[...]

> The bash test harness is the only reference we have for determining
> whether or not a particular version of bash -- about to be released
> as part of Solaris -- works, or doesn't work. Ignoring failures in
> the bash POSIX compatibility tests simply because "Milan Jurik
> doesn't like these changes" is not a viable option.

Well, no need to make it personal. I have to admit I do not know what
exactly causes your change to bash configuration. If customer wrote a
bash script using whatever was documented in bash man page (even if that
was not defensive way of writing scripts), can your change break his
script? If yes, it needs to be weighted the gain of being 100% harness
successful and have happy(-ier) customers. Would you mind describing
what is the impact of the change in question to real life scripts? It
seems that it might change behavior of 'echo' builtin?

At the end of the day, it's the customers happiness who pays us all, not
any harness test suite ... and surely opensource projects often do not
care too much :)

Thank you
-- 
        Vlad
_______________________________________________
userland-discuss mailing list
userland-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/userland-discuss

Reply via email to