[OpenIndiana-discuss] Locale questions in OI-151a7

[James Carlson]

On 03/13/13 10:40, Hans J. Albertsson wrote:
> Is this result really reasonable???
> I would have expected Å Ä Ö å ä ö to be classified as alpha in this
> locale....
> 
> 
> 
> 
> 
> hans at klaus%81%15:37> locale
> LANG=sv_SE.ISO8859-15
> LC_CTYPE="sv_SE.ISO8859-15"
> LC_NUMERIC="sv_SE.ISO8859-15"
> LC_TIME="sv_SE.ISO8859-15"
> LC_COLLATE="sv_SE.ISO8859-15"
> LC_MONETARY="sv_SE.ISO8859-15"
> LC_MESSAGES="sv_SE.ISO8859-15"
> LC_ALL=sv_SE.ISO8859-15
> hans at klaus%82%15:37> locale -k alpha
> alpha="A";"B";"C";"D";"E";"F";"G";"H";"I";"J";"K";"L";"M";"N";"O";"P";"Q";"R";\
> 
> "S";"T";"U";"V";"W";"X";"Y";"Z";"a";"b";"c";"d";"e";"f";"g";"h";"i";"j";"k";"l";"m";\
> 
> "n";"o";"p";"q";"r";"s";"t";"u";"v";"w";"x";"y";"z"
> hans at klaus%83%15:37>

Looking at the source code for locale.c, I see this:

#define CSSIZE  128
[...]
prt_ctp(char *name)
[...]
        for (i = 0; i < CSSIZE; i++) {

... so, it just prints the mappings for the first 128 bytes, regardless
of the type of character set.  This doesn't mean that the actual
isalpha() doesn't work and doesn't include the national language
character set code points you're interested in -- it almost certainly
does -- it just means that the 'locale' program is too weak to show them
to you.

-- 
James Carlson         42.703N 71.076W         <carlsonj at workingcode.com>