Hi Gary, * Ralf Wildenhues wrote on Thu, Sep 29, 2005 at 06:39:28PM CEST: > * Gary V. Vaughan wrote on Thu, Sep 29, 2005 at 06:32:32PM CEST: > > Ralf Wildenhues wrote: > > > > This patch is just to introduce LT_CONFIG_LTDL_DIR without regressions. > > The rest of the fixes for LT_WITH_LTDL are yet to be split into separate > > patches and posted for review. > > OK. Good point.
(There should be a way for the user to specify "I am also content with an older installed libltdl version"; for example, by allowing to test for a different library function.) * Gary V. Vaughan wrote on Fri, Sep 30, 2005 at 12:05:29PM CEST: > > Thanks for the re-review ;-) Okay to commit the attached? Yes, please commit, after accompanying the m4_pushdef with a m4_popdef at the end of the respective macro. Thank you! The BSD make failure should be fixed separately. There is some other stuff going on which I haven't understood fully yet, but let's fix that separately, too. (I'll post a report when I have sorted out that it's not again a failure of mine. :) > >Here's a possible reason: > >grep _LTDL_DIR Makefile configure > >| Makefile:subdirs = _LTDL_DIR > >| configure:ac_subdirs_all='_LTDL_DIR' > >| configure:subdirs="$subdirs _LTDL_DIR" > > Yep, that was the reason for all the pertinent failures. Fixed in > this version. While I can't see the difference in the patch versions that actually made this work, it seems to work now. :-)) > I've made your test case into a new Autotest case, which I'll post > separately. I'll soon post patches necessary to make the test work. > What is the status of this patch btw?: > > 277-gary-rename-remaining-troublesome-ltdl-apis.diff Since luckily that one is not part of the larger chain of patch dependencies (and it's nontrivial and introduces new code), I'd like to postpone it until the queue has been flushed a bit more.. Cheers, and thanks, Ralf
