RE: code pages I think
On 17 May 2001, at 13:17, Stuart Cox wrote:
> It's relevant because these guys are Euphoria developers and they're trying
to
> bridge the communications gap that is forced upon different language
> groups....
It's also helped when someone from Romania came into an irc channel i
haunt, and said something which would be abusive in english, but wasn't in
Romanian. The words he chose, or the translator he used, or something,
produced a word, altho it was incorrectly spelled for an expected english
word, it was correctly spelled for a Romanian word. Since i knew he was
from Romania, i was the only one who knew what he meant, but i don't
actually speak that language. Since i am working on some language
processing code, it helps to have a global perspective. With usa president
Bush promoting more nuclear power plants, for instance, i am looking to
Russian sites for old news (1957-58) on Sverdlovsk, Kyshtym, and Kamensk-
Uralskiy,, altho i still can't write them in Cyrillic.
> Code pages are arcane.
It also matters when a program written by a english-speaking programmer's
code that does heavy string manipulation encounters non-english bytes.
There is lots of code in the Eu archives that is written in a way that
inherently
excludes the use of Cyrillic, for instance, by typecasting out the upper 127-
255 chars, or ignoring them. I have no idea yet how to apply upper() and
lower() to non-english characters. This has affected my coding, and it's given
me more thoughts on items that must be considered in the future. Such as
internally translating the 8bits+codepage into unicode,, or whatever comes
around that's better,, or that i devise for my own use. 16-bit unicode (already
called obsolete) will of course break all my existing code, but maybe not by
much.
Kat
|
Not Categorized, Please Help
|
|
