Summary: | macos: emerge emacs-cvs/emacs-cvs-21.3.50-r1 fails from SSH key problem... | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Darin Morrison <dibrom> |
Component: | Current packages | Assignee: | osx porters <osx> |
Status: | VERIFIED FIXED | ||
Severity: | normal | CC: | emacs |
Priority: | High | ||
Version: | unspecified | ||
Hardware: | PPC | ||
OS: | All | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Bug Depends on: | 57677, 64319 | ||
Bug Blocks: |
Description
Darin Morrison
2004-09-16 23:03:48 UTC
After messing with the .ebuild some more, it appears the problem is due to a line break or something that occurs between the v and the 0= at the very end of the key. If I remove the whitespaces and bring it back to the same line, then the cvs checkout works. Does the cvs ebuild work after you changed it so? Last time I checked it all the CVS ebuild were broken due to the CVS version included in Mac OS X 10.3.5. cvs.eclass assumes >=cvs-1.11 and Mac OS X provides cvs-1.10 (so you cannot use CVS ebuilds without overwriting system /usr/bin/cvs) Yes, it works, as long as I apply the carbon patch at bug 64362. I tried once with just X/Gtk and it didn't work. I didn't try to make that work though since I wasn't particularly interested. Just for additional clarification, the original OS X 10.3.5 cvs ( Concurrent Versions System (CVS) 1.10 `Halibut' (client/server) ) is what I was using, and it seemed to checkout fine aside from the issue with the key. Right, it works. Lucid and Xaw3d interface run fine. I'm waiting for the dependency to be keyworded (the biggest one is gtk+2, which requires pango to be compiled with xft support) and the location of Emacs.app to be discussed in the next osx meeting. k, I managed to get emacs-cvs working on gtk+2. Once bug #57677 is fixed we can keyword gtk+2 and get emacs-cvs done. Added to ~ppc-macos, as we disabled gtk/gtk2 USE flags by use.mask for the moment. Closing out bugs that've been resolved for a while now... |