I am working on building a standalone version of libgnt. Currently the code for libgnt resides in the Pidgin repository. So as it stands we have two autogen.sh files. One in the toplevel of the source that has been there for a long time and recently we've added one in the libgnt folder to allow libgnt to be build standalone. We are essentially trying to make the libgnt folder extractable to be its own package so that distros can install it by itself without installing Pidgin
Some more info on the directories: toplevel/autogen.sh (calls 'automake --add-missing --copy') finch/libgnt/autogent.sh (calls 'automake --add-missing --copy') Here lies the problem. If I run toplevel/ag.sh it creates finch/libgnt/Makefile.in, as it always has and should. Within there it generates a variable called DIST_COMMON which doesn't reference the automake scripts such as 'install-sh'. If I then run finch/libgnt/ag.sh it will regenerate f/l/Makefile.in to where DIST_COMMON now contains 'install-sh' but it references the one in toplevel/ instead of one located in libgnt/. DIST_COMMON="..... ../../install-sh ...." On the other hand if I run finch/libgnt/ag.sh first, it generates Makefile.in with DIST_COMMON having 'install-sh' located in libgnt/ DIST_COMMON="...... install.sh ...." The latter behavior is what I want to get. Is there a way I can get the latter outcome by doing the former procedure while still preserving the latter procedure/outcome pair? I understand that this question would/might require looking at how we have this setup in our source tree. We currently don't have a tarball of it up as this feature hasn't come out in a release yet. Though if need be, you can find a tarball at http://cs.southern.edu/~club/pidgin.tar.gz. TIA, Eric -- http://aluink.blogspot.com -- "...indexable arrays, which may be thought of as functions whose domains are isomorphic to contiguous subsets of the integers." --Haskell 98 Library Report