- Tiny: less than 720 bytes minzipped including dependencies!
- Familiar: uses the same API as an ES6 Map/WeakMap
- Compliant: maintains all the invariants of Map/WeakMap including method return values and even iteration order!
- Correct: handles edge cases correctly
$ npm i mapamA bidirectional map is like a regular map except it preserves uniqueness of its values (in addition to its keys) and supports an inverse view, which is another bidirectional map containing and backed by the same entries, but with reversed keys and values.
For a weak bidirectional map specifically, its values (in addition to its keys) must be objects.
The API of one side/view of the bidirectional map is identical to the
Map/WeakMap API. Just use like a normal Map or WeakMap and call the
inverse map to get the other side of the bidirectional map!
import { BiMap, WeakBiMap } from 'mapam'
const biMap = new BiMap()
biMap.set(1, `one`)
biMap.set(2, `two`)
biMap.inverse().set(`three`, 3)
console.log([...biMap])
//=> [ [ 1, 'one' ], [ 2, 'two' ], [ 3, 'three' ] ]
console.log([...biMap.inverse()])
//=> [ [ 'one', 1 ], [ 'two', 2 ], [ 'three', 3 ] ]
console.log(biMap.get(2))
//=> two
console.log(biMap.inverse().get(`two`))
//=> 2
biMap.inverse().delete(`two`)
console.log(biMap.has(2))
//=> false
console.log([...biMap])
//=> [ [ 1, 'one' ], [ 3, 'three' ] ]
try {
// Throws!
biMap.set(10, `three`)
} catch (error) {
console.log(error.message)
//=> value already bound to another key
}
// Doesn't throw!
// Conveys intention of changing the value associated with the given key
biMap.inverse().set(`three`, 10)
// Also doesn't throw
biMap.set(10, `three`, { force: true })
const weakBiMap = new WeakBiMap()
const a = {}
const b = {}
weakBiMap.set(a, b)
console.log(weakBiMap.get(a) === b)
//=> true
console.log(weakBiMap.inverse().get(b) === a)
//=> trueSee the type definitions for more documentation.
| Category | mapam |
bimap@0.0.15 |
bim@1.3.3 |
mnemonist@0.38.3 |
@rimbu/bimap@0.7.2 |
|---|---|---|---|---|---|
| Iteration order | ✔️ | ❌ | ❌ | ❌ | ❌ |
| Negative zero | ✔️ | ❌ | ❌ | ❌ | ❌ |
Map (and many other "unordered" data structures) allow iteration in insertion
order so it's only logical for a bidirectional map to support iteration in a
logical order like insertion order.
const biMap = new BiMap()
biMap.set(1, `one`)
biMap.set(2, `two`)
biMap.set(3, `three`)
biMap.set(4, `four`)
biMap.inverse().set(`two`, -2)
biMap.delete(3)
console.log([...biMap])
// Correct behavior:
// mapam => [ [ 1, 'one' ], [ -2, 'two' ], [ 4, 'four' ] ]
// Incorrect behaviors:
// bimap@0.0.15 => Doesn't support iteration
// bim@1.3.3 => [ [ 1, 'one' ], [ 4, 'four' ], [ -2, 'two' ] ]
// mnemonist@0.38.3 => [ [ 1, 'one' ], [ 4, 'four' ], [ -2, 'two' ] ]
// @rimbu/bimap@0.7.2 => [ [ 1, 'one' ], [ 4, 'four' ], [ -2, 'two' ] ]A key in a Map can only occur once. But how is the key's uniqueness
determined? JavaScript's Map uses the
sameValueZero algorithm
when checking if two keys are equal. The algorithm considers +0 and -0 to be
equal, but they are actually two different values due to how
IEEE floating point numbers
work.
This means that Map coerces -0 to +0 for its keys, but not for its values. So
if a bidirectional map implementation uses Map internally, then it must either
coerce -0 to +0 for its values as well or
somehow broaden its keys to support
both -0 and +0.
const biMap = new BiMap()
biMap.set(`negative-zero`, -0)
biMap.set(`zero`, 0)
console.log(
biMap.get(`negative-zero`),
biMap.get(`zero`),
biMap.inverse().get(-0),
biMap.inverse().get(0),
)
// Correct behavior:
// mapam => -0 0 negative-zero zero
// Incorrect behaviors:
// bimap@0.0.15 => -0 0 negative-zero negative-zero
// bim@1.3.3 => undefined 0 zero zero
// mnemonist@0.38.3 => undefined 0 zero zero
// @rimbu/bimap@0.7.2 => -0 0 negative-zero negative-zeroStars are always welcome!
For bugs and feature requests, please create an issue.
For pull requests, please read the contributing guidelines.
This is not an official Google product.