[evolution-patches] Re: ctrl-shift-k behavior ...

[Michael Meeks <michael ximian com>]

Hi Michael,

	I guess most of this is lost on people due to the tri-anual evo-patches
submission approval process ;-> so I'll add Jeff in.

On Thu, 2004-02-05 at 00:53 +0800, Not Zed wrote:
> > > This will mess up enabling of the menu though.
> > 
> > 	Sure; but it seems to me few people look at that menu - and if c-s-k
> > still does something useful even when it's unread it's reasonable to
> > have it sensitive.
> 
> But it doesn't do anything useful if its read its a noop.  I dont get
> how it helps your problem either, since it will still timeout and mark
> it read anyway.

	It's not a no-op; if you hit C-S-k (and it's sensitive) on a mail that
is unread [ and you want it to remain unread - hence the
C-S-k ] it does:

emfv_popup_mark_unread(GtkWidget *w, EMFolderView *emfv)
{
	... no-op (re-mark unread) ...

	if (emfv->priv->seen_id) {
		g_source_remove(emfv->priv->seen_id);
		emfv->priv->seen_id = 0;
	}

	ie. it stops the timeout marking it as read.

	The current behavior is pain and sadness, since you move to a mail -
wish to leave it as 'unread' for future reference; hit C-S-k but it
still gets marked read; indeed you can only stop that behavior by
patiently waiting for it to be marked read; and then mark it unread
again.

	This is useful since often I want to mark something for future re-
reading, but want to skim it quickly now; and it also gives some
certainty to the process of switching mails that it is in fact correctly
marked and the timeout won't swallow it while switching or somesuch.

> > > On Tue, 2004-02-03 at 19:47 +0000, Michael Meeks wrote:
> > > > 	For some time I've been annoyed that if you move to a message and hit
> > > > ctrl-shift-k before the timeout it still times out, and sets it as read :-)
> > > > just had a quick hack, and it turns out this is down to the fact that
> > > > 'mark unread' is not sensitive when the mail is un-read - so we never hit
> > > > that code path; this tiny patch 'fixes' that; may I commit it ?
> > 
> > 	May I commit ?
> 
> No, because of the above, it should be fixed properly (otherwise we WILL
> get another bug report about the menu sensitisation).

	Well - it's pretty hard to even see the menu during that timeout; I had
to extend mine substantially to get that.

> This is oone of those bugs which keeps gets fixed and seems to break
> again when some seemingly unrelated bug gets fixed.

	I've never known this behave as I expected in evolution of any version.

	Regards,

		Michael.

-- 
 michael ximian com  <><, Pseudo Engineer, itinerant idiot