Re: [csswg-drafts] [css-grid] How to serialize the strings in grid-template-areas?

The CSS Working Group just discussed `How to serialize the strings in grid-template-areas?`, and agreed to the following:

* `RESOLVED: serialize grid-template-areas as simplified computed value for both specified and computed value`

<details><summary>The full IRC log of that discussion</summary>
&lt;dael> Topic: How to serialize the strings in grid-template-areas?<br>
&lt;dael> github: https://github.com/w3c/csswg-drafts/issues/3261<br>
&lt;dael> Rossen_: TabAtkins is IRC only. fantasai are you on?<br>
&lt;dael> ericwilligers: I can give background<br>
&lt;TabAtkins> Firefox preserves strings exactly. (as normal for strings) Other browsers serialize in simplified form.<br>
&lt;dael> ericwilligers: I wrote a wpt so serialization was similar to blink/edge/safari. FF was acting correctly, though, and the wpt was wrong.<br>
&lt;dael> ericwilligers: Question is what does group believe. I think spec is unclear<br>
&lt;dael> ericwilligers: We can defer if no one thought<br>
&lt;dael> heycam: If spec doesn't say else it's natural for strings to retain exact values. normalization inside string value feels a bit weird to me<br>
&lt;dael> ericwilligers: Other argument is 3 of 4 browsers normalize<br>
&lt;dael> florian: We have a micro language in the strings. It's not monolythic. normalization isn't crazy.<br>
&lt;dael> heycam: Did we consider if we want normalization only for computed values?<br>
&lt;dael> Rossen_: For computed values we're going to do simplified and for specified return actual string?<br>
&lt;dael> heycam: Feels more consistent with CSS<br>
&lt;dael> ericwilligers: I had same suggestion<br>
&lt;dael> Rossen_: This would require from current impl FF would need to change. I guess it's up to you if you're okay to change<br>
&lt;dael> heycam: Not a big deal to change. More about what makes sense<br>
&lt;dael> ericwilligers: If only computed value normalizes all browsers have to change. Everyone except FF changes specified<br>
&lt;dael> heycam: I'm arguing for what makes more sense rather from impl<br>
&lt;dael> Rossen_: I don't think anyone is aguing that we want to return simplied for specified.<br>
&lt;dael> ericwilligers: That's what 3 browsers do<br>
&lt;dael> Rossen_: For both speciffied and computed?<br>
&lt;dael> ericwilligers: Yes<br>
&lt;dael> Rossen_: So min implementation cost would be FF to return specified<br>
&lt;dael> ericwilligers: That was my original suggestion and original WPT<br>
&lt;dael> Rossen_: dbaron or heycam? What do you think?<br>
&lt;dael> dbaron: I think heycam had different opinion. Fromm change impl I think it's doable either way. For what behavior ought to be I think I'm more okay then heycam since it seems we're defining a different data type. And I'm okay saying this data type normalizes<br>
&lt;dael> heycam: I'm nto feeling strongly. Maybe even CSSOM guidelines about shortest values we can squint and make this fall under that guideline<br>
&lt;dael> Rossen_: Obj to serialize grid-template-areas as simplified computed value for both specified and computed value?<br>
&lt;dael> RESOLVED: serialize grid-template-areas as simplified computed value for both specified and computed value<br>
</details>


-- 
GitHub Notification of comment by css-meeting-bot
Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/3261#issuecomment-436828363 using your GitHub account

Received on Thursday, 8 November 2018 00:28:48 UTC