Re: Simple DBI handle caching.

...don't need or want the ache of defining a schema...

Schema generation with DBIC is generally trivial. Some special needs can definitely be a pain for a novice with the kit, I know. I'm not saying you need to use DBIC but it's really quite excellent and in the past your code style and mine have lined-up well so I recommend it to you.

Take a look at these goodies (assuming you haven't already)–

Comment on Re: Simple DBI handle caching.

Replies are listed 'Best First'.
Re^2: Simple DBI handle caching. by kyle (Abbot) on May 11, 2011 at 17:52 UTC
I've generated a schema from an existing database before, and I agree it's pretty easy. What it came up with was a tree full of files that described every field of every table in the database. I want to avoid that for a couple of reasons. The database I'm working with is old and huge and has been through a lot of bad design. I'd be surprised if the result "just works." The program I have is just one file, and I'd like to keep it that way. The environment I'm in is mostly "not Perl." I'm writing this as a sysadmin outside the support of the developer population, so I can expect future database changes behind my back. All of the work of my little program is done in one table. I have one SELECT and one UPDATE (each run many times with different parameters), and that's it. I'm not doing a ton of ugly stuff that DBIC would save me. Anyway, it looks like DBIx::Connector will do what I want, and my quick hack works too. If DBIx::Class can do the job without tangling me up in future work, I'd love to learn how. As an aside, I'm touched that you remember me after all my absence.	[reply]

Replies are listed 'Best First'.

Re^2: Simple DBI handle caching.
by kyle (Abbot) on May 11, 2011 at 17:52 UTC

I've generated a schema from an existing database before, and I agree it's pretty easy. What it came up with was a tree full of files that described every field of every table in the database. I want to avoid that for a couple of reasons.

The database I'm working with is old and huge and has been through a lot of bad design. I'd be surprised if the result "just works."
The program I have is just one file, and I'd like to keep it that way.
The environment I'm in is mostly "not Perl." I'm writing this as a sysadmin outside the support of the developer population, so I can expect future database changes behind my back.

All of the work of my little program is done in one table. I have one SELECT and one UPDATE (each run many times with different parameters), and that's it. I'm not doing a ton of ugly stuff that DBIC would save me.

Anyway, it looks like DBIx::Connector will do what I want, and my quick hack works too. If DBIx::Class can do the job without tangling me up in future work, I'd love to learn how.

As an aside, I'm touched that you remember me after all my absence.

[reply]