Keyboard question/change request

Troy tjk at tksoft.com
Tue Dec 6 04:12:57 PST 2005


Simon,

Thank you very much for your reply, and all the information.

I will investigate the sites you mentioned, and see if I
can generate the layouts we are planning. 

Five or six dead-keys seems enough in my mind as well. We have
discussed the issue but haven't decided on a specific limit 
ourselves yet. I would imagine it wouldn't be more than five.

One specific issue which I believe is still not doable is
two or more diacriticis which are the same. We have one
special case like this. Or maybe it is possible; I will
find out when I investigate the sites.

Again, thank you very much.



Troy




> 
> *Troy* wrote:
> 
> > Hi,
> >
> > I am a member of a workgroup which is designing a new keyboard
> > layout for Finland (and therefore also for Sweden). The purpose of the 
> > new keyboard is to enable the input
> > of a large range of latin based characters. Microsoft
> > has been involved in the work, so their support is built into the 
> > process.
> >
> > The keyboard is based largely on using dead-keys. The goal is generate 
> > an open ended group of Unicode characters, using Unicode's 
> > combinations to generate characters. The goal is to allow users to 
> > type any character, even ones
> > without a precomposed equivalent. We have used MES-2 as
> > a checklist for the important characters, but any Unicode character 
> > which can be generated from a combination of the listed diacritics and 
> > letters should be allowed to be generated using the keyboard.
> >
> > Because of the limited number of keys available on the keyboard,
> > we have been forced to add some uses for dead-keys which are not 
> > currently possible on MS Windows or X Windows (unless the capability has
> > been added to X Windows w/o me noticing). Specifically, current
> > dead-keys always combine with the following character, or then
> > generate the diacritic itself. Our new design requires that two
> > or more dead-keys can be used to generate a character with both
> > of those diacritics. (Some other special combinations are also
> > specified.) There are also some additional dead-keys which are
> > currently not allowed.
> >
> > I would like to know who is the person, or the persons, who
> > are responsible for keyboard input handling for X Windows?
> 
> Hello,
> 
> There is a transition in X.org from the older monolithic tree to a 
> modular version.
> In the new version, the keyboard input handling is being simplified 
> through the xkeyboard-config project, at
> http://www.freedesktop.org/wiki/Software_2fXKeyboardConfig
> You may also find some insight at
> http://www.livejournal.com/community/xkbconfig/
> a community about keyboard input support.
> 
> It appears your request is mostly related to adding the appropriate 
> entries to
> the "Compose" configuration file,
> http://cvs.freedesktop.org/xorg/xc/nls/Compose/en_US.UTF-8?view=markup
> (here you define the key sequences with dead keys that produce the 
> codepoinds you want)
> and then by adjusting the symbols configuration file for Finnish
> http://cvs.freedesktop.org/xlibs/xkbdesc/symbols/fi?view=markup
> you enable your specific keyboard layout.
> 
> As you can see, in the existing Finnish keyboard layout file there is 
> one layout that has dead keys.
> 
> In XOrg there is functionality to stuck up to five (or is it six?) 
> diacritics/accents/etc on a glyph, which
> would make happy all keyboard layout designers. Rumour has it that a 
> Greek was involved in this,
> as in ancient greek (polytonic) there are too many accents.
> Specifically, there are nine distinct "accents" and if we had to 
> associate with single keys, we would put in disuse
> a big area of the keyboard.
> One way to do it is associate three "dead_keys" to each physical 
> character (one by pressing the key as is, one with SHIFT and finally 
> with AltGr).
> 
> You want to stuck two (or more) dead keys to produce a character with 
> two "accents". You do this in the Compose file (en_US.UTF-8).
> In Modern Greek, the character ι (iota) can have either Î0x84, š or both Î0x85. 
> Therefore,
> 
> Î0x91. For Î0x84 (΀Î0x9fÎ0x9dÎ0x9fΣ) only:
> <dead_acute> <Greek_iota>	: "ί" U03AF # GREEK SMALL LETTER IOTA WITH TONOS
> 
> Î0x92. For š (DIALYTIKA) only: 
> <dead_diaeresis> <Greek_iota>	: "Ï0x8a" U03CA # GREEK SMALL LETTER IOTA WITH DIALYTIKA
> 
> C. For Î0x85 (DIALYTIKA AND TONOS) only: 
> <dead_acute> <dead_diaeresis> <Greek_iota>	: "Î0x90" U0390 # GREEK SMALL LETTER IOTA WITH DIALYTIKA AND TONOS
> <dead_diaeresis> <dead_acute> <Greek_iota>	: "Î0x90" U0390 # GREEK SMALL LETTER IOTA WITH DIALYTIKA AND TONOS
> 
> As you can see in case C, it is good to cover the permutations.
> 
> For an example of three "accents", enjoy the following :)
> 
> <dead_iota> <dead_grave> <U1f01>	: "áŸ0x83" U1F83 # GREEK SMALL LETTER ALPHA WITH DASIA AND VARIA AND YPOGEGRAMMENI
> 
> I hope this helps,
> Simos
> 
> >
> > FYI, the address for our workgroup is http://www.kotoistus.fi/
> >
> >
> >
> > Troy Korjuslommi                Tksoft Inc.
> > tjk at tksoft.com <http://lists.freedesktop.org/mailman/listinfo/xorg>
> >
> >
> 
> 




More information about the xorg mailing list