Grouping Data in TypeScript Array - arrays

I have JSON data that looks like this:
[
{
"id": 1,
"tags": [
"Test 1",
"Test 2",
"Test 3"
]
},
{
"id": 2,
"tags": [
"Test 2",
"Test 3",
"Test 4"
]
},
{
"id": 3,
"tags": [
"Test 3",
"Test 4"
]
}
]
I would like to transform this into data that looks like this:
[
{
"name": "Test 1",
"count": 1
},
{
"name": "Test 2",
"count": 2
},
{
"name": "Test 3",
"count": 3
},
{
"name": "Test 4",
"count": 1
}
]
I can think of some brute ways to do this, but I'm hoping there is something more performant and a little sexier? Possibly using .groupBy() or .reduce()?
Thanks for taking the time to check out my question.

I would:
parse the json
gather all tags in an array
count occurences using one of the approaches in Counting the occurrences / frequency of array elements
interface Item {
id: number,
tags: string[]
}
function countOccurences(a: string[]) {
return a.reduce(function (acc: {[key: string]: number}, curr: string) {
acc[curr] ??= 0;
acc[curr]++;
return acc;
}, {});
}
const data: Item[] = JSON.parse(json);
const tagOccurences = countOccurences(data.flatMap(o => o.tags))
Playground link

You can use reduce inside reduce to group the tags.
const array = [{
id: 1,
tags: ['Test 1', 'Test 2', 'Test 3'],
},
{
id: 2,
tags: ['Test 2', 'Test 3', 'Test 4'],
},
{
id: 3,
tags: ['Test 3', 'Test 4'],
},
];
const frequencies = Object.values(array.reduce((acc, curr) =>
curr.tags.reduce(
(nAcc, tag) => ((nAcc[tag] ??= {name: tag,count: 0}),nAcc[tag].count++,nAcc),
acc
), {}
));
console.log(frequencies);
In TypeScript:
const array = [{
id: 1,
tags: ['Test 1', 'Test 2', 'Test 3'],
},
{
id: 2,
tags: ['Test 2', 'Test 3', 'Test 4'],
},
{
id: 3,
tags: ['Test 3', 'Test 4'],
},
];
type Frequency = {
name: string,
count: number
}
const frequencies = Object.values(array.reduce((acc, curr) =>
curr.tags.reduce(
(nAcc, tag) => ((nAcc[tag] ??= {name: tag,count: 0}),nAcc[tag].count++,nAcc),
acc
), {} as Record<string, Frequency>
));
console.log(frequencies);
Playground

Using for...of iteration and a Map as a cache is a very straightforward approach... and sexy.
TS Playground
type TagsWithId = {
id: number;
tags: string[];
};
type TagCount = {
count: number;
name: string;
};
function verySexyTagCounter (input: TagsWithId[]): TagCount[] {
const map = new Map<string, number>();
for (const {tags} of input) {
for (const name of tags) {
map.set(name, (map.get(name) ?? 0) + 1);
}
}
return [...map.entries()].map(([name, count]) => ({name, count}));
}
const json = `[{"id":1,"tags":["Test 1","Test 2","Test 3"]},{"id":2,"tags":["Test 2","Test 3","Test 4"]},{"id":3,"tags":["Test 3","Test 4"]}]`;
const input: TagsWithId[] = JSON.parse(json);
const result = verySexyTagCounter(input);
console.log(result);

Related

How do i map through array to get array of object?

I have an array called data. How do i extract sub_data? Just need the sub_data part for each object.
const data = [
{
id: 1,
title: 'Logo'
sub_data: [
{
id: 2,
title: 'Company Logo'
},
{
id: 3,
title: 'Website Logo'
},
]
},
{
id: 2,
title: 'Brands'
sub_data: [
{
id: 25,
title: 'Company Brands'
},
{
id: 3,
title: 'Website Brands'
},
]
}
]
Example output will get two outputs because there is 2 objects:
const subData = [
{
id: 2,
title: 'Company Logo'
},
{
id: 3,
title: 'Website Logo'
},
]
const subData = [
{
id: 25,
title: 'Company Brands'
},
{
id: 3,
title: 'Website Brands'
},
]
Not very sure how to use the map function just to get sub_data in the correct structure
You can use flatMap to get sub_data in one array
const data = [
{
id: 1,
title: 'Logo',
sub_data: [
{
id: 2,
title: 'Company Logo'
},
{
id: 3,
title: 'Website Logo'
},
]
},
{
id: 2,
title: 'Brands',
sub_data: [
{
id: 25,
title: 'Company Brands'
},
{
id: 3,
title: 'Website Brands'
},
]
}
]
const result = data.flatMap(item => item.sub_data)
console.log(result)
If you want an array with the sub_data objects you can just map the original array:
const data = [
{
id: 1,
title: 'Logo',
'sub_data'
: [
{
id: 2,
title: 'Company Logo'
},
{
id: 3,
title: 'Website Logo'
},
]
},
{
id: 2,
title: 'Brands',
sub_data: [
{
id: 25,
title: 'Company Brands'
},
{
id: 3,
title: 'Website Brands'
},
]
}
]
const mappedData = data.flatMap(obj => obj.sub_data)
console.log(mappedData)
Another solution would be to use the .forEach function of javascript.
const subData = [];
data.forEach(item => subData.push(...item.sub_data))

