On Oct 25, 2010, at 2:19 PM, Mark Erbaugh wrote: > Does SA maintain a usable unique identifier for newly added child records > before the data is committed? > > I have a mapping of a one-many relationship using a foreign key. The detail > (many side) records are in an instrumented list. I need to relate the items > in this list to rows in an user interface object (a ttk.Treeview object). The > detail table has a primary key that is maintained by SA, but until the data > graph has been committed to the database, the corresponding fields detail > objects are None. > > I can't just use the position of the detail item in the instrumented list > because it is possible that the user can add and delete rows. The Treeview > object does not re-use row identifiers for rows that have been deleted, so > after the user has added and deleted detail rows, the Treeview rows will not > match the instrumented list rows.
the ORM uses Python object identity to maintain associations before foreign key / primary key identifiers are assigned. So you could either associate your user interface objects directly with the related objects, or with their identity via a dictionary, otherwise if you need something that is durable beyond the scope of a single Session you'd need to assign a unique identifier (uuid.uuid4() is a good choice for this) or issue flush() so that primary key ids are available. -- You received this message because you are subscribed to the Google Groups "sqlalchemy" group. To post to this group, send email to sqlalch...@googlegroups.com. To unsubscribe from this group, send email to sqlalchemy+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/sqlalchemy?hl=en.
