T
Tom
I’m trying to use a DSN-less connection to a SQLSERVER DB and am
experiencing problems. I can connect to all the tables I need and
most of the views. However, one view is giving me problems.
The view in question is a summary for use on a report – basically
totaling up hours and dollars from a bunch of detail records – in an
attempt to reduce the amount of data sent over the network. For some
unknown reason, I can’t set an index on the view, so whenever I try to
link to it programmatically I get the “Select Unique Record
Identifier” dialog box. I don’t want the end user to have to deal
with that.
So as a work around, I tried going to a pass-through query. Works
great, but since it’s a DSN-less connection I need to embed the
password in the ODBC connection string. That’s a non-starter.
I’d really rather not go to a DSN connection and changing the query to
do the totaling locally instead of on the server will likely end up
slowing things down unnecessarily.
So my questions are:
• How do I programmatically deal with the select unique record
identifier? If I can dismiss that programmatically, I should be good
to go.
• Alternately, how do I obscure the ODBC connection string on the pass-
through query?
• Any other ways to go about this?
Thanks
experiencing problems. I can connect to all the tables I need and
most of the views. However, one view is giving me problems.
The view in question is a summary for use on a report – basically
totaling up hours and dollars from a bunch of detail records – in an
attempt to reduce the amount of data sent over the network. For some
unknown reason, I can’t set an index on the view, so whenever I try to
link to it programmatically I get the “Select Unique Record
Identifier” dialog box. I don’t want the end user to have to deal
with that.
So as a work around, I tried going to a pass-through query. Works
great, but since it’s a DSN-less connection I need to embed the
password in the ODBC connection string. That’s a non-starter.
I’d really rather not go to a DSN connection and changing the query to
do the totaling locally instead of on the server will likely end up
slowing things down unnecessarily.
So my questions are:
• How do I programmatically deal with the select unique record
identifier? If I can dismiss that programmatically, I should be good
to go.
• Alternately, how do I obscure the ODBC connection string on the pass-
through query?
• Any other ways to go about this?
Thanks