Skip to main content

I'm trying to read annotation feature class from ArcGIS file geodatabase. For further processing, I need to check 'Split Complex Annotation' and 'Split Multi-part Annotation' in the reader.

ArcMapThe annotation in ArcMap looks like the above image. It's in vertical alignment: baseline, horizontal alignment: center.

While after importing from FME reader, it becomes to:

FmeAny safer partner could explain how FME calculate the anchor point of spitted text? seems the anchor point set to right bottom corner. And will the split consider the ArcGIS alignment setting?

Hi @ruochen​, looks like you're experiencing the default behaviour as mentioned in the documentation under Split Complex Annotation, "All the text will be bottom, left-justified without an X or Y offset."

I'm wondering if setting the geodb_h_align attribute (i.e. using an AttributeCreator) to center would do the trick for those specific annotation features.


@ruochen​ As @nampreetatsafe​ mentions this is the expected behaviour. Then FME splits complex annotations, it strives to preserve the location of the text. Other text placement characteristics are lost. This includes the original complex text geometry (which is actually a line) and that is converted to a point for each split text item. If you set the alignment to the original (center) then the text will shift.


@nampreetatsafe​  ​ @Mark Stoakes​  Thanks a lot for your reply. Setting geodb_h_align / geodb_v_align seems not work for me. These attributes are able to overwritten by AttributeManager, while have no effect on the geometry. By my guess, the ArcGIS geodb_xx attributes no longer link to geometry after importing from reader?

 

I'm still confusing about the mentioned "location of the text". By my test, a single line annotation feature with 'left' set in ArcMap could keep correct place after splitting.

The multiline annotation 'Melody Garden' (horizontal alignment: center, vertical alignment: top in ArcGIS) keeps it's location even though it's not set to 'left'. Would FME treat multiline text in a different way?

Multiline1multiline2


@markatsafe Is there a way to preserve the complex text geometry in the split text items? Or is there a way to write out the annotations to that they contain complex text geometry?

I’m running into an issue where my downstream program doesn’t properly read the split point text annotations.


Reply