[Bug 1095425] New: The ICU library ignores the users locale setting
![](https://seccdn.libravatar.org/avatar/3035b38ff33cf86f480bb169b8500b80.jpg?s=120&d=mm&r=g)
http://bugzilla.opensuse.org/show_bug.cgi?id=1095425 Bug ID: 1095425 Summary: The ICU library ignores the users locale setting Classification: openSUSE Product: openSUSE Tumbleweed Version: Current Hardware: Other OS: Other Status: NEW Severity: Normal Priority: P5 - None Component: Other Assignee: bnc-team-screening@forge.provo.novell.com Reporter: lopa@mailbox.org QA Contact: qa-bugs@suse.de Found By: --- Blocker: --- During the build of the ICU (International Components for Unicode) library, the preprocessor constant U_CHARSET_IS_UTF8 is set to 1 on Linux systems. This has the effect, that the users locale setting is ignored for all applications that are using this library, e.g. KDE and QT. It's not possible any more to work correctly with a different file system character set than UTF-8. Although UTF-8 is the recommended encoding, there are still reasons to use simple 8 bit encodings like ISO8859-x. And almost all parts of OpenSuse are still supporting such encodings, but this doesn't help because the ICU library returns hardcoded "UTF-8". During the package build, the preprocessor constant U_CHARSET_IS_UTF8 should be set to 0 explicitly. Then the users LANG or LC_CTYPE setting is used for character set conversions. Affected file in package icu-61.1-1.1: source/common/unicode/platform.h -- You are receiving this mail because: You are on the CC list for the bug.
participants (1)
-
bugzilla_noreply@novell.com