I'm just installing a Slack current box and noticed that the old warning about using UTF8 for the shell is still in place during the initial setup:

1. As this warning has been in place since I first started using Slackware, is this still a current issue after all these years?

2. What exactly are the possible problems with using UTF8 in the shell? What is likely to come and bite me in the backside if I use it - in practical terms? I have been selecting UTF8 during the setup on a number of servers over the years and can't say I've noticed any problem - but I also confess to not having much of an understanding of the wider issue.

I don't think it is an issue. I've always used UTF-8. Other distros use UTF-8 too.

1. Not a real issue, but I believe that some legacy applications still shipped in Slackware are not able to properly process UTF8-encoded data.

2. Nothing that I am aware of will bite you in the backside as far as I know. Some applications will work better (or at least faster) if you set your locale to C but that generally can be done at time of invoking them, like "LANG=C <application <arguments>. In other cases processing text files through iconv can help.

Unfortunately I do not have a list of applications that need a special care, but maybe this thread could be an occasion to begin one, like: application / issue or drawback / suggested workaround.

I really hope Slackware will use UTF-8 by default in the future but I guess that Pat would need help to do that, as an inventory of all applications that should need some care is yet to be established as far as I know, and then what to do with all these man pages using a legacy encoding...

Some of the older stuff does have problems with utf-8 sequences. 'elvis' (slackware's default vi implementation) is one of the more obvious ones. The ash shell is another.

I use UTF8. No issues, just some "cosmetic" annoyances -- the menus of some old GTK1 applications have very big text. I fixed this by starting these programs with custom LANG settings:

https://slackalaxy.wordpress.com/2012/06/10/utf-8/

1 members found this post helpful.

Thanks all for the replies - just one of those little niggly questions that's been on my mind for a while.

And I think Didier Spaier's idea above is a good one - to list which applications can cause issues with UTF8. It could be useful for any future decision on the mater - I guess.

Maybe this is the cause of my issue...

http://www.linuxquestions.org/questi...ue-4175533451/

GhostScript and GhostView are two that will make messy with UTF-8. Only ones I know about, though, and, because I have applications that require UTF-8, I just don't use either GS or GV much any more.

Might not be ghoscscript or gv themselves. I found that some fonts are incomplete in "ghostscript-fonts-std" packages. Some Cyrillic letters are incomplete.