I have a problem with the TMS dbgrid. A standard DBGrid is fine and I just want to check that the type of behaviour below doesn't happen with a DevEx grid before I start really shouting.

All responses welcomed.

> I've just been trying a few more tests with a DBISAM table.
> The base table has c150k records. If I set a filter on the
> table resulting in 51 records and then scroll down by holding
> the down arrow cursor key when the display reaches the last
> visible row it doesn't scroll the display, only the last row.
> Similarly scrolling upwards with the cursor key the selected
> row reaches the top of the display and doesn't scroll simply
> refreshes the row.
>
>
> I then tried using a canned query so that there would only be
> 51 records. This time the display scrolls BUT data is not
> displayed at random. Many of the cells simply contain one or
> two dots. The same is the case when trying to scroll through
> the full 150k records.
>
> The above only applies when PageMode := True. Setting it to
> False means that display is fine. Unfortunately in most cases
> that would not be an option.
>
> The only event set for the grid was the OnGetRecordCount and
> all of the properties were as default. I tried altering
> Datasettype and datasettypeauto but with no effect.
>
> Using Home/End and display was perfect, clicking on the
> scroll bar and that's OK, grabbing and dragging the scroll
> button and that's OK
>
> I've just dropped a DBNavigator on the form, linked it up and
> tried that. First/Last OK. Next/Prior disaster, even with a
> slow click of the button.
>
> More experimenting. It seems that the column the cursor is in
> refreshes OK, the others don't. Also with the screwed up
> display all I have to do is click on the scroll button, not
> move it just click it, and the display is refreshed correctly.
>
> Final test - set goRowSelect := True and everything refreshes
> properly.

Roy Lambert

Hi Roy,

> All responses welcomed.

I'm not sure if I'm following what you're after here, so I'll have a stab in
the dark. Ignore my response if it's totally irrelivant.

I can't say that I've experienced any of the problems you've mentioned
before with Quantumgrid. However - the Devex Quantumgrid is a much more
complicated Grid than your standard dbgrid. (Most of the properties are
different also, as well as the way it interacts). Don't expect that you can
change the Class of a Grid from a TDBGrid to a TcxDBGrid like you can with
other grids (such as woll2woll, rxDbGrid, jvdbGrid, etc). In fact, I don't
even think it has the same heirachy as other Grids off the top of my head.

From what I can gather, quantumgrid has it's own internal memory tables that
it loads all the records into first - before displaying to the user. This
gives the grid many extra features (such as the ability to sort, filter,
etc) without needing to modify the original dataset.

Therefore the problems you've suggested could possibly be eliminated with a
devex cxDBGrid anyway due to the way it interacts with the datasource as
opposed to other grids.

I can't comment on the TMS DBGrid, as I've never had any experience with it
before. It may not hurt to compare what you're describing with (what I would
guess would be) a similar kind of grid such as woll2woll's wwDBGrid, or the
jvDBGrid (part of the Jedi VCL Component library) for comparison.

Not sure if I'm on the same wavelength here or not, so ignore this if it's
just adding more confusion.

Cheers

Adam.

Adam


Thanks for the reply. TMS have done some checking, realised that there is a problem and sent me a fix. Apparently its something that crept in during the latest release.

Roy Lambert

Time checking time stamp...............12:46 on here at 10:51
uk time ?????????/


"Roy Lambert" <roy.lambert@skynet.co.uk> wrote in message
news:B4EA65A6-527C-4B33-87A2-53791162E107@news.elevatesoft.com...
> Adam
>
>
> Thanks for the reply. TMS have done some checking, realised that there is
> a problem and sent me a fix. Apparently its something that crept in during
> the latest release.
>
> Roy Lambert
>

Hi Roy,

> Thanks for the reply. TMS have done some checking, realised that there is
> a problem and sent me a fix. Apparently its something that crept in during
> the latest release.

Oh yes - those creeping problems can be the worst.  

Glad to hear it's all sorted out. BTW - I think xp is right - their's
another creepy here - timestamp for the last couple of posts have gone
coo-if.  (Either that, or their's a tare in the space/time continuum

Adam


OK lets see posted here 12:22 BST

Roy Lambert

Are you suggesting we may have gone through a worm hole
or something ?
If we can get the whole thing in reverse we could enter every
known lottery on the planet and make a killing.


"Adam H." <ahairsub4@rREMOVEMEspamSTOPPER.jvxp.com> wrote in message
news:191FEC84-3655-464B-B99C-37DC62C02722@news.elevatesoft.com...
> Hi Roy,
>
>> Thanks for the reply. TMS have done some checking, realised that there is
>> a problem and sent me a fix. Apparently its something that crept in
>> during the latest release.
>
> Oh yes - those creeping problems can be the worst.  
>
> Glad to hear it's all sorted out. BTW - I think xp is right - their's
> another creepy here - timestamp for the last couple of posts have gone
> coo-if.  (Either that, or their's a tare in the space/time continuum
>

xp


Warped but brilliant thinking <vbg>


Roy Lambert