[mapserver-dev] Call for a 5.6.1 release...

[Lime, Steve D (DNR)]

Tamas: Thanks for the patch, makes sense at first glance. Do you want to apply it or should I?

Steve

________________________________________
From: mapserver-dev-bounces at lists.osgeo.org [mapserver-dev-bounces at lists.osgeo.org] On Behalf Of Tamas Szekeres [szekerest at gmail.com]
Sent: Sunday, January 03, 2010 6:35 AM
To: Lime, Steve D (DNR)
Cc: mapserver-dev at lists.osgeo.org
Subject: Re: [mapserver-dev] Call for a 5.6.1 release...

Steve,

I've also added some fixes for 5.6, and I would also require further fixes with regards to the querymap drawing which would indeed validate a new release.

IMHO adding implicit layers to the map this way would bring in a couple of problems for the existing applications. It seems we make an assumption that the mapObj is destroyed after each drawQuery, however that's not the case with many of the mapscript applications. This approach would bring in extra unexpected layers for the users which are difficult to interpret.
There are some further issues with msCopyLayer which doesn't seem to work correctly when the destination have already been initialized (see the implementation of msCopyProjection, insertFeatureList, copying the classes etc). This would at least result in memory leaks and assertion failures.

As a quick solution I've created a patch (http://trac.osgeo.org/mapserver/attachment/ticket/3069/mapdraw.c.patch) to implement this kind of drawing without requiring to add the layer to the map.

Not strictly related to this problem, but in a further release we should also consider how we could add futher styles (like transparency or hatch patterns) to the highlighted features. The current approach provides only a couple of colors to be replaced temporarily with the highligted items and most of my current use cases require to copy the selected features to inline layers for a more advanced styling. For this reason it would be appropriate to define the query styles by using some 'template' layers (with no data connections) which could be referred from each queryable layers. I suspect this approach would work as far as msDrawShape(map, layer, &shape,...) would work if we specify such template layers as the second parameter.


Best regards,

Tamas



2010/1/2 Lime, Steve D (DNR) <Steve.Lime at state.mn.us<mailto:Steve.Lime at state.mn.us>>
Hi all: I just fixed a couple of inconvenient bugs in 5.6 that broke some point-based queries via the CGI. Without the fixes some
folks won't be able to upgrade. I've got a whole bunch of query test cases I need to get into the automated testing to avoid this
in the future. Are others sitting on changes in the 5.6 trunk that could go into 5.6.1?

Steve

BTW Ticket is http://trac.osgeo.org/mapserver/ticket/3246...
_______________________________________________
mapserver-dev mailing list
mapserver-dev at lists.osgeo.org<mailto:mapserver-dev at lists.osgeo.org>
http://lists.osgeo.org/mailman/listinfo/mapserver-dev