Thanks all for the comments.

For clarification the speed of the grid is fine for long datasets - lazy 
replication works well. The problem comes with large numbers of columns. To 
address that I'm planning on creating a grid which effectively "lazy 
replicates" the columns. Most of the time only a few columns are shown so there 
is no need to actually create the columns that cannot be seen (I've had some 
grids with 160+ columns and the current grid definitely struggles!).

To do the above probably requires thinking in terms of "gridColumn" components 
rather than "gridRows". This would also allow some neat functionality such as 
dragging and dropping columns for reordering. Given this, anything we do will 
be pretty orthogonal to the existing components.

Having said all that, before embarking on the development project, I wanted to 
be sure that there were no other third party components that can now be hooked 
in. If anyone has heard of something that would be great.

Cheers,

Andy

From: [email protected] 
[mailto:[email protected]] On Behalf Of Sebastian Wagner
Sent: 25 November 2009 15:54
To: Johannes Boesl
Cc: [email protected]
Subject: Re: [Laszlo-user] Looking for a faster grid

yes I have the same approach. You just re-use the views in a kind of 
view-pooling and thats it.
It is just not that easy to make a geneal component. The way the result-set of 
your datasource is kind of strict to the way the table needs it. And you cannot 
re-size the hole grid/table cause you would need to realculate the visible area 
and the number of views in the pool.

sebastian
2009/11/25 Johannes Boesl <[email protected]<mailto:[email protected]>>

My app used a custom table, too. My table implementation used text-views for 
each cell and reused those views on scrolling. So when you scrolled 10 lines 
down the top ten lines where removed and appended in the end. It worked okay 
but a rendering approach might be better I could imagine. I think this is the 
same method the current table uses when the data is lazily replicated.



--
Sebastian Wagner
http://www.webbase-design.de
http://openmeetings.googlecode.com
http://www.laszlo-forum.de
[email protected]<mailto:[email protected]>

No virus found in this incoming message.
Checked by AVG - www.avg.com
Version: 8.5.426 / Virus Database: 270.14.82/2525 - Release Date: 11/25/09 
07:31:00

Reply via email to