First, slang doesn't handle switching from non-unicode to unicode in the same process. Rather than find the magic internal variables to frob, just set to a unicode locale before we ever init slang. Second, there is some bizarre interaction going on so that when KMS initializes, some of the console font data is no longer correct. I could start digging into the kernel code to figure out WTF is going on here... or I could just reset the font whenever we change language. Third, we need the locale archive on stage1 so that wcwidth and friends actually calculate character width correctly. However, we don't want the entire glibc locale archive, so just create one with localedef that only consists of en_US.UTF-8. Adds 1.3MB to the installer initrd.img. Tested by building a new anaconda package and spinning a tree with pungi. Bill _______________________________________________ Anaconda-devel-list mailing list Anaconda-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/anaconda-devel-list
