I don't know about this particular problem, but I find in programming
in general it's always best to pass in the value of now as a
parameter, and pass it on to functions you are calling, thus avoiding
problems with mismatching values.
But as Andreas says the values should be the same within
You can allocate you're numbers from a sequence...
http://www.postgresql.org/docs/7.4/interactive/functions-sequence.html
Then you can peek at the sequence to see what was last allocated.
Exactly how you do it depends on your circumstances.
On 18 Nov 2005, at 13:26, Harald Armin Massa wrote:
So I have a scenario in which account creation at the application
layer generates a set of tables and indexes. The tables created have
foreign keys to tables in use (and frequently read from and written
to) by the rest of the application. Occasionally I was getting
deadlocks, and this definitel
The only linux app I can find that does the job is SugarCRM.
However, they use mysql as their backend. Might it be relatively
easy for me to change the code so it will work with
postgres? What's involved?
Rich, what's wrong with the app as it is? Why change anything?
---(en
Do other databases support both UTf8 and UTF16?
Oracle supports UTF-8, UTF-16 an some other special UFT encodings. I
think some of them are pre UTF-8 becoming ratified, hence they are
partially compatible.
It's an install time option for an Oracle database. ASCII databases
can be upgraded to
Try PEAR DB.
http://pear.php.net/package/DB
It's a database abstraction layer, which means that you can change DB
later, with less hasle.
Have a look at how phpPgAdmin works. Download the source and have a
look.
http://phppgadmin.sourceforge.net/
On 22 Apr 2005, at 20:04, [EMAIL PROTECTED] wro
> I usually put DDL statements in a transaction, for a couple of
> reasons: so that a mistake doesn't leave me with half-done work
> (any error will cause the entire transaction to roll back), and to
> make the changes atomic for the benefit of other transactions.
Can you do that in postgres? Wi
