This is the mail archive of the
cygwin-apps
mailing list for the Cygwin project.
Re: [PATCH] setup: replace radiobuttons on chooser
- From: Max Bowsher <maxb1 at ukf dot net>
- To: cygwin-apps at cygwin dot com
- Date: Wed, 15 Mar 2006 23:21:09 +0000
- Subject: Re: [PATCH] setup: replace radiobuttons on chooser
- Openpgp: id=C0F2C580
- References: <00ad01c64868$69542a40$a501a8c0@CAM.ARTIMI.COM>
- Reply-to: cygwin-apps at cygwin dot com
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Dave Korn wrote:
> On 14 March 2006 16:20, Dave Korn wrote:
>
>> On 14 March 2006 16:03, Max Bowsher wrote:
>
>>> Dave Korn wrote:
>
>>>> The radiobuttons for Prev, Keep, Curr, Exp at the top of the package
>>>> chooser should never have been radiobuttons. They don't select a
>>>> persistent mode, they perform an instantaneous action. Action buttons
>>>> should be pushbuttons. Hence this patch changes them to pushbuttons.
>>> They seem to have BOTH an instantaneous AND a persistent influence,
>>> actually - so I'd prefer to keep using controls which display the
>>> current state.
>> Huh? AYS? As far as I can see, there is code in
>> ChooserPage::OnMessageCmd that responds to clicking any given one by calling
>> keepClicked() or changeTrust(xxx), and there's nothing else anywhere else
>> that could query or in any way respond to the state of the buttons or that
>> tests which one is checked or anything. Why do you say they have a
>> persistent influence? I'm /very/ convinced that they change the state of
>> the selection when clicked and it's a one-shot, what bit of the code are
>> you looking at that suggests otherwise?
>
>
> Max, ping? Have you looked at this?
OK, now I have. The PickView object has a member variable deftrust,
which is used in numerous places, and set by the radio button change code.
> If I don't hear from you in the next
> few hours I'll check in part 2 of the patch on its own like you asked,
Could you post it to the list, first? To clarify, I haven't looked in
detail at part 2, I only noted that part 1 had a conceptual issue, and
part 2 sounded much less controversial, and could benefit from being
handled separately.
> but I'm still convinced part 1 is good as well.
Sorry, definitely looks like there is persistent state involved, which
action buttons would not be able to display.
Max.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2.1 (Cygwin)
iD8DBQFEGKFlfFNSmcDyxYARAnJ4AJ47UO9hP84t3c6Ah6RaOCOwe3+C7QCeKr65
cZoY2r9gUUtJLLwEn4oyXiA=
=vpwE
-----END PGP SIGNATURE-----