I have a dynamic workbench that gets passed in an encoded SQL statement to read data from an Oracle database and then pass that data to a local ArcGIS Geodatabase.
The problem I am having is that the data types in the schema are not being maintained when passed through to the ArcGIS Database.
The data coming in has all the correct schema i.e. dates, numeric, text, etc fields and nothing too complex of large.
The writer is using a dynamic schema setting as no single SQL query has the same data types.
There are a set of fields that need to be set to a specific data type and fortunately these are common across all datasets but these are also the fields that are not holding their data type.
For example I have a field called Owner1 and it needs to be a number fields as when populated it stores an ID that can be used to relate to another table but because this can sometimes be null ArcGIS treats it as a text field and as such the relate fails in ArcGIS.
In the simplist form it is just a data in => data out workbench.
I have tried many iterations for example
- reading just data and then feeding it to a schema scanner or
- reading data and schema
- and then exploding the schema list, manipulating the data type (fme_data_type), build the list and then passing the schema to the writer but the same issue is persistent.
The errors seems to be around the geometry being incorrect in the schema branch but it shouldn’t have any geometry as it is the schema.
WriteMergedLayers (WriterFactory): MULTI_WRITER: Geometry type is required for FILEGDB feature type '<Generic>' (WriteMergedLayers_0), but no geometry was specified in the schema feature using attributes named 'fme_geometry{n}'
I am using FME 2025.1
Is there a trick to get Oracle Data Types to map correctly to ArcGIS Data Types?
Has anyone had a similar issue or some suggestions.
Any help would be very much appreciated