Beefy Boxes and Bandwidth Generously Provided by pair Networks
There's more than one way to do things

DBI / DBD::PgPP vs. Pg

by DaWolf (Curate)
on May 16, 2005 at 19:16 UTC ( #457530=perlquestion: print w/replies, xml ) Need Help??
DaWolf has asked for the wisdom of the Perl Monks concerning the following question:


I've recently noticed that there is a module on CPAN called Pg.

I tipically use DBI and DBD::PgPP and on a very quick first glance DBI appears to be a more comprehensive work but then again Pg seems to be a more specific module, something specific tailored for Postgres.

So I was wondering: can anyone give me some thoughts about wich one is better? Maybe point some features that one have but another don't?


Replies are listed 'Best First'.
Re: DBI / DBD::PgPP vs. Pg
by jZed (Prior) on May 16, 2005 at 21:54 UTC
    There are three different modules and the one you should use is neither of the ones mentioned so far. Pg is a non-DBI interface. DBD::PgPP is a useful hack if you want to use DBI and can't install the real module. DBD::Pg is the best way to access PostgreSQL from DBI. It is updated and maintained by a team from PostgreSQL. It may be a bit harder to install than DBD::PgPP, but is much better.
Re: DBI / DBD::PgPP vs. Pg
by davidrw (Prior) on May 16, 2005 at 19:30 UTC
    Pg looks to be old (last update Apr 2000), so just off that DBD::PgPP is probably better. Do you have a specific reason for needing to use DBD::PgPP instead of DBD::Pg? I don't know for sure, but i would suspect that the latter is better (more robust & better supported), unless of course you need PgPP's ability to run where a PostgreSQL client library is not available.

      This logic, "looks to be old," therefore the other is "better," is flawed. There are plenty of modules on CPAN which do exactly what they are meant to do and have not required patches and thus remain "old." Another word for this is "stable."

      In this case, your invalid logic have lead you to the correct conclusion. So you're right in spite of yourself. Pg is a direct wrapper over the PostgreSQL C API. Recently, the PostgreSQL core team took over maintenance of DBD::Pg as well as Pg. They deprecated Pg in favor of DBD::Pg and it is the first and primary module you should use when interacting with PostgreSQL. The compatibility mode module DBD::PgPP was created by some other person and is useful if you can't use the standard DBD::Pg module.

Re: DBI / DBD::PgPP vs. Pg
by revdiablo (Prior) on May 16, 2005 at 20:15 UTC
    Pg seems to be a more specific module, something specific tailored for Postgres

    Sure, DBI is not "specifically tailored" for PostgreSQL, but DBD::PgPP is. The DBI simply specifies the interface; the DBD is where the rubber meets the road. The only thing that Pg could have is a more PostreSQL-friendly interface. Though it's been my experience that the interface enforced by DBI is quite flexible, and very nice. I've never had a problem using it with PostgreSQL.

    As far as specific things that Pg does behind the scenes that PgPP does not, I don't know. As davidrw indicated, Pg is significantly older. It probably hasn't seen much maintenance, and likely doesn't have any significant advantages in terms of implementation, but if it does, it would be possible to move those bits into DBD::Pg and DBD::PgPP, where applicable.

    So, my reply can be boiled down to: Pg might do some things better than DBD::PgPP, but there's no inherent reason that it should.

Log In?

What's my password?
Create A New User
Node Status?
node history
Node Type: perlquestion [id://457530]
Approved by xorl
and all is quiet...

How do I use this? | Other CB clients
Other Users?
Others exploiting the Monastery: (5)
As of 2017-10-22 08:40 GMT
Find Nodes?
    Voting Booth?
    My fridge is mostly full of:

    Results (272 votes). Check out past polls.