On Thu, 2008-06-05 at 05:22 -0400, Jakub Jelinek wrote:
On Thu, Jun 05, 2008 at 11:10:59AM +0200, Michael Schwendt wrote:
> The difference in the spelling of the codeset breaks Sylpheed (#450063)
> which only looks for codeset "UTF-8". Now where again is it defined
> that both are valid spellings for the codeset?
Then Sylpheed is just broken. It shouldn't look at the locale name,
but instead at nl_langinfo (CODESET) (the same can be queried from
locale -k LC_CTYPE | grep ^charmap=
).
And, no matter what spelling you use in the .* part of locale name
in environment, CODESET is the same:
LC_ALL=en_US.UTF-8 locale -k LC_CTYPE | grep charmap
charmap="UTF-8"
LC_ALL=en_US.utf8 locale -k LC_CTYPE | grep charmap
charmap="UTF-8"
LC_ALL=en_US.utf-8 locale -k LC_CTYPE | grep charmap
charmap="UTF-8"
LC_ALL=en_US.UTF_8 locale -k LC_CTYPE | grep charmap
charmap="UTF-8"
Jakub
Sounds sane, though I faced very similar problem with Evolution some
time back, when my e-mail written in Czech (with diacritic marks) was
delivered broken - because only Evolution was able to tell what utf8
charset actually is... Of course, manually setting it to UTF-8 fixed the
issue... I wondered whether the problem was due to wrong setting after
upgrading to Fedora 9 or something else, but I didn't pursued the issue
further.
Martin