Re: Fwd: Wired 4.09 p. 130: Lost in Translation

From: Martin J Duerst (mduerst@ifi.unizh.ch)
Date: Thu Aug 29 1996 - 09:09:47 EDT


Dan wrote:

>There is also an additional problem by having the combining character coming
>after the non-combinging, instead of before, as when reading interactively
>and a program want to act on a singel character it must use a timeout
>waiting for more bytes to decied when no more combining characters will
>arrive. If it had been defined that combining character came before the
>non-combining, parsing would be simpler - the non-combining ends a
>combining sequence.

In some cases, this is true. But in some cases, a program should act
on each character it sees, even if this may be followed by something
that combines with it. This also applies if combining characters come
before base characters. An example are dead keys. Many implementations
don't display the dead key when it is pressed, for convenience of
the implementation. However, for the user, it would be much more
convenient to see they accent after the dead key is pressed, and to
have it replaced by the combination after pressing the key of the
base character.

Regrads, Martin.



This archive was generated by hypermail 2.1.2 : Tue Jul 10 2001 - 17:20:31 EDT