Some spec issues

[Michael Watzek <mw...@spree.de>]

Hi Craig,

there might be some spec issues wrt version 2006-01-03. Please see below:

1) Lifecycle states in chapters 5.5.1 - 5.5.9 are grouped under 
"Datastore Transactions". Readers may be confused that there no such 
states for datastore transactions. Actually, the states applies for both 
transation types, right?

2) Chapter 5.6.1 specifies the definitions below. I wonder, if these 
definitions should be specified as assertions:
2.1) A persistent-nontransactional instance transitions to detached if a 
transaction is commited while the DetachAllOnCommit property is true.
2.2) A persistent-nontransactional instance transitions to 
persistent-nontransactional-dirty if a change is made outside a 
transaction while the NontransactionalWrite property is true.
2.3) The PersistenceManager must not hold a strong reference
to any persistent-nontransactional instance.

3) The state transition table in chapter 5.8 misses states 
detached-clean, detached-dirty, persistent-non-transactional-dirty, 
whereas the state interrogation table in chapter 7.4.6 considers these 
states. Furthermore, the state transition table misses operations 
detachCoyy and serialization.

4) The state transition table in chapter 5.8 appears under "Optimistic 
Transactions" in the table of contents. Readers may be confused that 
there is no state transition table for datastore transations. Actually, 
the table applies for both transation types, right?

Regards,
Michael
-- 
-------------------------------------------------------------------
Michael Watzek                  Tech@Spree Engineering GmbH
mailto:mwa.tech@spree.de        Buelowstr. 66
Tel.:  ++49/30/235 520 36       10783 Berlin - Germany
Fax.:  ++49/30/217 520 12       http://www.spree.de/
-------------------------------------------------------------------