A powerful JSON path processor with no third-party dependencies. Allows you to traverse JSON object trees with a simple dot-delimited path format e.g. "obj.name"
Irrelon Path is a JavaScript object manipulation library that uses dot notation to denote object key / value field locations within the object structure. It allows you to easily access, modify or remove data from an object at locations specified via a path string.
npm i @irrelon/path
- Simple Usage
- Escaping Fields with Periods
- Behaviour
- Default Values
- Most Common
- All Functions (Alphabetically)
- chop()
- clean()
- countLeafNodes()
- countMatchingPathsInObject()
- decouple()
- diff()
- diffValues()
- distill()
- down()
- escape()
- findOnePath()
- findPath()
- flatten()
- flattenValues()
- furthest()
- get()
- hasMatchingPathsInObject()
- isEqual()
- isNotEqual()
- join()
- joinEscaped()
- leafNodes()
- match()
- numberToWildcard()
- pop()
- pullVal()
- pullValImmutable()
- push()
- pushVal()
- pushValImmutable()
- set()
- setImmutable()
- shift()
- split()
- type()
- unSet()
- unSetImmutable()
- up()
- update()
- updateImmutable()
- values()
- wildcardToZero()
const {get} = require("@irrelon/path");
// Define an object in JSON
const obj = {
"users": {
"test1": {
"name": "My Test User"
}
}
};
// Grab data from the object via the path solver
const result = get(obj, 'users.test1.name');
console.log(result); // Logs: My Test User
Sometimes you want to access data where a field name has periods in it like this:
const obj = {
"users": {
"test@test.com": {
"name": "My Test User"
}
}
};
The user email address "test@test.com" contains a period that the path solver will interpret as a traversal indicator. If we try to ask the path solver to get the data in the key "test@test.com" it will look for a field called "test@test" with a sub-field "com".
To avoid this, escape the period using the escape() function:
const {get, escape} = require("@irrelon/path");
const result = get(obj, `users.${escape('test@test.com')}.name`);
console.log(result); // Logs: My Test User
If data or an object to traverse does not exist inside the base object, the path solver will return undefined and will NOT throw an error:
const {get} = require("@irrelon/path");
const obj = {
"foo": null
};
const result = get(obj, "foo.bar.one");
console.log(result); // Logs: undefined
When using get() you can specify a default value to return if the value at the given path is undefined.
const {get} = require("@irrelon/path");
const obj = {
"foo": null
};
const result = get(obj, "foo.bar.one", "My Default Value");
console.log(result); // Logs: My Default Value
Param | Type | Required | Default |
---|---|---|---|
path | String | true | none |
level | Number | true | none |
Chops a path
string down to the given level
. Given a path
string
like "foo.bar.ram.you.too", chop will remove any path parts below
the given level
. If we pass 2 as the level
with that given path
,
the result will be "foo.bar" as foo is level 1 and bar is level 2.
If the path
is shorter than the given level
, it is returned intact.
const {chop} = require("@irrelon/path");
const result = chop("foo.bar.one", 2);
console.log(result); // Logs: foo.bar
Param | Type | Required | Default |
---|---|---|---|
path | String | true | none |
Removes leading period (.) from string and returns new string.
const {clean} = require("@irrelon/path");
const result = clean(".foo.bar.one");
console.log(result); // Logs: foo.bar.one
Param | Type | Required | Default |
---|---|---|---|
obj | Object or Array | true | none |
Counts the total number of key leaf nodes in the passed object. Leaf nodes are any key that does not have a value of object or array.
const {countLeafNodes} = require("@irrelon/path");
const result = countLeafNodes({"foo": {"bar": null}, "moo": true});
console.log(result); // Logs: 2
Param | Type | Required | Default |
---|---|---|---|
testKeys | Object or Array | true | none |
testObj | Object or Array | true | none |
Tests if the passed object has the paths that are specified and that
a value exists in those paths and if so returns the number matched.
The output includes matchedKeys
with an object where the same structure
exists as the testObj
where each leaf node key will be a boolean that
describes if the leaf node exists in the testKeys
object.
