tests: more tests for schema (#166)

Signed-off-by: Henry Schreiner <henryschreineriii@gmail.com>

Author: henryiii

.pre-commit-config.yaml

@@ -73,7 +73,7 @@ repos:
       - id: check-jsonschema
         name: Validate Histogram examples
         args: [--schemafile, src/uhi/resources/histogram.schema.json]
-        files: ^tests/resources/.*\.json
+        files: ^tests/resources/valid/.*\.json
 
   - repo: https://github.com/henryiii/validate-pyproject-schema-store
     rev: 2025.02.24

docs/serialization.md

@@ -93,7 +93,10 @@ Or with code:
 ```python
 import uhi.schema
 
-uhi.schema.validate("some/file.json")
+with filename.open(encoding="utf-8") as f:
+    data = json.load(f)
+
+uhi.schema.validate(data)
 ```
 
 Eventually this should also be usable for JSON's inside zip, HDF5 attributes,

src/uhi/schema.py

@@ -1,10 +1,11 @@
 from __future__ import annotations
 
+import functools
 import json
 import sys
+from collections.abc import Callable
 from pathlib import Path
-
-import fastjsonschema
+from typing import Any
 
 if sys.version_info < (3, 9):
     import importlib_resources as resources
@@ -13,17 +14,47 @@
 
 histogram_file = resources.files("uhi") / "resources/histogram.schema.json"
 
-with histogram_file.open(encoding="utf-8") as f:
-    histogram_schema = fastjsonschema.compile(json.load(f))
+__all__ = ["histogram_file", "validate"]
+
+
+def __dir__() -> list[str]:
+    return __all__
+
+
+@functools.lru_cache(maxsize=None)
+def _histogram_schema() -> Callable[[dict[str, Any]], None]:
+    import fastjsonschema
+
+    with histogram_file.open(encoding="utf-8") as f:
+        return fastjsonschema.compile(json.load(f))  # type: ignore[no-any-return]
+
+
+def validate(data: dict[str, Any]) -> None:
+    """Validate a histogram object against the schema."""
+    validate = _histogram_schema()
+    validate(data)
+
+
+def main(*files: str) -> None:
+    """Validate histogram files."""
+    import fastjsonschema
 
+    retval = 0
 
-def validate(path: str | Path) -> None:
-    path = Path(path)
-    with path.open(encoding="utf-8") as f:
-        example = json.load(f)
+    for file in files:
+        with Path(file).open(encoding="utf-8") as f:
+            data = json.load(f)
+        try:
+            validate(data)
+        except fastjsonschema.JsonSchemaValueException as e:
+            print(f"ERROR {file}: {e.message}")  # noqa: T201
+            retval = 1
+        else:
+            print(f"OK {file}")  # noqa: T201
 
-    histogram_schema(example)
+    if retval:
+        raise SystemExit(retval)
 
 
 if __name__ == "__main__":
-    validate(*sys.argv[1:])
+    main(*sys.argv[1:])

tests/resources/invalid/metadata_nested.error.txt

@@ -0,0 +1 @@
+data.one.metadata.one must be valid exactly by one definition

tests/resources/invalid/metadata_nested.json

@@ -0,0 +1,7 @@
+{
+  "one": {
+    "metadata": { "one": { "two": 2 } },
+    "axes": [],
+    "storage": { "type": "int", "values": [1] }
+  }
+}

tests/resources/invalid/missing_axis.error.txt

@@ -0,0 +1 @@
+data.one must contain ['axes'] properties

tests/resources/invalid/missing_axis.json

@@ -0,0 +1,5 @@
+{
+  "one": {
+    "storage": { "type": "int", "values": [1] }
+  }
+}

tests/resources/invalid/missing_storage.error.txt

@@ -0,0 +1 @@
+data.one must contain ['storage'] properties

tests/resources/invalid/missing_storage.json

@@ -0,0 +1,5 @@
+{
+  "one": {
+    "axes": []
+  }
+}

tests/resources/valid/noaxes.json

@@ -0,0 +1,6 @@
+{
+  "one": {
+    "axes": [],
+    "storage": { "type": "int", "values": [1] }
+  }
+}

tests/resources/reg.json renamed to tests/resources/valid/reg.json

@@ -1,11 +1,38 @@
 from __future__ import annotations
 
+import json
+import re
 from pathlib import Path
 
+import fastjsonschema
+import pytest
+
 import uhi.schema
 
 DIR = Path(__file__).parent.resolve()
 
+VALID_FILES = DIR.glob("resources/valid/*.json")
+INVALID_FILES = DIR.glob("resources/invalid/*.json")
+
+
+@pytest.mark.parametrize("filename", VALID_FILES, ids=lambda p: p.name)
+def test_valid_schemas(filename: Path) -> None:
+    with filename.open(encoding="utf-8") as f:
+        data = json.load(f)
+    uhi.schema.validate(data)
+
+
+@pytest.mark.parametrize("filename", INVALID_FILES, ids=lambda p: p.name)
+def test_invalid_schemas(filename: Path) -> None:
+    with filename.open(encoding="utf-8") as f:
+        data = json.load(f)
+
+    try:
+        errmsg = filename.with_suffix(".error.txt").read_text(encoding="utf-8").strip()
+    except FileNotFoundError:
+        errmsg = "NO ERROR MESSAGE FILE FOUND"
 
-def test_example_1() -> None:
-    uhi.schema.validate(DIR / "resources/reg.json")
+    with pytest.raises(
+        fastjsonschema.exceptions.JsonSchemaException, match=re.escape(errmsg)
+    ):
+        uhi.schema.validate(data)