Best Practices on Insert....

[drbigfresh]

I need to insert 20 rows every 30 seconds, and I need to make sure that what I am inserting is unique... Luckily each piece of data has a unique GUID I can check against.

 

I know that I can check to see if there are duplicates by doing a select while I am doing the insert, but as the data grows, I imagine it will prove to be inefficient.

 

I was thinking I could make the GUID field in the database a guid or primary key, and then the database would throw an exception when I was doing the insert (if there was a duplicate)...I would then just need to find a way to trap the error so the rest of the insert can continue...

 

Anyone have an opinion on any of this or some guidance?

 

Thanks!

Scott.

 

 

 

 

[bubblegum.anarchy]

This is the way I usually check for duplicates:

$result = mysql_query("INSERT INTO temp SET content = 'variance'") or mysql_errno() == 1062 or tigger_error(mysql_error(), E_USER_ERROR);

if (mysql_errno() == 1062) // handle duplicate
   ...
else
   ...

[fenway]

How about INSERT INGORE?