FreeTDS Development Group

[Dan Mohn <mohnd AT egolfscore.com>]

James K. Lowden wrote:
> Yes, you're on the right track.  You might want to just apply
> frediano's patch, posted at 11:07 AM Monday.  But if you want to just
> hack the code,  I think are two critical places:
> 
>   File      Function               Line
> 4 odbc.c    _SQLAllocStmt          1441 stmt->attr.query_timeout = 0; 
> 5 odbc.c    _SQLExecute            3032 tds->query_timeout = 0;

So I downloaded the most recent nightly build this morning.  Then in
odbc.c I changed the above timeouts to 20 and re-built the driver.

I swung for the fences and went right to my RealBasic test application,
and YES! it timed out.

Jonathan wrote;
>> Unfortunately, I don't think there's a RB workaround, since any 
property set using the RB SetConnectionAttribute method: <<

I think the "timeout" in RealBasic is for the connection timeout.  The
timeout for a query seems to be set in SetStmtAttribute, which RealBasic
doesn't seem to have the ability to call.

>> I have filed a bug report - Dan might want to sign on: <<

I shall, thanks.

ZIGLIO, Frediano, VF-IT wrote:
> I didn't committed this patch cause
> - I added a "default_query_timeout" on HDBC. Is this required??
> - is the way query_timeout is reset to default correct??
> - should we handle case when statement is created before connection??

James K. Lowden wrote:
>> For me to answer these questions, I will have to study the
documentation. I won't be able to do that until next week.  Perhaps
someone else knows, or would like to research it? <<

I'm not sure exactly what is being asked , but given a bit more
direction, I can try and research "it".

Dan