PDO will only abstract the API for us.  It will not help us abstract query 
syntax or solve the fact that there is no such thing as "SQL", just vaguely 
similar languages called MySQL, PostgreSQL, SQLite, Oracle, etc.  

On Sunday 11 November 2007, Ken Rickard wrote:
  

I would agree with Khalid, actually.  Having rock-solid code on a
popular platform seems ideal, at least until something like PDO is
ready to handle database abstraction for us.

I only installed pgSQL to be a "responsible" module maintainer after
my requests for pgSQL testers went unanswered.

At work, we're entirely MySQL.

- Ken Rickard

On Nov 11, 2007 7:30 PM, Khalid Baheyeldin <kb@2bits.com> wrote:
    

On Nov 11, 2007 5:17 PM, Karoly Negyesi <karoly@negyesi.net> wrote:
      

In my opinion, any patch should be committable to core once it works
on mysql and has a decent hope (use common sense) to work on
postgresql. Once it's committed those who care about postgresql, if
they so want to, can test and if there is a need, fix it.

Prime example: http://drupal.org/node/146466 this is the most
important patch we have currently as it makes Drupal search speedy and
nice. And it is more or less on hold just because noone is sure
whether the postgresql update works or not.
        

I know many will not like what I will say, but  I have to say it.

A prime example of where MySQL works fine to solve an issue with a few
lines is this issue http://drupal.org/node/83738. All the huge changes
and jumping through hoops is because of accomodating PostgreSQL

We introduced schema changes just because PostgreSQL cannot do case
insensitive matching by default, while MySQL works fine. For the sake of
5% (or 1%) of the sites, we are increasing complexity.

      

Regards,

Karoly Negyesi
        

--
Khalid M. Baheyeldin
2bits.com
http://2bits.com
Drupal optimization, development, customization and consulting.