Re: User documentation was: [Auto-config for collision column]

2006-09-03 Thread haleh mahbod

Wiki seems to be good place for in progress information.  It makes sense to
move the documenation to the website once it is matured.  What do others
think?

Haleh


On 9/1/06, Kevin Williams [EMAIL PROTECTED] wrote:


Yes.  Eventually, we will need some type of user-oriented documentation
for all provided features.  I am not sure what form this should
take...the Wiki, a Users Guide?  We should probably have similarly
structured user documentation for the three subprojects.  Any ideas or
precedence for this?

--Kevin

Luciano Resende wrote:

 Hi Kevin

   This sounds like a good aproach for one of the ways to support OCC in
 DAS. Should we also have this aproach documented in our Wiki as a
 reference
 ?

 - Luciano

 On 8/30/06, Kevin Williams [EMAIL PROTECTED] wrote:


 Currently, in the absence of otherwise conflicting, explicit
 configuration, we assume that any DO property named id will map to a
 primary key column in the database.  Likewise a property name of the
 form x_id maps to a foreign key column that references some row in
 table x.

 I would like to add another auto configuration convention for the RDB
 DAS for collision properties.  So, any DO property with the name
 occ_count will be assumed to map to an integer table column that is
 used for optimistic concurrency control.  This column will then be used
 to overqualify any related UPDATE statements for that table.  In the
 future, we may want to add support for occ_timestamp, etc.

 Any comments?

 Thanks,

 --Kevin


 -
 To unsubscribe, e-mail: [EMAIL PROTECTED]
 For additional commands, e-mail: [EMAIL PROTECTED]







-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]




User documentation was: [Auto-config for collision column]

2006-09-01 Thread Kevin Williams
Yes.  Eventually, we will need some type of user-oriented documentation 
for all provided features.  I am not sure what form this should 
take...the Wiki, a Users Guide?  We should probably have similarly 
structured user documentation for the three subprojects.  Any ideas or 
precedence for this?


--Kevin

Luciano Resende wrote:


Hi Kevin

  This sounds like a good aproach for one of the ways to support OCC in
DAS. Should we also have this aproach documented in our Wiki as a 
reference

?

- Luciano

On 8/30/06, Kevin Williams [EMAIL PROTECTED] wrote:



Currently, in the absence of otherwise conflicting, explicit
configuration, we assume that any DO property named id will map to a
primary key column in the database.  Likewise a property name of the
form x_id maps to a foreign key column that references some row in
table x.

I would like to add another auto configuration convention for the RDB
DAS for collision properties.  So, any DO property with the name
occ_count will be assumed to map to an integer table column that is
used for optimistic concurrency control.  This column will then be used
to overqualify any related UPDATE statements for that table.  In the
future, we may want to add support for occ_timestamp, etc.

Any comments?

Thanks,

--Kevin


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]









-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]



Auto-config for collision column

2006-08-30 Thread Kevin Williams
Currently, in the absence of otherwise conflicting, explicit 
configuration, we assume that any DO property named id will map to a 
primary key column in the database.  Likewise a property name of the 
form x_id maps to a foreign key column that references some row in 
table x. 

I would like to add another auto configuration convention for the RDB 
DAS for collision properties.  So, any DO property with the name 
occ_count will be assumed to map to an integer table column that is 
used for optimistic concurrency control.  This column will then be used 
to overqualify any related UPDATE statements for that table.  In the 
future, we may want to add support for occ_timestamp, etc.


Any comments?

Thanks,

--Kevin


-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]