On Mon, Aug 12, 2024 at 11:31:34AM GMT, Richard W.M. Jones wrote: > On Mon, Aug 12, 2024 at 11:41:22AM +0200, Antonio Caggiano wrote: > > Add INSTALL_OCAMLLIB parameter for allowing ocaml install to a user > > defined path. If not defined, fallback to `ocamlc -where`. > > Can't the usual destdir mechanism be used? > > > Signed-off-by: Antonio Caggiano <quic_acagg...@quicinc.com> > > --- > > m4/ocaml.m4 | 3 +++ > > ocaml/Makefile.am | 14 +++++++------- > > 2 files changed, 10 insertions(+), 7 deletions(-) > > > > diff --git a/m4/ocaml.m4 b/m4/ocaml.m4 > > index fddd6a0c2..91896f386 100644 > > --- a/m4/ocaml.m4 > > +++ b/m4/ocaml.m4 > > @@ -17,6 +17,9 @@ AC_DEFUN([AC_PROG_OCAML], > > OCAMLVERSION=`$OCAMLC -v | sed -n -e 's|.*version* *\(.*\)$|\1|p'` > > AC_MSG_RESULT([OCaml version is $OCAMLVERSION]) > > OCAMLLIB=`$OCAMLC -where 2>/dev/null || $OCAMLC -v|tail -1|cut -d ' ' > > -f 4` > > + if test "x$INSTALL_OCAMLLIB" = "x"; then > > + INSTALL_OCAMLLIB=$OCAMLLIB > > + fi > > AC_MSG_RESULT([OCaml library path is $OCAMLLIB]) > > > > AC_SUBST([OCAMLVERSION]) > > Note this is a copy-in file from another project, I don't know if we > can really modify it here. (Unfortunately I'm not sure of the > upstream status ...) > > Eric, any idea about this? >
Which upstream? The file was added in commit acf90002 in 2009, and the git log does not state what the upstream source would be (only that there is supposed to be an ocaml.m4 man page somewhere?) Okay, a quick google search found: https://manpages.ubuntu.com/manpages/mantic/man1/ocaml.m4.1.html which says upstream is: https://forge.ocamlcore.org/projects/ocaml-autoconf/ and sure enough, Fedora 40 has the ocaml-autoconf-1.1-40.fc40.noarch package available for use, and which appears newer than the copy we are using (I think it was copied from ocaml-autoconf-1.0), although upstream still carries only a 2009 copyright date, so it's hard to tell if upstream has any recent maintenance. The beauty of open source is that we can fork the file to meet our needs; the curse of open source is that if we fork and don't push the improvements upstream, no one else will benefit. But given how little upstream seems to change, I don't know that there are many users in the first place. It may also be possible to accomplish what you want by adding your code to configure.ac after the call to AC_PROG_OCAML, rather than modifying the imported file. -- Eric Blake, Principal Software Engineer Red Hat, Inc. Virtualization: qemu.org | libguestfs.org _______________________________________________ Libguestfs mailing list -- guestfs@lists.libguestfs.org To unsubscribe send an email to guestfs-le...@lists.libguestfs.org
