Question: why layout values are rounded?

[twop]

Just spent an hour debugging why the leaf node I measured is set to 63.0 when my measure function returned 63.375

When I looked at the sources I found the explicit rounding in the algo.rs

fn round_layout(
    nodes: &mut [NodeData],
    children: &[sys::ChildrenVec<NodeId>],
    root: NodeId,
    abs_x: f32,
    abs_y: f32,
) {
    let layout = &mut nodes[root].layout;
    let abs_x = abs_x + layout.location.x;
    let abs_y = abs_y + layout.location.y;

    layout.location.x = sys::round(layout.location.x);
    layout.location.y = sys::round(layout.location.y);

    layout.size.width = sys::round(abs_x + layout.size.width) - sys::round(abs_x);
    layout.size.height = sys::round(abs_y + layout.size.height) - sys::round(abs_y);

    for child in &children[root] {
        Self::round_layout(nodes, children, *child, abs_x, abs_y);
    }
}

I couldn't find any mentions why the rounding is needed, thus I'm curious. It seems to be 100% intentional. Maybe add an explanation in the code annotations?

[alice-i-cecile]

This is an inherited code base. Unless I can find directions in the spec to do this or behavior from Chrome that matches, I'm going to call this a bug.

[twop]

One other observation:
From the MeasureFunc I returned (63.75, 10.5) but got back (63,10). Note the rounding of 10.5 down to 10, which feels a bit weird

Red box is the result of the layout, note that the text goes out of bounds because of the rounding.

[TimJentzsch]

I also can't find an explanation on why this was implemented in the related commits

• d9e8609
• 1e0eda9

@mockersf also edited parts of that code once in this commit, maybe they know more?

[mockersf]

Browsers round fractional pixels. Some up, some down.

http://www.simonbattersby.com/blog/browsers-and-fractional-pixels/

This lib is made to work like Chrome so it's rounding like it.

[alice-i-cecile]

Browsers round fractional pixels. Some up, some down.

Ah, of course.

My instinct is that I'd like the ability to control whether or not we reproduce Chrome's bugs / quirks. But I'm not quite sure what design we should use for that. I'm not super keen on piping a config enum through everything, but a feature flag also seems very iffy.

Anyways, I'm going to close this out as "not a bug" for now, and we can open up a more focused issue for the "compatibility mode" question as it comes up.