On Tue, 5 Jun 2001 20:37:41 -0700, Justin Erenkrantz wrote: >On Wed, Jun 06, 2001 at 03:33:46AM -0000, [EMAIL PROTECTED] wrote: >> bjh 01/06/05 20:33:46 >> >> Modified: uri Makefile.in >> Log: >> Need to specify libtool objects so libtool is used to compile. >> >> Revision Changes Path >> 1.5 +1 -1 apr-util/uri/Makefile.in >> >> Index: Makefile.in >> =================================================================== >> RCS file: /home/cvs/apr-util/uri/Makefile.in,v >> retrieving revision 1.4 >> retrieving revision 1.5 >> diff -u -r1.4 -r1.5 >> --- Makefile.in 2001/05/22 01:22:05 1.4 >> +++ Makefile.in 2001/06/06 03:33:46 1.5 >> @@ -4,7 +4,7 @@ >> # bring in rules.mk for standard functionality >> @INCLUDE_RULES@ >> >> -gen_uri_delims_OBJECTS = gen_uri_delims.o >> +gen_uri_delims_OBJECTS = gen_uri_delims.lo >> gen_uri_delims: $(gen_uri_delims_OBJECTS) >> $(LINK) $(EXTRA_LDFLAGS) $(gen_uri_delims_OBJECTS) $(EXTRA_LIBS) > >Roy initially pointed out that we shouldn't need libtool to compile >gen_uri_delims - which is why it was gen_uri_delims.o. Is this not >the case on OS/2?
No, it's not. If linking with libtool, objects must be compiled with libtool too. This is the case everywhere else in the build. >gen_uri_delims is only needed to produce the uri_delims.h file. That >object file doesn't need to be slurped into the libaprutil.a file. >I think Roy suggested ditching slurping in all .lo files and going >for a more manual approach. Might be a good idea, though excluding a few explicitly wouldn't be hard. -- ______________________________________________________________________________ | Brian Havard | "He is not the messiah! | | [EMAIL PROTECTED] | He's a very naughty boy!" - Life of Brian | ------------------------------------------------------------------------------
