At 6:37 AM -0500 12/17/04, Michael G Schwern wrote: > > > And while we're here, distdir is broken. $(MAKE) has no meaning on >> VMS; what you want is $(MMS)$(MMSQUALIFIERS). There are various >> examples in MM_VMS.pm. > >I changed the code to define a MAKE macro on VMS set to >$(MMS)$(MMSQUALIFIERS). At least I thought I did. The end of >MM_VMS->constants() is where its set and the middle of MM_Unix->constants() >should be where its printed out. Make sure you have the latest version >out of the repository. > >Tomorrow I'll start debugging things on VMS and Win9x, I haven't looked at >it at all so there's likely to be stupid mistakes. For now though I go to >sleep, so the changes are done for the night. > >It might be useful to look at the diff between 6.25 (revision 2146) and now >to see what got deleted in MM_VMS so you know where to look. Here's the >list of deleted subroutines.
snip I will try to take a look at this sometime, or better yet, read through a generated descrip.mms pretty carefully to see what it gave us. The attached patch against 2177 is necessary to even get it to run. Even after that for some reason EXE_FILES is not getting vmsified properly, so it fails its own build with: %MMK-F-CANTUPD, cannot update target BIN/INSTMODSH - sources unknown I can't see what it is and I'm out of time to tinker at the moment. >PS Does Subversion run on VMS? Has not been ported yet as far as I know. It should be doable since Apache 2.0 has been ported and whatever that library is called that Subversion uses must've been ported with it. While we are tearing the guts out of MakeMaker, any chance of getting manifind reimplemented with a case blind hash? There is an item in RT for this and I've just never had the time to do it. -- ________________________________________ Craig A. Berry mailto:[EMAIL PROTECTED] "... getting out of a sonnet is much more difficult than getting in." Brad Leithauser
mm_unix.patch
Description: Binary data
