[notting@apone: ~]$ alacarte Traceback (most recent call last): File "/usr/bin/alacarte", line 36, in ? main() File "/usr/bin/alacarte", line 32, in main app = MainWindow(datadir, version, sys.argv) File "/usr/lib/python2.4/site-packages/Alacarte/MainWindow.py", line 45, in __init__ self.editor = MenuEditor() File "/usr/lib/python2.4/site-packages/Alacarte/MenuEditor.py", line 36, in __init__ self.__loadMenus() File "/usr/lib/python2.4/site-packages/Alacarte/MenuEditor.py", line 42, in __loadMenus self.applications.path = os.path.join(util.getUserMenuPath(), self.applications.tree.get_menu_file()) File "/usr/lib/python2.4/site-packages/Alacarte/util.py", line 132, in getUserMenuPath os.makedirs(menu_dir) File "/usr/lib64/python2.4/os.py", line 159, in makedirs mkdir(name, mode) OSError: [Errno 20] Not a directory: '/home/devel/notting/.config/menus' [notting@apone: ~]$ rpm -q alacarte alacarte-0.10.0-1.fc6
does not happen here. What is ~/.config/menus, if not a directory ?
[notting@devserv: ~]$ cd ~/.config -bash: cd: /home/devel/notting/.config: Not a directory The parent dir isn't there...
if I rm -rf ~/.config and run alacarte, it works fine...
Whoops, that's what I get for commenting at midnight. .config is a regular file.
Hardly a blocker bug, unless some other tool we ship creates a $HOME/.config file. Of course, alacarte could handle this nicer...
What does it use this for - what is it storing? Is this part of a xdg-ish standard, or alacarte-specific?
part of the basedir spec, which talks about things like XDG_DATA_DIRS, XDG_DATA_HOME, etc
Do you have any idea how the file got created? Was it some kernel .config file laying around? Maybe we should pop up a dialog and say something like "In order to successfully edit menus, there needs to be a folder in your home directory called '%s'. Unfortunately, there is a file with that name in the way. Would you like to rename the file to '%s'? [Quit] [Rename Conflicting File] well something roughly like that anyway.
Yeah, it was a random kernel .config file. I dunno, I suppose this is probably an upstream bug at this point.
ah looks like it already got fixed upstream too.