fme-form
I am reading some feature types in FME, doing some processing and writing as ESRI shape file format. Is there a way I could delete all the attributes from a table that are not necessery?? I need my shape files as small as possible (file size). I know python caller can probably do something like this but maybe there is an easier way.
This message:
During translation, some features were read that did not match a reader feature type in the workspace. This can happen if the reader dataset is changed, or a reader feature type removed or renamed.
...should have nothing to do with how the data is written, just what is read. This FME Evangelist post should explain that part of the process.
With the attributes, one possible problem is that you aren't over-writing the original Geodatabase. ie once you run the workspace w/o creating the attributes, then future translations will never add them unless you overwrite the Geodatabase or at least drop the table. FME can create a table, but not modify one.
So, check in the Navigator window for a parameter to delete the geodatabase first, or open the Feature Type (table) properties dialog and click the Parameters tab for a parameter to drop the table first. Hopefully this will solve the problem.
I don't think the NeighborFinder will work with the raw data because it treats the point cloud as an entire feature, so it only locates the centre point regardless. It seems that you will have to convert the point cloud geometry to something else.
As @alexgleith said, you could convert it to a surface and use draping. I would suggest using the PointCloudCoercer transformer to convert it to a set of point (or multi-point) features. Use a CoordinateExtractor to fetch the Z value for each and a 2DForcer to flatten the data.
Then you could use the NeighborFinder to get what you need.
Amendment: Having thought about it, there is a quicker method.
Create a buffer around your point feature using the Bufferer transformer. Set it to 100m (for example) in the same way you would for the NeighborFinder. Now use this buffer to clip the Point Cloud with the Clipper transformer.
What this will do is drastically reduce the amount of point cloud you are working with. From potentially billions of points you could be down to just a handful. Then using the above PointCloudCoercer/CoordinateExtractor/NeighborFinder method on the output from the Clipper should produce a result way, way, way, quicker.
I created an example...
...and placed the fmw file at: https://dl.dropboxusercontent.com/u/4200566/Community/PointCloudNearestNeighbor.fmw
Best Answer
So what's happening here is that you are using what we call "Dynamic" mode. It means that FME automatically outputs exactly the same data that is input (in terms of attributes). The idea is so that you can create a workspace that converts any source dataset, regardless of schema (data model).
Using the AttributeRemover won't help because the output schema is already being set by the input. The reason you get no output at all is because one attribute provides the name of the feature type to write and you're probably removing that attribute. The log would probably say something like "x features didn't match a known feature type and were discarded".
So... do you need to have a dynamic solution? If the source data is always the same in terms of layers and attributes, then the answer is no. What you should do is recreate the workspace without dynamic mode turned on. Then you can manually edit the output feature types to have only the attributes you need.
But... if you do need a dynamic solution then it becomes more complicated. That's because FME needs to know what schema to write, but doesn't know where to get that information, which could vary from run to run anyway. You need to tell FME what to write - for example, you could set up a text file with a list of attributes and use that.
I won't go into the dynamic side too much, because I'm hoping that you don't necessarily need to use it. But if you do, then I can point you to some useful resources that will help.
Oh... please don't resort to the PythonCaller! There should be a non-programmatic way to handle this scenario!