third_party/WebKit/LayoutTests/paint/invalidation/table-two-pass-layout-overpaint-expected.html - Issue 2670603002: Avoid pathological layout on nested percent height tables

Unified Diff: third_party/WebKit/LayoutTests/paint/invalidation/table-two-pass-layout-overpaint-expected.html

Issue 2670603002: Avoid pathological layout on nested percent height tables (Closed)

Patch Set: bug 687061 Created 3 years, 10 months ago

Use n/p to move between diff chunks; N/P to move between comments. Draft comments are only viewable by you.

Jump to:

View side-by-side diff with in-line comments

« no previous file with comments | « third_party/WebKit/LayoutTests/fast/css-intrinsic-dimensions/height-css-tables-expected.html ('k') | third_party/WebKit/LayoutTests/paint/invalidation/table-two-pass-layout-overpaint-expected.txt » ('j') | third_party/WebKit/PerformanceTests/Layout/nested-percent-height-tables.html » ('J')
Expand Comments ('e') | Collapse Comments ('c') | Hide Comments ('s')

Index: third_party/WebKit/LayoutTests/paint/invalidation/table-two-pass-layout-overpaint-expected.html

diff --git a/third_party/WebKit/LayoutTests/paint/invalidation/table-two-pass-layout-overpaint-expected.html b/third_party/WebKit/LayoutTests/paint/invalidation/table-two-pass-layout-overpaint-expected.html

index 013c37972c68d297293d4cad466e57579afd37df..58a291a3a0bb3734d209d6c714213f2248581d40 100644

--- a/third_party/WebKit/LayoutTests/paint/invalidation/table-two-pass-layout-overpaint-expected.html

+++ b/third_party/WebKit/LayoutTests/paint/invalidation/table-two-pass-layout-overpaint-expected.html

@@ -1,7 +1,7 @@

<html>

<head>

<style>

- table { background-color: lightblue; height: 100px; }

+ table { background-color: lightblue; height: 99px; }

mstensho (USE GERRIT) 2017/02/06 21:11:07 This seems wrong. The table in the test really is

mstensho (USE GERRIT) 2017/02/07 19:52:33 This is still unresolved.

rhogan 2017/02/07 20:27:42 Without my CL the rendered height on the table in

mstensho (USE GERRIT) 2017/02/07 21:09:10 Oh dear. I had no idea it was this broken. Thanks

Oh dear. I had no idea it was this broken. Thanks for the explanation. Yeah, attacking that FIXME sounds like the proper way to fix this (and it might become more urgent with this fix landed... then again, nobody has been complaining loudly about us losing a couple of pixels either, so...). So the test used to pass by accident. It's confusing that the test and the ref don't use the same table height, though. I think you should either: A: Keep the ref at 100px and change it to also use two-pass layout (by changing the table width, or by setting display:none on some new sibling of #target, or whatever), and document why you do this. or: B: document why it has to be 99px. A is silly, but, on the other hand, the specified height mismatch between the test and the ref is hacky. We probably need a bug report for the FIXME at this point, so that we can refer to it.

td { width: 100px; height: 50px; }

div#target { height: 25px; width: 50px; background-color: lightgreen; }

</style>