Hi

I have looked around on the net and even put some output in the hibernate source to try and understand how the name for a foreign key is generated for example "fkb96df9aa146a5373".

Our problem is that when we deploy with hbm2ddl auto set to update duplicated foreign key constraints with different ID's but are mapping the same columns. We wanted to understand how the name was generated so that each time the same name would be generated hence not creating duplicate constraints

Thanks for any help in advance.

Cheers

Daniel Lees
(Senior Software Engineer)

I think i might have found something, is it:

hexString of a hash of the table name plus the hexString of the sum of the hashes of all the columns in the parent table? The bit of code below?

while ( iterator.hasNext() ) {
result += iterator.next().hashCode();
}

( Integer.toHexString( name.hashCode() ) + Integer.toHexString( result ) )


If this is the case does any1 know why when a new column is added there is the need to create a new name for an existing foreign key constraint? I would have thought that if the column has no relevance to the existing constraints then no new name would be needed?

Cheers

Daniel Lees

Hi leesy84,

if you don't like these funny names, give foreign key constraints a self defined name; i.e:



Carlo