Skip to main content
Hi everyone,

 

 

If I'm using the Dynamic Schema to write out, can I create a sub-folder based on input formats or elsewhere?  I've got a workbench I want to publish to FME server, it's quite basic. It takes shp files, runs them through a GeometryFilter and then adds appropritate statistics, i.e. for points it add the coordinates, for polygons it adds areas, etc.

 

 

They write out to three different destination folders using dynamic schema:

 

 

\\data\\Tasks\\Geometry_Statistics\\Writer\\Points

 

\\data\\Tasks\\Geometry_Statistics\\Writer\\Polylines

 

\\data\\Tasks\\Geometry_Statistics\\Writer\\Polygons

 

 

These all appear as published parameters in FME server, what I'd like to do is have it so users can just specify:

 

 

\\data\\Tasks\\Geometry_Statistics\\Writer\\

 

 

with FME doing the rest, i.e. dynamically creating the sub-folders.  I've worked out you can link the destination to a parameter, but I can't see how you can make that parameter kind of dynamic, i.e. choose between geometry and therefore folder names.  I don't know how to pass this information onto the writer.

 

 

The other option is that I use the fanout but I'm not entirely sure how to do it that way either!  I can concatenate a folder address, but not sure how to add this to the destination direction parameter (if at all possible).

 

 

Thanks.
Hi,

 

 

I believe the easiest is to use a fanout dataset:

 

 

 

 

You can either expose and use "fme_geometry" from the reader directly or use an AttributeValueMapper to transform the value into something more suitable.

 

 

David
I didn't know you could fanout and use the dynamic schema at the same time, you learn something new every day!  Thanks (again) David.

Reply