Update postcss-selector-parser to the latest version 🚀

[greenkeeper]

Version 4.0.0 of postcss-selector-parser was just published.

Dependency	postcss-selector-parser
Current Version	3.1.1
Type	dependency

The version 4.0.0 is not covered by your current version range.

If you don’t accept this pull request, your project will work just like it did before. However, you might be missing out on a bunch of new features, fixes and/or performance improvements from the dependency update.

It might be worth looking into these changes and trying to get this project onto the latest version of postcss-selector-parser.

If you have a solid test suite and good coverage, a passing build is a strong indicator that you can take advantage of these changes directly by merging the proposed change into your project. If the build fails or you don’t have such unconditional trust in your tests, this branch is a great starting point for you to work on the update.

---

Commits

The new version differs by 36 commits.

• faf2eb2 v4.0.0
• e725bdd v4.0.0-rc.1
• 0456064 Handle nodes with instance properties set to null.
• 27a3162 Make typescript definitions compatible with noImplicitAny.
• 9f68704 v4.0.0-rc.0
• 9aa8646 Test set Attribute#quoteMark
• 3c52465 Unescape namespace in all node types.
• e20cfbc don't run these tests that fail on node v6.
• a38fe31 Make it easier to fix lint errors.
• 382eb52 Tests for new code in Attribute.
• 1b95958 Fix some failing tests.
• 0c60825 Add some missing test cases.
• 718d865 Address code review comments.
• 3d79ea5 Handle uppercase case insensitivity attribute flag.
• 7389f5a Handle Comments with whitespace around combinators.

There are 36 commits in total.

See the full diff

FAQ and help

There is a collection of frequently asked questions. If those don’t help, you can always ask the humans behind Greenkeeper.

---

Your Greenkeeper bot 🌴

[alexander-akait]

@ai
I don't know what is wrong with postcss-selector-parser contributor, after @ben-en left as, no one good release, only regressions/bugs/incompatibility, i confirm what prettier also can't update to latest postcss-selector-parser (also cssnano) because they break some places without changelog notes or other information, also @chriseppstein don't try to notify/upgrade postcss-selector-parser in big projects (as do @ben-en), also he is very unresponsive for OSS project (I know all people have life aside, but if you can't have time on this just let it be done by others. Words Send PR is not acceptable, many issue need discussion/triangle and testing in other projects and one response in mouth is bad).

I admit that this is not the first statement of incompetence current contributor. But understand me, I am correcting many problems in top projects based on postcss, and every new release breaks a lot of thing. I'm just afraid to update this in prettier, because it is do prettier worse. Also you can look on CI and see memory leak. 3 and 4 broken.

Maybe we can release new major postcss version with support selector parser/value parser? I know it is a lot of work, but it is salvation for a lot of project.

Thanks!

[chriseppstein]

The pre release has been out for three weeks. There was an extensive write up on the changes and a twitter announcement of the release candidates. I fixed all the regressions that were reported.

https://github.com/postcss/postcss-selector-parser/releases/tag/v4.0.0-rc.0

postcss/postcss-selector-parser#133

It’s a major version release. It didn’t break anything of yours. And it’s not my job to send you pull releases.

File bugs for issues and I’ll look into them.

Also, I have issues with the way you spoke to me and about me in this post, but I’m not going to address those right now.

[alexander-akait]

@chriseppstein That is, you openly declare that you do not care about the fate of projects using your package, correctly?

[chriseppstein]

No that’s not at all what I said.

[alexander-akait]

@chriseppstein I tried for a long time to stay calm and try to be loyal in everything. But such a disrespectful attitude towards projects that use your code for me is not acceptable IMHO. And for me it is still not clear why this is difficult, as it is difficult to solve problems with supporting projects for which you do not have enough time (like cssnano).

Anyway thanks for your work. Looks fork is only good solution.

@stylelint/contributors In near future i will fork and fix memory leak and regression problems. I propose to unite around postcss parser for selectors and values with prettier. Because he has same errors/problems.

[alexander-akait]

@chriseppstein

It’s a major version release. It didn’t break anything of yours. And it’s not my job to send you pull releases.

Yes, not your work, this is called responsiveness and mutual help

[chriseppstein]

No need to fork, you can have the project. Clearly, I’m not the maintainer this project needs. Who should I give commit and npm publish permissions to?

