[freetds] SQLFetch broken in HEAD?

ZIGLIO, Frediano, VF-IT Frediano.Ziglio at vodafone.com
Mon Jul 3 11:27:46 EDT 2006


> 
> > From: ZIGLIO, Frediano, VF-IT
> > Sent: Monday, July 03, 2006 8:19 AM
> > 
> > > Frediano Ziglio wrote:
> > > > > I added a new unit test, src/odbc/unittests/rpc.c.  I
> > > think it's coded
> > > > > correctly, but SQLFetch returns SQL_NO_DATA_FOUND the
> > > first time it's
> > > > > called.  TDSDUMP shows the rows coming back, and
> > > > > src/odbc/unittests/getdata.c also fails to return rows
> > > (but the test
> > > > > passes).
> > > > > 
> > > > > I'm using unixODBC 2.2.11.
> > > >
> > > > SQLFetch return SQL_NO_DATA cause there are no rows...  "select
> > > > name from sysobjects where 0=1" return 0 rows.
> > > 
> > > I'm sorry, I should have looked at getdate.c more closely.
> > > 
> > > Could you take a look at my new rpc.c and tell me if it's coded
> > > correctly?
> > 
> > Perhaps you forgot to attach updated file? CVS version is still
> > 1.1.
> 
> No, I didn't forget.  Friday was the first time I added it to the unit
> tests.  
> 
> I'm interested to know if it's right and if it works for you.  :-)
> 

Well... it depends on what is the expected behavior... written as it's
written currently it write

  error: expected at least 15 rows to be returned, found 0

and it exit with 1. Well, this is the correct behavior for ODBC, but I
think that intention for any tests should be return 0 for expected
behavior...

freddy77




More information about the FreeTDS mailing list