It sounds as if at least part of the problem is the character map that
Rgui is using on your W98 machine.
BTW, I don't think they are `corrupted', just not treated as being in
the encoding you intended (which has nothing to do with locales per se
under Windows). Rgui does not itself handle encodings, so it accepts
whatever keycode it gets and prints/plots that code in the current
font.
You do know that MicroSoft no longer supports W98?
On Wed, 7 Jan 2004, Petr Pikal wrote:
Dear all
I am using two computers, one with Windows2000 and the other one
with W98 both have the same version (precompiled binary) R 1.8.1 and
I have experienced a slight problem with text used in plotting on
W98 machine.
When I try to write some local characters into R console
????????? (not sure if it is OK on your computers)
I strongly suspect they are not: I cannot display ISO-8859-2.
in W2000 **everything** is OK but in W98 some of the characters
(1,3,4,5) are changed (and the same is when I try to annotate a plot
with them). When I copy it from R console to some editor (e.g.
Notepad) the characters are again OK and when I copy the characters
from editor back to console they are again corrupted.
[1] "LC_COLLATE=Czech_Czech Republic.1250;LC_CTYPE=Czech_Czech
Republic.1250;LC_MONETARY=C;LC_NUMERIC=C;LC_TIME=Czech_Czec h
Republic.1250"
is same on both and I have checked Rprofile and my profile on both
computers and it seems to be the same.
I can put Czech characters on finished plots with some other program
(GIMP) but I prefer to do all possible annotation in R.
Please, can you give me some hint where to look for possible
solution on W98 machine setting.
Thank you
Petr Pikal
petr.pikal at precheza.cz