[Biblical-languages] Unicode proposals for apparatus symbols, Biblical Hebrew

Peter Kirk peter.r.kirk at ntlworld.com
Thu Jun 26 14:50:30 EDT 2003


Peter_Constable at sil.org wrote:

>Peter Kirk wrote on 06/26/2003 12:58:13 PM:
>
>  
>
> <snip>
>
>>>But it's unlikely that rendering systems and fonts will make any order 
>>>display correctly.
>>>
>>>      
>>>
>>I thought Unicode specified that canonically equivalent text should be 
>>displayed the same. Is this correct?
>>    
>>
>
>I can't recall for certain now what it does or doesn't say, though 
>obviously the intent is that different canonically equivalent sequences 
>ought to be display the same. But, for font developers, the number of 
>permutations to deal with when you've got four or five marks in different 
>combining classes (120 permutations for five) is a lot, and not trivial to 
>implement. I had told Joan that all possible orders should be supported in 
>Ezra, but she and Ralph balked at that, as have other implementers.
>
Just for clarification here: The final draft of The Unicode Standard 
4.0, section  3.2 point C9 (p.59), states: "Ideally,  an implementation 
would always interpret two canonical-equivalent character sequences 
identically. There are practical circumstances under which 
implementations may reasonably distinguish them" (from 
http://www.unicode.org/book/preview/ch03.pdf). So Joan and Ralph are 
probably within their rights, just. However, from point C10, any other 
process may replace character sequences by their canonical equivalents 
and this does not count as modification of the text. So some other 
process may normalise or un-normalise the text without even saying so. 
This makes it very risky to rely on a particular ordering as input to a 
rendering engine, especially if that is a non-normalised order.

<snip>

-- 
Peter Kirk
peter.r.kirk at ntlworld.com
http://web.onetel.net.uk/~peterkirk/




More information about the Biblical-languages mailing list