[css-align-3][css-multicol] How is the last-baseline of a multi-col determined? #7639

bfgeek · 2022-08-23T19:54:55Z

Is it the last-baseline of the first fragmentainer or the last?

Today in an inline-context the baseline is always synthesized from the block-end border-box edge. (We likely aren't restricted by compat here however).

tabatkins · 2022-08-29T21:09:43Z

Four options that seem to maybe make sense here are:

Last baseline of the last column
- logically makes the most sense, but least likely to be near the bottom of the container
Last baseline of the first column
- as likely to be relevant as any other filled column, and easiest to talk about
Lowest of the last baselines of all the columns
- Gets a real baseline, not susceptible to large-box-moved-to-next-column fragmentation quirks.
No last baseline; synthesize
- simple, current behavior, but doesn't line up text at all despite the author clearly desiring that

Note: A regular block box (that contains text) uses its last baseline. (Defined in 2.1 for inline blocks, at least, which was the only context the question made sense back then.)

We suspect (3) will give the best behavior; in simple cases it and (2) will be more or less identical, but (3) works better in more cases.

/cc @MurakamiShinyu for opinions

css-meeting-bot · 2022-09-15T19:28:50Z

The CSS Working Group just discussed How is the last-baseline of a multi-col determined?, and agreed to the following:

RESOLVED: (and vice-versa for the last baseline)
RESOLVED: for multicolumns, use the lowest last baseline of all columns as its last baseline

The full IRC log of that discussion

<Rossen_> Topic: How is the last-baseline of a multi-col determined?
<fremy> RESOLVED: (and vice-versa for the last baseline)
<Rossen_> github: https://github.com//issues/7639
<fremy> iank_: for multicolumns elements, we currently take the baseline of the first column
<fremy> iank_: and implementations don't look for a last baseline
<fremy> iank_: maybe we should?
<fremy> iank_: but it's still an option not to, like we do today
<TabAtkins> q+
<fremy> iank_: we can take the last baseline of the first column
<fremy> iank_: or of the last column
<fremy> iank_: or the lowest of all the columns
<fremy> dbaron: exporting the baseline of a multicolumn sounds scary
<fremy> dbaron: especially if you have balancing going on
<dbaron> s/the baseline/the last baseline/
<fremy> iank_: if you use the last column and you balance, it will often work
<fremy> iank_: but without balancing, you can end up somewhere random in the height
<Rossen_> q?
<fremy> fantasai: that seems the worst indeed
<fremy> fantasai: I think the best would be the max
<fremy> TabAtkins: fantasai just said what I was about to say
<fantasai> s/the max/the max, and second best would be the last line of the first column/
<fremy> iank_: the only caveat is that we probably don't want to include spanned items
<fremy> fantasai: why not?
<fremy> fantasai: if sounds like an rare case, but why not?
<fremy> iank_: (nodding)
<fremy> fantasai: I think checking all columns is better
<rachelandrew> +1 to checking all columns
<fremy> fantasai: in case you have a large image that doesn't fit at the bottom of the first column
<fremy> iank_: I'm fine with that
<fremy> iank_: we have some compat constraints for inline-block
<fremy> iank_: but here it's ok
<fremy> fantasai: proposed resolution is to use the lowest last baseline of all columns
<fremy> Rossen_: any objection?
<fremy> RESOLVED: for multicolumns, use the lowest last baseline of all columns as its last baseline
<TabAtkins> ScribeNick:
<TabAtkins> ScribeNick: TabAtkins

MurakamiShinyu · 2022-09-16T01:13:11Z

/cc @MurakamiShinyu for opinions

Sorry for not giving my opinion. I had no idea of inline multi-column use cases, but I agree that the worst is "Last baseline of the last column", and the best is "Lowest of the last baselines of all the columns". So I am content with the resolution.

See: w3c/csswg-drafts#7639 Effectively we need to propagate the last baseline from the lowest of all fragments. This does change the baseline of a multi-col within a inline-block context. Bug: 885175 Change-Id: I8385f89cbe0bb3cc89fc972546f56ddd9ae391bd

mstensho · 2022-10-07T09:04:41Z

Would this be a good opportunity to also fix first-baseline, so that we're consistent about this?

I agree with the resolution, so that last-baseline becomes the lowest last baseline of all columns. But it's quite different from what we do for first-baseline. Why not also make the first-baseline of a multicol container the same as the highest first-baseline of all columns?

Granted, it's very likely that the first column has a good first-baseline, much more so than expecting the last column to have a good last-baseline (the last column may only be partially filled, for instance), but it's not guaranteed. Also, consistency.

bfgeek · 2022-10-07T18:07:26Z

@mstensho Split off into new issue.

