295571 - Populating a select box using onmousedown not longer works.

Status: RESOLVED FIXED

(Core :: Layout: Form Controls, defect)

Description

[Jon Levell]

I work on a complicated intranet web app that contains a lot of select boxes. If
these are populated in the usual way it makes navigating around our interface
very unwieldy so we only populate the select box options "Just in Time" (see the
URL for an example).

This works brilliantly upto and including Firefox 1.0.4 but on the trunk, this
seems to be broken. I'm guessing this bug will be marked as invalid but were
that to be the case I think it would be very limiting for future web apps with
complicated user interfaces.

Comment 1

[Bob Clary [:bc] (inactive)]

Tested with a Deer Park 1, 20050525 winxp build. When you tab to the select and
press the arrow down key repeatedly, the select box is populated transparently
to the user. When you click on the select, only the initial value is displayed
until you release the mouse and click the select again.

It appears that the select is not redrawn when the new options are added.
Not js engine, over to Layout Form Controls.

Comment 2

[Andrew Schultz]

this regressed between linux trunk 2005042805 and 2005042901, probably bug 240276.

Comment 3

[Caleb]

I believe that this regression should be fixed for 1.8b3 or atleast 1.8b4.

Reproduced on Windows XP aswell, setting OS -> All, blocking1.b3 -> ?

Comment 4

[Robert O'Callahan (:roc) (email my personal email if necessary)]

I think everything's OK except the dropdown widget isn't being resized.

Comment 5

[Robert O'Callahan (:roc) (email my personal email if necessary)]

Attachment: fix

The bug is that the combobox frame doesn't allow its dropdown listbox's view to
resize during reflow. Usually this is OK and a performance optimization because
we resize and reposition the view when we show the dropdown. Of course, when
the dropdown is already showing during reflow, we need to let the view resize.
(I don't know how this worked before!) This patch is simple and does that.

Comment 6

[Mats Palmgren (inactive)]

*** Bug 298198 has been marked as a duplicate of this bug. ***

Comment 7

[Robert O'Callahan (:roc) (email my personal email if necessary)]

Comment on attachment 186565 [details] [diff] [review]
fix

Simple fix to a regression that hurts dynamic Web sites

Comment 8

[chris hofmann]

Comment on attachment 186565 [details] [diff] [review]
fix

a=chofmann

Comment 9

[Robert O'Callahan (:roc) (email my personal email if necessary)]

checked in

Comment 10

[jag (Peter Annema)]

It doesn't look like backing this out helped btek Tp, though unfortunately Neil
backed out his change so we only get to see one cycle's worth of data.