https://bugs.kde.org/show_bug.cgi?id=384321
Bug ID: 384321 Summary: Mess with encoding in kpmcore Product: partitionmanager Version: unspecified Platform: Other OS: Linux Status: UNCONFIRMED Severity: normal Priority: NOR Component: general Assignee: andr...@stikonas.eu Reporter: pali.ro...@gmail.com Target Milestone: --- Created attachment 107670 --> https://bugs.kde.org/attachment.cgi?id=107670&action=edit $ git grep -E 'Latin1|Utf8' Look at attched output from command: git grep -E 'Latin1|Utf8' There is a big mess with encodings, on some places is used Latin1 on other places UTF-8 without reason. For converting filename stored in char* to QString should be used function QFile::decodeName(). And for reverse operation -- conversion of filename from QString to char* should be used function QFile::encodeName() http://doc.qt.io/qt-5/qfile.html#decodeName http://doc.qt.io/qt-5/qfile.html#encodeName For converting other locale/system dependent settings (like env variables, command line arguments or stdin/stdout), QString methods toLocal8Bit() and fromLocal8Bit() should be used. http://doc.qt.io/qt-5/qstring.html#toLocal8Bit http://doc.qt.io/qt-5/qstring.html#fromLocal8Bit Methods toUtf8() resp. toLatin1() should be used only in case callee C function has defined API that char* must be in UTF-8 resp. in Latin1. For ASCII strings, both UTF-8 and Latin1 encoding doing exaclty same things. But problem happen for non-ASCII strings, when Unicode internationalization is widely used. Properly written Unicode applications should not have this problem. Please recheck where UTF-8 resp. Latin1 functions are correctly used and where decodeName() and fromLocal8Bit() should be instead. -- You are receiving this mail because: You are watching all bug changes.