Known Issue: Error when user attribute contains fme_ or _ prefix

Symptom

When I have user attributes that contain the prefix fme_ or <format>_ the Reader/Writer/Transformer does not return the expected output. In some cases, the attributes may not be present, or they are not processed.

Cause

Format attributes are specific to a formats schema, and these are denoted by a prefix with that format type. There are also specific format attributes perceived by FME with a similar fme_ prefix. FME treats these attributes differently from user attributes as they have a set purpose with their values defined by processes in FME.

Therefore when user attributes are assigned the same naming convention, this can cause certain transformers to misinterpret what they are and exclude them from processing.

For readers, in some cases, we know exactly which attributes are of which type (user, format, fme_). This is common when we have well known or stored schema formats. However, in schema-less formats (e.g. Excel, CSV, etc.) FME will scan the data features to determine schema and may make assumptions about which attributes fall into which category. This can result in attributes not being listed on the feature type once the reader has been added to the canvas.

Known transformers to be affected:

• ListIndexer

• BulkAttributeRemover

• SchemaMapper

• Excel Reader

Note: If you encounter any issues with reserved words or transformers, please comment below so this article can be kept up to date.

Resolution

The fme_ / <format>_ prefix should be reserved for internal use by FME only and should not be used by the author when naming user attributes, as this could potentially have an adverse impact on the translation. Users should also err on the side of caution using the <format>_ prefix for the same reason.

For more information on Format attributes, see the Documentation.

keyword(s): reserved word restricted word