Content: Missing items from CSS_Ref

Created on 4 Jan 2021 · 11Comments · Source: mdn/content

As raised in the sprints repo https://github.com/mdn/sprints/issues/2068 there are various missing properties in the CSS Reference at https://developer.mozilla.org/en-US/docs/Web/CSS/Reference

The pages for the missing properties exist on MDN.

I'm not sure how this is generated, but I can go fix the missing items if I'm pointed in the right direction.

Source

rachelandrew

Most helpful comment

Thanks @sideshowbarker @rachelandrew @chrisdavidmills! I just reviewed/merged @sideshowbarker's https://github.com/mdn/yari/pull/2338, which should take care of adding these CSS items with an experimental status. I'm also going to publish a new npm package for mdn-data, which should take care of things like the missing user-select (see https://github.com/mdn/sprints/issues/2068).

escattone on 7 Jan 2021

👍4

All 11 comments

As far as I can see from looking at the https://github.com/mdn/content/blob/main/files/en-us/web/css/reference/index.html source, the content of that article, none of it is generated; instead it’s added to manually.

sideshowbarker on 4 Jan 2021

it's the property names in the index I'm talking about, sorry if that wasn't clear. They are generated by the macro {{CSS_Ref}}

rachelandrew on 4 Jan 2021

it's the property names in the index I'm talking about, sorry if that wasn't clear.

Thanks, I see now — sorry for having been daft

From looking into the sources, here’s a short summary of how it seems to work:

https://github.com/mdn/data/tree/master/css is the upstream source it’s generated from
https://www.npmjs.com/package/mdn-data is the downstream resource it’s directly taken from
https://www.npmjs.com/package/mdn-data was last published 2 months ago, so there’s a lag
https://github.com/mdn/data/tree/master/css items without "status": "standard" don’t get picked up

So, taking user-select as an example, it’s in https://github.com/mdn/data/blob/e545c59014e0760af428145b5f56125423fa77f9/css/properties.json#L8976-L8991 — but per https://github.com/mdn/data/blame/e545c59014e0760af428145b5f56125423fa77f9/css/properties.json#L8989 and https://github.com/mdn/data/commit/65ef6ada8c5504389905b26c270ce8c2015835b7, was nonstandard until Nov 23rd.

So now it would show up in the index — after a new https://www.npmjs.com/package/mdn-data is released — but because the current downstream package is two months behind the changes to the https://github.com/mdn/data upstream source, it’s not showing up yet.

Taking a look at a couple others: aspect-ratio and appearance don’t show up because per https://github.com/mdn/data/blob/master/css/properties.json#L1948-L1963 and https://github.com/mdn/data/blob/master/css/properties.json#L1932-L1947, they’re both marked "status": "experimental". If they get marked standard, they’ll show up in the index.

https://github.com/mdn/yari/blob/56db834cc548d63dbd6fcd7132893ecb08db05b9/kumascript/macros/CSS_Ref.ejs#L11-L13 is what shows it only grabs "status": "standard" items.

sideshowbarker on 5 Jan 2021

@sideshowbarker I think there is a case for including the experimental items in this list. Some of them have a lot of developer interest, and there aren't a huge number of them (I was involved in a project last year to figure out which we should remove the flag from). I feel that if there is enough reason for us to document them there is reason to expose them in the list.

Pinging @chrisdavidmills for his opinion on that.

rachelandrew on 5 Jan 2021

👍1

I think there is a case for including the experimental items in this list… I feel that if there is enough reason for us to document them there is reason to expose them in the list.

100% agreed, for exactly that same line of thinking

Pinging @chrisdavidmills for his opinion on that.

I’ve also gone ahead and raised https://github.com/mdn/yari/pull/2338 with the change to the macro source that’d be needed to enable this. That allows testing it locally. So with that in place in my local environment, I’ll generate a diff for https://developer.mozilla.org/en-US/docs/Web/CSS/Reference output (in a few hours), so we can see exactly what all it would add that’s not included now.

sideshowbarker on 5 Jan 2021

@sideshowbarker I think there is a case for including the experimental items in this list. Some of them have a lot of developer interest, and there aren't a huge number of them (I was involved in a project last year to figure out which we should remove the flag from). I feel that if there is enough reason for us to document them there is reason to expose them in the list.

Pinging @chrisdavidmills for his opinion on that.

Yes, agreed.