XLSX to OFN support draft

outputToRDF.py

@@ -2,12 +2,14 @@
 from rdflib.namespace import RDF, OWL, RDFS, SKOS, DCTERMS, TIME
 from ofnClasses import ClassType, RPPType, Relationship, Term, TermClass, Trope, VocabularyType, Vocabulary
 from datetime import datetime
+from urllib.parse import urlparse
 
 # TODO: Better IRI generation (PN_LOCAL)
 # TODO: Code lists
 # TODO: Error handling
 # TODO: Input validation
-
+# TODO: check for duplicate values
+# TODO: search for terms within file is IRI is not provided
 # TODO: Write to file
 
 
@@ -17,6 +19,19 @@ def getRDFoutput(graph: Graph, outputLocation: str):
         outputFile.write(graph.serialize(format=format))
 
 
+def testInputString(string: str) -> bool:
+    return string is not None and string.strip() != ""
+
+
+def getURIRefOrLiteral(string: str):
+    result = URIRef(string)
+    try:
+        result.n3()
+        return result
+    except Exception:
+        return Literal(string)
+
+
 def convertToRDF(vocabulary: Vocabulary, defaultLanguage: str, outputFile: str):
     graph = Graph()
 
@@ -50,22 +65,30 @@ def convertToRDF(vocabulary: Vocabulary, defaultLanguage: str, outputFile: str):
         graph.add((termIRI, SKOS.inScheme, vocabularyIRI))
         # prefLabel
         for lang, name in term.name.items():
-            graph.add((termIRI, SKOS.prefLabel, Literal(name, lang)))
+            if testInputString(name):
+                graph.add((termIRI, SKOS.prefLabel, Literal(name, lang)))
         # description
         for lang, name in term.description.items():
-            graph.add((termIRI, DCTERMS.description, Literal(name, lang)))
+            if testInputString(name):
+                graph.add((termIRI, DCTERMS.description, Literal(name, lang)))
         # definition
         for lang, name in term.definition.items():
-            graph.add((termIRI, DCTERMS.definition, Literal(name, lang)))
+            if testInputString(name):
+                graph.add((termIRI, SKOS.definition, Literal(name, lang)))
         # relation
         for relation in term.related:
-            graph.add((termIRI, DCTERMS.relation, URIRef(relation)))
+            if testInputString(relation):
+                graph.add((termIRI, DCTERMS.relation,
+                          getURIRefOrLiteral(relation)))
         # conformsTo
-        if term.source is not None:
-            graph.add((termIRI, DCTERMS.conformsTo, URIRef(term.source)))
+        if testInputString(term.source):
+            graph.add((termIRI, DCTERMS.conformsTo,
+                      getURIRefOrLiteral(term.source)))
         # exactMatch
         for equivalent in term.equivalent:
-            graph.add((termIRI, SKOS.exactMatch, URIRef(equivalent)))
+            if testInputString(equivalent):
+                graph.add((termIRI, SKOS.exactMatch,
+                          getURIRefOrLiteral(equivalent)))
         # shared in PPDF
         if term.sharedInPPDF is not None:
             graph.add((termIRI, URIRef("https://slovník.gov.cz/agendový/104/pojem/je-sdílen-v-propojeném-datovém-fondu"), Literal(
@@ -74,17 +97,21 @@ def convertToRDF(vocabulary: Vocabulary, defaultLanguage: str, outputFile: str):
         # term types (& broader)
         if isinstance(term, Term):
             for broader in term.subClassOf:
-                graph.add((termIRI, SKOS.broader, URIRef(broader)))
+                if testInputString(broader):
+                    graph.add((termIRI, SKOS.broader,
+                              getURIRefOrLiteral(broader)))
         if isinstance(term, TermClass):
