Re: Debian readline/libedit breakage - Mailing list pgsql-hackers

From Stephen Frost
Subject Re: Debian readline/libedit breakage
Date
Msg-id 20110217032739.GO4116@tamriel.snowman.net
Whole thread Raw
In response to Re: Debian readline/libedit breakage  (Bruce Momjian <bruce@momjian.us>)
List pgsql-hackers
* Bruce Momjian (bruce@momjian.us) wrote:
> Joshua D. Drake wrote:
> > > * Tom Lane (tgl@sss.pgh.pa.us) wrote:
> > > > In particular, getting rid of use of OpenSSL would not be sufficient
> > > > to satisfy the most rabid GPL partisans that we were in compliance.
> > >
> > > I've never heard anyone argue that position, don't believe anyone would,
> > > and certainly don't agree with it.
> >
> > Yeah I am not sure I can buy Tom's argument here. The GPL is fairly
> > clear and is compatible with the BSD/Postgres license.
>
> Compatible if you want the result to be GPL, yes.  I hardly see that as
> desirable.  I am unclear what is being said above.

The comment regarding compliance, to me anyway, came across as meaning
that community PG wouldn't be compliant with the GPL if it linked with
libreadline and not with OpenSSL.  Perhaps I misunderstood.

If the concern is that EDB's (or anyone else's) modified PG is linked
and distributed with libreadline, then yes, there is room to be
concerned about the GPL and I'd recommend they not do that.  Of course,
I'd also recommend they not link with OpenSSL either, given the
advertising clause in that license.  Thankfully, there's a clear option
for dealing with libreadline- distribute libedit and let users
LD_PRELOAD (or maybe distribute a community/OSS psql, if it's
unmodified..).  There isn't an option for dealing with the OpenSSL
problem currently. :(
Thanks
    Stephen

pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: Unexpected page allocation behavior on insert-only tables
Next
From: Greg Stark
Date:
Subject: Re: Debian readline/libedit breakage