Add more tests for ref resolution

Ensure that the `$id` and retrieval URI handling are both done
correctly. JSON Schema specifies that `$id` has precedence, so one of
the tests confirms that preference.

Author: sirosen

tests/acceptance/test_remote_ref_resolution.py

@@ -37,11 +37,8 @@
 }
 
 
-@pytest.mark.parametrize("check_passes", (True, False))
-@pytest.mark.parametrize("casename", ("case1", "case2"))
-def test_remote_ref_resolution_simple_case(
-    run_line, check_passes, casename, tmp_path, monkeypatch
-):
+@pytest.fixture(autouse=True)
+def _mock_schema_cache_dir(monkeypatch, tmp_path):
     def _fake_compute_default_cache_dir(self):
         return str(tmp_path)
 
@@ -51,6 +48,10 @@ def _fake_compute_default_cache_dir(self):
         _fake_compute_default_cache_dir,
     )
 
+
+@pytest.mark.parametrize("check_passes", (True, False))
+@pytest.mark.parametrize("casename", ("case1", "case2"))
+def test_remote_ref_resolution_simple_case(run_line, check_passes, casename, tmp_path):
     main_schema_loc = "https://example.com/main.json"
     responses.add("GET", main_schema_loc, json=CASES[casename]["main_schema"])
     for name, subschema in CASES[casename]["other_schemas"].items():
@@ -74,3 +75,69 @@ def _fake_compute_default_cache_dir(self):
         assert result.exit_code == 0, output
     else:
         assert result.exit_code == 1, output
+
+
+# this test ensures that `$id` is preferred for the base URI over
+# the retrieval URI
+@pytest.mark.parametrize("check_passes", (True, False))
+def test_ref_resolution_prefers_id_over_retrieval_uri(run_line, tmp_path, check_passes):
+    main_schema = {
+        "$id": "https://example.org/schemas/main.json",
+        "$schema": "http://json-schema.org/draft-07/schema",
+        "properties": {
+            "title": {"$ref": "./title_schema.json"},
+        },
+        "additionalProperties": False,
+    }
+    title_schema = {"type": "string"}
+
+    retrieval_uri = "https://example.org/alternate-path-retrieval-only/schemas/main"
+    responses.add("GET", retrieval_uri, json=main_schema)
+    responses.add(
+        "GET", "https://example.org/schemas/title_schema.json", json=title_schema
+    )
+
+    instance_path = tmp_path / "instance.json"
+    instance_path.write_text(json.dumps({"title": "doc one" if check_passes else 2}))
+
+    result = run_line(
+        ["check-jsonschema", "--schemafile", retrieval_uri, str(instance_path)]
+    )
+    output = f"\nstdout:\n{result.stdout}\n\nstderr:\n{result.stderr}"
+    if check_passes:
+        assert result.exit_code == 0, output
+    else:
+        assert result.exit_code == 1, output
+
+
+@pytest.mark.parametrize("check_passes", (True, False))
+def test_ref_resolution_does_not_callout_for_absolute_ref_to_retrieval_uri(
+    run_line, tmp_path, check_passes
+):
+    retrieval_uri = "https://example.net/schemas/main"
+
+    main_schema = {
+        "$id": "https://example.net/schemas/some-uri-which-will-never-be-used/main.json",
+        "$schema": "http://json-schema.org/draft-07/schema",
+        "$defs": {"title": {"type": "string"}},
+        "properties": {
+            "title": {"$ref": f"{retrieval_uri}#/$defs/title"},
+        },
+        "additionalProperties": False,
+    }
+
+    # exactly one GET to the retrieval URI will work
+    responses.add("GET", retrieval_uri, json=main_schema)
+    responses.add("GET", retrieval_uri, json={"error": "permafrost melted"}, status=500)
+
+    instance_path = tmp_path / "instance.json"
+    instance_path.write_text(json.dumps({"title": "doc one" if check_passes else 2}))
+
+    result = run_line(
+        ["check-jsonschema", "--schemafile", retrieval_uri, str(instance_path)]
+    )
+    output = f"\nstdout:\n{result.stdout}\n\nstderr:\n{result.stderr}"
+    if check_passes:
+        assert result.exit_code == 0, output
+    else:
+        assert result.exit_code == 1, output