>>>>> "Mario" == Mario Rodriguez <biomates at telefonica.net> writes:
Mario> El s?b, 01-06-2013 a las 07:40 -0700, Raymond Toy escribi?:
>>
>> $ LANG=en at piglatin maxima-local
>> Maxima branch_5_30_base_101_g67286ae_dirty http://maxima.sourceforge.net
>> usingway Isplay CMU Common Lispway snapshot-2013-05 (20D Unicodeway)way
>> Istributedday underway ethay GNU Ublicpay Icenselay. Eesay ethay ilefay COPYING.
>> Edicatedday otay ethay emorymay ofway Illiamway Elterschay.
>> Ethay unctionfay ugbay_eportray() ovidespray ugbay eportingray informationway.
>> (%i1) ? mnewton
>>
Mario> So, this is piglatin. Nothing to do with the latin I learnt at school.
Nothing like the latin I learned in school either.
http://en.wikipedia.org/wiki/Pig_Latin
But the rules are simple so a machine can easily do the translation.
>>
>>
>> Erethay areway alsoway omesay inexactway atchesmay orfay `mnewtonway'.
>> Ytray `?? mnewtonway' otay eesay emthay.
>>
>> Evaluationway ooktay 0.1900 econdssay (0.1900 elapsedway) usingway 8.539 MB.
>> (%o1) true
>>
>>
>> Still needs work because I just hacked things together.
>>
>> I think this also shows a bug in the (default) info system. I think
>> that if we can't find the info files for a language, we should just
>> default and return the english version. It's better than nothing. :-)
Mario> I haven't compiled Maxima with option --enable-lang-es-utf8 for a long
Mario> time. This is what I get with current git sources and command
Mario> ./configure --with-clisp --enable-lang-es-utf8
[snip[
Mario> It can't detect that coding is utf8. Finally, this is a partial output
Mario> of command locale:
Mario> $ locale
Mario> LANG=es_ES.UTF-8
Mario> LANGUAGE=es_ES:en
Mario> Perhaps intl::*locale* should be equal to LANG, and not to LANGUAGE ?
I've noticed this too. The code for this is in setlocale intl.lisp.
It searches environment variables in the order LANGUAGE, LC_ALL,
LC_MESSAGES, and then LANG. I do not know what the correct order
should be or if a more complicated algorithm should be used.
Ray