Mark Pizzolato wrote: > ... Running these needs some sort of scripting paradigm to load, start and > examine the results of such tests.
Over a decade ago I wrote a Tcl "expect" script to install ITS on a SIMH simulated KS-10 from "install tapes". It isn't pretty (I had to look for input echo in order not to lose input characters in some places), but it works, or worked, I haven't run it in a while! ftp://ftp.ultimate.com/pdp10/its-install Python, Ruby and Perl now have expect-like modules, so there's a wider range of choices for scripting language-- the key thing is that simh has to be run under a pseudo-terminal (pty)-- pipes will not do! I'm a big believer in regression tests (ever since I joined the DEC FORTRAN-10/20 project, which used SNOBOL4 to parse test results). My SNOBOL4 in C project(*) has a regression test suite, and every bug I fix, and most features I add get a test. Phil (*) http://www.snobol4.org/csnobol4 _______________________________________________ Simh mailing list [email protected] http://mailman.trailing-edge.com/mailman/listinfo/simh
