Profile pr script

.github/workflows/check-profile-pr.yml

@@ -0,0 +1,20 @@
+name: Check added profile link
+
+on:
+  pull_request:
+    types: [opened, synchronize, reopened]
+    paths:
+      - 'scripts/profile_urls.txt'
+  workflow_dispatch:
+jobs:
+  check:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+      - name: Install dependencies
+        run: python3 -m pip install -r requirements.workflow.txt
+      - name: Run script
+        run: |
+          cd scripts
+          python check_profiles.py

.gitignore

@@ -1,2 +1,3 @@
 # Environments
 .env
+venv/

requirements.workflow.in

@@ -0,0 +1 @@
+rdflib==7.4.0

requirements.workflow.txt

@@ -0,0 +1,10 @@
+#
+# This file is autogenerated by pip-compile with Python 3.12
+# by the following command:
+#
+#    pip-compile --output-file=requirements.workflow.txt requirements.workflow.in
+#
+pyparsing==3.2.5
+    # via rdflib
+rdflib==7.4.0
+    # via -r requirements.workflow.in

scripts/check_profiles.py

@@ -0,0 +1,42 @@
+import re
+
+from hashlib import md5
+from rdflib import Graph, URIRef, Literal, XSD, RDF, OWL
+from urllib.parse import urljoin
+
+
+g = Graph()
+profile_class = URIRef("http://www.w3.org/ns/dx/prof/Profile")
+
+with open("profile_urls.txt", "r") as file:
+    profile_urls = [line.strip() for line in file if line.strip()]
+
+for url in profile_urls: 
+    g.parse(url, format="json-ld", publicID=urljoin(url, '.'))
+
+datetime_pattern = re.compile(r"^-?\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}(Z|[+-]\d{2}:\d{2})?$")
+date_pattern = re.compile(r"^-?\d{4}-\d{2}-\d{2}$")
+time_pattern = re.compile(r"^\d{2}:\d{2}:\d{2}(Z|[+-]\d{2}:\d{2})?$")
+
+# Add type to datetime, date and time data
+for s, p, o in g.triples((None, None, None)):
+    typed_o = None
+
+    if datetime_pattern.match(o):
+        typed_o = Literal(o, datatype=XSD.dateTime)
+    elif date_pattern.match(o):
+        typed_o = Literal(o, datatype=XSD.date)
+    elif time_pattern.match(o):
+        typed_o = Literal(o, datatype=XSD.time)
+
+    if typed_o is not None:
+        g.add((s, p, typed_o))
+        g.remove((s, p, o))
+
+    if p == RDF.type and o == profile_class:
+        id = md5(s.encode('utf-8')).hexdigest()
+        new_s = URIRef(f"http://example.org/data/profile/{id}") # TODO: Change URI
+        g.add((new_s, RDF.type, o))
+        g.add((new_s, OWL.sameAs, s))
+
+ttl_data = g.serialize(format="turtle")

scripts/profile_urls.txt

@@ -0,0 +1,5 @@
+https://trefx.uk/5s-crate/0.4/ro-crate-metadata.json
+https://www.researchobject.org/workflow-run-crate/profiles/0.5/process_run_crate/ro-crate-metadata.json
+https://www.researchobject.org/workflow-run-crate/profiles/0.5/workflow_run_crate/ro-crate-metadata.json
+https://about.workflowhub.eu/Workflow-RO-Crate/ro-crate-metadata.json
+https://www.researchobject.org/workflow-run-crate/profiles/0.5/provenance_run_crate/ro-crate-metadata.json