forked from mirrors/gecko-dev
71aa26835f
The regressing bug regressed some calc() expressions like the one in the
test-case:
max-width: calc(100% - <some-fixed-length>);
Where floating point precision basically made us lose a little bit of
precision here and there.
Before bug 1841612, this worked because we'd operate in fixed point for
such simple expressions at least. Things could easily go equally south
for more complex things including e.g. rem().
Rounding seems like the better default for calc() since we basically
have no idea of what the output is going to be, and makes the answer
more likely to be precise.
Note that this only affects actual mixed calc() expressions: If things
resolve to a percentage we keep existing behavior.
Differential Revision: https://phabricator.services.mozilla.com/D202746
37 lines
939 B
HTML
37 lines
939 B
HTML
<!doctype html>
|
|
<!--
|
|
Any copyright is dedicated to the Public Domain.
|
|
http://creativecommons.org/publicdomain/zero/1.0/
|
|
-->
|
|
<meta charset="utf-8">
|
|
<title>CSS Test: calc rounding doesn't accumulate a lot of error.</title>
|
|
<link rel="author" href="mailto:mats@mozilla.com" title="Mats Palmgren">
|
|
<link rel="author" href="mailto:emilio@crisal.io" title="Emilio Cobos Álvarez">
|
|
<link rel="help" href="https://bugzilla.mozilla.org/show_bug.cgi?id=1323735">
|
|
<link rel="help" href="https://drafts.csswg.org/css-values/#funcdef-calc">
|
|
<link rel="match" href="calc-rounding-001-ref.html">
|
|
<style>
|
|
body {
|
|
background: #f3f5f6;
|
|
}
|
|
|
|
div {
|
|
font-size: 15px;
|
|
width: 401px;
|
|
margin: 20px;
|
|
background: #fff;
|
|
display: flex;
|
|
flex-wrap: wrap;
|
|
}
|
|
|
|
b {
|
|
height: 50px;
|
|
background: red;
|
|
width: calc((100% - 4.5em) / 4); /* .5em gutters */
|
|
}
|
|
|
|
b:not(:last-child) {
|
|
margin-right: 1.5em;
|
|
}
|
|
</style>
|
|
<div><b></b><b></b><b></b><b></b></div>
|