How will new validations be decided?

[louiss0]

I'm the kind of person that likes to ask for new features that I believe are beneficial to developers.
But I have noticed that it's difficult to get maintainers to add features to libraries even though it's beneficial and would be
appreciated.

If I wanted to ask for a new transformation or validation how would you decide whether it's worth adding?

[fabian-hiller]

If it is generally useful, needed by many or corresponds to a standard, for example like ISO timestamp or UUID, I am open for any ideas. Due to the modular design it has no impact on existing users if we extend the functionality here.

A validation of credit cards would be cool for example. I just added a function today with isLuhnAlgo that could be useful for this. On the other hand, I'm cautious about things like phone number, as it will be impossible to cover every country and variant.

[louiss0]

Do you happen to have a validator that validates characters in a string but ignores spaces? I think that's useful and be the default. What if there was a validator that validated characters based on punctuation?

[fabian-hiller]

Do you have a concrete example how code with such a validation looks like? Basically anything can be validated using regex, custom or a custom function. Valibot does not limit this.

[louiss0]

Do you have a concrete example how code with such a validation looks like? Basically anything can be validated using regex, custom or a custom function. Valibot does not limit this.

I'm sorry I don't know the library so I'll have to use a regular function.

Characters based on only letters not spaces

 
function numOfLettersInThisStringIsEqualToTheNumber(string:string, number:number) {

const  trimmedStringLength = string.trim().length 

if(trimmedStringLength  !== number)

throw new Error("The characters in this string are invalid")
}

[fabian-hiller]

You can already do this easily with Valibot:

import { email, endsWith, string, toTrimmed } from 'valibot'; // 0.57 kB

const EmailSchema = string([toTrimmed(), email(), endsWith('@example.com')]);

More info here: https://valibot.dev/guides/pipelines/

[louiss0]

I was looking for the number of letters in the string in that example.

I think you meant

import { size string, toTrimmed } from 'valibot'; // 0.57 kB

const numOfCharsSchema = string([toTrimmed(), size(60) ]);

Either way letters is not enough punctuation would also have to be taken to account.

The user would have to I think write this if I'm correct.

import { size() string, toTrimmed } from 'valibot'; // 0.57 kB

const numOfCharsSchema = string([toTrimmed(),   size()]);

[fabian-hiller]

For the validation of a length Valibot provides length, minLength and maxLength. Note that toTrimmed will change the string. If you don't want this, you can use custom to write your own validation logic.

import { custom, length, string, toTrimmed } from 'valibot';

const numOfCharsSchema1 = string([toTrimmed(), length(60)]);
const numOfCharsSchema2 = string([custom((input) => input.trim().length === 60)]);

More info here: https://valibot.dev/guides/pipelines/

[louiss0]

Got it

[louiss0]

For the validation of a length Valibot provides length, minLength and maxLength. Note that toTrimmed will change the string. If you don't want this, you can use custom to write your own validation logic.

import { custom, length, string, toTrimmed } from 'valibot';

const numOfCharsSchema1 = string([toTrimmed(), length(60)]);
const numOfCharsSchema2 = string([custom((input) => input.trim().length === 60)]);

More info here: https://valibot.dev/guides/pipelines/