On Tue, Dec 8, 2009 at 9:46 AM, Armando Di Cianno <[email protected]> wrote: > irb(main):068:0> system(cmd) > Waiting for Emacs... > => true > ... i.e., it works. > > irb(main):069:0> launch(cmd) > sh: emacsclient --alternate-editor emacs foo: command not found > => true > ... i.e. it failed to launch.
Thanks, this is a bug indeed. It is solved in this commit: http://github.com/sunaku/wmiirc/commit/c824cc5888153fc1f51e2aa79b42ed6c60e706d1 > ... so, I guess I'm asking what the intention of the "command = > words.shelljoin" in the launch method was supposed to really do? > Would removing the line, which does work, be a good idea overall, or > am I missing the real intention you had in mind? Basically, I wanted to avoid shell quoting issues (which arise when manually constructing a shell command) by making the method-call arguments into the command's ARGV. See the examples section I added in the launch() method documentation (at the URL shown above). Cheers.
