[Date Prev][Date Next][Thread Prev][Thread Next][Thread Index]

RE: [XaraXtreme-dev] Text tool shortcuts

We've done some tweaking in this area (Luke knows the details) and might
have changed the behaviour from Xtreme.

I think that both Keypress and Char events are needed and I suggest that
the text tool should claim keypress events even when it does nothing
with them (although it may choose to pass some crucial ones on).


> -----Original Message-----
> From: owner-dev@xxxxxxxxxxxxxxxx 
> [mailto:owner-dev@xxxxxxxxxxxxxxxx] On Behalf Of Martin Wuerthner
> Sent: 21 July 2006 11:04
> To: dev@xxxxxxxxxxxxxx
> Subject: [XaraXtreme-dev] Text tool shortcuts
> I have had a look at text tool shortcuts, which currently do 
> not work and I have found the following:
>  (1) A keypress like "arrow left" moves the caret AND is handled 
>      by the hotkeys system, too, so it causes a nudge operation 
>      as well. So, if you move the caret left in text and go to 
>      the Edit menu, it says "Undo nudge".
>  (2) Ctrl+letter shortcuts (e.g., Ctrl-W) are not working at all.
>  (3) Even if Ctrl-W (swap case) was correctly recognized it would 
>      not work because an underlying function to swap case is not 
>      yet implemented.
> (2) is caused by wxWidgets setting the virtual key of 
> Ctrl-letter combinations to the control code (i.e., 1 for 
> Ctrl-A) instead of the uppercase letter, which is what the 
> code expects. I have fixed that in the Keypress constructor. 
> I also have a fix for (3).
> I found the problem that causes (1) but I am wondering how to 
> best deal with it. The problem is that the application gets 
> THREE key messages for a keypress like "arrow left":
>       IsChar() IsRelease()
> 1st    FALSE     FALSE
> 2nd    TRUE      FALSE
> 3rd    FALSE     TRUE
> The text tool responds to presses with IsChar=TRUE and 
> IsRelease=FALSE while the hotkey system responds to presses 
> with IsChar=FALSE and IsRelease=FALSE. So, the text tool 
> ignores the 1st message, which the hotkey system happily 
> processes. Then, the text tool processes the second message. 
> Surely, something is wrong there. What is the point in having 
> two "press" messages for each keypress? Is this deliberate?
> Martin