[alexander-akait]

@chriseppstein postcss-selector-parser already in postcss org - i.e. not require npm/commit permissions, just be a little more responsive for major updates, i will create new issues when start solve problem in prettier/stylelint (or somebody other). I'm not trying to condemn you. Also do pull requests rc release in major projects allow to solve a lot of problems for many people.

Off-top: cssnano is stale and will be great if you give @ai and @evilebottnawio (me) npm/commit permissions.

[chriseppstein]

@evilebottnawi Being in the github org doesn't grant npm permissions. I have given the following publish privileges to @ai. He can grant them to you or someone else of his choosing.

+ ai (postcss-selector-parser)
+ ai (cssnano-util-get-match)
+ ai (cssnano-util-get-arguments)
+ ai (cssnano-util-raw-cache)
+ ai (cssnano-util-same-parent)
+ ai (postcss-colormin)
+ ai (postcss-convert-values)
+ ai (postcss-discard-duplicates)
+ ai (postcss-discard-comments)
+ ai (postcss-discard-empty)
+ ai (postcss-discard-overridden)
+ ai (postcss-minify-font-values)
+ ai (postcss-discard-unused)
+ ai (postcss-minify-selectors)
+ ai (postcss-normalize-charset)
+ ai (postcss-normalize-unicode)
+ ai (postcss-normalize-string)
+ ai (postcss-normalize-url)
+ ai (postcss-normalize-whitespace)
+ ai (postcss-reduce-initial)
+ ai (postcss-reduce-idents)
+ ai (css-size)
+ ai (cssnano)
+ ai (cssnano-preset-advanced)
+ ai (cssnano-preset-default)
+ ai (postcss-merge-idents)
+ ai (postcss-merge-longhand)
+ ai (postcss-merge-rules)
+ ai (postcss-minify-gradients)
+ ai (postcss-minify-params)
+ ai (postcss-normalize-display-values)
+ ai (postcss-normalize-positions)
+ ai (postcss-normalize-repeat-style)
+ ai (postcss-normalize-timing-functions)
+ ai (postcss-ordered-values)
+ ai (postcss-reduce-transforms)

error: (postcss-svgo)
error: (postcss-zindex)

It looks like ben didn't give me permissions for postcss-svgo and postcss-zindex so I can't get anyone access to those, I'm sure he can add whoever is going to maintain cssnano.

Please have whoever takes over the projects remove me from the project once they're on board. I'm available to walk them through the code changes I made to postcss-selector-parser.

I can't grant any commit privileges to cssnano because ben didn't give me admin access. You'll need to contact him directly.

Let me know if there's anything else that you need.

[alexander-akait]

@chriseppstein Awesome! Very thanks!

[ai]

@evilebottnawi the current plan of taking selector/value parser to PostCSS was just coping postcss-selector/value-parsers API to the core.

So we can’t just do it if we had a problem with this side parsers :(. Let’s first fix the problem.

Also. I understand your feelings. But don’t press to @chriseppstein too much. He was locked in big css-block development. Yeap, maybe 3 projects maintenance (css-black, postcss-selector-parser, and cssnano) is too much. But we all can do such mistake.

I see only that he took too many projects to maintain. But nobody took the project to maintain expect him. So don’t push him.

Let’s better think how to solve the problems. What exactly problem do we have right now?

[alexander-akait]

@ai Issue already created with example postcss/postcss-selector-parser#139

[ai]

So we have only memory leaks problem right now?

[alexander-akait]

@ai no, looks like tokenizer broken, because this in infinity loop with throw memory problem, maybe issue header is misleading, i think first it is memory leak, but after correct issue and add example

[ai]

OK, do you have plans to send PR?

We spoke with Chris and 4.0 will be on him. But then we will start to move selectors parser to the core. Because it requires fewer tokens compare to value parser.

[alexander-akait]

@ai don't have time right now, if it is not fixed in near week, i will do this

[jeddy3]

It looks like I'm a little late to this thread. I've been catching up on a couple of week's worth of stylelint issues in the last couple of days.

It sounds like a long-term way forward has been proposed i.e. to move the selector parser and then the value parser to PostCSS. That conversation is continuing in postcss/postcss#1145. In the short term, it sounds like @chriseppstein has kindly contributed another PR to align stylelint with the changes in postcss-selector-parser. And it looks like @ntwb has offered to review that PR when he has time.

