From c554cbd69f9b786a5d72c9bb51be27386501cabb Mon Sep 17 00:00:00 2001
From: Sergei Maertens <sergei@maykinmedia.nl>
Date: Wed, 18 Dec 2024 16:43:44 +0100
Subject: [PATCH] :white_check_mark: Run XSD validation in eherkenning metadata
 test

---
 tests/test_eherkenning_metadata.py | 13 +++++++++++++
 tests/utils.py                     | 22 ++++++++++++++++++++++
 2 files changed, 35 insertions(+)

diff --git a/tests/test_eherkenning_metadata.py b/tests/test_eherkenning_metadata.py
index c2fe4d9..44742c2 100644
--- a/tests/test_eherkenning_metadata.py
+++ b/tests/test_eherkenning_metadata.py
@@ -1,3 +1,5 @@
+from pathlib import Path
+
 from django.test import TestCase
 
 import pytest
@@ -12,6 +14,13 @@
 )
 
 from .mixins import EherkenningMetadataMixin
+from .utils import validate_against_xsd
+
+_repo_root = Path(__file__).parent.parent.resolve()
+
+SAML_METADATA_XSD = (
+    _repo_root / "digid_eherkenning" / "xsd" / "saml-schema-metadata-2.0.xsd"
+)
 
 NAME_SPACES = {
     "md": "urn:oasis:names:tc:SAML:2.0:metadata",
@@ -162,6 +171,10 @@ def test_generate_metadata_all_options_specified(self):
         self.eherkenning_config.save()
 
         eherkenning_metadata = generate_eherkenning_metadata()
+
+        with self.subTest("passes XSD validation"):
+            validate_against_xsd(eherkenning_metadata, SAML_METADATA_XSD)
+
         self.assertEqual(eherkenning_metadata[:5], b"<?xml")
         entity_descriptor_node = etree.XML(eherkenning_metadata)
 
diff --git a/tests/utils.py b/tests/utils.py
index 199e168..b24b7e4 100644
--- a/tests/utils.py
+++ b/tests/utils.py
@@ -1,5 +1,10 @@
 from base64 import b64encode
+from functools import lru_cache
 from hashlib import sha1
+from io import BytesIO
+from pathlib import Path
+
+from lxml import etree
 
 SAML_NAMESPACES = {
     "samlp": "urn:oasis:names:tc:SAML:2.0:protocol",
@@ -19,3 +24,20 @@ def create_example_artifact(endpoint_url, endpoint_index=b"\x00\x00"):
     message_handle = b"01234567890123456789"  # something random
 
     return b64encode(type_code + endpoint_index + source_id + message_handle)
+
+
+@lru_cache
+def _load_schema(path: Path):
+    with path.open("r") as infile:
+        return etree.parse(infile)
+
+
+def validate_against_xsd(xml: bytes, xsd_schema: Path) -> None:
+    """
+    Validate the XML against a schema.
+
+    See https://lxml.de/validation.html
+    """
+    xmlschema = etree.XMLSchema(_load_schema(xsd_schema))
+    doc = etree.parse(BytesIO(xml))
+    xmlschema.assertValid(doc)