Bug 4032

Created attachment 1876 [details]
workaround for wrong modifier detection

This looks like the place where space is assigned the erratic dual purpose behavior.

Presumably this should be exclusive. That is if mod_pressed is true the mod binding should be handled and the non-mod binding not.

Created attachment 1877 [details]
make space action unambiguous

This will however disable the default action because mod_pressed is obviously always true.

ok, it's way too broken.

The mod_pressed is set properly. However, displaying the message in message view breaks out if summaryview->displayed != summaryview->selected. When the message is already displayed the handling falls through to scrolling which has both mod_pressed and !mod_pressed branches.

What is space supposed to do, exactly?

Can one of the mod_pressed or !mod_pressed branch be moved into SHIFT handling?

If it does not make sense maybe the whole overengineered switches like 
 if (!mimeview_scroll_page(messageview->mimeview, TRUE))
   summary_select_prev_unread(summaryview);
can be dropped?

Can't the key combination do just one thing?

Created attachment 1878 [details]
make space key bindings unambiguous

Created attachment 1879 [details]
make space handling in summaryview unambiguous

Existing handling:

Shift-space: mimeview_scroll_page UP
Mod1-Shift-space: mimeview_scroll_page UP
Mod1-space: mimeview_scroll_page UP || summary_select_prev_unread
space: display message || mimeview_scroll_page DOWN || summary_select_next_unread

New handling:

Mod1-space: next unread
Mod1-Shift-space: prev unread

space: display message || scroll DOWN
Shift-space: display message || scroll UP

Created attachment 1880 [details]
make space handling in summaryview unambiguous

Fix the modifier handling

Created attachment 1881 [details]
fix shift handling in messageview

Another place where the modifier handling is not exactly nice and easy to explain.

I suppose if these bindings *were* documented people would complain they do not work as described.

Rather than being "completely fucked up", as you call it, it can actually be a convenient way to step through new/unread mails in a folder. It opens a msg, scrolls through the msg, and then opens the next new or unread msg.

It however means that if you press space one more time than required you jump to some random place in the inbox and there is no way to get back to what you were reading.

I call that fucked up interface.

(In reply to comment #9)
> It however means that if you press space one more time than required you
> jump to some random place in the inbox and there is no way to get back to
> what you were reading.
> 
> I call that fucked up interface.

No, it's not random. It goes to the next unread or new msg.

(In reply to comment #10)
> (In reply to comment #9)
> > It however means that if you press space one more time than required you
> > jump to some random place in the inbox and there is no way to get back to
> > what you were reading.
> > 
> > I call that fucked up interface.
> 
> No, it's not random. It goes to the next unread or new msg.

Which has nothing to do with the message I am viewing or even the part of the summary I am viewing.

So to put it another way - how do I display the message selected in the summary *without* the possibility of the selection randomly jumping somewhere completely else?

PageUp / PageDown ?

(In reply to comment #12)
> PageUp / PageDown ?

Those work once the message is displayed. I was asking about loading the selected message into the messageview.

press the 'Enter' for that.

(And please stop re-opening this bug item)