130388 - Plus/Minus keys on numeric keypad won't expand/contract branches in trees

Status: NEW

(Core :: XUL, enhancement, P4)

Description

[Edward Mendelson]

In most Windows apps, when you see a plus sign next to a collapsed branch of a
tree, you can press the Plus key (either the one next to Backspace or the one on
the number pad) to expand that branch, or press the Minus key to close it again.
These keys don't work in the Mozilla 2002031104 History tree (under Windows).  

One advantage of Mozilla is that it supports the keyboard so well - can we get
this working here?

Thanks!

Edward Mendelson
Contributing Editor
PC Magazine

Comment 1

[Jesse Ruderman]

Tested several apps on Windows 98:

mIRC 6 prefs                  no
Netscape 4 prefs              only number pad +,-
Windows Explorer folder tree  only number pad +,-
Winamp prefs                  only number pad +,-
IE 6 security settings        only number pad +,-
IE 6 advanced prefs           only number pad +,-
IE 6 favorites sidebar        only number pad +,-

All of these trees, as well as Mozilla's, support the left and right arrow keys 
for collapsing and expanding branches.  Most support number pad +,- but none 
support -/=/+ in the main part of the keyboard.

This bug isn't specific to history; it also happens in prefs and in the mail 
folder pane.  -> Tree widget.

Comment 2

[Edward Mendelson]

Jesse is right: other apps' trees only respond to the Plus/Minus keys on the
numberpad.  I was wrong to mention the Plus/Minus keys near the Backspace -
other apps don't respond to those. 

So a bugfix would cause the trees in history, etc., to respond to the Plus/Minus
keys on the numberpad, but NOT the Plus/Minus keys near the backspace.

Thanks again...

Edward Mendelson
Contributing Editor
PC Magazine

Comment 3

[Kyle Yuan]

This bug is a part of bug 135749, but much easier to fix. So I want to put a 
patch here to let "+", "-" work first, then study how to support "*" and 
Ctrl+"+" later.

Comment 4

[Kyle Yuan]

Attachment: patch

just copy the implementation of "vk_left" and "vk_down" but change event from
"keypress" to "keydown", because event handler does not send "keypress" event
when the vk_add was pressed.

Comment 5

[Kyle Yuan]

Hi, Jesse, I found Mozilla maps many keys code to NS_VK_SUBTRACT:
1. Windows, Numpad '-'(code 109) and '-' near the backspace (code 189);
2. GTK, GDK_minus, GDK_KP_Subtract and GDK_underscore

So it is why Numpad '-' and '-' near the backspace do the same work.
I'm afraid I can not change this way :(

Comment 6

[Kyle Yuan]

akkana, what do you think about comment 5?

Comment 7

[Akkana Peck]

Mapping GDK_underscore to VK_SUBTRACT sounds wrong, but looking at
nsIDOMKeyEvent.idl and nsGUIEvent.h I don't see an NS_VK or DOM_VK code for
underscore.  I wonder what we're supposed to use for that?  What do the Windows
and Mac event handling code do for the underscore keys?

Distinguishing keypad minus from regular minus looks like the same issue:
neither the DOM key event model, nor the derived Mozilla model, offer separate
symbols to use for this.

Joki, what is the W3C position on this -- are we not supposed to be able to
distinguish the two '-' keys, or use an underscore at all?  Can we add DOM_VK
symbols for non-keypad dash and underscore?  Should we add them in the mozilla
event model even if the dom doesn't distinguish them?  (Probably not, since then
people writing DOM-compatible event handling code would get a different result
in mozilla than in apps which strictly followed the dom spec.)  Also adding jst
in case he knows what the W3C is doing for dom events.

Comment 8

[Kyle Yuan]

Windows mapping code is:
http://lxr.mozilla.org/seamonkey/source/widget/src/windows/nsWindow.cpp#2613

GTK mapping code is:
http://lxr.mozilla.org/seamonkey/source/widget/src/gtk/nsGtkEventHandler.cpp#168
http://lxr.mozilla.org/seamonkey/source/widget/src/gtk/nsGtkEventHandler.cpp#185
http://lxr.mozilla.org/seamonkey/source/widget/src/gtk/nsGtkEventHandler.cpp#209

If we can not distinguish keypad minus from regular minus, how can we implement 
the feature described in this bug?

Comment 9

[Kyle Yuan]

I want to take this bug...

Comment 10

[Kyle Yuan]

cc bryner...

Comment 11

[Kyle Yuan]

*** Bug 146726 has been marked as a duplicate of this bug. ***

Comment 12

[Serge Gautherie (:sgautherie)]

(see comment 1)
Summary: "history tree" should be replaced by "XUL trees".

Comment 13

[Serge Gautherie (:sgautherie)]

See my bug 147348 comment 7 (and 5),
for confirmation on Mozilla v1.3b/W2K (and v1.2b/W95).

There has been nearly no activity for 9 months: any chance to have some for v1.a ?

Comment 14

[Serge Gautherie (:sgautherie)]

Of course, I meant "v1.4a" in my previous comment :->

Comment 15

[Kyle Yuan]

Serge, the fix for this bug is quite simple, but bug 153344 and bug 154207 block
us, nobody wants to take a big risk for this minor UI enhancement :(

Comment 16

[Serge Gautherie (:sgautherie)]

[Mozilla/5.0 (Windows; U; Win95; en-US; rv:1.4a) Gecko/20030401]

Bug still there.

Comment 17

[Stefan Borggraefe]

*** Bug 216425 has been marked as a duplicate of this bug. ***

Comment 18

[Aaron Leventhal]

*** Bug 187202 has been marked as a duplicate of this bug. ***

Comment 20

[BugBot [:suhaib / :marco/ :calixte]]

The bug assignee didn't login in Bugzilla in the last 7 months.
:enndeakin, could you have a look please?
For more information, please visit auto_nag documentation.