Consider removing lodash

larowlan → credited rikki_iki → .

Crediting rikki for advice

We're using it only in 2 .ts files. Removing this to avoid relying on unmaintained software is important. Thanks for spotting this!

Working on this

I need some guidance on it. One lodash function we use is debounce. Are we supposed to create a new custom debounce function?

Yes. https://github.com/you-dont-need/You-Dont-Need-Lodash-Underscore?tab=rea...

Just as a reminder and for it it helps, Drupal already ships a debounce function. The library can be found on core/drupal.debounce, on core.libraries.yml.

Thank you. That helps. Will send an initial MR soon.

@sanket.tale How is it going? 😊

Hi @wim-leers

I am a beginner. Tried to remove lodash dependency completely. But getting errors. Will try to solve it.

Thanks for getting this going! 😊

It looks like camelCase, kebabCase, isEmpty and debounce are the 4 lodash functions we use (so far).

I don't think there is value is moving camelCase, kebabCase into our codebase vs importing them.

The 'native' isEmpty solution given on you-dont-need/You-Dont-Need-Lodash-Underscore is not as full featured as the one from lodash which supports object, collection, map, or set (vs native which supports only objects and arrays).

The debounce method in Drupal core is based on underscore and which is even more out of date than lodash! Lodash's debounce is significantly more feature rich and controllable.

Given that lodash is static (has no external dependencies), is battle-hardened and is used sparingly I am afraid that I don't see the value in duplicating functions from it into our code base (which would also mean bringing along their tests) vs just importing them.

I apologise to those of you who have contributed here on not reviewing and closing this sooner.