https://bugzilla.novell.com/show_bug.cgi?id=761974 https://bugzilla.novell.com/show_bug.cgi?id=761974#c0 Summary: YaST doesn't display CJK characters in its package summary/descriptions right. Classification: openSUSE Product: openSUSE 12.2 Version: Factory Platform: All OS/Version: openSUSE 12.1 Status: NEW Severity: Normal Priority: P5 - None Component: YaST2 AssignedTo: bnc-team-screening@forge.provo.novell.com ReportedBy: i@marguerite.su QAContact: jsrain@suse.com Found By: --- Blocker: --- Created an attachment (id=490595) --> (http://bugzilla.novell.com/attachment.cgi?id=490595) random strings in YaST2 User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/536.6 (KHTML, like Gecko) Chrome/20.0.1090.0 Safari/536.6 SUSE/20.0.1090.0 Hi, YaST2 team At first, I know it's not right to input CJK characters into Summary/Description fields in a RPM spec file. I should use "Summary(zh_CN)" and "%description -l zh_CN", although YaST2 don't know how to deal with them. But what if a CJK project has no proper english project name? eg, this package: fcitx-chewing in Factory. it has two Chinese characters in its default description.( not %description -l zh_CN) I did so because actually no Chinese is familiar with its english name "chewing". While It's Chinese name is known to everybody. because in fairy tale, it's the name of the creator of Chinese characters, "仓颉". and considering no other users except Chinese will use this package(it's an locale-specific input method), I have to make it precise. so I included its Chinese name too. then the problem happens: "those two characters were displayed as random strings no human can read." as I said, it's an out-range problem. but as we know, spec file and changelog have to be in UTF-8 format. If not, OBS will prompt a warning message in rpmlint.log. That's a common situation not only to Chinese but also to Russian. because some of our developers have no English names, if they write their names in Chinese or Russian, they have to save the related files in ISO-9*** or GBK, then OBS warns. But if they save that files in the "right" format, UTF-8, their names are displayed as machine codes. It's sad, because we can't force one to pick an English name or change the way he writes his name, although it's the simplest solution. as I know, UTF-8 should display everything right. because its universal. so I hope you can dig into this, and find a fix or workaround. then: 1. it'll help a lot to OBS packaging and to our end user experience. because in newbies' eyes, any such random strings will be supposed as "OS not good". that's why we "foreign" communities package packages in pure English. but we do expect one day we can sign our names in local language (because actually no end user cares your name that much, they just need you email) or include local explanations into locale specific packages. 2. it'll help a lot to translation teams. Package translations will only applies on certain locales. But some users prefer English locale to learn English or something, but project names are the ones no need to learn. and those local characters in plain description or summary will mess our translation system. if I'm the packager myself, I know what the random string means, but if I'm not or even a translator from other locales, I don't know what they're. then what can I do? cut them off? or copy the same meaningless random string to my version? but like I said, it's optional the thing I want to draw you attention is: YaST2 or our whole system (because I update translation pot today, and that won't need YaST2) can't recognize any "other characters" in spec file/changelog which is in UTF-8 format(in which case it should recognize) a small "bug" may indicate a big one. so I opened this bug report. Marguerite Reproducible: Always -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.