MAY NOT BE INFINITE RECURSION SAFE.
const {countMatchingPathsInObject} = require("@irrelon/path");
const result = countMatchingPathsInObject({
"moo": true
}, {
"foo": {
"bar": null
},
"moo": true
});
console.log(result);
Outputs:
{
"matchedKeyCount": 1,
"matchedKeys": {
"foo": {
"bar": false
},
"moo": true
},
"totalKeyCount": 2
}
Param | Type | Required | Default |
---|---|---|---|
obj | Object or Array | true | none |
options | Object | false | {} |
options.immutable | Boolean | false | false |
If options.immutable
=== true then return a new de-referenced
instance of the passed object/array. If immutable is false
then simply return the same obj
that was passed. The returned
instance is NOT deeply immutably cloned because we recurse through
object trees and only immutably clone when making changes. This is
useful so you can instantly compare two states with a strict
equality check such as when using setImmutable().
const {decouple} = require("@irrelon/path");
const obj = {"foo": true};
const result = decouple(obj);
console.log(result);
console.log(result === obj);
Outputs:
{
"foo": true
}
false
Param | Type | Required | Default |
---|---|---|---|
obj1 | Object or Array | true | none |
obj2 | Object or Array | true | none |
options | Object | false | {} |
options.strict | Boolean | false | false |
options.maxDepth | Number | false | Infinity |
options.basePath | String | false | "" |
options.exclusive | Boolean | false | false |
If you need to know the description of the differences rather than only the paths that contain differences, use
diffValues()
instead.
Compares two objects / arrays and returns the differences as an array of paths to the different fields.
Fields are considered "different" if they do not contain equal
values. The equality check is either strict or non-strict based
on the options.strict
argument.
It is important to understand that this function detects differences between field values, not differences between object structures. For instance if a field in obj1 contains
undefined
and obj2 does not contain that field at all, it's value in obj2 will also technically be equal toundefined
so there would be no difference detected.
const {diff} = require("@irrelon/path");
const obj1 = {
"user": {
"_id": 1,
"firstName": "Jimbo",
"lastName": "Jetson"
}
};
const obj2 = {
"user": {
"_id": "1", // Notice string instead of numerical _id
"firstName": "James", // We also changed the name from "Jimbo" to "James"
"lastName": "Jetson"
}
};
const resultArr1 = diff(obj1, obj2, {basePath: "", strict: false}); // Non-strict equality check
const resultArr2 = diff(obj1, obj2, {basePath: "", strict: true}); // Strict equality check
console.log(resultArr1); // Logs: ["user.firstName"]
console.log(resultArr2); // Logs: ["user._id", "user.firstName"]
Param | Type | Required | Default |
---|---|---|---|
obj1 | Object or Array | true | none |
obj2 | Object or Array | true | none |
options | Object | false | {} |
options.strict | Boolean | false | false |
options.maxDepth | Number | false | Infinity |
options.basePath | String | false | "" |
options.exclusive | Boolean | false | false |
Compares two objects / arrays and returns the differences as an object where keys are paths to values that differ and the values are objects describing the difference, including the values and types.
Fields are considered "different" if they do not contain equal
values. The equality check is either strict or non-strict based
on the options.strict
argument.
It is important to understand that this function detects differences between field values, not differences between object structures. For instance if a field in obj1 contains
undefined
and obj2 does not contain that field at all, it's value in obj2 will also technically be equal toundefined
so there would be no difference detected.
const {diffValues} = require("@irrelon/path");
const obj1 = {
"user": {
"_id": 1,
"firstName": "Jimbo",
"lastName": "Jetson"
}
};
const obj2 = {
"user": {
"_id": "1", // Notice string instead of numerical _id
"firstName": "James", // We also changed the name from "Jimbo" to "James"
"lastName": "Jetson"
}
};
const resultArr1 = diffValues(obj1, obj2, {basePath: "", strict: false}); // Non-strict equality check
const resultArr2 = diffValues(obj1, obj2, {basePath: "", strict: true}); // Strict equality check
console.log(resultArr1); // Logs: {"user.firstName": {val1: "Jimbo", val2: "James", type1: "string", type2: "string", difference: "value"}}
console.log(resultArr2);
/*
Logs:
{
"user._id": {
val1: 1,
val2: "1",
type1: "number",
type2: "string",
difference: "type"
},
"user.firstName": {
val1: "Jimbo",
val2: "James",
type1: "string",
type2: "string",
difference: "value"
}
}
*/
Param | Type | Required | Default |
---|---|---|---|
obj | Object or Array | true | none |
pathArr | Array | true | none |
Gets the values of the paths in pathArr and returns them as an object with each key matching the path and the value matching the value from obj that was at that path.
const {distill} = require("@irrelon/path");
const obj = {
"user": {
"firstName": "Jim",
"lastName": "Jones",
"age": 22
}
};
const result = distill(obj, [
"user.firstName",
"user.lastName"
]);
console.log(result);
Outputs:
{
"user.firstName": "Jim",
"user.lastName": "Jones"
}
Param | Type | Required | Default |
---|---|---|---|
path | String | true | none |
levels | Number | false | 1 |
Returns the given path after removing the first leaf from the path. E.g. "foo.bar.thing" becomes "bar.thing".
