Quantcast

Zend\Db For Beta3 Timeline

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Zend\Db For Beta3 Timeline

ralphschindler
Hi all,

In trying to nail down a suitable set of goals for delivering Zend\Db
for beta 3, I have looked at the RFC and want to propose what I think is
doable and an associated timeline.  If beta3 is 4-6 weeks away, here are
the aspects I feel should be within reach of delivering:

   * The Adapter infrastructure, and integration with:
     Mysql, Postgres, Sql Server, Sqlite, and Db2 for i5

   * The Resultset infrastructure capable of iterating results from
     the above platforms satisfying a minimal API consistently.
     This means that all platforms should conform to a minimal API
     (when iterating), but can have additional features specific to
     the platform.

   * Testing infrastructure.  This goes without saying, but while
     we are developing the above, we need to be testing.  Since there
     are very specific needs for testing this codebase against the
     variety of backends, this solution must be well thought out
     and will probably take a good bit of time

   * Finally, a portion of the Zend\Db\Sql component.  I would like
     to target 3 distinct platforms for Sql abstraction:  Mysql,
     Sql Server and Db2 for i5.  These 3 servers represent a good
     bit of the kinds of variety one would need to address when
     developing an abstraction layer for issuing valid statements
     to a database server.

   * Note: in all of these infrastructures, cross cutting concerns
     should be addressed (like logging, plugins - more than likely
     through the use of the EventManager)

Once these things are mostly in place, a solid infrastructure should be
available for the community to start developing along side me, if they
are so inclined.  In fact, much of the above has already been started
and gives me a head start on the 4-6 timeline for delivering these.  It
might be the case that we get a bit more done, but the above list
represents an attainable set of goals within this timeline.

After that, I would then suggest addressing the following:

   * All of the remaining vendor specific adapters
   * The remaining result iterators
   * The remaining sql platform abstractions
   * Then, higher order APIs:
     * TableGateway
     * RowGateway
     * ActiveRecord (if Autor is so inclined)
     * ..., other APIs

What do you guys thing?

-Ralph

--
List: [hidden email]
Info: http://framework.zend.com/archives
Unsubscribe: [hidden email]


Loading...