[XeTeX] On combining diacritics again

John Was john.was at ntlworld.com
Mon Jun 11 23:18:45 CEST 2007


I hasten to add that my suggestion is no more than a quick workaround if you 
need to get output and are stuck!  VOLT (which I've never looked at but 
ought to) seems to me to be the correct way to go for a proper solution. 
That said, it's surprising how often a quick superimposition of rarely 
combining characters can save the day - and at least the ugly \cs names 
{\amacronacute} or whatever one invents are very easily 
search-and-replaceable once one has found an elegant solution.  Of course, I 
come from an age when the only way to get on Old English thorn was to type 
p, backspace, and then type b....

Best


John

----- Original Message ----- 
From: "Florian Grammel" <grammel at gmx.net>
To: "Unicode-based TeX for Mac OS X and other platforms" <xetex at tug.org>
Sent: Monday, June 11, 2007 10:04 PM
Subject: JunkEmail: Re: [XeTeX] On combining diacritics again


>
> Am 11.06.2007 um 16:52 schrieb Nikola Lecic:
>
>> You want to say that you need language/character/diacritics set that
>> is not covered by any advanced font (or perhaps unknown to Unicode)?
>
> Actually, yes.
>
> On the one hand Unicode is still under development and thus many
> historical scripts and additions to scripts are still missing. On the
> other hand Unicode will never cover *everything* one might want to
> print (in our case books on and transcriptions of medieval
> manuscripts) because of the UTC's strict definition of a "character".
> Thus we have Unicode-compatible fonts and use (some would say "hack")
> the PUA to find space for characters that aren't covered.
>
> cf. http://www.mufi.info/
>
>
>
>
>
> Am 11.06.2007 um 22:02 schrieb Jonathan Kew:
>
>> I don't believe this is possible at the TeX macro level, except
>> through incredibly complex/ugly schemes that would involve making the
>> *base characters* active and then programming them to look ahead for
>> diacritics. You really, really don't want to even consider such a
>> thing.
>>
>> You can't do it just by programming the diacritics, because you can't
>> "look back" at the preceding character.
>
> I was afraid that the answer would be something in this direction,
> but it's good to have it confirmed.
>
>>
>>> If none of the experts can be tempted to waste some of their spare
>>> time to solve this probably not too trivial problem -- I still will
>>> be looking for a handy way to adjust the placement without fiddling
>>> with the font itself. I would prefer not to alter the text itself,
>>> but define the changed placement (vertically or horizontally
>>> respectively) for a given pair of glyphs once and for all in the
>>> preamble.
>>>
>>>
>>> I would be very grateful for any help to get my diacritics to behave!
>>
>> As I understand it, you want your text to contain simple Unicode
>> sequences of <base, diacritic(s)>, and have the diacritic(s)
>> positioned properly. The appropriate way to solve this is by building
>> the behavior into the font. For Latin-script text and all common (and
>> many uncommon) diacritics, fonts like Doulos SIL, Charis SIL, or
>> Microsoft's new Core fonts in Vista provide this support.
>>
>> However, you mentioned needing to use PUA characters from a special
>> font. Is this a font that you own and can modify, or can you persuade
>> the font developer to work on it?
>
> Not really -- that's the reason I ask:
> We just have user-licenses for the font, so we cannot modify it
> ourselves. The developer is a professional (who has to make a living)
> and we have used our funds to pay him to add really many glyphs.
> These can form loads of rare but possible combinations, so doing this
> once and for all would be a lot of work anyway. Thus I'd prefer an
> imperfect but working system as Cocoa's...
>
>> If so, the solution is to use VOLT
>> (Microsoft's
>
> ... isn't there a Mac version? ;)
>
>> OpenType font development tool, available free) to add
>> attachment points and "mark" and "mkmk" features to the font, so that
>> the ICU OpenType engine will automatically position the diacritics
>> properly. It's not very hard to do this -- easier (I would say) than
>> advanced TeX macro programming! -- and it will work transparently
>> without the risk of conflicting with other macros or markup in your
>> document.
>
> Good to know for another time, when I won't need my PUA...
> So I'll look closer into John Was' and John Smith's suggestions now.
>
> Thanks a lot for your help everybody!
>
> Best regards,
> Florian.
>
> _______________________________________________
> XeTeX mailing list
> postmaster at tug.org
> http://tug.org/mailman/listinfo/xetex
> 


More information about the XeTeX mailing list