[unixODBC-support] second opinion on quiktest.c
sharkey at netrics.com
Mon Feb 21 21:51:37 GMT 2005
> Given that you return SQL_IC_LOWER, then yes, your driver is correct in
> what its doing.
> It does hi-light the fact that a good test to cover multiple drivers is
> not a trivial matter.
The test seems pretty buggy, though.
BuildInsertStmt() looks like it's inserting one row per supported data
type, and then one extra row using SQLParamData and SQLPutData. Then
TestLikeQuery() executes a query designed to retrieve all rows, and
checks that the number of rows is equal to the number of types. This
should fail with all datasources. I don't see how this test made it
out the door.
Anyway, I'll try and put together a patch once I'm done working through
the remaining bugs. I don't know if you have contacts at MS and can
push some of these fixes through to them.
More information about the unixODBC-support