October 30, 2002 at 1:07 pm
Does anyone have an explanation to why subscriber tables in TX-Rep don't have primary keys? They have unique clustered indexes instead. Is there any difference between a primary key and a unique clustered index?
Remember this when a developer tells you it will just be temporary. Temporary = Permanent.
October 30, 2002 at 5:40 pm
A primary key cannot have a null, a unique index can have one. Either can be clustered.
Andy
October 31, 2002 at 2:36 pm
Thanks Andy, I guess that confuses me even more as to why replication wouldn't promulgate the primary key definition to the subscriber tables. I would think the last thing replication could want is to be able to insert a null into the primary key field.
Remember this when a developer tells you it will just be temporary. Temporary = Permanent.
October 31, 2002 at 5:24 pm
Its a good question, one that Im planning on digging into it a little (along with about a 100 other things, list keeps getting longer!), be interesting to see.
One other advantage, sort of, is that you can set a unique index to ignore duplicate keys, cannot do that with a primary key.
Andy
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply