Skip to main content
I have encounter a weird problem. I use point dataset in FFS format and region dataset in MapInfo TAB format inside a pointOnAreaOverlayer.

 

 

I have noticed that some (not all) points receive the Format Attribute named mapinfo_type from the region dataset even if this attribute is not exposed in the Reader.

 

 

Then, if I want to write the points in a MapInfo TAB writer, the writer is confuse because it consider those points as area and cannot write it down because there is only 1 vertex in the feature.

 

 

Is it normal (I dont think so) that the pointOnAreaOverlayer pass the Format Attributes of the features to the other? It's even more weird that only some points receive this format attribute.

 

 

Can someone help me with this issue?

 

 
Hi Butron,

 

Have you tried prefixing the attributes before the POAO? or removing the format attributes?
Hi Itay, you are right. This is how to prevent the problem. I still wonder why the format attributes are passed over in the POAO even if they are not exposed, why only some features receive those format attributes, and why I encounter this problem only since I have installed FME 2014 SP4.
Hi Burton,

 

These are answers only Safe can give, but if i am correct the format attributes are part of FME's datamodel and therefore always available, even when not exposed.
I have an uncanny feeling that spatialrelating is somehow somwhat bugged in fme2014.

 

 

Yes availabe, we understand that. But why incorporated even when not exposed? And why sometimes and not always? or always not?

 

Seem bugged to me.

 

 

I too encouter previously not seen problems since we switched to 2014 (only since a couple months now).

 

 

Like spatial relator: i expose fme_feature_type on the supplier and not requesteor...but somehow it insist on merging the non exposed requestor fme_featuretype whilst ignored the one i want..
I dont know if this would make sense to someone, but I tryed to put a geometryValidator on the areas before the PoAo and the problem stopped. I thought the problem could be in an invalid geometry but curiously all areas was in the valid output of the geometryValidator (no invalid or repaired geometries).

Reply