[PATCH] winecfg: Update changed state on associations checkbox click.

[Nikolay Sivov]

Signed-off-by: Nikolay Sivov 
---
 programs/winecfg/theme.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/programs/winecfg/theme.c b/programs/winecfg/theme.c
index 1934538f44..5304f7240c 100644
--- a/programs/winecfg/theme.c
+++ b/programs/winecfg/theme.c
@@ -1320,6 +1320,7 @@ ThemeDlgProc (HWND hDlg, UINT uMsg, WPARAM wParam, LPARAM lParam)

case IDC_ENABLE_FILE_ASSOCIATIONS:
                             update_mime_types(hDlg);
+                            SendMessageW(GetParent(hDlg), PSM_CHANGED, 0, 0);
                             break;
                     }
                     break;

[Dmitry Timoshkov]

Nikolay Sivov  wrote:
>                      case IDC_ENABLE_FILE_ASSOCIATIONS:
>                          update_mime_types(hDlg);
> -
>                        SendMessageW(GetParent(hDlg), PSM_CHANGED, 0, 0);
>                        break;

There is no need for that, update_mime_types() already takes care about
the change. There are other handlers that don't send PSM_CHANGED too.

[Nikolay Sivov]

On 8/17/2018 3:21 AM, Dmitry Timoshkov wrote:
> Nikolay Sivov  wrote:
> >                      case IDC_ENABLE_FILE_ASSOCIATIONS:
> >                          update_mime_types(hDlg);
> > -
> >                        SendMessageW(GetParent(hDlg), PSM_CHANGED, 0, 0);
> >                        break;
>
> There is no need for that, update_mime_types() already takes care about
> the change. There are other handlers that don't send PSM_CHANGED too.

Takes care about what? It's still possible to cancel the change. For
consistency all of them should send that.

[Dmitry Timoshkov]

Nikolay Sivov  wrote:
> > >                      case IDC_ENABLE_FILE_ASSOCIATIONS:
> > >                          update_mime_types(hDlg);
> > > -
> > >                        SendMessageW(GetParent(hDlg), PSM_CHANGED, 0, 0);
> > >                        break;
> >
> > There is no need for that, update_mime_types() already takes care about
> > the change. There are other handlers that don't send PSM_CHANGED too.
>
> Takes care about what?

About the change.
> It's still possible to cancel the change.

No, the things like check boxes don't need the cancelling.
> For consistency all of them should send that.

I tend to disagree, and since there are other switches that behave same
way as my patch does there are others who also do. Anyway if you don't
like this you may want to start with actually adding support for cancelling
the change, in current state your patch is useless.

[Nikolay Sivov]

On 8/17/2018 7:35 AM, Dmitry Timoshkov wrote:
> Nikolay Sivov  wrote:
> > > >                      case IDC_ENABLE_FILE_ASSOCIATIONS:
> > > >                          update_mime_types(hDlg);
> > > > -
> > > >                        SendMessageW(GetParent(hDlg), PSM_CHANGED, 0, 0);
> > > >                        break;
> > >
> > > There is no need for that, update_mime_types() already takes care about
> > > the change. There are other handlers that don't send PSM_CHANGED too.
> >
> > Takes care about what?
>
> About the change.
> > It's still possible to cancel the change.
>
> No, the things like check boxes don't need the cancelling.

Have you actually tried? set_reg_key() does not write to registry.
> > For consistency all of them should send that.
>
> I tend to disagree, and since there are other switches that behave same
> way as my patch does there are others who also do. Anyway if you don't
> like this you may want to start with actually adding support for cancelling
> the change, in current state your patch is useless.

Cancelling is already supported.

[Dmitry Timoshkov]

Nikolay Sivov  wrote:
> > > It's still possible to cancel the change.
> >
> > No, the things like check boxes don't need the cancelling.
>
> Have you actually tried? set_reg_key() does not write to registry.
> > > For consistency all of them should send that.
> >
> > I tend to disagree, and since there are other switches that behave same
> > way as my patch does there are others who also do. Anyway if you don't
> > like this you may want to start with actually adding support for cancelling
> > the change, in current state your patch is useless.
>
> Cancelling is already supported.

Urgh, my bad. My apologies for the confusion. Your patch works as intended.