The reason for this change is that symbolSizes and maxSymbolSize
on the instance will be wrong as soon as the resolution changes
and cached tiles are used. It turned out that the approach used
now has several advantages: smaller symbolSizes objects, no need
to merge symbolSizes objects, and cache management for free (no
risk of memory leaks). Note that the symbolSizes and
maxSymbolSize for each tile are not strictly tile specific -
they represent the rendering pass that created the tile. This
has no negative side effects, and it has the advantage that
there is not a single additional loop needed to create these
structures.
With this change, hit detection for lines and points gets very
accurate, because the vector renderer instance keeps track of
line widths and point symbol sizes. After doing a bbox query in
the RTree, returned lines and points are evaluated against the
thresholds of their line width or symbol size. The KML example
with its different symbolizers now has getFeatureInfo too to
show this in action.
This adds a parser (read/write) for GML v2 and v3. GML v3 is limited to the
simple features profile of GML 3.1.1, just like OpenLayers 2 was. This will
be the basis for the WFS parser, but it only makes sense to continue this work
once feature modification (insert, update, delete) is in place in ol3. So the
WFS parser will be another pull request.
Without this we get the following warning from the compiler:
JSC_EXPORTED_FUNCTION_UNKNOWN_RETURN_TYPE. Unable to determine return type for exported function ol.control.Control.prototype.handleMapPostrender at ../src/ol/control/control.js line 79 : 51
I'm not sure why explicitely specifying the return type is required here.
If we use ol.control.Control.prototype.handleMapPostrender = goog.nullFunction the API doc doesn't show the mapEvent parameter of the handleMapPostRender function.
Now that we correctly export the ol.animation.*, ol.easing.* and ol.coordinate.* symbols we can make the generate-exports.py script work for more cases.
This .exports file made goog.require be exported as a null function. This was needed to be able to run the ol3 examples uncompiled against the ol.js build. Now that host-examples target removes the goog.require statements from the examples' js files (74b8fea6) we don't need to export goog.require anymore.
This also involves some modifications to the xmleql functionality now that
it was tested with a real-life case (KML). Also, some of the input KML files
needed to be changed since we currently cannot roundtrip 100% of the input.
