114580 - commonDialog is slow as molasses

Status: RESOLVED FIXED

(SeaMonkey :: UI Design, defect)

Description

[Joe Hewitt (gone)]

Ever try calling alert() in rapid succession, and notice that the dialog takes
forever to appear? I have.

commonDialog.xul needs to get spanked.

Comment 1

[Joe Hewitt (gone)]

Working on this... just by changing collapsed="true" to hidden="true" on the
optional container boxes, I save about 60% of window open time.  Hyatt reminded
me that doing this is tricky because that will cause the textbox/checkbox
bindings to load async when they are shown, so I have to be sure to unset hidden
from an inline function call instead of from onload.

Comment 2

[Joe Hewitt (gone)]

Attachment: patch

This patch cleans up the commonDialog xul and js, changes collapsed to hidden
on the textbox containers, converts to use the dialog tag, and updates the
dialog tag to support extra buttons.

Comment 3

[Joe Hewitt (gone)]

Attachment: improved patch

Comment 4

[Joe Hewitt (gone)]

Attachment: patch 3.0

Comment 5

[jag (Peter Annema)]

r=/sr=jag with the bug pointer.

Comment 6

[Brian Ryner (not reading)]

Comment on attachment 63887 [details] [diff] [review]
patch 3.0

r=bryner

Comment 7

[Christopher Blizzard (:blizzard)]

a=blizzard on behalf of drivers for the cropping fixes

Comment 8

[Asa Dotzler [:asa]]

looks like this is checked in. can it be resolved?

Comment 9

[[e-mail] (gone - use [e-mail] instead)]

Asa: only dialog.xml was checked in, the rest of the patch hasn't landed yet.

Comment 10

[David Baron :dbaron: (⌚️UTC-4, no longer working on Mozilla)]

Removing the 0.9.8+ since my memory of the discussion was that we only wanted
part of the patch for 0.9.8 (and thought the rest could wait), and it looks like
part of the patch was checked in (I'm assuming the part we decided).  If I got
confused here and there's still something that needs to go in for 0.9.8, please
get it in sooner rather than later.

Comment 11

[Joe Hewitt (gone)]

almost forgot about this - landed on trunk now

Comment 12

[Brian Ryner (not reading)]

This checkin caused the regression in bug 123216.