ADO DataView Templates

rbuser

Getting this to work is most frustrating.
Documentation on DataViews is limited at best, and the tech-tips/DADE
threads appear to be either linking incorrectly, or are limited.

I cannot understand why the dbADO module does not contain the required
TdaADOTemplateDataView
TdaADOQueryTempateDataView
ancestor classes

While past threads have indicated this is "easy", in fact it is not
completely clear as to what exactly is required and what is not.

So, my question:
Has someone [Nard ?] written these base classes for ADO custom dataviews and
have them working correctly ?
This would save untold aggravation, and allow me to get on with the "job"

Your help appreciated.
Phil

nardmoseley

If you cannot see the DADE tech tips articles in your newreader then
configure it to load old threads.

I converted the example to ADO, you can download it from

http://www.digital-metaphors.com/tips/daADOTemplateDataView.zip



--
Nard Moseley
Digital Metaphors Corporation
http://www.digital-metaphors.com

rbuser

Man thanks.
That was a huge help. I see now what I was doing wrong. DM should add
DataView base classes to the standard code base when developing a new
plugin.

I had a few problems with exceptions on designing a saved report. This was
related to properties of my JIT pipelines. How these are streamed and
restored inside the DADE is a little magic.

Also, it seems that you cannot connect pipeline event handlers in the
constructor of the Dataview, instead it seems that
ConnectPipelinesToData
is the correct place to connect all the event handlers for JIT pipelines
that are contained within a DataViewQuery, and saved as a DADE definition
within the report.

It remains a little mysterious as to how the JIT traversal is controlled.
Steppinh thru with debugger seems like a lot of RecordIdx = 0 being called
in a GetFieldValue

Now I can see how to added custom fields as well in the Dataview, one of
which is a large graphic that is constructed in real time during data
traversal.

nardmoseley

Thanks for the feedback.

The report engine is responsible for DataPipeline traversal. The engine may
call methods such as DataPipeline.First multiple times. The engine does not
know whether it is traversing a DBPipeline, JITPipeline, etc, it calls
methods of the DataPipeline ancestor class. The JITPipeline implements these
methods to manage its RecordIndex.


--
Nard Moseley
Digital Metaphors Corporation
http://www.digital-metaphors.com