Angular filter array inside array

I'm working in an Angular 9 app and I need to filter an array based on another array inside each object. To understand me here is an example of the array
const products = [
{
name: 'Product 1',
origins: [
{ quantity: 1, name: 'Pack 1' },
{ quantity: 1, name: 'Pack 2' },
]
},
{
name: 'Product 2',
origins: [
{ quantity: 1, name: 'Pack 1' },
{ quantity: 1, name: 'Pack 2' },
]
},
{
name: 'Product 3',
origins: [
{ quantity: 1, name: 'Inventory' },
{ quantity: 1, name: 'Pack 5' },
]
}
]
So I got a filter input which has to filter the products by a coincidence on the name of the product or one or more origin's name.
For example, if I type "2" the result array must be:
products = [
{
name: 'Product 1',
origins: [
{ quantity: 1, name: 'Pack 2' },
]
},
{
name: 'Product 2',
origins: [
{ quantity: 1, name: 'Pack 2' },
]
}
]
Because the character "2" is in the name of origin of Product 1 and Product 2, also is present in the name "Product 2"
I tried many things to do this but the result array always modifies my original array when I put this in a pipe
<input type="text" [(ngModel)]="searchtext">
<div *ngFor="let p of (saleProducts | filter : searchtext); let i = index">
{{ p.name }}
<div *ngIf="p.origins.length > 0">
<div *ngFor="let o of p.origins">
{{ o.name }}
</div>
</div>
</div>
What is the best (simple and optimized) way to filter this using a pipe without modifying the original array?
below a pipe implementation to with a default key origins:
import { Pipe, PipeTransform } from '#angular/core';
#Pipe({
name: 'filter'
})
export class FilterPipe implements PipeTransform {
transform(value: any[], searchName: string, key = 'origins'): any[] {
const products = []
value.forEach(product => {
const matches = product[key].filter(({ name }) => name === searchName)
if (matches.length) {
products.push({ ...product, [key]: matches })
}
})
return products
}
}

Angular 2+ compare differences between 2 arrays

The arrays I have
const users = [
{ id: 1, name: "field 1" },
{ id: 2, name: "field 2" },
{ id: 3, name: "field 3" },
{ id: 4, name: "field 4" },
];
const onlineUsers = [
{ id: 1, name: "field 1" },
{ id: 3, name: "field 3" }
];
I would like to find the online and offline ones by comparing the two series
I want to do:
const userLists = [
{ id: 1, name: "field 1", online: true },
{ id: 2, name: "field 2", online: false },
{ id: 3, name: "field 3", online: true },
{ id: 4, name: "field 4", online: false },
];
Using Array.map and Array.some
const users = [
{ id: 1, name: "field 1" },
{ id: 2, name: "field 2" },
{ id: 3, name: "field 3" },
{ id: 4, name: "field 4" },
];
const onlineUsers = [
{ id: 1, name: "field 1" },
{ id: 3, name: "field 3" }
];
var retVal=users.map(u=>{
var isOnline=onlineUsers.some(ou=> ou.id==u.id);//this will check if onlineUsers have some record with given userid
return {...u,online:isOnline}
})
console.log(retVal)
You can just traverse through the user list and you can find out the onlineuser using find and just push it in the onlineuserList.
const users = [
{ id: 1, name: "field 1" },
{ id: 2, name: "field 2" },
{ id: 3, name: "field 3" },
{ id: 4, name: "field 4" },
];
const onlineUsers = [
{ id: 1, name: "field 1" },
{ id: 3, name: "field 3" }
];
const userLists = [];
users.forEach(user => {
if(onlineUsers.find(q => q.id == user.id)){
userLists.push({
id: user.id,
name: user.name,
online: "true"
})
}
else{
userLists.push({
id: user.id,
name: user.name,
online: "false"
})
}
})
console.log(userLists);
a bit faster approach using Array.indexOf() and JSON.strigify()
const onlineUsers = JSON.stringify([
{ id: 1, name: "field 1" },
{ id: 3, name: "field 3" }
]);
const userList = users.map(user =>
({
...user,
online: onlineUsers.indexOf(JSON.stringify(user)) > -1
})
);
OR if you neither want to change original onlineUsers array nor declare another variable:
const userList = users.map(user =>
({ ...user, online: JSON.stringify(onlineUsers).indexOf(JSON.stringify(user)) > -1 })
);

