Merge pull request #1186 from ontodev/issue1127-invalidxref

Fixing the invalid xref check
ontodev · May 1, 2024 · e76c16e · e76c16e
2 parents c1ad167 + 6fda119
commit e76c16e
Show file tree

Hide file tree

Showing 5 changed files with 7 additions and 4 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -19,6 +19,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - [`merge`] and 'annotate' operations '--annotate-defined-by' excludes reserved OWL 2 vocabularies [#1171]
 - Handle IRIs that are not entities in export [#1168]
 - Fix integration tests [#1181]
+- Invalid Xrefs test has been fixed to recognise invalid CURIEs correctly [#1127]
 
 ## [1.9.5] - 2023-09-20
 

diff --git a/docs/examples/report.tsv b/docs/examples/report.tsv
@@ -2,6 +2,8 @@ Level	Rule Name	Subject	Property	Value
 ERROR	missing_ontology_description	http://purl.obolibrary.org/obo/uberon.owl	dc:description	
 ERROR	missing_ontology_license	http://purl.obolibrary.org/obo/uberon.owl	dc:license	
 ERROR	missing_ontology_title	http://purl.obolibrary.org/obo/uberon.owl	dc:title	
+WARN	invalid_xref	BFO:0000051	oboInOwl:hasDbXref	BFO-:0000051
+WARN	invalid_xref	BFO:0000062	oboInOwl:hasDbXref	:0000062
 WARN	missing_definition	BFO:0000050	IAO:0000115	
 WARN	missing_definition	BFO:0000051	IAO:0000115	
 WARN	missing_definition	BFO:0000063	IAO:0000115	

diff --git a/docs/examples/uberon_fragment.owl b/docs/examples/uberon_fragment.owl
@@ -262,7 +262,7 @@
     <owl:ObjectProperty rdf:about="http://purl.obolibrary.org/obo/BFO_0000051">
         <rdf:type rdf:resource="http://www.w3.org/2002/07/owl#TransitiveProperty"/>
         <rdfs:label rdf:datatype="http://www.w3.org/2001/XMLSchema#string">has_part</rdfs:label>
-        <oboInOwl:hasDbXref rdf:datatype="http://www.w3.org/2001/XMLSchema#string">BFO:0000051</oboInOwl:hasDbXref>
+        <oboInOwl:hasDbXref rdf:datatype="http://www.w3.org/2001/XMLSchema#string">BFO-:0000051</oboInOwl:hasDbXref>
         <oboInOwl:id rdf:datatype="http://www.w3.org/2001/XMLSchema#string">has_part</oboInOwl:id>
         <oboInOwl:shorthand rdf:datatype="http://www.w3.org/2001/XMLSchema#string">has_part</oboInOwl:shorthand>
         <oboInOwl:hasOBONamespace rdf:datatype="http://www.w3.org/2001/XMLSchema#string">uberon</oboInOwl:hasOBONamespace>
@@ -276,7 +276,7 @@
     <owl:ObjectProperty rdf:about="http://purl.obolibrary.org/obo/BFO_0000062">
         <rdf:type rdf:resource="http://www.w3.org/2002/07/owl#TransitiveProperty"/>
         <rdfs:label rdf:datatype="http://www.w3.org/2001/XMLSchema#string">preceded_by</rdfs:label>
-        <oboInOwl:hasDbXref rdf:datatype="http://www.w3.org/2001/XMLSchema#string">BFO:0000062</oboInOwl:hasDbXref>
+        <oboInOwl:hasDbXref rdf:datatype="http://www.w3.org/2001/XMLSchema#string">:0000062</oboInOwl:hasDbXref>
         <obo:IAO_0000115 rdf:datatype="http://www.w3.org/2001/XMLSchema#string">X preceded_by Y iff: end(Y) before_or_simultaneous_with start(X)</obo:IAO_0000115>
         <oboInOwl:hasExactSynonym rdf:datatype="http://www.w3.org/2001/XMLSchema#string">is preceded by</oboInOwl:hasExactSynonym>
         <oboInOwl:shorthand rdf:datatype="http://www.w3.org/2001/XMLSchema#string">preceded_by</oboInOwl:shorthand>

diff --git a/docs/report_queries/invalid_xref.md b/docs/report_queries/invalid_xref.md
@@ -11,7 +11,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 SELECT DISTINCT ?entity ?property ?value WHERE {
   VALUES ?property {oboInOwl:hasDbXref}
   ?entity ?property ?value .
-  FILTER (!regex(?value, "^[a-z|A-Z|_|\\-|0-9]*:(?!$)\\S*$"))
+  FILTER (!regex(?value, "^[A-Za-z_][A-Za-z0-9_.-]*[A-Za-z0-9_]:[^\\s]+$"))
   FILTER (!isBlank(?entity))
 }
 ORDER BY ?entity

diff --git a/robot-core/src/main/resources/report_queries/invalid_xref.rq b/robot-core/src/main/resources/report_queries/invalid_xref.rq
@@ -10,7 +10,7 @@ PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
 SELECT DISTINCT ?entity ?property ?value WHERE {
   VALUES ?property {oboInOwl:hasDbXref}
   ?entity ?property ?value .
-  FILTER (!regex(?value, "^[a-z|A-Z|_|\\-|0-9]*:(?!$)\\S*$"))
+  FILTER (!regex(?value, "^[A-Za-z_][A-Za-z0-9_.-]*[A-Za-z0-9_]:[^\\s]+$"))
   FILTER (!isBlank(?entity))
 }
 ORDER BY ?entity