See: w3c/csswg-drafts#7639 Effectively we need to propagate the last baseline from the lowest of all fragments. This does change the baseline of a multi-col within a inline-block context. Bug: 885175 Change-Id: I8385f89cbe0bb3cc89fc972546f56ddd9ae391bd

See: w3c/csswg-drafts#7639 Effectively we need to propagate the last baseline from the lowest of all fragments. This does change the baseline of a multi-col within a inline-block context. Bug: 885175 Change-Id: I8385f89cbe0bb3cc89fc972546f56ddd9ae391bd Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3931189 Commit-Queue: Morten Stenshorne <mstensho@chromium.org> Reviewed-by: Morten Stenshorne <mstensho@chromium.org> Cr-Commit-Position: refs/heads/main@{#1056833}

See: w3c/csswg-drafts#7639 Effectively we need to propagate the last baseline from the lowest of all fragments. This does change the baseline of a multi-col within a inline-block context. Bug: 885175 Change-Id: I8385f89cbe0bb3cc89fc972546f56ddd9ae391bd Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3931189 Commit-Queue: Morten Stenshorne <mstensho@chromium.org> Reviewed-by: Morten Stenshorne <mstensho@chromium.org> Cr-Commit-Position: refs/heads/main@{#1056833} NOKEYCHECK=True GitOrigin-RevId: bc669f613244300d069571e47dc3e89b16871937

…ne to lowest of all fragments, a=testonly Automatic update from web-platform-tests [last-baseline] Set multicol last-baseline to lowest of all fragments See: w3c/csswg-drafts#7639 Effectively we need to propagate the last baseline from the lowest of all fragments. This does change the baseline of a multi-col within a inline-block context. Bug: 885175 Change-Id: I8385f89cbe0bb3cc89fc972546f56ddd9ae391bd Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3931189 Commit-Queue: Morten Stenshorne <mstensho@chromium.org> Reviewed-by: Morten Stenshorne <mstensho@chromium.org> Cr-Commit-Position: refs/heads/main@{#1056833} -- wpt-commits: 5e9bb795bc7c57a07ee258f362e50c4014108530 wpt-pr: 36316

…7639

fantasai · 2023-01-20T00:17:38Z

This is edited in, and the latest text should soon be appearing here: https://drafts.csswg.org/css-align-3/#baseline-export
Please let us know if anything seems incorrect.

bfgeek added css-align-3 Current Work css-multicol-1 Current Work labels Aug 23, 2022

bfgeek added the Agenda+ TPAC label Sep 1, 2022

astearns added this to 3:30-4:45 in TPAC Thursday 2022 Sep 11, 2022

astearns moved this from 3:30-4:45 to 11:45-12:30 in TPAC Thursday 2022 Sep 15, 2022

astearns removed the Agenda+ TPAC label Sep 15, 2022

mozilla-apprentice mentioned this issue Sep 15, 2022

[css-align-3][css-multicol] How is the last-baseline of a multi-col determined? mozilla/wg-decisions#893

Open

chromium-wpt-export-bot mentioned this issue Oct 6, 2022

[last-baseline] Set multicol last-baseline to lowest of all fragments web-platform-tests/wpt#36316

Merged

bfgeek mentioned this issue Oct 7, 2022

[css-align-3][css-multicol-1] First baseline of a multicol #7856

Open

fantasai added Closed Accepted by CSSWG Resolution Needs Edits labels Oct 24, 2022

fantasai added a commit that referenced this issue Jan 19, 2023

[css-align-3] Define multicol baselines as highest/lowest set. #7856 #…

c385021

…7639

fantasai removed the Needs Edits label Jan 19, 2023

fantasai added a commit that referenced this issue Jan 20, 2023

[css-align-3] Use alignment, not dominant, baseline. #7856 #7639

3325874

fantasai added the Commenter Response Pending label Jan 20, 2023

rachelandrew removed the css-multicol-1 Current Work label Apr 5, 2024

rachelandrew added the css-multicol-2 label Apr 5, 2024

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[css-align-3][css-multicol] How is the last-baseline of a multi-col determined? #7639

[css-align-3][css-multicol] How is the last-baseline of a multi-col determined? #7639

bfgeek commented Aug 23, 2022

tabatkins commented Aug 29, 2022 •

edited by fantasai

css-meeting-bot commented Sep 15, 2022

MurakamiShinyu commented Sep 16, 2022

mstensho commented Oct 7, 2022

bfgeek commented Oct 7, 2022

fantasai commented Jan 20, 2023

[css-align-3][css-multicol] How is the last-baseline of a multi-col determined? #7639

[css-align-3][css-multicol] How is the last-baseline of a multi-col determined? #7639

Comments

bfgeek commented Aug 23, 2022

tabatkins commented Aug 29, 2022 • edited by fantasai

css-meeting-bot commented Sep 15, 2022

MurakamiShinyu commented Sep 16, 2022

mstensho commented Oct 7, 2022

bfgeek commented Oct 7, 2022

fantasai commented Jan 20, 2023

tabatkins commented Aug 29, 2022 •

edited by fantasai