Remove fme_text_size, fme_rotation and fme_text_string when writing dynamically in ORACLE

When you ‘open’ your writer in your workspace are those attributes listed?
Like this:
 

Sorry for the late reply. I’m using a FeatureWriter because I need to add another transformer to its ‘synthesis’ output port. I tried using a Writer just to see, but those unwanted attributes aren’t listed. 

I’m not sure whether this is due to the Geopackage format of the source data or the Oracle format of the destination database (using the dynamic method), which might be causing issues with the data being written correctly.

When you ‘open’ your writer in your workspace are those attributes listed?
Like this:
 

 

Hi ​@aliciad 

Are you able to see these attributes in the user artibutes or format attributes in the feature writer?

If possible, would you be able to share a workspace template with feature caching enabled with us?

No, I can only see them once the data has been imported into the Oracle database.
I finally managed to delete them using the ‘attributes to delete’ option in FeatureWriter (using the names defined by Oracle, i.e. HEIGHT, ROTATION and TEXTSTRING, rather than the format attribute name (fme_…)), but I still don’t understand why they were created automatically in Oracle.

In my external schema (CSV file), I do not define a geometry_type (as this does not work; I get an error stating that the geometry I entered is invalid, even though my geometry was ‘fme_point’). Could the fact that I haven’t defined this geometry and have asked FeatureWriter to set GEOM = all in the attribute parameters be the cause?

When I write the data by unchecking the dynamic point and therefore define the attributes in the FeatureWriter, these 3 attributes are not written. 

There are therefore several possibilities: 
- As the external schema does not define the geometry, Oracle may be creating these three columns in anticipation
- Could the dynamic use in a FeatureWriter instead of a standard writer be causing this? (I need a feature writer because I need an output port to add transformers after the writing).

Here are parameters for the featurewriter

I’ll try to send the dynamic workspace as soon as possible (probably not this week). The transformers I’ve used are simple (just a bulkattributerenamer, an attributetrimmer, a counter and a timestamper). 

In my external schema (CSV), I have fme_feature_type (table name), name (of the attributes), fme_data_type (for the attributes) and order (of the attributes in each table). 

In the reader (nothing out of the ordinary, everything was read via a single port and fme_feature_type is exposed)

Hi ​@aliciad 

Is it possible that the Geopackage you are reading from contains those attributes in the schema? For example, if the Geopackage file was created using FME, those attributes may have been included.

If you add a Geopackage reader in Workbench to read that source file you are using, do those unwanted attributes show up in the reader's user attributes list? 

Hello,

These attributes are not present when I use a reader. I don't have any information about how the geopackage was created.
 

Here is some additional information:

When I export the data to GDB using exactly the same parameters (except for the “spatial definition” in the attribute settings, which is set to “from the first feature” instead of “all”—a setting I cannot change in the Oracle FeatureWriter), the unwanted attributes are not written.