It's inspiring to see an international team of passionate contributors. I believe most of us contribute to stylelint entirely in our free time. I think it's important to remember to respect the effort we all put in.

stylelint recently passed 2 million npm downloads a month. I think we can all take a moment to congratulate each other for passing that milestone. This feels like an especially good time to thank contributors of the packages that stylelint builds upon, particularly contributors who work on the low-level libraries. stylelint would not be possible without them.

[chriseppstein]

@jeddy3 Thanks, parsers are tricky and I've recently taken over the project from the previous maintainer and I had no experience with the code's internals when I did so. It's challenging to take over a project of this size (postcss-selector-parser gets 9 million downloads/mo) with no overlap and no ability to even get code-reviews from the previous developer. There is no other developer collaborating with me on the selector parser. It's just me, brand new, and doing my best. I've tried to move the project forward, there were a number of bugs affecting projects like this one that had been ignored because the internals didn't support fixing them. On top of this, I've also been working on the two projects that I have as my primary job responsibility and could only devote a few days a month to this project, so I have used a lot of my personal time on nights and weekends. Because of my limited time, I had to rely on feedback from the community through the release candidate process to let me know whether I had introduced material regressions. I don't know why greenkeeper doesn't create a PR for release candidates. I did it this time. I do not commit to doing this for future releases.

I woke up after working hard on releasing some other projects the day before and after pulling an all-nighter the night before, to @evilebottnawi calling me incompetent. At the time, I said I was not going to address that, because I did not think that I could do so without losing my temper and I did not want to further escalate the situation. But I want you to know, @evilebottnawi, that your words were hurtful and insensitive to my hard work, even if that effort was flawed, attacking me over it was unprofessional. The whole experience has left me with a very sour taste in my mouth. It makes me want to avoid collaborating with you and the projects that you are involved with. I know that you said "I'm not trying to condemn you." but, this isn't an apology and it's not consistent with what I read. I'm not interested in litigating what happened, I believe I have created a go-forward plan that will work better. If you intend to keep collaborating with me on any project, I expect you to do so professionally. If you cannot do this then on the next occurrence, I will block you, without further warning and you will get no further replies from me.

Now. let's all get back to working on making awesome CSS tools. 🍻

[ben-en]

Whoever tagged me was seriously off base, i'm not even involved with the project. @chriseppstein i wouldn't take it too personally, nor would i hesitate to use the block button if it happens again. My sympathies.

[chriseppstein]

@ben-en They meant ben-eb but he is not involved anymore and none of this has anything to do with him so I'm glad he wasn't looped in.

[alexander-akait]

@chriseppstein I apologize for being rude and somewhat unrestrained. I really did not want to offend you in any way.

[greenkeeper]

Version 5.0.0-rc.2 just got published.

Update to this version instead 🚀

Commits

The new version differs by 19 commits.

• 51879ab v5.0.0-rc.2
• 4a1b32d Fix regressions in stylelint.
• c8b8cdf Make test for nonstandard attribute flags more clear as to the current behavior.
• e2f5ffb Update stylelint submodule.
• eaa5def Relax node engine requirement for now.
• 479d8a7 Merge pull request #140.
• 746e4ba Touch up the CHANGELOG.
• bac536d Drop support for node 4 & 5, node < 6 is EOL'd and no longer maintained.
• fdfeed3 Run stylelint's tests during our CI tests.
• 71e1f1e package.json: Use prepare instead of prepublish.
• 38d1663 Update Changelog for 5.0.0.rc.0
• 9c7ba24 Update docs for descendant combinators.
• 2a00f62 API for getting the node at specific source location.
• b964b3d Utility for stripping comments.
• b4c3e57 Add entries for 4.0 to the changelog.

There are 19 commits in total.

See the full diff

[alexander-akait]

@chriseppstein let's do this first in prettier and then i will do this here 👍

[greenkeeper]

Version 5.0.0-rc.3 just got published.

Update to this version instead 🚀

Commits

The new version differs by 2 commits.

• 4c6dfdb Upgrade dev dependencies to reduce security warnings.
• 18ecae9 Make sure there's a token after the combinator.

See the full diff

[hudochenkov]

Closing this, as it was superseded by #3284.