Re: [Evolution-hackers] Status bar

[Not Zed <notzed ximian com>]

On Fri, 2003-12-05 at 22:57, ERDI Gergo wrote:
> Hi,
>
> > I think you should call it 'on-url', its really only proxying the
> > gtkhtml event anyway.
>
> OK I changed its name, but it's not just a proxy since the signature
> changes from void () (const char*) to void () (const char*, const char*)

That's still, essentially, a proxy.  and to be honest 'url hover' just sounds 'gay'.

> > Actually i dont understand why you are exporting this signal anyway?
> > 
> > You have two places the status is set, one from the signal callback in
> > em-folder-view via the <status> tag, and the other from the callback
> > in mail-component via activity client status.
>
> I thought I already explained it.
> The problem is that there are two competing ways to set the statusbar, and
> parts of Evolution use both.
> THe main shell uses its own Evolution::Component interface, thus when the
> mail component is embedded in the main window, it has to use this.
> On the other hand, Bonobo proper uses the Bonobo::UI interface. This is
> the interface used when you double-click a mail message and it is opened
> in a separate window.
What has 'bonobo proper' to dowith evolution?  If the evolution::component interface suffices?

I still don't grok it.

> > as i've statued beofre, i dont understand this logic.  why not set it
> > to "" intsead of NULL, so you don't have to do this messy stuff?
>
> like I said, look at the implementation of bonobo_ui_component_set_status:
>
>         if (text == NULL ||
>             text [0] == '\0') {
>                 /*
>                  * FIXME: Remove what was there to reveal other msgs
>                  * NB. if we're using the same UI component as the view
>                  * was merged in with, this will result in us loosing our
>                  * status bar altogether - sub-optimal.
>                  */
>                 bonobo_ui_component_rm (component, "/status/main", opt_ev);
Wow, that seems rather 'fucked'.  but sure ok.  i must've missed the last time you said it before.

> so that's why. I realise that there has been some problems in the minor
> details in the patch (I haven't done any straight-C GTK+ hacking since
> ages), but please assume that I know what I'm doing :)


i'll review the patch when on-hours start again.

 Z