experimental: Build pydantic from schema

gen.sh

@@ -0,0 +1,20 @@
+#!/usr/bin/env bash
+
+datamodel-codegen \
+  --disable-timestamp \
+  --enable-version-header \
+  --enum-field-as-literal one \
+  --field-constraints \
+  --input $1 \
+  --input-file-type "jsonschema" \
+  --output-model-type pydantic_v2.BaseModel \
+  --snake-case-field \
+  --strict-nullable \
+  --target-python-version 3.8 \
+  --use-default-kwarg \
+  --use-double-quotes \
+  --use-schema-description \
+  --use-standard-collections \
+  --use-subclass-enum \
+  --use-title-as-name \
+  --use-union-operator

pyproject.toml

@@ -35,6 +35,7 @@ dependencies = [
     "numpy",
     "pandas",
     "pyarrow",
+    "pydantic>=2.0.0",
     "PyYAML",
     "xtgeo>=2.16",
 ]