2009-12-04 18:40 UTC+0200 Mindaugas Kavaliauskas (dbtopas/at/dbtopas.lt)
* harbour/contrib/rddads/ads1.c
! added comment and workaround to invalid AdsGotoRecord( hTable, 0 )
behavior
This commit is contained in:
@@ -17,6 +17,11 @@
|
||||
past entries belonging to author(s): Viktor Szakats.
|
||||
*/
|
||||
|
||||
2009-12-04 18:40 UTC+0200 Mindaugas Kavaliauskas (dbtopas/at/dbtopas.lt)
|
||||
* harbour/contrib/rddads/ads1.c
|
||||
! added comment and workaround to invalid AdsGotoRecord( hTable, 0 )
|
||||
behavior
|
||||
|
||||
2009-12-04 15:55 UTC+0200 Mindaugas Kavaliauskas (dbtopas/at/dbtopas.lt)
|
||||
* harbour/src/rdd/dbf1.c
|
||||
! fixed variable length field setting after DBCREATE()
|
||||
|
||||
@@ -881,7 +881,15 @@ static HB_ERRCODE adsGoTo( ADSAREAP pArea, ULONG ulRecNo )
|
||||
ulRecNo = 0;
|
||||
u32RetVal = AdsGotoRecord( pArea->hTable, 0 );
|
||||
}
|
||||
if( u32RetVal != AE_SUCCESS )
|
||||
|
||||
/* Usually AdsGotoRecord( *, 0 ) is valid call and returns no error,
|
||||
* but if previous error was AE_INVALID_RECORD_NUMBER, the following
|
||||
* AdsGotoRecord( *, 0 ) returns the same error. I'm not sure if this
|
||||
* AdsGotoRecord( *, 0 ) will position to phantom record and return
|
||||
* empty fields. I hope I have not made any regressions in code, since
|
||||
* it replicates previous behaviour. [Mindaugas]
|
||||
*/
|
||||
if( u32RetVal != AE_SUCCESS && u32RetVal != AE_INVALID_RECORD_NUMBER )
|
||||
{
|
||||
commonError( pArea, EG_CORRUPTION, ( HB_ERRCODE ) u32RetVal, 0, NULL, EF_CANDEFAULT, NULL );
|
||||
return HB_FAILURE;
|
||||
|
||||
Reference in New Issue
Block a user