On Fri, 7 Sep 2001, Robert Alexander wrote:
| Hi Claudio,
|
| I think having the functionality of stored procedures would be a
| wonderful addition to MySQL. Let me be the first to say a hearty
| 'Thank you!' for your very generous offer to devote your vacation time
| to this task.
I definately agree, and I'm reading all those other emails wondering
whether people are crazy?! This guy is saying that he wants to devote
considerable time in making stored procedures, and people start to argue
against him with sarcasm and everyhting.
|
| I like Oracle's PL/SQL. I've used it a fair amount, and it get's the
| job done. PL/SQL = Procedural SQL -- SQL plus procedural logic. This
| has the advantage of being a language that uses the SQL that we all
| already know and love, plus enough procedural things -- loops,
| conditionals, etc. -- to implement logic. It's more a programming
| language than just a query language. This is a good thing.
One LARGE point in PL is that, yes, it's Oracle's, and they have a large
industy momentum and are kind of the "defacto" industry standard. MySQL
could "steal" a lot of momentum from Oracle if you could port your
application to MySQL by just swapping the backend of your application. And
this is definately not negative for a Open Source project. (I'd like to
see MySQL be a _real_ alternative to the big-dudes. Let the OS community
dominate the whole serverside market totally!)
Oracle's PL/SQL would be a _huge_ benefit for MySQL, I believe.
"Transactions, subqueries and stored procudures NOW!"
|
| That being said, I don't think re-implementing PL/SQL is necessarily
| the way to go for MySQL. There may indeed be proprietary issues
| (though I'm not sure) and I think it'd be possible to take a more
| 'open source' approach.
The proprietary issues might of course be a huge problem. Anyone knows
anything about this?
|
| Let's remember what we want to accomplish with stored procs; that is,
| the ability to store server-side logic where we can implement such
| things as business rules, and ensure consistency in the way
| 'front-ends' of whatever flavour (and probably written by different
| programmers) perform transactions with the dataset.
Also, remember that _speed_ is a _big_ thing here.
If your application needs to "grep" through a large amount of data and
compare lots of rows from different tables in an elaborate way, then you'd
have to transfer immense amounts of data through the TCP stack (and maybe
even though a couple of ethernet cards and a cable!) or anyways between
different tiers and data representations in your system (think e.g.
Java?). This is done _much_ more effeciently with a couple of stored
procedures.
--
Mvh,
Endre
---------------------------------------------------------------------
Before posting, please check:
http://www.mysql.com/manual.php (the manual)
http://lists.mysql.com/ (the list archive)
To request this thread, e-mail <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
