On Wed, May 28, 2008 at 1:14 PM, Rob Edwards <[EMAIL PROTECTED]> wrote:
>>> That said, I find it's substantially more
>>> complicated (versus complex) and difficult to use, esp if you have an SQL
>>> background. You have to really jump through hoops to get some basic (in SQL)
>>> stuff done, and only really recently the docs have got to the point where
>>> you don't have to be a core developer to make full use of it.
>>
>
> Although I dont  have much of a background in SQL I couldn't disagree
> more, I think it makes everything a lot easier. A lot of thought has
> obviously gone in to the interface to make it really intuitive, a
> great improvement on CDBI.

It definitely is a big improvement. I'm not knocking it hard, I use it
in a project and don't regret it. Once you're into the territory of
complex queries though it gets ugly, fast.

I'd be curious the kinds of things you're doing with DBIC. Obviously I
can't speak for anyone else but I suspect if you got good at SQL you'd
find the same thing. While I appreciate it's possible to get DBIC to
do it in a somewhat clunky way (bind => [$arg1, $arg2] is hardly too
literate) it's not pretty or elegant. It _is_ pretty & elegant for
quite a decent subset of SQL though. As I say I am impressed & it's
not a big ding or anything.

I don't understand the objection to having SQL in code. If I pull out
data out of a hash I don't necessarily have to abstract that into some
file elsewhere. Why should I do that if I'm pulling data out of a
database?

P

>
> I hate SQL amongst code unless its performance critical (at my work
> its generally isn't) I like to keep it abstracted where ever possible.
>
> Rob
> _______________________________________________
> BristolBathPM mailing list
> [email protected]
> http://mailman.bristolbath.org/mailman/listinfo/bristolbathpm
>
_______________________________________________
BristolBathPM mailing list
[email protected]
http://mailman.bristolbath.org/mailman/listinfo/bristolbathpm

Reply via email to