On Tue, 20 Feb 2001 09:52:01 -0800 (PST) Richard Soderberg <[EMAIL PROTECTED]> wrote: > Hmm. I could live with that. Makes finding them in the bugdb real easy, > too. Is that easily doable? > > <flamebait> > Perhaps another well-known multiple of 3? >:) > </flamebait> There's more to consider + Would it be helpful to have smokers `register' their testing machines, so they don't have to send in the configuration time after time. + Would it be helpful to have a stampfile other than .patch that indicates the patchlevel that *does* matter for testing (document patches won't change test results) to prevent useless computer cycles and test reports. Test reports should than mention this patch number in the header like SMOKE=8876 OK 5.7.1 on HP-UX 11.00 Dperlio Dthreads Dithreads D64bitall > On Tue, 20 Feb 2001, Stephen P. Potter wrote: > > > Lightning flashed, thunder crashed and Kevin Meltzer <[EMAIL PROTECTED]> whis > > pered: > > | > > generated with the help of perlbug 1.33 running under perl v5.7.0. > > > > Would it be worth changing perlbug so that if it is a smoke test, it > > automatically goes to [ID: $DATE.999], so that all the reports can get > > added to the same ID? > > > > -spp > > -- H.Merijn Brand using perl5.005.03 on HP-UX 10.20, HP-UX 11.00, AIX 4.2, AIX 4.3, DEC OSF/1 4.0 and WinNT 4.0 SP-6a, often with Tk800.020 and/or DBD-Unify ftp://ftp.funet.fi/pub/languages/perl/CPAN/authors/id/H/HM/HMBRAND/ Member of Amsterdam Perl Mongers (http://www.amsterdam.pm.org/)
