postcss/postcss

Transforming CSS with JS plugins

Zgjim Gjonbalaj

@zgjimgjonbalaj

I thought postcss-media-queries output to standard CSS media queries? May have misunderstood there. I did attempt to use a mixin but may revisit back again. Thanks for your help!

Andrey Sitnik

@ai

@zgjimgjonbalaj yeap, it is converted to standard CSS media queries

CSS Custom Media was created according to CSS draft from CSSWG, where was suggested official new syntax to CSS

so it is not some custom sytax sugar, but how CSSWG saw the future of CSS

but at the end, CSSWG changed mind about CSS Custom Media and now nobody is sure that this draft will be part of CSS

Zgjim Gjonbalaj

@zgjimgjonbalaj

Ah i see, back to the drawing board! I have not been following their recommendations/suggestions and appreciate the information. I think i'm going to go that route of just creating a custom mixin with @mixin-content and follow that.

JimMorrison723

@JimMorrison723

Hello!
I want to create a stylelint plugin/rule for my project that uses scss files, and i need postcss for that. Is this the right place to ask questions about that? (reading back a little - it might be..)
I want to get an imported file parse - i mean if any declarations has mixins in it, i want it to be there (so basically -> read file -> make css -> and use that information in the file i am currently linting).
I use fs to read the file, but i am lost the "make css" part, that i can use.
Can anyone help me here? Kinda lost in the docs :/

🐙 Julien Kirch 🐙

@archiloque_twitter

Hello, there is a long-standing bug in Safari that require adding line-height: 0vw; to the pages body for responsive fonts size, see for example seaneking/postcss-responsive-type#15 or twbs/rfs#14 . When I try to use it with cssnano the vwis stripped, which makes sense for all other cases but this one. Is there a workaround for this case and/or should I open a ticket for this ? (I checked the project's tickets and doc and found nothing mentioning it, sorry if it's documented somewhere)

Andrey Sitnik

@ai

@archiloque_twitter cssnano is a modular minimizer. You can find plugin, which minimize units and disable it

🐙 Julien Kirch 🐙

@archiloque_twitter

@ai Disabling the convertValues plugin worked, we should be fine with the few things that are unoptimized because of this, thanks !

🐙 Julien Kirch 🐙

@archiloque_twitter

And should I open an issue about this to ask for an exception for this case to the plugin ?

Andrey Sitnik

@ai

@archiloque_twitter yeap, you can try to open an issue

🐙 Julien Kirch 🐙

@archiloque_twitter

Thanks

Marcello

@marcebdev

So I've set up postcss-preset-env running with the default stage 2 its working properly for chrome and firefox transpiling down but Safari is still not working... is there something else I need to be doing?

Andrey Sitnik

@ai

@marcebdev what CSS feature do not work in Safari?

Marcello

@marcebdev

@ai I'm trying to use the gap property which atm only really works on safari... I've setup postcss-preset-env so it's working on chrome correctly now (before it wasn't) but still remain broken on Safari

The default browser list is supposed to support all browsers from what I can see so not sure

Andrey Sitnik

@ai

show your code

do you use it with flexbox or with grid?

Marcello

@marcebdev

I'm using it with flexbox, I know it only has default support on newer Firefox versions from caniuse but the whole point of postcss-env-preset is supposed to be able to do this, no?

.frame {
  display: flex;
  column-gap: 30px;
}

as I mentioned it wasn't working on Chrome before either, hence why I added preset-env but even with the plugin Safari is not playing nice

Andrey Sitnik

@ai

I afraid that it is a limit of gap polyfill

it works well to polyfill gap for grid

@jonathantneal am I right? should we show a warning on gap and display: flex if we have a browser without gap support for flexbox?

Marcello

@marcebdev

Ah ok, appreciate the insight... I guess for now I will work on making margins work. At least I know I'm not going crazy trying to figure out what is wrong

@ai A warning would be awesome imo luckily I double-checked on safari or I wouldn't have known

Gabo

@GaboPlums_twitter

Hello... I'm a new developer

Andrey Sitnik

@ai

hi :)

Gabo

@GaboPlums_twitter

I wanna learn more about css

At this moment I'm working with tailwind framework

Is very easy and usefull

Andrey Sitnik

@ai

I agree

Gabo

@GaboPlums_twitter

@ai How long have you been working with tailwind?

Andrey Sitnik

@ai

I do not use Tailwind CSS

But I heard the same feelings from the collegues which I trust

Gabo

@GaboPlums_twitter

It sounds great ... you are frontend or full stack?

Andrey Sitnik

@ai

I was a fullstack (Ruby), but not a frontend

But I can do Node.js :) Just do not have a project to be a fullstack

Gabo

@GaboPlums_twitter

Are Ruby and Node.js similar?

Andrey Sitnik

@ai

nope :) very different

Jonathan Neal

@jonathantneal

@marcebdev, the issue with flexbox gap is, as Andrey pointed out, that there is no straightforward way to do it.

One plugin in pursuit of this is https://github.com/limitlessloop/flex-gap-polyfill

The issues are those opened by the author, who is using Custom Properties and margin to simulate the gap. This introduces a number of limitations on the implementation.

With limited support in browsers at the time of this writing, the polyfill would need to be very robust. https://caniuse.com/#feat=flexbox-gap

Jonathan Neal

@jonathantneal

I would recommend you sooner use CSS Grid and the Autoprefixer polyfill. At least then you’re targeting your polyfill squarely at IE11, which reduces your test cases to a single browser.

Joseph Kaptur

@josephkaptur

Hello! I think I've found an edge-case bug in PostCSS. I'm excited to fix it, but I figured I would start here to make sure it really is a bug.

If you have an AST where a node root has child nodes [a, b, c], and you have other nodes d and e, then calling a.replaceWith(a, d, e) results in root having the children [d, e, b, c], not [a, d, e, b, c], as I would expect

Andrey Sitnik

@ai

@josephkaptur Yeap, it is a bug. Send PR to ose branch.

Joseph Kaptur

@josephkaptur

Thanks very much. Should I create an issue on GitHub as well, or just send the PR?

Andrey Sitnik

@ai

@josephkaptur just PR

Where communities thrive

People

Repo info

Activity