Performance issues when have multiple entries

[minhtranite]

Currently I have a project with about 15 webpack entries. Each entry have own stylesheet file. And here is webpack build time:

• With stylelint-webpack-plugin:

webpack building...
webpack built 832650538a0ff1527328 in 21824ms

webpack building...
webpack built 832650538a0ff1527328 in 13755ms

• Without stylelint-webpack-plugin

webpack built 832650538a0ff1527328 in 8009ms
webpack building...
webpack built 832650538a0ff1527328 in 194ms
webpack building...
webpack built 832650538a0ff1527328 in 204ms

[mciastek]

I have encountered same issue. Without stylelint-webpack-plugin build takes approx. 300ms. With stylelint-webpack-plugin it takes 10x slower, approx. 3s. What cause that performance issue?

[renatorib]

Same issue here.

Without plugin: 500ms
With plugin: 3800ms

[JaKXz]

Thanks for reporting this! Would anyone like to take a crack at a PR / investigating? This may also be an upstream issue with stylelint itself.

[sergesemashko]

It seems like the performance issue here is with stylelint itself. I ran just stylelint task alone and had the approximately the same linting time.
As a workaround:

1. Use lintDirtyFilesOnly: true flag, which makes plugin to lint only touched files. The PR with flag was merged into master, but not released yet.
2. (Needs work) Skip linting of good files and lint only failed files, which I suggested in another issue. Basically, it is a cache implementation.

Of course, fundamentally, performance issue should be solved on stylelint level, because it has the biggest impact.
There are some thoughts about that in the comment.

[beckend]

How about forking the stylelint process instead?

[JaKXz]

@beckend would you be willing to make a PR that demonstrates what you're suggesting? I'd be interested in something like that if we could write simple integration tests for it :)

[shellscape]

Closing as stale due to age and inactivity. If there is still interest in this issue, please do submit a pull request.