On Thu, 2007-03-01 at 21:31 +0000, Peter wrote: > This logic is wrong. I've posted a patch previously. > The assignment of INTTLOOLIZE and LIBTOOLIZE are reversed. > A TRUE condition will assign $bindir whereas it should not. > It should be the other way around.
GAR. Even worse, that code was totally broken. Yes, you are correct, the logic was reversed -- however, it never could have worked with anything but a GARNOME provided tool. > @if which intltoolize > /dev/null; then \ > INTLTOOLIZE=$(bindir)/intltoolize; \ > else \ > INTLTOOLIZE=$(shell which intltoolize > /dev/null 2>&1); \ Redirecting both STDOUT and STDERR to the big bitbucket results in assigning an empty string always for the 'else' case. If the fixup tool actually has been called, it definitely was the one in our $prefix. I had a look at your previously posted patch, and that indeed works. You removed that bad redirection, too. :) However, IMHO we should default to the GARNOME built tool, if it exists, and fall back to the system one otherwise. If neither one exists, just freak out and stop the build. :) That's exactly what the new code does. Also, that old hack has been unnecessarily complicated and confusing. To fix this, I hardly did anything but remove a huge mess. See the diff. [1] All we need to do is just call the tool we want. The $PATH already takes care about it for us. In gar.conf.mk the $PATH is set to explicitly use the GARNOME $prefix first, then whatever the $PATH has been before. Thus we default to an executable in our $prefix if it exists, and try the system ones otherwise. Peter, thanks again for pointing this out and repeatedly poking. :) guenther [1] http://svn.gnome.org/viewcvs/garnome/trunk/gar.lib.mk?r1=4586&r2=4777 -- char *t="[EMAIL PROTECTED]"; main(){ char h,m=h=*t++,*x=t+2*h,c,i,l=*x,s=0; for (i=0;i<l;i++){ i%8? c<<=1: (c=*++x); c&128 && (s+=h); if (!(h>>=1)||!t[s+h]){ putchar(t[s]);h=m;s=0; }}} -- garnome-list mailing list [email protected] http://mail.gnome.org/mailman/listinfo/garnome-list