const {down} = require("@irrelon/path");
const result = down("user.friends.0.firstName");
console.log(result);
Outputs:
"friends.0.firstName"
Param | Type | Required | Default |
---|---|---|---|
path | String | true | none |
Escapes any periods in the passed string so they will not be identified as part of a path. Useful if you have a path like "domains.www.google.com.data" where the "www.google.com" should not be considered part of the traversal as it is actually in an object like:
{
"domains": {
"www.google.com": {
"data": "foo"
}
}
}
Usage:
const {escape} = require("@irrelon/path");
const result = escape("www.google.com");
console.log(result);
Outputs:
"www\\.google\\.com"
Finds the first item that matches the structure of query
and returns the path to it
const {findOnePath} = require("@irrelon/path");
const myDataArray = [{
"profile": {
"id": 1,
"name": "Ron Swanson"
}
}, {
"profile": {
"id": 2,
"name": "April Ludgate"
}
}];
// Find the object that has a key "profile"
// with a object that has a key "_id" that
// has a value 1, and return the path to it
const result1 = findOnePath(myDataArray, {
profile: {
_id: 1
}
});
console.log(result1); // Logs: "0"
// Find the object that has a key "_id" that
// has a value 1, and return the path to it
const result2 = findOnePath(myDataArray, {
_id: 1
});
console.log(result2); // Logs: "0.profile"
See the unit tests for findOnePath() for many more examples of usage.
Param | Type | Required | Default | Description |
---|---|---|---|---|
source | * | true | none | The source to test. |
query | * | true | none | The query to match. |
Finds all items in source
that match the structure of query
and
returns the path to them as an array of strings.
const {findPath} = require("@irrelon/path");
const myData = {
"profile": {
"id": 1,
"name": "Ron Swanson",
"data": {
"mobile": "+001293284732"
}
}
};
const result = findPath(myData, {
data: {
"mobile": "+001293284732"
}
});
console.log(result);
Output:
{
"match": true,
"path": [
"profile"
]
}
Param | Type | Required | Default | Description |
---|---|---|---|---|
obj | Object or Array | true | none | The object to scan. |
Takes an object and finds all paths, then returns the paths as an array of strings.
const {flatten} = require("@irrelon/path");
const myData = {
"profile": {
"id": 1,
"name": "Ron Swanson"
}
};
const result = flatten(myData);
console.log(result);
Output:
[
"profile.id",
"profile.name",
"profile"
]
Param | Type | Required | Default | Description |
---|---|---|---|---|
obj | Object or Array | true | none | The object to scan. |
Takes an object and finds all paths, then returns the paths as keys and the values of each path as the values.
const {flattenValues} = require("@irrelon/path");
const myData = {
"profile": {
"id": 1,
"name": "Ron Swanson"
}
};
const result = flattenValues(myData);
console.log(result);
Output:
{
"profile": {
"id": 1,
"name": "Ron Swanson"
},
"profile.id": 1,
"profile.name": "Ron Swanson"
}
Param | Type | Required | Default | Description |
---|---|---|---|---|
obj | Object or Array | true | none | The object to operate on. |
path | String | true | none | The object to operate on. |
Given object obj
and a path
, determines the outermost leaf node
that can be reached where the leaf value is not undefined.
const {furthest} = require("@irrelon/path");
const myData = {
"profile": {
"id": 1,
"name": "Ron Swanson"
}
};
const result = furthest(myData, "profile.id.bson");
console.log(result);
Output:
"profile.id"
Param | Type | Required | Default |
---|---|---|---|
obj | Object or Array | true | none |
path | String | true | none |
defaultValue | Any | false | undefined |
Gets a value from the obj
at the given path
and if no value exists for
that path, returns defaultValue
if one was provided.
const {get} = require("@irrelon/path");
const obj = {
"foo": null
};
const result1 = get(obj, "foo");
const result2 = get(obj, "foo.bar.one", "My Default Value");
console.log(result1); // Logs: null
console.log(result2); // Logs: My Default Value
If you want to access elements of an array, simply use the element index as part of your path e.g.
const {get} = require("@irrelon/path");
const obj = {
"myArr": [
"hello",
{
"bar": "goodbye"
}
]
};
const result1 = get(obj, "myArr.0"); // hello
const result2 = get(obj, "myArr.1.bar"); // goodbye
Param | Type | Required | Default |
---|---|---|---|
obj | Object or Array | true | none |
path | String | true | none |
value | Any | true | none |
Sets a value
in the obj
at the given path
.
If the given path doesn't exist in the target object it will be created by making each non-existent path part a new object.
const {set, get} = require("@irrelon/path");
const obj = {
"foo": null
};
const result1 = get(obj, "foo.bar"); // Currently: undefined
set(obj, "foo.bar", "hello");
const result2 = get(obj, "foo.bar");
console.log(result1); // Logs: undefined
console.log(result2); // Logs: hello
This is a helper function that calls
set()
with immutable flag switched on.
