912949 - Layout doesn't get adjusted correctly when resizing the window

Status: RESOLVED INVALID

(Core :: Layout, defect)

Description

[Tobi Reif]

User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7; rv:23.0) Gecko/20100101 Firefox/23.0 (Beta/Release)
Build ID: 20130814063812

Steps to reproduce:

Load http://tobireif.com/ . Drag the right edge of the window to the left (quickly) (eg and back).


Actual results:

Sometimes, the outer circles of the "stamp" (the layout item which includes the mailto link) become very much disaligned with the rest of that area.


Expected results:

The layout (including the circles) should stay intact during and after resize. (Just FYI: works in Chrome).
Perhaps call the resize event one last time right at the end of each resize phase?

Comment 1

[Loic]

Attachment: FF26-screenshot.png

Comment 2

[Tobi Reif]

It might be that the fix is not layout related, but related to events.

(I meant: Perhaps *fire* the resize event one last time right at the end of each resize phase.)

Comment 3

[Boris Zbarsky [:bzbarsky]]

We fire resize events any time a resize happpens; there can't be resizing without a resize event firing later.

It looks like the scripts on the page which manually reposition everything just randomly forget to do it sometimes.... unless this is a graphics invalidation bug with retained layers or something?

Comment 4

[Alice0775 White]

Attachment: semi-reduced zip, open index.html

Comment 5

[Mats Palmgren (inactive)]

Fwiw, if I change:
$(window).resize(function() {
  positionAContactCircle(contactCircle1);
  positionAContactCircle(contactCircle2);
  positionContact();
});

to:
$(window).resize(function() {
  positionContact();
  positionAContactCircle(contactCircle1);
  positionAContactCircle(contactCircle2);
});

then it seems to work as intended.

It looks like positionAContactCircle() is using the current position of
"contact", which is updated in positionContact()...

Comment 6

[Tobi Reif]

Thanks Mats!

Sorry guys for posting an issue whose resolution is not one to take place in the FF code. It worked in Chrome, so I thought my JS code seems to at least potentially be OK (it is possible to run it and get the intended layout), and posted the issue.

When I change the position of the resize functions it works reliably. Strange that it worked in Chrome with the wrong sequence of positioning func calls.

The fixed code is at http://tobireif.com/ .

Closing as "resolved" and "invalid".

Comment 7

[Loic]

Another example that shows Chrome is not automatically the reference for code rendering. :D

Comment 8

[Boris Zbarsky [:bzbarsky]]

It's possible that Chrome fires an extra resize event even when the size has not changed...  or something.