Re: unsigned characters

[Olli Saarela]

John W. Eaton wrote:
> On 30-May-2007, Olli Saarela wrote:
>
> | Proper handling of national
> | characters in Octave might be a larger project, though. The data type 
> | used is signed 8-bit which cannot hold the necessary codes:
> | octave.exe:23> double(char(246))
> | ans = -10
> | This should return 246 (the code for letter ö).
>
> Yes, the problem is that Octave uses std::string and the charMatrix
> and charNDArray objects use char instead of unsigned char.  Fixing
> this will touch many files.  I suppose we need to do it eventually,
> but I don't see it happening before 3.0 unless people want to delay
> the release for the time it will take to resolve all the little
> problems that a large change like this is likely to create.
>
> jwe

I see this as a low-priority item, even though I'm a daily user of 
umlaut characters. Please don't delay 3.0 for it.

As a workaround,
   double(str)+128*(str<0)
has been needed only in a couple of places in my code. Pictures aimed 
for publication I generate with script files anyway (to make the results 
repeatable), and
   plot(rand(5,1));
   title('Characters with diacritics: åäöÅÄÖ');
works when executed from a script.

Olli