tpf@jdyx.UUCP (Tom Friedel) (10/21/90)
Can someone point me to references for answers to the following questions. I see there is started code for handling Compose sequences in Xlib; What is the best way for a server writer to implement the functionality of compose keys. What is the best way to handle dead key's which I understand can be pressed and released (unlike modifier keys) before the associated key is pressed thanks, tom friedel Tom Friedel JDyx Enterprises (404) 320-7624 tpf@jdyx.UUCP -- Tom Friedel JDyx Enterprises (404) 320-7624 tpf@jdyx.UUCP Unix BBS: (404) 325-1719 <= 2400 ; (404) 321-5020 >= 2400 "Live simply, so that others may simply live."
mouse@LARRY.MCRCIM.MCGILL.EDU (10/22/90)
> I see there is started code for handling Compose sequences in Xlib; > What is the best way for a server writer to implement the > functionality of compose keys. (Is that a question or a statement?) The server writer shouldn't. Compose-character is done on the client side of the client/server split. > What is the best way to handle dead key's which I understand can be > pressed and released (unlike modifier keys) before the associated > key is pressed (Dead key's *what*? A possessive needs something to possess.) Again, the server has no business knowing that a given key is supposed to be a dead key. As I understand it there is currently no Xlib support for dead keys. (There are no keysyms for them either, which IMO is a serious omission that should be rectified at the next opportunity, whenever that is.) der Mouse old: mcgill-vision!mouse new: mouse@larry.mcrcim.mcgill.edu
brossard@sasun1.epfl.ch (Alain Brossard EPFL-SIC/SII) (10/22/90)
In article <9010220805.AA11311@Larry.McRCIM.McGill.EDU>, mouse@LARRY.MCRCIM.MCGILL.EDU writes: > > I see there is started code for handling Compose sequences in Xlib; > > What is the best way for a server writer to implement the > > functionality of compose keys. > > The server writer shouldn't. Compose-character is done on the client > side of the client/server split. Indeed, all the magic resides in Xlib. One of the advantage of using Open Windows 2.0 is that Sun has completed the Compose part of Xlib. > > > What is the best way to handle dead key's which I understand can be > > pressed and released (unlike modifier keys) before the associated > > key is pressed > > Again, the server has no business knowing that a given key is supposed > to be a dead key. As I understand it there is currently no Xlib > support for dead keys. (There are no keysyms for them either, which > IMO is a serious omission that should be rectified at the next > opportunity, whenever that is.) Sun does provide support for Compose keys, and in the MIT's distrib there is a keysym: Multi-Key. For example <Compose>'e would give you ``i'' for those who use an appropriate font with an 8 bit OS, assuming this character makes it through the news software... :-(. On the other hand der Mouse is right in that there is no support for "dead" keys per se meaning typing 'e and expect to see the appropriate letter with an ' above it. But there is support to compose a sequence of keys into a single character (eg: <Compose>co for the copyright symbol). Well there will be for everybody on R5??? > der Mouse -- Alain Brossard, Ecole Polytechnique Federale de Lausanne, SIC/SII, EL-Ecublens, CH-1015 Lausanne, Suisse brossard@sasun1.epfl.ch
mouse@LARRY.MCRCIM.MCGILL.EDU (10/24/90)
> Indeed, all the magic resides in Xlib. One of the advantage of using > Open Windows 2.0 is that Sun has completed the Compose part of Xlib. On the other hand, they have completed it in their way. If your way happens to disagree with theirs, tough. Now if you MIT X, on the other hand, you can do anything you like to it. UTSL, y'know. > [Will there be support for compose and dead keys] for everybody on > R5??? Well, at least one implementation exists. I know a person who implemented compose-character processing for MIT X11R4. He gave me his implementation to put up for anonymous ftp; it can be gotten from 132.206.1.1, in X/justin-compose. If you add a few keysyms for dead-acute, dead-grave, dead-circumflex, etc, it does dead keys as well (or so he tells me - I haven't tried it myself). It is not an optimal implementation, but it's sure a lot better than nothing. der Mouse old: mcgill-vision!mouse new: mouse@larry.mcrcim.mcgill.edu
elric@imryrr.Eng.Sun.COM (Rick Heli) (10/31/90)
In article <1990Oct22.124125@sasun1.epfl.ch> writes: >In article <9010220805.AA11311@Larry.McRCIM.McGill.EDU>, mouse@LARRY.MCRCIM.MCI LL.EDU writes: >> Again, the server has no business knowing that a given key is supposed >> to be a dead key. As I understand it there is currently no Xlib >> support for dead keys. (There are no keysyms for them either, which >> IMO is a serious omission that should be rectified at the next >> opportunity, whenever that is.) The OpenWindows version 2 Xlib defines the three dead keys which appear on Sun keyboards in X11/Sunkeysym.h. These have now been registered with the Consortium and will in the future appear in XKeysymDB. > Sun does provide support for Compose keys, and in the MIT's distrib >there is a keysym: Multi-Key. For example <Compose>'e would give >you ``i'' for those who use an appropriate font with an 8 bit OS, >assuming this character makes it through the news software... :-(. It didn't, at least not to me. To explain, I believe this meant <Compose> ' e generates the e-acute character (code 233). > On the other hand der Mouse is right in that there is no support >for "dead" keys per se meaning typing 'e and expect to see >the appropriate letter with an ' above it. But there is support >to compose a sequence of keys into >a single character (eg: <Compose>co for the copyright symbol). This is correct. However, on most Sun type 4 keyboards -- US and UK versions excepted -- there are dead keys (aka "floating accent keys") which do have this functionality supported in OpenWindows version 2. Rick Heli Window Systems Platforms -- Rick Heli Internet: rheli@sun.COM