Re: [csswg-drafts] [css-grid][cssom] Disallow repeat() syntax in grid-template-rows/columns resolved values

The Working Group just discussed `Disallow repeat() syntax in grid-template-rows/columns resolved values`, and agreed to the following resolutions:

* `RESOLVED: Specify the current behavior in all the browsers except Edge. Just don't use repeat() in grid serialization`

<details><summary>The full IRC log of that discussion</summary>
&lt;dael_> Topic: Disallow repeat() syntax in grid-template-rows/columns resolved values<br>
&lt;dael_> github: https://github.com/w3c/csswg-drafts/issues/2427<br>
&lt;fantasai> ScribeNick: fantasai<br>
&lt;fantasai> TabAtkins: So we resolved to serialize out the repeat()s that were specified,<br>
&lt;fantasai> TabAtkins: but actually you can't do that as fantasai points out in https://github.com/w3c/csswg-drafts/issues/2427#issuecomment-377357237<br>
&lt;fantasai> TabAtkins: So we need to choose some different option<br>
&lt;fantasai> TabAtkins: There are a few possible options<br>
&lt;fantasai> TabAtkins: First is to reverse previous resolution: don't use repeat() in serialization of gCS<br>
&lt;fantasai> TabAtkins: This is very simple and straightforward<br>
&lt;fantasai> TabAtkins: downside is it can potentially produce very long values for grid-template*<br>
&lt;fantasai> TabAtkins: if you do something like grid-template-rows: repeat(10000, 1px)<br>
&lt;fantasai> TabAtkins: Second option is to compress any adjacent tracks that have the same track size (and line names)<br>
&lt;fantasai> astearns: Clarification on 2nd option, is that only when specified value was repeat() or anytime?<br>
&lt;fantasai> TabAtkins: Anytime<br>
&lt;fantasai> TabAtkins: more complicated variants are to track which tracks came from a repeat(), and then collapse them if possible<br>
&lt;fantasai> (These options are summarized in https://github.com/w3c/csswg-drafts/issues/2427#issuecomment-377357237 )<br>
&lt;fantasai> TabAtkins: The Igalia folks don't like tracking which things were in repeat() originally<br>
&lt;fantasai> TabAtkins: seems like too much complexity for little gain<br>
&lt;fantasai> TabAtkins: I think the best thing to do is to not serialize repeat(). The only issue is a blow-out of string sizes if you have long ones<br>
&lt;fantasai> TabAtkins: but there is a cap on the number of tracks so it won't be too crazy<br>
&lt;fantasai> florian: I think I'd go with non-collapsing things<br>
&lt;fantasai> florian: Seems to me that there are a lot of corner cases<br>
&lt;dael> florian: I think I'd go with non-collapsing things as well. As i'm looking through there seems to be lots of corner cases. I'm not sure they'll all be fine.<br>
&lt;dael> florian: Given there's limited value le's skip the pain.<br>
&lt;fantasai> ScribeNick: dael_<br>
&lt;fantasai> ScribeNick: dae<br>
&lt;fantasai> ScribeNick: dael<br>
&lt;dael> florian: Given that you get things like calc that are somewhat the same. I'd rather just not go there.<br>
&lt;dael> astearns: Only edge rep is melanierichards . Since Edge is the browser that retains repeat() and thought we should. I'd like to get an Edge opinion.<br>
&lt;dael> fantasai: frremy says still siding on the don't use repeat() sidde<br>
&lt;dael> TabAtkins: Rossen wants repeat() and frremt would rather not.<br>
&lt;dael> astearns: Since frremy commented on the issue my Edge concern is satisfied.<br>
&lt;dael> astearns: Other comments on if we should deal with repeats or throw them out?<br>
&lt;dael> astearns: Prop: Specify the behavior in all the browsers except Edge. just don't use repeat() in grid serialization<br>
&lt;dael> RESOLVED: Specify the current behavior in all the browsers except Edge. Just don't use repeat() in grid serialization<br>
</details>


-- 
GitHub Notification of comment by css-meeting-bot
Please view or discuss this issue at https://github.com/w3c/csswg-drafts/issues/2427#issuecomment-378781275 using your GitHub account

Received on Wednesday, 4 April 2018 23:58:24 UTC