Summary: | Audacity WORKS with wxGTK compiled against GTK2 | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Axel Gneiting <axelgneiting> |
Component: | Current packages | Assignee: | Jeremy Huddleston (RETIRED) <eradicator> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | blkdeath, dju, k, radek, sound |
Priority: | High | ||
Version: | unspecified | ||
Hardware: | x86 | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- |
Description
Axel Gneiting
2003-11-04 16:20:52 UTC
that depends if wxGTK is compiled with or without unicode support, with unicode emabled audacity WILL break Hm, why don't you check for unicode and gtk2 then? because due to historical reasons there a a lot of ppl out there who dont have unicode inUSE but still have wxGTK compiled with unicode enabled Due we have any universal means of checking with what USE flags a installed package is compiled? As in usually check the contents of /var/db/pkg/<package>/<app>/USE In that case we could just replace if [ `use gtk2` ]; with something similar to if [ `depend-use wxGTK gtk2` ]; I wonder what happens if I wanted to install audacity when I have compiled wxGTK with USE="gtk2 unicode". I bet I can't without manual interference due to the odd if [ `use gtk2` ] workaround. So is this problem with gtk2, unicode, or unicode+gtk2? If it's centred around unicode, could we check instead for unicode support? IME, users are migrating from gtk->gtk2, so to me it makes little sense to make gtk2 inhibit the use of a (perfectly good) ebuild. unicode is a "sub option" of gtk2. so that means : if you have wxGTK + gtk2 you can choose the unicode flavour, or non-unicode flavour. if you have wxGTK + gtk1, then unicode has no effect. so it the audacity ebuild should check unicode rather than gtk2. basically check if 'wx-config --libs' has 'gtk2u'. *** Bug 40300 has been marked as a duplicate of this bug. *** new 1.1.3 and 1.2.0_pre3 ebuilds are now in portage. They correctly check for wxGTK compiling without UNICODE rather than relying on the use flag. Check out bug # 62002 ... test, if you can. |