deep-iterator iterates recursively through objects, arrays, sets, maps, generic iterables, or any mix of them.
- several search algorithms: bfs, dfs pre/post order
- circular references detection
- can iterate either all the nodes or just the leaves
- can skip iteration of specified objects
npm install deep-iterator
import deepIterator from 'deep-iterator';
// commonJS : var deepIterator = require('deep-iterator').default;
const obj = {a: 1, b: [2, 3]};
for (let {value} of deepIterator(obj)) {
console.log(value);
}
/* outputs:
{a: 1, b: [2, 3]} (the root) - use the onlyLeaves option to remove
1
[2, 3] - use the onlyLeaves option to remove
2
3
*/
const deepArray = [[1, 2], [3, 4]];
const it = deepIterator(deepArray, {onlyLeaves: true});
for (let {parent, key} of it) {
parent[key]++;
}
// deepArray === [[2, 3], [4, 5]]
For each iteration, deep-iterator yields a node object containing the following members :
value
: the value of the nodeparent
: the parent of the nodekey
: the key of the node in its parentpath
: an array containing all the keys from the root down to the nodeparentNode
: the parent node objectisCircular()
: true if the node is a circular referenceisLeaf()
: true if the node is a leaftype
: a string representing the type of the node. Possible values are:Null
,Undefined
,Boolean
,String
,Symbol
,Date
,RegExp
,Function
,GeneratorFunction
,Promise
,Array
,Set
,Map
,UserDefinedIterable
: an iterable that is not an array, map, set or string,NonIterableObject
: an object that doesn't implement the iterator protocol
Additionally, the node object contains helpers methods to determine its type, in the form of
"isType()".
Examples : isMap()
, isNonIterableObject()
, ...
Remarks on the node object :
- parent[key] lets you update the current node
- key is undefined if the node is the root or the iterated collection is a user defined iterable
- path and type are lazy evaluated : not using them will improve performance, i. e. :
for (let {value, parent, key} of it) => path is not evaluated
for (let {value, parent, key, path} of it) => path is evaluated
Default options are :
const iterator = deepIterator(obj, {
search: 'dfsPreOrder', // DFS algorithm / parent before child
onlyLeaves: false, // all nodes are iterated
circularReference: 'leaf', // circular references are treated as leaves (not recursively iterated)
iterateOverObject: true, // non-iterable objects are iterated through their properties
skipIteration : node => false // no node are skipped
});
accepted values: 'dfsPreOrder'(default), 'dfsPostOrder', 'bfs'
Specifies the iteration algorithm:
- dfsPreOrder: Depth First Search, parent before child
- dfsPostOrder: Depth First Search, child before parent
- bfs: Breadth First Search
accepted values: false (default) or true
Iterates all the nodes or only leaves
accepted values: 'leaf', 'throw', 'noCheck'
Handles the circular references detection :
- leaf: circular references are treated as leaves (no further iteration)
- throw: circular references throw an error
- noCheck: circular detection is off (this improves performances)
accepted values: true(default) or false
Enables iteration over non-iterable objects or not.
accepted values: callback: node => boolean
Specifies whether a recursive iteration is performed on the node or not.
- if the callback returns true : the node is not iterated itself and is considered as a leaf
- if the callback returns false : the node is deeply iterated
Note: skipIteration is not called on leaves.
Example : Skip iteration of all arrays :
const it = deepIterator(obj, {skipIteration: node => Array.isArray(node.value)});