Merge branch 'main' of https://github.com/svdimchenko/pydantic-glue

Author: svdimchenko

.pre-commit-config.yaml

@@ -1,6 +1,6 @@
 repos:
   - repo: https://github.com/pre-commit/pre-commit-hooks
-    rev: v4.6.0
+    rev: v5.0.0
     hooks:
       # Identify invalid files
       - id: check-ast
@@ -27,42 +27,15 @@ repos:
         args: [--markdown-linebreak-ext=md]
       - id: end-of-file-fixer
 
-  - repo: https://github.com/PyCQA/autoflake
-    rev: v2.3.1
+  - repo: https://github.com/astral-sh/ruff-pre-commit
+    rev: v0.9.4
     hooks:
-      - id: autoflake
-
-  - repo: https://github.com/asottile/pyupgrade
-    rev: v3.15.2
-    hooks:
-      - id: pyupgrade
-        args:
-          - '--py39-plus'
-
-  - repo: https://github.com/pycqa/isort
-    rev: 5.13.2
-    hooks:
-      - id: isort
-        name: isort
-        args:
-          - '.'
-
-  - repo: https://github.com/psf/black
-    rev: 24.4.2
-    hooks:
-      - id: black
-
-  - repo: https://github.com/pycqa/flake8
-    rev: 7.0.0
-    hooks:
-      - id: flake8
-        additional_dependencies:
-          - 'Flake8-pyproject~=1.1'
-        args:
-          - '.'
+      - id: ruff
+        args: [ --fix ]
+      - id: ruff-format
 
   - repo: https://github.com/pre-commit/mirrors-mypy
-    rev: v1.3.0
+    rev: v1.14.1
     hooks:
       - id: mypy
         args:
@@ -74,7 +47,7 @@ repos:
         exclude: ^tests/
 
   - repo: https://github.com/igorshubovych/markdownlint-cli
-    rev: v0.40.0
+    rev: v0.44.0
     hooks:
     - id: markdownlint
       args:

pydantic_glue/cli.py

@@ -1,6 +1,7 @@
+"""CLI to convert pydantic class to AWS Glue schema."""
+
 import json
 import logging
-import os
 import sys
 from argparse import ArgumentParser
 from dataclasses import dataclass
@@ -15,6 +16,8 @@
 
 @dataclass
 class Arguments:
+    """CLI Arguments."""
+
     module_file: str
     class_name: str
     output_file: str
@@ -23,6 +26,7 @@ class Arguments:
 
 
 def parse_args(argv: list[str]) -> Arguments:
+    """Parse CLI arguments."""
     parser = ArgumentParser()
     parser.add_argument("-f", dest="source_file", required=True, type=str, help="Path to the python file")
     parser.add_argument("-c", dest="class_name", required=True, type=str, help="Python class name")
@@ -49,9 +53,10 @@ def parse_args(argv: list[str]) -> Arguments:
 
 
 def cli() -> None:
+    """CLI entry point."""
     args = parse_args(sys.argv[1:])
-    sys.path.append(os.path.dirname(args.module_file))
-    imported = __import__(os.path.basename(args.module_file))
+    sys.path.append(Path(args.module_file).parent.as_posix())
+    imported = __import__(Path(args.module_file).name)
 
     model = getattr(imported, args.class_name)
     input_schema = json.dumps(model.model_json_schema(by_alias=args.json_schema_by_alias))
@@ -60,7 +65,7 @@ def cli() -> None:
     output = json.dumps(
         {
             "//": f"Generated by pydantic-glue at {datetime.now(tz=timezone.utc)}. DO NOT EDIT",
-            "columns": {k: v for (k, v) in converted},
+            "columns": dict(converted),
         },
         indent=2,
     )

pydantic_glue/errors.py

@@ -0,0 +1,16 @@
+"""Errors."""
+
+
+class ObjectWithoutPropertiesError(Exception):
+    def __init__(self) -> None:
+        super().__init__("Object without properties or additionalProperties can't be represented")
+
+
+class UnknownTypeError(Exception):
+    def __init__(self, type_name: str) -> None:
+        super().__init__(f"Unknown type: {type_name}")
+
+
+class GlueMapWithoutTypesError(Exception):
+    def __init__(self) -> None:
+        super().__init__("Glue Cannot Support a Map Without Types")

pydantic_glue/handler.py

@@ -1,25 +1,29 @@
+"""Convert Json schema to glue."""
+
+from __future__ import annotations
+
 from typing import Any, Union
 
 import jsonref
 
+from pydantic_glue.errors import GlueMapWithoutTypesError, ObjectWithoutPropertiesError, UnknownTypeError
 
-def dispatch(v: dict[str, Any]) -> str:
-
-    glue_type = v.get("glue_type", None)
 
-    if glue_type is not None:
+def dispatch(v: dict[str, Any]) -> str:  # noqa: PLR0911
+    """Dispatch json schema element as glue type."""
+    if glue_type := v.get("glue_type"):
         return str(glue_type)
 
     if "anyOf" in v:
-        return handle_union(v)
+        return _handle_union(v)
 
     t = v["type"]
 
     if t == "object":
-        return handle_object(v)
+        return _handle_object(v)
 
     if t == "array":
-        return handle_array(v)
+        return _handle_array(v)
 
     if t == "string":
         if v.get("format") == "date-time":
@@ -37,53 +41,55 @@ def dispatch(v: dict[str, Any]) -> str:
     if t == "number":
         return "float"
 
-    raise Exception(f"unknown type: {t}")
+    raise UnknownTypeError(t)
 
 
-def handle_map(o: dict[str, Any]) -> str:
+def _handle_map(o: dict[str, Any]) -> str:
     t = o["additionalProperties"]
     res = dispatch(t)
     return f"map<string,{res}>"
 
 
-def handle_union(o: dict[str, Any]) -> str:
+def _handle_union(o: dict[str, Any]) -> str:
     types = [i for i in o["anyOf"] if i["type"] != "null"]
     if len(types) > 1:
         res = [dispatch(v) for v in types]
         return f"union<{','.join(res)}>"
     return dispatch(types[0])
 
 
-def map_dispatch(o: dict[str, Any]) -> list[tuple[str, str]]:
+def _map_dispatch(o: dict[str, Any]) -> list[tuple[str, str]]:
     return [(k, dispatch(v)) for k, v in o["properties"].items()]
 
 
-def handle_object(o: dict[str, Any]) -> str:
+def _handle_object(o: dict[str, Any]) -> str:
     if "additionalProperties" in o:
         if o["additionalProperties"] is True:
-            raise Exception("Glue Cannot Support a Map Without Types")
-        elif o["additionalProperties"]:
+            raise GlueMapWithoutTypesError
+        if o["additionalProperties"]:
             if "properties" in o:
-                raise NotImplementedError("Merging types of properties and additionalProperties")
-            return handle_map(o)
+                msg = "Merging types of properties and additionalProperties"
+                raise NotImplementedError(msg)
+            return _handle_map(o)
 
     if "properties" not in o:
-        raise Exception("Object without properties or additionalProperties can't be represented")
+        raise ObjectWithoutPropertiesError
 
-    res = [f"{k}:{v}" for (k, v) in map_dispatch(o)]
+    res = [f"{k}:{v}" for (k, v) in _map_dispatch(o)]
     return f"struct<{','.join(res)}>"
 
 
-def handle_array(o: dict[str, Any]) -> str:
+def _handle_array(o: dict[str, Any]) -> str:
     t = dispatch(o["items"])
     return f"array<{t}>"
 
 
-def handle_root(o: dict[str, Any]) -> list[tuple[str, str]]:
-    return map_dispatch(o)
+def _handle_root(o: dict[str, Any]) -> list[tuple[str, str]]:
+    return _map_dispatch(o)
 
 
 def convert(schema: str) -> Union[list[Any], list[tuple[str, str]]]:
+    """Convert json schema to glue."""
     if not schema:
         return []
-    return handle_root(jsonref.loads(schema))
+    return _handle_root(jsonref.loads(schema))

pyproject.toml

@@ -21,46 +21,72 @@ jsonref = "^1.1.0"
 pydantic = "^2.7.1"
 
 [tool.poetry.group.dev.dependencies]
-autoflake = "^2.3.1"
 pytest = "^8.2.0"
+cli-test-helpers = "^4.1.0"
 mypy = "^1.10.0"
-flake8 = "^7.0.0"
-black = "^24.4.2"
-isort = "^5.13.2"
 pre-commit = "^3.7.0"
-cli-test-helpers = "^4.1.0"
-
-[tool.autoflake]
-recursive = true
-in-place = true
-remove-all-unused-imports = true
-remove-unused-variables = true
+ruff = "^0.9.4"
+pytest-cov = "^6.0.0"
 
-[tool.black]
+[tool.ruff]
 line-length = 120
-target-version = ['py39', 'py310', 'py311', 'py312']
-include = '\.pyi?$'
+indent-width = 4
+target-version = "py39"
 
-[tool.flake8]
-files = '.*\.py'
-max-line-length = 120
-exclude = ['.git', '.eggs', '__pycache__', 'venv', '.venv']
+[tool.ruff.lint]
+select = ["ALL"]
 ignore = [
-    # space before: (needed for how black formats slicing)
-    'E203',
-    # line break before binary operator (needed for how black formats long lines)
-    'W503'
+	# The following rules may cause conflicts when used with the formatter
+	"COM812",
+	"ISC001",
+
+	# General ignores
+	"ANN204", # Missing return type annotation for special method `__init__`
+	"D107",   # Missing docstring in `__init__`
+	"D203",   # one-blank-line-before-class
+	"D213",   # multi-line-summary-second-line
+	"G004",   # Logging statement uses f-string
+	"UP007",  # Use `X | Y` for type annotations
 ]
 
-[tool.isort]
-profile = 'black'
-src_paths = ['athena_udf', 'test']
+fixable = ["ALL"]
+unfixable = []
+dummy-variable-rgx = "^(_+|(_+[a-zA-Z0-9_]*[a-zA-Z0-9]+?))$"
 
-[tool.pytest.ini_options]
-testpaths = [
-    "tests"
+[tool.ruff.lint.per-file-ignores]
+"**/{tests}/*" = [
+	"ANN001",  # Missing type annotation for function argument
+	"ANN201",  # Missing return type annotation for public function
+	"D100",    # Missing docstring in public module
+	"D101",    # Missing docstring in public class
+	"D103",    # Missing docstring in public function
+	"S101",    # Use of `assert` detected
+	"S603",    # `subprocess` call: check for execution of untrusted input
+]
+"__init__.py" = ["D104"]  # Missing docstring in public package
+
+"pydantic_glue/errors.py" = [
+	"D101",    # Missing docstring in public class
 ]
 
+[tool.ruff.lint.flake8-type-checking]
+runtime-evaluated-base-classes = ["pydantic.BaseModel"]
+
+[tool.ruff.lint.mccabe]
+max-complexity = 15
+
+[tool.ruff.format]
+quote-style = "double"
+indent-style = "space"
+skip-magic-trailing-comma = false
+line-ending = "auto"
+docstring-code-format = false
+docstring-code-line-length = "dynamic"
+
+[tool.pytest.ini_options]
+testpaths = ["tests"]
+addopts = "--cov"
+
 [tool.mypy]
 ignore_missing_imports = true
 strict = true

tests/data/__init__.py

@@ -1,13 +1,12 @@
 import json
 from pathlib import Path
-from unittest import TestCase
 
 from cli_test_helpers import shell
+
 from pydantic_glue.cli import parse_args
 
 
 def test_parse_args():
-
     args = parse_args(["-f", "foo/bar/file.py", "-c", "Test"])
     assert args.module_file == "foo/bar/file"
     assert args.class_name == "Test"
@@ -17,7 +16,6 @@ def test_parse_args():
 
 
 def test_parse_args_schema_by_name():
-
     args = parse_args(["-f", "foo/bar/file.py", "-c", "Test", "--schema-by-name"])
     assert args.module_file == "foo/bar/file"
     assert args.class_name == "Test"
@@ -30,11 +28,11 @@ def test_cli():
     shell("pydantic-glue -f tests/data/input.py -c A -o tests/tmp/actual.json")
     actual = json.loads(Path("tests/tmp/actual.json").read_text())
     expected = json.loads(Path("tests/data/expected.json").read_text())
-    TestCase().assertDictEqual(actual["columns"], expected["columns"])
+    assert actual["columns"] == expected["columns"]
 
 
 def test_cli_schema_by_name():
     shell("pydantic-glue -f tests/data/input.py -c A -o tests/tmp/actual_by_name.json --schema-by-name")
     actual = json.loads(Path("tests/tmp/actual_by_name.json").read_text())
     expected = json.loads(Path("tests/data/expected_by_name.json").read_text())
-    TestCase().assertDictEqual(actual["columns"], expected["columns"])
+    assert actual["columns"] == expected["columns"]