[Dbix-class] Paging all re nested sets

[Moritz Onken]

Am 05.04.2008 um 18:00 schrieb Sebastian Willert:
> Unfortunately it's not as easy as this. Off the top of my head:
> transactional isolation just prevents dirty reads wrt. this problem,
> i.E. reading an inconsistent state.

Understood. But the transaction could include a read which checks  
whether the tree is still what it used to be and then update. This way  
the first part of the transaction (read) has to wait for a concurrent  
update. Am I wrong?

If so, a workaround could be some sub-classes which specify the dbms  
specific commands (DBIC::(Tree::)Lock::PostgreSQL?).