[lazarus] TDBGrid question

[Jeroen van Iddekinge]

Hi,

If I simulate what ShiftBuffersUp and ShiftBufferDown do, I think
thats wrong

First what shiftbuffersup does:

For example
Buffer: 0123456 (numbers ar original content of buffer before call to shift buffersup )
procedure is called with the following:
Offset=2
Distance=3
Buffercount=7:
FirstMove
move1:
 "Move(FBuffers[Offset],Temp^,MoveSize);"
 Temp=234 (temp containts the old buffer 2,3 and 4)
 buffer=unchanged      

move2: 
     Move(FBuffers[Offset+Distance],FBuffers[Offset],
     (FBufferCount-Distance-Offset)*SizeOf(Pchar));
               =5                        =2          (2)*sizeof()
      Temp=unchanged
       buffer=01563456
move3:
   Move(Temp^,FBuffers[FBufferCount-Distance-Offset],MoveSize);
                         2                             (3)*sizeof
   temp=unchanged
   buffer=0123456 <==final. 

If i look at the result i find it verry strange...

So far as I understand what procedure  ShiftBuffersDown does, it rotate all the buffers. Thats compleetly different what ShiftBuffersUp does. For 
rotation you don't need an offset parameter, so why does ShiftBuffers have a 'offset' parameter?


Greez,

   jeroen van Iddekinge

--------- Original Message ---------

DATE: Mon, 19 May 2003 10:55:53
From: Michael Van Canneyt <michael.vancanneyt at wisa.be>
To: lazarus at miraclec.com
Cc: 

>
>
>On Mon, 19 May 2003 hernan.et at pg.com wrote:
>
>>
>>
>> Hi Jesus,
>>
>> I have tried using TDBGrid and it work quite well. But I noticed that I cannot
>> scroll backward when my table contains more than 10, (my table contains 56
>> rows). I have tried to track it down and it seems that the problem is somewhere
>> in TDataset or its parent(s). Can someone have a look  on this?
>
>I've had this issue in the past, and I suspect has something to do with the buffer
>scrolling. I have not yet been able to pinpoint the exact error, however.
>
>The procedure you want is in dataset.inc, line 1634
>
>Procedure TDataset.ShiftBuffers (Offset, Distance : longint);
>
>Subprocedure 'shiftbuffersdown'.
>
>I currently do not have enough time to look at it, but if someone else
>has, the above info may help to pinpoint the exact problem.
>
>Michael.
>
>_________________________________________________________________
>     To unsubscribe: mail lazarus-request at miraclec.com with
>                "unsubscribe" as the Subject
>   archives at http://www.lazarus.freepascal.org/mailarchives
>