Update README.md to reference gulp-pug-lint2

[jwalton]

This PR updates README.md to suggest gulp-pug-lint2 instead of gulp-pug-lint.

gulp-pug-lint hasn't seen any updates in 2 years. It:

• forces you to use pug-lint@2.1.2 which doesn't support es6 syntax, despite two PRs to try to update package dependencies, both of which are well over a year old.
• doesn't have support for failing on errors despite a PR from April.
• reads the .pug template files from disk even though the contents of the file are already in memory file.contents, which makes it slow.

gulp-pug-lint2, on the other hand, uses peer dependencies so you can use whatever pug-lint you want to use. It also users puglint.checkString() to check the contents in the vinyl File buffer provided to us by gulp, instead of going back to the disk a second time. Finally, gulp-pug-lint2 has tests which test actually linting a file, which gulp-pug-lint does not.

[adrienverge]

Looks OK to me.

@ilyakam what do you think?

[ilyakam]

Hi @adrienverge,

Thank you for the opportunity to chime in. I recommend merging my PR #69 instead of this one for a few reasons. First, I created gulp-pug-linter because I've also noticed that gulp-pug-lint has not been updated for a long time already. As mentioned in my PR, _gulp-puglinter is under active development, it reads all configuration files, and it comes with 100% code coverage. Now that it has been out for 19 months, you can see that it's doing pretty well for itself, even without your endorsement:

(Source)

I can't tell for certain, and @jwalton - please correct me if I'm wrong, but I think that what lead to #139 is that:

1. A developer tries to lint his pug files and starts with pug-lint
2. Knowing that he's using gulp, he performs a simple in-browser search for gulp in this repo's README and comes across your recommendation for the outdated gulp-pug-lint
3. After a few unsuccessful attempts at making it work, he realizes that the plugin is not that complicated and can be made to work rather quickly. In @jwalton's case, it took him an hour from the time of the fork to his release of version 1.0.0
4. This PR gets created, and when looking at all open PRs, another "Update README.md" catches his eye, leading to where we are now

What's more, @jwalton was kind enough to offer an improvement for gulp-pug-linter, which I am planning to thoroughly review and benchmark by Sunday night PDT. I guess ultimately that's what it boils down to:

• Should this project continue endorsing the outdated gulp-pug-lint?
• Should it endorse the more up-to-date version of gulp-pug-lint2?
• Or should it endorse a wholly different approach with gulp-pug-linter?

Again, I recommend going with the third option, by merging PR #69.

[jwalton]

Yeah, @ilyakam has it exactly right; I created gulp-pug-lint2 because it was really easy, and I was too lazy to check to see if someone else had already done it. I like the name gulp-pug-linterbetter, even, and Ilya has already spent the time to get all that code coverage done (which I haven't had a chance to yet).

I'm perfectly happy to mark gulp-pug-lint2 as deprecated and point to Ilya's project. I just want gulp-pug-lint out of the README.md before we end up with a third creatively named project. :P

[adrienverge]

@ilyakam @jwalton Thanks for these detailed explanations. Let's merge #69 if no one has objections!