From: Sean Kamath <kamath@ge...> - 2004-01-22 06:51:56
So Jason just added me as a developer, and I haven't got a good feel
for how you all distribute work. I'm assuming "if you got the time,
have at it" is the general rule.
Since I'm sick, I'm not at work, and I figure "I got time". :-)
There are two outstanding bugs I think might be related, or at least
fixed in the same region of code: shift-tab not working, and
I grabbed the rfbproto.pdf spec file, and note that there are two
things it says:
1) Servers should ignore lock keysyms such as CapsLock and NumLock
where possible. Instead they should interpret each character-based
keysym according to its case.
2) There is no universal standard for backward tab in the X Window
System. On some systems shift+tab gives the keysym ISO Left Tab,
on others it gives a private BackTab keysym and on others it gives
Tab and applications tell from the shift state that it means
backward-tab rather than forward-tab. In the RFB protocol the
latter approach is preferred. Viewers should generate a shifted Tab
rather than ISO Left Tab. However, to be backwards-compatible with
existing viewers, servers should also recognise ISO Left Tab as
meaning a shifted Tab.
For the latter, when you hit Shift-Tab, the mac returns
NSBackTabCharacter and that is what is sent.
1) ignore capslock/numlock, and check caplock/numlock when sending
characters, and send pre-shifted characters. This kinda sucks
because I can't send a shift, then the character. A quick check
using X11, and it appears that the CAPSLOCK key only affects a-z.
So I could just map a-z to A-Z. Thoughts?
2) on NSBackTabCharacter, send a NSTabCharacter (I did this, and I
think it works, because you'll only get an NSBackTabCharacter if
someone hits Shift-Tab, and so the Shift will have already been
These will both be minor changes to just RFBConnection.m, I think. Do
I checkin to send patches?
> So Jason just added me as a developer, and I haven't got a good feel
> for how you all distribute work. I'm assuming "if you got the time,
> have at it" is the general rule.
That is indeed the rule. There's way more stuff that needs doing than
there is time to do it in, so any contribution you want to make is
> There are two outstanding bugs I think might be related, or at least
> fixed in the same region of code: shift-tab not working, and
I agree, but there are actually quite a few that I think are related.
The ones you mention, and bugs numbered:
If it helps, the original release of Chicken had explicit support for
umlauts, which I guess I broke at some point.
> These will both be minor changes to just RFBConnection.m, I think. Do
> I checkin to send patches?
It's up to you - you can cvs commit your patches, send 'em to the list,
or submit 'em as patches on the sourceforge site. I'll get emailed on
any of the three actions and look 'em over, so from my point of view,
they're all equivalent.