Param | Type | Required | Default |
---|---|---|---|
obj | Object or Array | true | none |
path | String | true | none |
value | Any | true | none |
Sets a value
in the obj
at the given path
in an immutable way
and returns a new object. This will not change or modify the existing
obj
.
Keep in mind that references to objects that were not modified by the operation remain the same. This allows systems like React to appropriately act on changes to specific data rather than re-rendering an entire DOM tree when one sub-object changes.
If the given path doesn't exist in the target object it will be created by making each non-existent path part a new object.
const {setImmutable, get} = require("@irrelon/path");
const obj = {
"foo": {
"bar": "goodbye",
"subBar": {
"somethingElse": true
}
},
"otherObj": {
"enabled": true
}
};
const result1 = get(obj, "foo.bar"); // Currently: goodbye
const newObj = setImmutable(obj, "foo.bar", "hello");
// Original object remains unmodified (will still be "goodbye");
const result2 = get(obj, "foo.bar");
// New object has new value of "hello"
const result3 = get(newObj, "foo.bar");
console.log(result1); // Logs: goodbye
console.log(result2); // Logs: goodbye
console.log(result3); // Logs: hello
// Objects that did not have any modifications remain the same
// and still share a reference in memory
console.log(obj.otherObj === newObj.otherObj); // Logs: true
// Objects that did have modifications will not be the same
console.log(obj.foo === newObj.foo); // Logs: false
// Child objects of modified parents will still have references
// to the original since the child object wasn't modified directly
console.log(obj.foo.subBar === newObj.foo.subBar); // Logs: true
Param | Type | Required | Default |
---|---|---|---|
obj | Object or Array | true | none |
path | String | true | none |
Deletes a key from an object by the given path.
const obj = {
"foo": {
"bar": [{
"moo": true,
"baa": "ram you"
}]
}
};
console.log(obj.foo.bar[0].baa); // Logs: ram you
unSet(obj, "foo.bar.0.baa");
console.log(obj.foo.bar[0].baa); // Logs: undefined
console.log(obj.foo.bar[0].hasOwnProperty("baa")); // Logs: false
Param | Type | Required | Default |
---|---|---|---|
obj | Object or Array | true | none |
basePath | String | true | "" |
updateData | Object or Array | true | none |
options | Object | false | {} |
options.immutable | Boolean | false | false |
options.strict | Boolean | false | false |
options.ignore | RegExp | false | false |
Sets a single value on the passed object and given path. This will directly modify the "obj" object.
If you need immutable updates, use
updateImmutable()
instead.
const obj = {
"foo": {
"bar": [{
"moo": true,
"baa": "ram you"
}]
}
};
console.log(obj.foo.bar[0].baa); // Logs: ram you
// Calling this function with a basePath as an empty string
// will operate directly on the passed `obj` instead of a
// sub-object of `obj`.
update(obj, "", {
"foo.bar.0.baa": "hello I've been updated",
"and.so": "have I!"
});
console.log(obj.foo.bar[0].baa); // Logs: hello I've been updated
console.log(obj.and.so); // Logs: have I!
The update() and updateImmutable() functions have their signature changed to include a base path in the arguments. If migrating from a previous version you can simply add an empty string as the basePath argument to have the functions operate in the same way as before e.g.
update(obj, updateObj);
updateImmutable(obj, updateObj);
update(obj, "", updateObj);
updateImmutable(obj, "", updateObj);
The basePath argument was added so that you can target a path within
the passed obj
to receive the update e.g.
const obj = {subObj: {}};
update(obj, "subObj", {"foo": true});
The update above will modify obj.subObj.foo
to equal true
.
There was a bug in the get() function that would return an incorrect value when a non-object was passed to get data from and a path was passed e.g.
get("foo-im-not-an-object", "some.path.to.get.data.from"); // Version 2.x returned "foo-im-not-an-object"
In version 3.x, this call will return undefined
as expected.
Version 1.x exported a class that you could instantiate. Version 2.x exports an object with all available functions. You can require version 2.x either all at once (all functions) or you can destructure to require only the functions you need. This change is primarily to support tree shaking, as well as move to a more functional programming style, albeit not pure functional style :)
Version 2.x is a breaking change from version 1.x and you will need to migrate your code to work with the new version. Migration is fairly simple and instead of using an instance of the 1.x class, you simply require the parts of the library you need e.g.
// DON'T DO THIS !!!!!!!!!!!
const Path = require("irrelon-path");
const pathSolver = new Path();
const a = {hello: {foo: true}};
const b = pathSolver.get(a, "hello.foo"); // b === true
// DO THIS :)
const {get} = require("@irrelon/path");
const a = {hello: {foo: true}};
const b = get(a, "hello.foo"); // b === true