From: | Magnus Hagander <magnus(at)hagander(dot)net> |
---|---|
To: | Greg Sabino Mullane <greg(at)turnstep(dot)com> |
Cc: | pgsql-bugs(at)postgresql(dot)org |
Subject: | Re: [Fwd: DBD::Pg on HP-UX 11.31 64bit] |
Date: | 2010-12-19 13:21:42 |
Message-ID: | AANLkTinywO6Auqpb78AjcjetNJSMTWN5-Sg0MwDYut+y@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-bugs |
On Sun, Dec 19, 2010 at 13:42, Greg Sabino Mullane <greg(at)turnstep(dot)com> wrote:
>
>> How much are the postgres people willing to find a solution that would
>> force postgres to optional use BSD type sockets?
>
> FWIW, my long-term plan is to have DBD::Pg provide its own libpq. Not
> sure if this would enable you to solve the problem or not, but
> I suspect it might. Actually removing the libpq dependency is awaiting
> copious amounts of free time on my part, or funding from a sponsor.
> But it's on the todo list. :)
Spinning off into the off-topic based on the question - if you do
this, please make it *optional*. Unless you plan to actually implement
all the libpq functionality and also shoulder the burden to release a
new version whenever a new version of libpq is out. Things like
kerberos/gssapi and cert authentication that actually work the same
way as others...
Might be worth looking at what the ODBC folks did though - they run
the actual protocol for the queries, but they use libipq for
connection setup and authentication (if found - they'll fallback on
doing simple auth etc if libpq isn't there)
--
Magnus Hagander
Me: http://www.hagander.net/
Work: http://www.redpill-linpro.com/
From | Date | Subject | |
---|---|---|---|
Next Message | Peter Speck | 2010-12-19 17:27:47 | BUG #5794: 'explain' fails, but executing sql is ok. |
Previous Message | Greg Sabino Mullane | 2010-12-19 12:42:31 | Re: [Fwd: DBD::Pg on HP-UX 11.31 64bit] |