refactor: use zod in builders

[Qjuh]

Please describe the changes this PR makes and why it should be merged:

Replaces shapeshift with zod and zod-validation-error in @discordjs/builders.

Status and versioning classification:

• Code changes have been tested against the Discord API, or there are no code changes
• I know how to update typings and have done so, or typings don't need updating
• This PR changes the library's interface (methods or parameters added)

[vercel]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

2 Skipped Deployments

Name	Status	Preview	Comments	Updated (UTC)
discord-js	⬜️ Ignored (Inspect)	Visit Preview		Aug 23, 2024 7:36pm
discord-js-guide	⬜️ Ignored (Inspect)	Visit Preview		Aug 23, 2024 7:36pm

[codecov]

Codecov Report

Attention: Patch coverage is 94.80969% with 15 lines in your changes missing coverage. Please review.

Project coverage is 49.12%. Comparing base (e2e71b4) to head (914cf54).

Files	Patch %	Lines
...ers/src/components/selectMenu/ChannelSelectMenu.ts	16.66%	5 Missing ⚠️
...src/components/selectMenu/MentionableSelectMenu.ts	20.00%	4 Missing ⚠️
...ilders/src/components/selectMenu/RoleSelectMenu.ts	33.33%	2 Missing ⚠️
...ilders/src/components/selectMenu/UserSelectMenu.ts	33.33%	2 Missing ⚠️
...ilders/src/components/selectMenu/BaseSelectMenu.ts	85.71%	1 Missing ⚠️
packages/builders/src/util/validation.ts	88.88%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##             main   #10117       +/-   ##
===========================================
+ Coverage   35.45%   49.12%   +13.66%     
===========================================
  Files         228       93      -135     
  Lines       14319    10339     -3980     
  Branches     1254      692      -562     
===========================================
+ Hits         5077     5079        +2     
+ Misses       9198     5257     -3941     
+ Partials       44        3       -41     

Flag	Coverage Δ
builders	91.27% <94.80%> (-0.31%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[vladfrangu]

Also, this needs a rebase

[favna]

Builders used to use Zod but there were many many complaints about it. That was after /builders initially used the ow library as well. It was changed to the at the time freshly made shapeshift. One of the reasons for us to even make shapeshift was improvements to /builders.

I can reason this is related to confusing errors (the actual linked issue is a duplicate). I very strongly recommend you instead contribute to Shapeshift by fulfilling this issue for which most work is already finished in this pr and only unit tests need to be updated. Once that issue is resolved custom messages can be assigned in /builders and get very clear errors.

Also relevant:

• refactor: switch to /builders Embed #7067

[didinele]

This has all already been discussed in internals, albeit rather scattered, see:

• https://discord.com/channels/222078108977594368/999796393810067607/1201209906112503818

there's probably more if you search for "shapeshift".

[favna]

Copying my message from Discord to here after reading that conversation:

So I understand the argument of "it's still a draft", but there's also a good reason for that, there's simply still tasks to do. I've buggered @vladfrangu and @kyranet for (checks date of PR) 1,5 years now to help me with the PR because I'm inclined to think that I overcorrected in the source code as evident from what I see get returned in the unit tests, but I've never received any. Ergo, PR is stuck.

Adding to it, technically I could finish that PR in a couple of days if I really sit down and just finish it. I've just been lacking the motivation to do so. Once it's finished though, adding custom messages is a singe.

[favna]

Dumping a comment on here as well as follow-up to my reply above, the shapeshfit version with custom messages has been released. We can review where we want to take this PR further now.

[didinele]

I'm personally of the opinion that we should go further with zod after all.

I understand that an effort was made for that shapeshift release, but it's too little too late, on top of other accumulating issues. I know we keep on going back and forth on this, but the argument of "just wait until zod has an issue and we decide to switch back" really falls apart considering zod is a massive project that iterates quicker than shapeshift ever could - on top of, at least at this time, to my knowledge, having no big issues for us, while shapeshift has been problematic in various ways since the moment we introduced it.

Resolved.

[Qjuh]

Superseded by #10448