Filter Array of Object from another Array of Object

Currently I'm filtering data based from questions that have checked property value equals to true..
const data = [
{Name: "foo", X1: "1", X2: "1", Other: "Test1"},
{Name: "bar", X1: "2",X2: "2",Other: "Test2"},
{Name: "test",X1: "2",X2: "3",Other: "Test3"}
];
const questions = [{rows: {text: "Text 1", checked: true,fields: "1",column: "X1"}
}, {rows: {text: "Text 2", checked: true,fields: "2",column: "X1"}
}, {rows: {text: "Text 3", checked: false,fields: "1",column: "X2"}
}, {rows: {text: "Text 4", checked: false,fields: "2",column: "X2"}
}, {rows: {text: "Text 5", checked: false,fields: "3",column: "X2"}
}];
console.log(questionArr);
// console.log(dataArr);
const res = data.filter(d => questions.find(f => d[f.rows.column] === f.rows.fields && f.rows.checked));
which works but does not work when filtering the actual data below. I think there's a slight difference between the question object and the actual question object below.. What should be my filter code when accessing these kind of structure ?
I think this is what you're looking for. I matched the data structure to the image in your question. Let me know if I missed something.
const data = [
{ Name: "foo", X1: "1", X2: "1", Other: "Test1" },
{ Name: "bar", X1: "2", X2: "2", Other: "Test2" },
{ Name: "test", X1: "2", X2: "3", Other: "Test3" }
];
const questions = [
{ rows: [{ text: "Text 1", checked: true, fields: "2", column: "X1" }] },
{ rows: [{ text: "Text 2", checked: true, fields: "2", column: "X1" }] },
{ rows: [{ text: "Text 3", checked: false, fields: "1", column: "X2" }] },
{ rows: [{ text: "Text 4", checked: false, fields: "2", column: "X2" }] },
{ rows: [{ text: "Text 5", checked: false, fields: "3", column: "X2" }] }
];
const result = data.filter(function(item){
return questions.some(function(question){
return question.rows.some(function(row){
return (row.checked && item[row.column] === row.fields);
});
});
});
console.log(result);
The compact version
const result = data.filter((item) => questions.some((question) => question.rows.some((row) => (row.checked && item[row.column] === row.fields))));
With perf in mind
const data = [
{ Name: "foo", X1: "1", X2: "1", Other: "Test1" },
{ Name: "bar", X1: "2", X2: "2", Other: "Test2" },
{ Name: "test", X1: "2", X2: "3", Other: "Test3" }
];
const questions = [
{ rows: [{ text: "Text 1", checked: true, fields: "2", column: "X1" }] },
{ rows: [{ text: "Text 2", checked: true, fields: "2", column: "X1" }] },
{ rows: [{ text: "Text 3", checked: false, fields: "1", column: "X2" }] },
{ rows: [{ text: "Text 4", checked: false, fields: "2", column: "X2" }] },
{ rows: [{ text: "Text 5", checked: false, fields: "3", column: "X2" }] }
];
const result = {};
for(let a = 0, b = data.length; a < b; a++){
const item = data[a];
for(let c = 0, d = questions.length; c < d; c++){
const rows = questions[c].rows;
for(let e = 0, f = rows.length; e < f; e++){
const row = rows[e];
if(row.checked && item[row.column] === row.fields){
result[item.Name] = item;
break;
}
}
}
}
// this could be replaced with Object.values(result);
const matches = [];
for(let match in result){
matches.push(result[match]);
}
// not supported by IE yet
// Object.values(result);
console.log(matches);

Grouping objects with lodash or ES5

I'm a javascript noobie, I started to use the library lodash and I'm having some problems. Can someone help me? :/
Input:
var a = [{group: 'A', title: 'test1', value: 'test2'},
{group: 'A', title: 'test2', value: 'test3'},
{group: 'B', title: 'test3', value: 'test4'},
{group: 'B', title: 'test1', value: 'test2'},]
Output:
var a = {A: [
{ title: 'test1', value: 'test2'},
{ title: 'test2', value: 'test3'}
],
B: [
{ title: 'test1', value: 'test2'},
{ title: 'test2', value: 'test3'}
],}
You can group with _.groupBy() and then use _.mapValues() on each group to remove the group property from the objects by mapping them to a new object using _.omit().
const a = [{group: 'A', title: 'test1', value: 'test2'},
{group: 'A', title: 'test2', value: 'test3'},
{group: 'B', title: 'test3', value: 'test4'},
{group: 'B', title: 'test1', value: 'test2'}
];
const result = _(a)
.groupBy('group') // group by the group prop
.mapValues((group) => group.map((o) => _.omit(o, 'group'))) // remove the group prop from each item
.value();
console.log(result);
<script src="https://cdnjs.cloudflare.com/ajax/libs/lodash.js/4.17.4/lodash.min.js"></script>
The lodash documentation for group by is pretty explicit you can find it here.
In your case it would look something like this.
_.groupBy(a, 'group')
You can also do it yourself pretty easily as well using the reduce function. It would look something like this:
let groups = a.reduce((grouping, currentItem) => {
if (!grouping[currentItem.group]) {
grouping[currentItem.group] = []
}
grouping[currentItem.group].push(currentItem);
return grouping;
}, {});
A working example can be found here.
In both cases the result would be an object with the groups as property names.
{
"A": [{
"group": "A",
"title": "test1",
"value": "test2"
}, {
"group": "A",
"title": "test2",
"value": "test3"
}],
"B": [{
"group": "B",
"title": "test3",
"value": "test4"
}, {
"group": "B",
"title": "test1",
"value": "test2"
}]
}

Resources