ArcPro 2.9.3 Unable to Detect Geometry Field

592
10
07-20-2022 11:59 AM
Labels (1)
by Anonymous User
Not applicable

Encountered a problem exporting features from an Oracle SDE geodatabase feature class. The feature class contained an erroneous SHAPE field, field type long integer, entirely NULL values. The feature class geometry was stored in SHAPE_1. When selecting and exporting features to a file geodatabase feature class, ArcGIS Pro 2.9.3 appeared to pull the NULL values from SHAPE (long) and returned a table with NULL geometries.

ArcMap 10.8.2 is able to recognize the correct geometry field and the same extraction was completed successfully.

Obviously, an erroneous SHAPE field is bad data management, but this appears to be a bug.

Tags (2)
0 Kudos
10 Replies
DanLee
by Esri Regular Contributor
Esri Regular Contributor

Not sure what tool you used to export features.  There is no reason for this to happen. If you think it's a bug, the best is to contact Esri Tech Support and log a BUG.  🙂

https://support.esri.com/en/

 

0 Kudos
by Anonymous User
Not applicable

Feature Class to Feature Class

0 Kudos
DanLee
by Esri Regular Contributor
Esri Regular Contributor

I tested Feature Class To Feature Class with an input that has a different name for the geometry (Shape) field.  The tool worked.  It's better you log a BUG with tech support so our team can take a closer look with your data.  Thanks.

0 Kudos
by Anonymous User
Not applicable

I've logged a bug, thanks Dan.

Did your test feature class also include a SHAPE field that is not a geometry and exclusively NULL values? If not, the test only covers half of the conditions I encountered.

0 Kudos
DanLee
by Esri Regular Contributor
Esri Regular Contributor

Great. Do you have the BUG number?  Thanks!

I was unable to add a SHAPE field because it's a reserved field name by the system. I am not sure how your data got such field.

DanLee_0-1658442612841.png

 

0 Kudos
by Anonymous User
Not applicable

Just received case #03109753.

We haven't got to the bottom of how that happened either...

0 Kudos
DanLee
by Esri Regular Contributor
Esri Regular Contributor

It seems that your issue has been followed up by Esri Support analyst. Hope that is helpful.

Dan

0 Kudos
by Anonymous User
Not applicable
It was suggested the geometry was corrupted in the original database, which does not appear to be the case. Esri Support was not able to reproduce my issue and closed the ticket. I understand this is an unusual case and we have not determined how the superfluous SHAPE field was generated to begin with.
0 Kudos
DanLee
by Esri Regular Contributor
Esri Regular Contributor

Yeah, it's hard to investigate without a repro case.  Hope you will be able to determine how the problematic SHAPE field was generated in early processing.

0 Kudos