john13@garfield.UUCP (John Russell) (02/25/88)
In article <1026@radio.toronto.edu> cks@radio.UUCP (Chris Siebenmann) writes: > > I would argue against this because it does something that's already >easy to do (check for RAWKEY code == 0x5f) and because there are only >31 Intuition message classes possible, and I'd like to preserve as >many as possible for future message classes. Is it actually necessary to impose a limit of distinct bit positions? For the window IDCMPFlags I can see, but isn't all Intuimessage processing done with case CLASS1: case CLASS2: etc which would work just as well with 1,2,3,4... as with 1,2,4,8...? If so (although it's entirely possible I'm overlooking something obvious) then there might be future I-Flags such as LISTEN_FOR_WEIRD_KEYS which would only take up 1 bit in the flags structure but which would cause your window to receive HELPKEY, F1KEY, CURSORUPKEY, etc. Would this be possible (desirable, compatible, insert your own adjective)? John -- "Fanaticism is all right... as long as you're ALONE! HAHAHAHA!" -- Pat Robertson shares a gem of wisdom told to him by Richard Nixon, and thus becomes the first politician to whom I can honestly apply the term "scares the willies out of me"
kenchiu@phoenix.Princeton.EDU (Kenneth Chiu) (02/27/88)
In article <4531@garfield.UUCP> john13@garfield.UUCP (John Russell) writes: >Is it actually necessary to impose a limit of distinct bit positions? For the >window IDCMPFlags I can see, but isn't all Intuimessage processing done with > >case CLASS1: >case CLASS2: This is something I've wondered about. Sometimes I use (flag & CLASS1), but I've never noticed more than one class being set at a time. So could we actually have 2^32 different classes? Or is there some funniness that I've never run across. Ken Chiu
cks@radio.toronto.edu (Chris Siebenmann) (02/29/88)
In article <4531@garfield.UUCP> john13@garfield.UUCP (John Russell) writes: ... > Is it actually necessary to impose a limit of distinct bit positions? For the > window IDCMPFlags I can see, but isn't all Intuimessage processing done with > > case CLASS1: > case CLASS2: > > etc which would work just as well with 1,2,3,4... as with 1,2,4,8...? The problem is things like ModifyIDCMP() and NewWindow structures, to which you supply OR-ed together flags. In addition, the application I'm working on now uses message masks to decide what routine to dispatch a given message to, and Intuition presumably uses the mask internally to decide whether to send a window a given event. You might be able to do it for new message classes, but it would probably take too much effort to be really useful. -- You're a prisoner of the dark sky/The propeller blades are still And the evil eye of the hurricane's/Coming in for the kill Chris Siebenmann {allegra,mnetor,decvax,pyramid}!utgpu!radio!cks cks@radio.toronto.edu or ...!utgpu!{chp!hak!ziebmef,ontmoh}!cks