[BUG] tailwind/no-custom-classname taking a TON of time

[Nick2bad4u]

Describe the bug
I think the rule is getting stuck on trying to scan every single html file in the project and getting stuck.

tailwind/no-custom-classname taking a TON of time on ONLY .tsx files:

The full timing=1 run I posted here because it was too big: Full Timing=1 Run Gist

Rule	Time (ms)	Relative
tailwind/no-custom-classname	273354.598	98.8%
prettier/prettier	990.519	0.4%
import-x/namespace	475.004	0.2%
etc/no-deprecated	463.620	0.2%
react-compiler/react-compiler	148.570	0.1%
sonarjs/deprecation	146.075	0.1%
tailwind/enforces-negative-arbitrary-values	116.024	0.0%
no-explicit-type-exports/no-explicit-type-exports	112.823	0.0%
tailwind/classnames-order	110.018	0.0%

To Reproduce
Run linter on this file, or download my repo and run the linter like I do:

git+https://github.com/Nick2bad4u/Uptime-Watcher

https://github.com/Nick2bad4u/Uptime-Watcher/blob/3b32a211babfe6619907908d2eeef41e1f619517/src/App.tsx

Steps to reproduce the behavior:

            "tailwind/no-custom-classname": [
                "warn",
                {
                    skipClassAttribute: true,
                },
            ],

Full config: https://github.com/Nick2bad4u/Uptime-Watcher/blob/9dbf4d850905b21a77f321c09938d9552bb6ac84/eslint.config.mjs
Expected behavior
A clear and concise description of what you expected to happen.

Shouldn't take 5 minutes :D

Screenshots
If applicable, add screenshots to help explain your problem.

Environment (please complete the following information):

• OS: win11 latest
• Softwares + version used:
  Version: 1.103.2 (user setup)
  Commit: 6f17636121051a53c88d3e605c491d22af2ba755
  Date: 2025-08-20T16:45:34.255Z
  Electron: 37.2.3
  ElectronBuildId: 12035395
  Chromium: 138.0.7204.100
  Node.js: 22.17.0
  V8: 13.8.500258-electron.0
  OS: Windows_NT x64 10.0.22631

Additional context
Add any other context about the problem here.

eslint config file or live demo
By providing a link to a live demo, a demo video or a github repo fixing the issue will be easier.

[kachkaev]

Possible duplicate of #42 and #276. See #276 (comment) for a potential workaround

[Nick2bad4u]

Possible duplicate of #42 and #276. See #276 (comment) for a potential workaround

It's a dupe but those are closed.

I've kind of narrowed it down with the eslint timing and debug to having a ton of .html files in the repo.

The rule seems to try and scan every single .html file on the workspace no matter if they are ignored or not.

I ran into this after Stryker made a ton of files in the repo and immediately this bug hit me. Even though these files were on my eslint and git ignore, I can see from eslint debug that it's still scanning html in those folders.

Solutions:

1. Force users to define a search location
2. Respect and parse .gitignore and eslint ignores from config and .eslintignore
3. Only search in eslint includes space. For example, I have this rule setup under a custom files block, it should only look for html in those locations. Instead it tries from the workspace root it seems to scan all HTML

[uu29]

I'm experiencing the same issue, but my repository doesn't have many HTML files.
I'm attaching the result of running ESLint on a specific file for reference.

[Nick2bad4u]

I'm experiencing the same issue, but my repository doesn't have many HTML files.
I'm attaching the result of running ESLint on a specific file for reference.

The comment on the thread above has a working solution.