https://bugs.kde.org/show_bug.cgi?id=396435
--- Comment #20 from Mariusz Glebocki <m...@arccos-1.net> --- (In reply to Martin Hostettler from comment #19) > This seems to be quite close to what vte/gnome-terminal uses. But not quite > the same. Surrogates technically should be invalid; they are intended to be combined in pairs (high+low) when using UTF16 to form a valid UTF32 code point. Decoder does the conversion, so valid pairs are passed as another code point. I made regional indicator letters wide manually because of vim, and width in noto sans color emoji and emoji one. glibc 2.27 treat them as narrow, so I'm not sure which width should we choose. > vte combines hangul clusters so it's correct for them. It seems konsole does > so too so maybe it should also list them as combining like vte? I was thinking about this, also with emoji skin tones. Vim and glibc 2.27 treat them as separate and narrow, so I opted for compatibility rather than good look. I don't know Korean nor its input methods, but I read they are almost never used in normal text, as everything is already present in composed form as separate code points. I think it should not cause problems in real world use cases. -- You are receiving this mail because: You are watching all bug changes.
