From: "Branko Čibej" <[EMAIL PROTECTED]> Sent: Monday, February 05, 2001 1:08 PM > [...] the other half of my argument still stands: allowing keyword > expansion in *.ds[pw]'s is a bad idea. I agree it's a potentially troublesome problem, for .dsp and .mak files. I'm willing to accept the -ko flag on all .dsw, .dsp and .mak files to prevent CVS substitution from occuring. I'm unwilling to make these files binary with the -kb flag, as that reduces the cvs commit email reporting to something like: Index: FooFile.dsp =================================================================== RCS file: /home/cvs/scratch/FooFile.dsp,v retrieving revision 1.1 retrieving revision 1.2 diff -u -r1.1 -r1.2 Binary files /tmp/cvsnfHdy81050 and /tmp/cvsIHawC81050 differ Note the total absence of +/- details. That's the part I found unacceptable. The only remaining question ... why do we need the -ko flag in the first place? I've grepped for the '[' symbol in our .dsp and .mak files, which triggers the substitution, and I found no occurances. If you can point out the man page that details something I've overlooked, please do so. Bill
