svn commit: r326554 - in head: . usr.bin/sponge usr.bin/sponge/tests usr.bin/tee

[Alexey Dokuchaev]

[ Charset UTF-8 unsupported, converting... ]
On Tue, Dec 05, 2017 at 01:06:40PM -0800, Conrad Meyer wrote:
> On Tue, Dec 5, 2017 at 12:16 PM, Cy Schubert wrote:
> > In message [...], Eitan Adler writes:
> >> I'm not near my commit box right now so someone will have to revert
> >> it. What's funny is that I ran a complete 'make buildworld
> >> buildkernel' last night with this patch.
> >>
> >> Will respond to the rest of the thread later[.]
> >
> > No problem. r326575 addresses the issue.
> 
> Not fully -- it doesn't address that the code is functionally broken or
> that the change touched things it didn't need to.  I think a revert is
> still called for, when you can reach a machine with commit access.

While it was already reverted, all the input so far raises a question of
should it be readded even if/when fixed, since it apparently can be done
with existing shell tools, in a better way [1].

I'd like us to have a smaller, orthogonal base, with tools not repeating
themselves.

./danfe

[1] https://lists.freebsd.org/pipermail/svn-src-head/2017-December/107001.html