Eric Blake <ebb9 <at> byu.net> writes:

> I've got a pending patch series to add m4_set_* where I wanted to use
> m4_join, but not have the empty argument omitted.  I'm borrowing this from
> the m4 manual, and committing.

It turns out that m4_joinall is quadratic on m4 1.4.x (just as any other 
recursive algorithm on $@), so I didn't end up using it in my m4_set 
implementation after all.  But it is still a useful interface, and with the $@ 
algorithm fix m4 1.6, m4_joinall becomes linear, so I won't revert it.

-- 
Eric Blake





Reply via email to