>>Based on different scenarios, different approaches possible to show something is slow or fast. I say, regarding to original title, if it's desired to hilite current row then dynamiccolors are sufficient and flexible enough both in respect to visual fx and speed. Secondly, regarding the other reply to Arnon, never be so sure that something can't be done. Details are in a thread months ago but as an overview clue a parallel related sql ordered table would supply the way for that white-pale green. Now I ask, how would you handle not only colors but fonts, currentcontrol, inputmask etc dynamically ? Dynamic*color is not only accessory properties of grid. Definitely David Frankenbach's shape approach is an state-of-art approach. I appreciate it but here dynamic*colors are sufficient IMHO.
>>Cetin
>--- reply ---
>Cetin,
>
>Exactly what are we arguing about? If you say that manipulating dynamic properties to display background colors is a faster method than the approach David F. developed, I disagree. Expecially in large grids, his approach is noticeably faster. If you are saying that manipulating background colors using dynamicbackcolor is fast enough and flexible enough for most situations, I completely agree -- it is.
So we agree.
>
>My statement that one cannot display alternating green/white bars if the unless the record source for the grid has alternating odd and even records is true. Of course you can fix this by arranging the record source table in odd and even records -- but unless you do this, you cannot display alternating color rows in the grid. Since I don't want to do this, that ends the discussion for me.
You don't touch the original table, just sql, ordered by your index to a cursor and relate. Then get recnos from cursor.
>
>Any decent, experienced programmer can make anything work given enough time to work with it. VFP's grids are deficient when it comes to direct control over their display. Other grid controls have .currentrow and .currentcolumn properties which make manipulating row and column display properties very easy. But this is a minor, minor convern. The dynamic properties approach elected by M$ is a very good approach to solving a lot of display problems, and I like it very much. But it does not solve every conceivable problem. And in the ideal world we would have direct display control as well as dynamic properties. Add that to my wish list!!
Sure many wishlist about grids. I really prefer the browse valid still and mimic it (original browse concept was more powerfull than a grid).
>
>Anyway, I see no reason to continue to discuss this. You are right, dynamic properties are fast enough in most situations. And I'm right, David's approach is faster at displaying background colors than dynamic property manipulation in all of the uses I have made of if. So now we are both right. It was not my intention to suggest that dynamic properties be avoided, but merely to point Colin Magee to a different approach in response to his original question.
>
>Regards,
>
>JME
Yes we all agree. Only for that particular situation I found dynamic* enough and it was only s-l-o-w that was annoying for me.
Thanks
Cetin
