>>Does it have a .net data provider for using ADO .net ?
Not that I see in the beta documentation.
It may not matter that much depending on exactly what you want to do. It is
really easy to get U2 data and easily put it into an ADO.NET dataset
(especially if you use the U2 XML support - ~6 lines of .NET code after you
get the data from a unisubroutine method). It takes a little more effort to
get the data back into U2 from an ADO.NET dataset.
Stop reading here unless you want my .02 on what would really help for
better U2/ADO.NET integration.
What would really be helpful IMHO has more to do with improving the XML
support rather than with UO.NET support for ADO.NET. As far as UniData is
concerned (I haven't tried UniVerse), if you have an empty field, you end up
with something like this:
<ROOT>
<FILE_record>
<_ID>ITEM_ID</_ID>
<DESC>SOME DESCRIPTION</DESC>
<MVDATA_MV>
<MVDATA></MVDATA>
</MVDATA_MV>
</FILE_record>
</ROOT>
I can see how you may need to have it look like this in certain situations,
however, there should be an option to remove the empty fields (unless it is
among other MVs or SMVs that do have data, to keep associations in order).
I'm not sure if it has something to do with "NULL" vs. "Empty" fields, but
it is annoying to deal with. I have written programs to strip out the
problem so that you end up with:
<ROOT>
<FILE_record>
<_ID>ITEM_ID</_ID>
<DESC>SOME DESCRIPTION</DESC>
</FILE_record>
</ROOT>
It would be really nice to have this option already built-in instead of
having to hack at the XML output. It makes populating datagrids through
ADO.NET much easier when you have multiple bands of data.
I know that many people have suggested the DOWNLOAD tool (and it is a fine
tool)as a replacement, but it didn't work the way I had hoped it would (I
don't recall the reason why though).
Regards,
Jim
-------
u2-users mailing list
[EMAIL PROTECTED]
To unsubscribe please visit http://listserver.u2ug.org/
