cvs commit: src/tools/regression/usr.bin/sed regress.s3.out regress.s4.out regress.s5.out regress.sg.out regress.sh

[Peter Wemm]

"Daniel C. Sobral" wrote:
> Mike Barcroft wrote:
> > Tony Finch <fanf at FreeBSD.org> writes:
> > 
> >>fanf        2003/06/04 08:26:24 PDT
> >>
> >>  FreeBSD src repository
> >>
> >>  Modified files:
> >>    tools/regression/usr.bin/sed regress.sh 
> >>  Added files:
> >>    tools/regression/usr.bin/sed regress.s3.out regress.s4.out 
> >>                                 regress.s5.out regress.sg.out 
> >>  Log:
> >>  Add tests for the behaviour of substitution when the regex can match
> >>  a zero-length string. This shows bugs in the s///g and s///2+ cases.
> 
> Let me add my thanks too. This kind of bug is the worst kind possible. 
> Particularly when you _don't_ detect it, and end up with flawed data.

In this case however, the cure is worse than the disease.  World wont build
anymore since sed is now broken.  'make world' seems to be a better
basic regression test in this instance.

Cheers,
-Peter
--
Peter Wemm - peter at wemm.org; peter at FreeBSD.org; peter at yahoo-inc.com
"All of this is for nothing if we don't go to the stars" - JMS/B5