|
|
Dear all
I am trying to make a function that iterates through the elements of a table and adds them to an ontology saved locally. I know this code can and will be improved, but at the moment the following issue is my most pressing problem.
My issue is that at every iteration (i.e. update of the table), the functions makes the urn of the elements appearing in protégé longer (repeats the name of the ontology as additional subfolder) and I do not understand why.
So if the first urn is "urn:/Ontology.owl/Ontology.owl/Ontology.owl" (why 3 times?), when I add a "second level" or execute the function a second time (whcih should not change the ontology), the urn becomes "urn:/Ontology.owl/Ontology.owl/Ontology.owl/Ontology.owl/Ontology.owl/Ontology.owl" with the repeated elelments copied.
Thank you very much for the help!
Here my code:
def addElementsfromXLS(ontology, xlsfile, column1, column2="", column3=""):
onto_path.append(ontologyLocation) #since the owl file is saved locally, the path is "C:\\Users\\...
ontoname = ontology+".owl"
onto = get_ontology(ontoname).load()
excel= pd.read_excel(xlsfile)
with onto:
column1Head = "Titel" #can get dynamically, but currently the head is non descriptive
NewClass = types.new_class(column1Head, bases = (Thing,), )
for i in range(1,len(excel)):
if excel.loc[i, column1] not in ["0", " "]:
textCell = excel.loc[i, column1]
NewClass = types.new_class(textCell, bases = (onto[column1Head], ))
else:
pass
# If no third column is given, it is assumed that column1 is parent of column2
if column3 == "":
if column2 != "":
column2Head = "Title2"
NewClass = types.new_class(column2Head, bases = (Thing,), )
for i in range(1,len(excel)):
textCell = excel.loc[i, column2]
if excel.loc[i, column1] not in ["0", " "]:
parent = excel.loc[i, column1]
NewClass = types.new_class(textCell, bases = (onto[column2Head], onto[parent]),)
else:
NewClass = types.new_class(textCell, bases = (onto[column2Head],))
else:
pass
else:
for i in range(1,len(excel)):#### TO BE COMPLETED, not relevant for the moment as I am still working on only two columns
pass
onto.save()
|
Locked
