Re: Patch for DBD::Pg pg_relcheck problem - Mailing list pgsql-interfaces

From Tom Lane
Subject Re: Patch for DBD::Pg pg_relcheck problem
Date
Msg-id 20819.1039477650@sss.pgh.pa.us
Whole thread Raw
In response to Re: Patch for DBD::Pg pg_relcheck problem  (Ian Barwick <barwick@gmx.net>)
Responses Re: Patch for DBD::Pg pg_relcheck problem  (Ian Barwick <barwick@gmx.net>)
List pgsql-interfaces
Ian Barwick <barwick@gmx.net> writes:
> So out with the candle wax and pins ;-). Am I right
> in thinking that the string returned by SELECT version()
> starts with the word "PostgreSQL" followed by:
>   a space;
>   a single digit indicating the major version number;
>   a full stop / decimal point;
>   a single digit indicating the minor version number;
> and either "interim release" number (e.g. ".1" in the case of 7.3.1), or
> "devel", "rc1" etc. ?

I would recommend parsing it the same way pg_dump does, see
_check_database_version().  Looks like it's basically skip-11-chars-
and-sscanf.  Note the lack of assumptions about numbers of digits.

In the next protocol version update (hopefully 7.4) I would like to see
the basic version string (eg, "7.3.1" or "7.4devel") delivered to the
client automatically during connection startup and then available from a
libpq inquiry function.  This would eliminate the need to call version()
explicitly and to know that you must skip "PostgreSQL " in its output.
However, it will only help for clients/libraries that are willing to
deal exclusively with 7.4-or-newer backends, so it will take a few
releases to become really useful.
        regards, tom lane


pgsql-interfaces by date:

Previous
From: Ian Barwick
Date:
Subject: Re: Patch for DBD::Pg pg_relcheck problem
Next
From: "Godson Retna"
Date:
Subject: FATAL 1: Password authentication failed