Find less common string characters in Cell Ontology

uncommon_ontology_chars.py

"""Find less common string characters in Cell Ontology

Prints characters that aren't an ASCII letter, number, hyphen, space, or
underscore that appear among standard labels in Cell Ontology (CL).
This snippet helps get testing input for more robust ontology handling.

Tested on Python 3.10.9, without dependencies.  Output as of 2024-08-02:
.=/[:]–><';,()+
"""

import xml.etree.ElementTree as ET
import re
import urllib.request
from pathlib import Path

# Fetch ontology file if not already cached
if not Path('cl.owl').is_file():
    # Official Cell Ontology OWL URL, per
    # https://www.ebi.ac.uk/ols4/ontologies/cl
    cl_url = 'http://purl.obolibrary.org/obo/cl.owl'
    with urllib.request.urlopen(cl_url) as f:
        cell_ontology_owl = f.read()

    with open('cl.owl', 'wb') as f:
        f.write(cell_ontology_owl)

# Parse ontology XML
tree = ET.parse('cl.owl')
root = tree.getroot()
namespaces = {'rdfs': 'http://www.w3.org/2000/01/rdf-schema#'}
labels = root.findall('.//rdfs:label', namespaces)

# Find uncommon characters, dedup them
finds = list(map(lambda label: re.findall('[^A-Za-z0-9-_ ]', label.text), labels))
filtered_finds = list(filter(lambda match: len(match) > 0, finds))
raw_unique_finds = set([char for find in filtered_finds for char in find])
unique_finds = "".join(list(raw_unique_finds))

print('unique_finds')
print(unique_finds)