Re: bcp_done always returns error if no data was inserted

Further investigation I found an example in the distribution of FreeTDS 
that shows the way to handle no records being bcp'd
Before calling bcp_done you determine if any rows were processed if none 
do not execute the bcp_done command
Forgot to mention that we are using SYBASE.and it does work with no 
records being processed..
I will look into version 0.92

Thanks


Example Found: $1$DGA1802:[FREETDS-0_91.SRC.APPS]DATACOPY.C

        if (rows_read) {
                gettimeofday(&batch_start, 0);

                ret = bcp_done(dbdest);

                gettimeofday(&batch_end, 0);
                elapsed_batch = elapsed_batch +
                        ((double) (batch_end.tv_sec - batch_start.tv_sec) 
+
                         ((double) (batch_end.tv_usec - 
batch_start.tv_usec) / 1000000.00)
                        );
                if (ret == -1) {
                        fprintf(stderr, "bcp_done failed.  \n");
                        return FALSE;
                } else {
                        rows_done += ret;
                }
        }



Angelo Coccimiglio




From:   "Craig A. Berry" <[email protected]>
To:     FreeTDS Development Group <[email protected]>, 
Date:   12/05/2014 04:17 AM
Subject:        Re: [freetds] bcp_done always returns error if no data was 
inserted
Sent by:        "FreeTDS" <[email protected]>




> On Dec 3, 2014, at 10:30 AM, Angelo Coccimiglio <[email protected]> 
wrote:
> 
> Hello,
> 
> I'm new to freetds... running version FREETDS-0_91

Try 0.92 and see if there's any difference.

> Whenever an attempt to bcp data within a C program and there are no 
> records
> the bcp_done routine returns -1 indicating an error... If there are 
> records
> available to bcp works fine..

I haven't been able to find any documentation indicating what bcp_done 
should do if there are no records.  There's quite likely an oversight for 
that case in the FreeTDS library but if there's any chance you could try 
your code with a Microsoft or Sybase library and confirm it works there 
that would help.  Or post a small reproducer.

> 
> Is there anything I should be looking for or setting up..
> Here is part of the log wher bcp_done is called
> 
> 
1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.TDS]UTIL.C;1:156:Changed 

> query state from PENDING to DEAD
> 
$1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.DBLIB]DBLIB.C;46:326:dblib_release_tds_ctx(1)
> 
$1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.DBLIB]DBLIB.C;46:6321:dbfreebuf(5DD490)
> 
$1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.DBLIB]DBLIB.C;46:4984:dberrhandle(0)
> 
$1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.DBLIB]DBLIB.C;46:4984:dberrhandle(0)
> 
$1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.DBLIB]DBLIB.C;46:5003:dbmsghandle(0)
> 
$1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.DBLIB]DBLIB.C;46:5003:dbmsghandle(180298)
> 
$1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.DBLIB]DBLIB.C;46:1689:dbresults(5B9F40)
> 
$1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.DBLIB]DBLIB.C;46:1695:dbresults: 

> dbresults_state is 3 (_DB_RES_NEXT_RESULT)
> 
$1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.TDS]TOKEN.C;1:540:tds_process_tokens(5BA7A0, 

> 7ACFBA58, 7ACFBA50, 0x6914)
> 
$1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.TDS]TOKEN.C;1:543:tds_process_tokens() 

> state is COMPLETED
> 
$1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.DBLIB]DBLIB.C;46:1716:dbresults() 

> tds_process_tokens returned 2 (TDS_NO_MORE_RE?
>                        result_type TDS_DONE_RESULT
> 
$1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.DBLIB]DBLIB.C;46:1678:dbresults 

> returning 2 (NO_MORE_RESULTS)
> 
$1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.DBLIB]BCP.C;3:2251:bcp_done(5C5E00)
> 
$1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.TDS]BULK.C;1:886:tds_bcp_done(5C6620, 

> 7ACFBA50)
> 
$1$DGA1808:[TSG.ACOCCIMIGLIO.VMS_1.FREETDS-0_91.SRC.TDS]NET.C;1:741:Sending 

> packet
> 0000 0f 01 00 08 00 00 00 00-                        |........|
> 
> 
> thanks
> 
> Angelo Coccimiglio
> _______________________________________________
> FreeTDS mailing list
> [email protected]
> http://lists.ibiblio.org/mailman/listinfo/freetds

________________________________________
Craig A. Berry
mailto:[email protected]

"... getting out of a sonnet is much more
 difficult than getting in."
                 Brad Leithauser

_______________________________________________
FreeTDS mailing list
[email protected]
http://lists.ibiblio.org/mailman/listinfo/freetds