?) -> *;
.
Type signature: () -> cellx.ObservableList;
.
Type signature: () -> Array;
.
Type signature: () -> string;
.
No vulnerabilities found.
Reason
no binaries found in the repo
Reason
0 existing vulnerabilities detected
Reason
Found 0/25 approved changesets -- score normalized to 0
Reason
0 commit(s) and 0 issue activity found in the last 90 days -- score normalized to 0
Reason
no SAST tool detected
Details
Reason
no effort to earn an OpenSSF best practices badge detected
Reason
security policy file not detected
Details
Reason
license file not detected
Details
Reason
branch protection not enabled on development/release branches
Details
Reason
project is not fuzzed
Details
Score
Last Scanned on 2025-02-03
The Open Source Security Foundation is a cross-industry collaboration to improve the security of open source software (OSS). The Scorecard provides security health metrics for open source projects.
Learn More