Validating CSV column names

[albamoralest]

I am querying a CSV and misspelt a column name. SPARQL anything will not throw a warning or error about it.

E.g
CSV columns' names: types, entity

Example of query:

SERVICE x-sparql-anything: {
BIND ( CONCAT ( "../path/", ?_fileId , ".csv" ) AS ?location ) .
fx:properties fx:location ?location ; fx:csv.headers "true" .
[] xyz:types ?typ ;
xyz:entty ?ent .
...

ERROR misspeld name
xyz:entty

SPARQL Anything does not throw any error
using the sparql-anything-0.8.1 version

[enridaga]

Normally, RDF graphs are schema-less, so the query engine does not make any assumption on existing, non-existing, or constrained properties, classes, or data types. However, this is not true for FX graphs (generated applying Facade-X to non-RDF resources). Maybe it is worth considering raising warnings if there are schema elements mentioned that do not exist in the original source.