[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [STUMP] [PATCH] Clean up run-or-raise, add run-or-pull.

From: Shawn Betts
Subject: Re: [STUMP] [PATCH] Clean up run-or-raise, add run-or-pull.
Date: Fri, 24 Apr 2009 09:42:32 -0700
User-agent: Gnus/5.11 (Gnus v5.11) Emacs/22.2 (gnu/linux)

John Li <address@hidden> writes:

> Note1: run-or-pull just does move-window-to-group and pull-window. Are
> these too low-level, and do they always work?

When in doubt, check if the function is exported. If it is it's
considered public so yer totally safe. However, it's doubtful that the
function is exported because lots of public functions are not. So when
in double-doubt check where the function is used. If it's used in
command (or is a command), it's prolly safe to use.

This is an ongoing and interesting problem with stumpwm that I've
never really had in other programs. There seems to be a constant
ambiguity about what are "internal" functions and what is part of an
unpublished API, of sorts.

I think the ultimate way out is to just giv'r and use whatever
function appears to make sense. With an adequately plump test suite,
changes to such supporting functions would cause some tests to
fail. So all that we need to do is write a bunch of tests. I actually
started doing this with a test framework I didn't like, so it didn't
go very far. But I've recently been farting around with 5am, which
seems much less cumbersome, so perhaps I'll port the handful of tests
over to that. With that in place, hopefully ppl will just start
magically writing awesome tests because fuck is it ever a pain in the

> Note2: pull-window only makes sense for tiled groups. For float
> groups, it'd make sense to at least move the window to the group, but
> this doesn't do that, bleh.

There are probably a number of functions that could be extended to
work with other kinds of groups. For now it seems reasonable that
pull-window only works on tiling groups.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]