BUG: A missing space between an ID and the following operator (':', ':>') causes and error in Rosetta Editor
JWT007 opened this issue · 3 comments
Problem:
If a whitespace is omitted for example between an ci ID _ and the following ':' an error is raised in Rosetta. It is probably consuming the ':' as part of the ID.
The same thing occurs in a vocabulary between an aspect/concept and a 'super' (for example 'concept Cola:> Drink').
Note: Missing whitespace between the operator (':', ':>') and the following IRI does not cause an error.
Expected behavior:
I would expect the parser to ignore whitespace and be able to deal with this for example: "ci SquarePart1:ShapedPartInstance"
I debugged this issue, and it turns out that when the colon is attached to the id the tokenizer gets confused between two terminal rules: ID and ABBREV_IRI. The former is the right one to choose in this case, but it ends up choosing ABBREV_IRI, which is defined as ID:ID. Based on this, it thinks the token is ill-formed and it complains. I will look to see if there is a way to disambiguate the terminals definition.
I think I have a fix for this one. It will make it into the next patch release of the OML server, then will update OML Theia to use it.
This has been fixed in the latest OML release (v0.9.0)