From: Craig A. Berry [craigberry@mac.com] Sent: Sunday, February 25, 2001 9:20 PM To: Nelson Ingersoll Cc: Everhart, Glenn (FUSA) Subject: Re: RCS 5.7 (nei) for Spring 2001 CD Nelson, Sorry not to test this when you first sent it my way; the queue gets a little long sometimes. I've done a build now with the command file you supplied using Compaq C V6.2-007 on OpenVMS Alpha V7.2-1. I get a relatively clean build except for the QUESTCOMPARE1 informationals that you note. These can be suppressed either at the command line or by adding the right magic to an include file that gets included everywhere. The latter is what I did in diffutils, specifically: /* disable informational message about time_t declarations in C 6.2 and later */ #if __DECC_VER > 60200000 # pragma message disable (QUESTCOMPARE1) #endif Either approach is cheating and the sources should really be fixed, but it's not always obvious what to do. In some cases a simple cast would be appropriate, but in some cases not. Note that there is a utime() replacement in our GNU patch for VMS which we lifted from the Perl source (and it'd already been around before then). I didn't see any test suite but the images appear to be functional. At 12:41 PM -0700 2/16/01, Nelson Ingersoll wrote: >Glenn, > > Well, here it is. I think it is complete. I've put the object >files in. Even tested the BUILD_RCS.COM procedure in the >[.RCS57_NEI.SRC] directory. Hopefully, anyone with a modicum of >VMS experience can install and use this version of RCS. > > If you can, and I realize you just might be busier than a one >legged man in a butt kicking contest, please test the build and read >as much of the verbiage as you can stand. I've never had the >pleasure of submitting something like this to the world before. >Mistakes are expected; but, that don't mean I have to make them or >like them when I do. > > I copied Craig on this as well since he helped me over a big >hurdle with his GNU diff for VMS. If he has a chance maybe he can >test it too. > >- Nelson ... -- ____________________________________________ Craig A. Berry mailto:craig.berry@psinetcs.com