- From: Mats Palmgren <mats@mozilla.com>
- Date: Wed, 2 Dec 2015 21:32:04 +0100
- To: www-style@w3.org
- Cc: Daniel Holbert <dholbert@mozilla.com>
https://drafts.csswg.org/css-grid/#valdef-repeat-auto-fill I have a few questions on auto-fill/fit. 1. On 'auto-fit' it says "except that after grid item placement any empty repetitions are dropped". It's not clear how tracks are "dropped". For example, how many columns does this grid have? <div style="display: grid; grid-template: repeat(auto-fit,100px) / 10px; width: 200px;"> <item style="grid-column:2"></item> </div> The text says we should "drop" empty tracks; column 1 is empty, column 2 is not empty; so we drop column 1, and the answer is that the grid has one column. When I asked Daniel Holbert what he thought, he said "they probably only meant for empty tracks to be dropped from the end", which makes sense in retrospect. If so, then I misunderstood what the text meant here so this definitely needs clarification in the spec text. (FWIW, I actually implemented the drop-empty-repeat-tracks- everywhere for 'auto-fit' and it's about a dozen extra lines of code compared to drop-empty-repeat-tracks-from-the-end version (in Gecko). In case you've considered adding an 'auto-compact' repeat keyword or something ;-) .) 2. The "Otherwise, the specified track list repeats only once." at the end of 'auto-fill' is slightly ambiguous for what it means for 'auto-fit'. I'm guessing that 'auto-fit' can remove that last track too if it's empty? Perhaps you could explicitly call that out under 'auto-fit', with for example: "This means that 'auto-fit' can result in zero tracks if all of them are empty." 3. I can't find a formal definition of what an "empty track" is. For example, "repeat(auto-fit,[a] 100px [b])" with a grid-aligned abs.pos. child placed between a and b - is that track empty? I assume it is, because of the general "such elements do not take up space or otherwise participate in the layout of the grid" note in 10.1, but it would be good to have a formal definition of it. Thanks, Mats
Received on Wednesday, 2 December 2015 20:32:36 UTC