From: ningersoll@codenet.doh.net Sent: Friday, February 18, 2000 7:44 PM To: Info-VAX@Mvb.Saic.Com Subject: Re: Problem using RCS 5.7 and diffutils 2.7 Update: Trying to debug a program when it redefines SYS$OUTPUT is tough. My work is not yet finished; but I have discovered the following: 1) In the RCSKIT conf.h file DIFF_L should apparently be set to 0 (zero). Doing that removed the error I saw and reported. 2) Unfortunately, VMS is a fixed case system. In spite of the drivel Compaq (nee Digital) Support has told me to wit: "Everyone knows UNIX is lower case.", UNIX is not. In VMS v7.0 the default case has apparently been changed from UPPER to lower. So VMS now, instead of munging "-m" into "-M", mungs "-M" into "-m". Ya gotta love VMS Engineering. At least they're consistently irrational! Alas the system() call made in RCSUTIL/runv() apparently also suffers this case munging behavior. I am trying to see if I can quote the appropriate 'co' switches in the rcsdiff.c code. So far I have not made it work right. More as I figure it out. - Nelson ... (Don't get me wrong, I really like VMS. But dangit, sometimes it's more like a lover than a friend. It's flowers or no dance dude!) On Fri, 18 Feb 2000 18:00:37 GMT, ningersoll@codenet.doh.net (Nelson E. Ingersoll) wrote: > > I am trying to install RCS v5.7 on an OpenVMS 7.1 system using DEC >C v5.6-003. Using the excellent information from Arne Vajhøj I was >able to get both RCS and DIFFUTILS (diff & diff3) to compile and link. >(Obviously if they compile and link they MUST be good. Right? ) > > I can use the ci command. However, when I try to use the rcsdiff >command I get the following output: > >MAPLE$ rcsdiff ci.c >=================================================================== >RCS file: RCS/ci.c_v >retrieving revision 1.1 >diff -r1.1 ci.c >dsa16:[cam.][ningersoll.gnu.rc.src]diff.exe;2: invalid option -- . >dsa16:[cam.][ningersoll.gnu.rc.src]diff.exe;2: Try >`dsa16:[cam.][ningersoll.gnu.rc.src]diff.exe;2 --help' for more >information. >%NONAME-E-NOMSG, Message number 00000002 > > > For the life of me I am not sure what diff is telling me. But it >looks like it is telling me it doesn't understand the '-r1.1' option. >I searched a bit through the net, especially comp.os.vms, and have not >seen a reference to this problem. As I am not a UNIX diff expert I >don't know what the -r1.1 option is for. Were it an RCS command, >-r1.1 would be telling the RCS program to fetch and use revision 1.1. > > Can anyone offer any insight into this problem and what might be >done about it? > >- Nelson ...