             graph.add((termIRI, RDF.type, OWL.Class))
-            if term.ais is not None:
+            if testInputString(term.ais):
                 graph.add((termIRI, URIRef(
-                    "https://slovník.gov.cz/agendový/104/pojem/údaje-jsou-v-ais"), URIRef(term.ais)))
-            if term.agenda is not None:
+                    "https://slovník.gov.cz/agendový/104/pojem/údaje-jsou-v-ais"), getURIRefOrLiteral(term.ais)))
+            if testInputString(term.agenda):
                 graph.add((termIRI, URIRef(
-                    "https://slovník.gov.cz/agendový/104/pojem/sdružuje-údaje-vedené-nebo-vytvářené-v-rámci-agendy"), URIRef(term.agenda)))
+                    "https://slovník.gov.cz/agendový/104/pojem/sdružuje-údaje-vedené-nebo-vytvářené-v-rámci-agendy"), getURIRefOrLiteral(term.agenda)))
             for broader in term.subClassOf:
-                graph.add((termIRI, RDFS.subClassOf, URIRef(broader)))
+                if testInputString(broader):
+                    graph.add((termIRI, RDFS.subClassOf,
+                              getURIRefOrLiteral(broader)))
         if isinstance(term, TermClass) and term.type is ClassType.OBJECT:
             graph.add((termIRI, RDF.type, URIRef(
                 "https://slovník.gov.cz/veřejný-sektor/pojem/typ-objektu-práva")))
@@ -93,19 +120,24 @@ def convertToRDF(vocabulary: Vocabulary, defaultLanguage: str, outputFile: str):
                 "https://slovník.gov.cz/veřejný-sektor/pojem/typ-subjektu-práva")))
         if isinstance(term, Trope) or isinstance(term, Relationship):
             for broader in term.subClassOf:
-                graph.add((termIRI, RDFS.subPropertyOf, URIRef(broader)))
+                if testInputString(broader):
+                    graph.add((termIRI, RDFS.subPropertyOf,
+                              getURIRefOrLiteral(broader)))
         if isinstance(term, Trope):
             graph.add((termIRI, RDF.type, OWL.DatatypeProperty))
             if term.target is not None:
-                graph.add((termIRI, RDFS.domain, URIRef(term.target)))
+                graph.add(
+                    (termIRI, RDFS.domain, getURIRefOrLiteral(term.target)))
             if term.datatype is not None:
-                graph.add((termIRI, RDFS.range, URIRef(term.datatype)))
+                graph.add(
+                    (termIRI, RDFS.range, getURIRefOrLiteral(term.datatype)))
         if isinstance(term, Relationship):
             graph.add((termIRI, RDF.type, OWL.ObjectProperty))
             if term.domain is not None:
-                graph.add((termIRI, RDFS.domain, URIRef(term.domain)))
+                graph.add(
+                    (termIRI, RDFS.domain, getURIRefOrLiteral(term.domain)))
             if term.range is not None:
-                graph.add((termIRI, RDFS.range, URIRef(term.range)))
+                graph.add((termIRI, RDFS.range, getURIRefOrLiteral(term.range)))
 
         # RPP public/private types
         if term.rppType is not None:
@@ -118,4 +150,4 @@ def convertToRDF(vocabulary: Vocabulary, defaultLanguage: str, outputFile: str):
                 graph.add((termIRI, RDF.type, URIRef(
                     term.rppPrivateTypeSource)))
 
-        getRDFoutput(graph, outputFile)
+    getRDFoutput(graph, outputFile)

requirements.txt

@@ -1,12 +1,15 @@
 blinker==1.8.2
 click==8.1.7
 colorama==0.4.6
+et_xmlfile==2.0.0
 Flask==3.0.3
 Flask-Cors==4.0.1
 gunicorn==22.0.0
 itsdangerous==2.2.0
 Jinja2==3.1.4
 MarkupSafe==2.1.5
+openpyxl==3.1.5
 packaging==24.0
-setuptools==69.2.0
+pyparsing==3.2.0
+rdflib==7.1.1
 Werkzeug==3.0.3