> Hi Jody,
>
> I spent a little while on this tonight and made quite a bit of
> progress. I haven't committed anything so far because its not yet
> buildable.
I completely understand; I have actually obtained aaime like patience with
starting and stopping work in a non buildable state. After "mastering" local
branches with git. It is how he was able to keep from going insane on all the
MapLayer work. I know you have not made the transition yet ... but I can
confirm that it is useful.
> Another IRC chat would be good tomorrow or whenever (I'm having an
> early mark tonight). Meanwhile, here are some questions to prime the
> chat:
>
Tomorrow is a good time for an IRC chat; I actually have some work using
gt-swing for a couple of days and may have some useful feedback for you.
> 1. I'm using the Layer user data map to hold "selected" status but I
> think this would be better handled by just having a boolean field as
> is done per "visible". Having to check if the key exists prior to
> testing the flag is a bit cumbersome.
I am not sure I quite understand; if I remember JListModel we should be
recording the layer as selected in a separate data structure? I don't mind if
we need to add something; I would like to arrange to have the JMapPane use the
usermap for something as an example for those coding after you ... it would be
good to have an example showing that they do not need to fork JMapPane and
supporting data structures and can instead use the UserData map.
> 2. I'm not sure whether the aspect ratio calculations are best kept in
> the map pane or moved into the viewport. A related question is whether
> the viewport's screen area is intended to include slack space or not ?
I understand that the rendering code uses a special visitor to figure out the
current symbol size; and performs its own "slack space" around the target
image.... I can be wrong on this front (indeed I may be as a patch on this
topic was recently applied to udig ....).
My own feeling on aspect ratio calculations is that they are very easy to get
"out of sync" as such I would like them in the viewport ... so we can do them
in one spot for reuse in a number of contexts.
> 3. Some functions, such as tracking the mouse cursor world position,
> presently rely on getting the AffineTransform from the map pane. I'd
> like to talk about the best way to proceed with these, whether the
> viewport should cache the transforms etc.
I would like to move the AffineTransform into the viewport; and make sure the
mouse tracking tools have access to the viewport.
In a very real sense the viewport represents our data model for "the screen";
as such it makes sense the mouse tracking tools need to have access to the data
structure.
Cheers,
Jody
------------------------------------------------------------------------------
All of the data generated in your IT infrastructure is seriously valuable.
Why? It contains a definitive record of application performance, security
threats, fraudulent activity, and more. Splunk takes this data and makes
sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-d2d-c2
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel
