Use patched version of jsonschema

.pre-commit-config.yaml

@@ -9,19 +9,31 @@ repos:
       - id: check-xml
       - id: check-yaml
       - id: end-of-file-fixer
+        exclude: courts_schema_types_autogenerated\.py
       - id: forbid-submodules
       - id: mixed-line-ending
       - id: no-commit-to-branch
       - id: trailing-whitespace
 
+  # There is a bug with upstream which means the imports are
+  # not sorted -- this should do for now
+  # - repo: https://github.com/camptocamp/jsonschema-gentypes
+  - repo: https://github.com/dxw/jsonschema-gentypes
+    rev: patch-1
+    hooks:
+      - id: jsonschema-gentypes
+        files: src/ds_caselaw_utils/data/schema/courts.schema.json
+
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.6.7
+    rev: v0.6.8
     hooks:
       - id: ruff
         args:
           - --fix
           - --exit-non-zero-on-fix
+        exclude: courts_schema_types_autogenerated\.py
       - id: ruff-format
+        exclude: courts_schema_types_autogenerated\.py
 
   - repo: https://github.com/pre-commit/mirrors-mypy
     rev: v1.11.2
@@ -33,7 +45,7 @@ repos:
         args:
           - --strict
         files: ^(src/ds_caselaw_utils|scripts)
-        exclude: test_
+        exclude: (?:test_|courts_schema_types_autogenerated\.py)
 
   - repo: https://github.com/pre-commit/mirrors-mypy
     rev: v1.11.2
@@ -71,7 +83,7 @@ repos:
         exclude: courts\.md
 
   - repo: https://github.com/commitizen-tools/commitizen
-    rev: v3.29.0
+    rev: v3.29.1
     hooks:
       - id: commitizen
       - id: commitizen-branch

jsonschema-gentypes.yaml

@@ -0,0 +1,22 @@
+headers: >
+  # Automatically generated file from a JSON schema
+# Used to correctly format the generated file
+# callbacks:
+#   - - black
+#   - - isort
+generate:
+  - # JSON schema file path
+    source: src/ds_caselaw_utils/data/schema/courts.schema.json
+    # Python file path
+    destination: src/ds_caselaw_utils/courts_schema_types_autogenerated.py
+    # The name of the root element
+    root_name: RawCourtRepository
+    # Argument passed to the API
+    api_arguments:
+      additional_properties: Only explicit
+    # Rename an element
+    name_mapping: {}
+    # The minimum Python version that the code should support. By default the
+    # currently executing Python version is chosen. Note that the output
+    # may require typing_extensions to be installed.
+    python_version: "3.9"

src/ds_caselaw_utils/courts.py

@@ -8,6 +8,7 @@
 from datetime import date
 from typing import NewType, Optional
 
+from courts_schema_types_autogenerated import RawCourt, RawCourtRepositoryData, RawJurisdiction
 from ruamel.yaml import YAML
 
 CourtCode = NewType("CourtCode", str)
@@ -16,14 +17,14 @@
 
 
 class Jurisdiction:
-    def __init__(self, data):
+    def __init__(self, data: RawJurisdiction):
         self.code: JurisdictionCode = data["code"]
         self.name = data["name"]
         self.prefix: str = data.get("prefix")
 
 
 class Court:
-    def __init__(self, data) -> None:
+    def __init__(self, data: RawCourt) -> None:
         self.code: CourtCode = CourtCode(data["code"])
         self.name: str = data["name"]
         self.grouped_name: str = data.get("grouped_name") or data["name"]
@@ -112,7 +113,7 @@ class CourtNotFoundException(Exception):
 
 
 class CourtsRepository:
-    def __init__(self, data):
+    def __init__(self, data: RawCourtRepositoryData):
         self._data = data
         self._byParam = {}
         self._byCode = {}
@@ -223,6 +224,6 @@ def get_listable_tribunals(self) -> list[Court]:
 yaml = YAML()
 datafile = pathlib.Path(__file__).parent / "data/court_names.yaml"
 with open(datafile) as f:
-    court_data = yaml.load(f)
+    court_data: RawCourtRepositoryData = yaml.load(f)
 
 courts = CourtsRepository(court_data)

src/ds_caselaw_utils/courts_schema_types_autogenerated.py

@@ -0,0 +1,86 @@
+# Automatically generated file from a JSON schema
+
+
+from typing import List, Required, TypedDict, Union
+
+
+class RawCourt(TypedDict, total=False):
+    """ Raw Court. """
+
+    code: Required[str]
+    """
+    pattern: ^[A-Za-z]{2,}(-[A-Za-z0-9]+)*$
+
+    Required property
+    """
+
+    name: Required[str]
+    """ Required property """
+
+    grouped_name: str
+    param: str
+    """ pattern: ^[a-z]{2,}(?:/[a-z0-9]+)?$ """
+
+    extra_params: List["_RawCourtExtraParamsItem"]
+    ncn: str
+    link: Required[str]
+    """
+    format: uri
+
+    Required property
+    """
+
+    start_year: int
+    """ minimum: 1066 """
+
+    end_year: int
+    listable: Required[bool]
+    """ Required property """
+
+    selectable: Required[bool]
+    """ Required property """
+
+    jurisdictions: List["RawJurisdiction"]
+
+
+RawCourtRepository = List["_RawCourtListItem"]
+"""
+Raw Court List.
+
+A list of courts
+"""
+
+
+
+class RawJurisdiction(TypedDict, total=False):
+    """ Raw Jurisdiction. """
+
+    prefix: Required[str]
+    """ Required property """
+
+    name: Required[str]
+    """ Required property """
+
+    code: Required[str]
+    """ Required property """
+
+
+
+_RawCourtExtraParamsItem = str
+""" pattern: ^[a-z]{2,}(/[a-z]+)?$ """
+
+
+
+class _RawCourtListItem(TypedDict, total=False):
+    name: Required[str]
+    """ Required property """
+
+    display_name: Required[Union[str, None]]
+    """ Required property """
+
+    is_tribunal: Required[bool]
+    """ Required property """
+
+    courts: Required[List["RawCourt"]]
+    """ Required property """
+

src/ds_caselaw_utils/data/schema/courts.schema.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://json-schema.org/draft/2020-12/schema",
   "$id": "https://schema.caselaw.nationalarchives.gov.uk/courts.schema.json",
-  "title": "Courts",
+  "title": "Raw Court List",
   "description": "A list of courts",
   "type": "array",
   "items": {
@@ -20,6 +20,7 @@
         "type": "array",
         "items": {
           "type": "object",
+          "title": "Raw Court",
           "properties": {
             "code": {
               "type": "string",
@@ -65,6 +66,7 @@
             "jurisdictions": {
               "type": "array",
               "items": {
+                "title": "Raw Jurisdiction",
                 "type": "object",
                 "properties": {
                   "prefix": {