OpenMetadata/scripts/datamodel_generation.py
Pere Miquel Brull d8e2187980
#15243 - Pydantic V2 & Airflow 2.9 (#16480)
* pydantic v2

* pydanticv2

* fix parser

* fix annotated

* fix model dumping

* mysql ingestion

* clean root models

* clean root models

* bump airflow

* bump airflow

* bump airflow

* optionals

* optionals

* optionals

* jdk

* airflow migrate

* fab provider

* fab provider

* fab provider

* some more fixes

* fixing tests and imports

* model_dump and model_validate

* model_dump and model_validate

* model_dump and model_validate

* union

* pylint

* pylint

* integration tests

* fix CostAnalysisReportData

* integration tests

* tests

* missing defaults

* missing defaults
2024-06-05 21:18:37 +02:00

84 lines
4.0 KiB
Python

# Copyright 2021 Collate
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
# http://www.apache.org/licenses/LICENSE-2.0
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""
This script generates the Python models from the JSON Schemas definition. Additionally, it replaces the `SecretStr`
pydantic class used for the password fields with the `CustomSecretStr` pydantic class which retrieves the secrets
from a configured secrets' manager.
"""
import datamodel_code_generator.model.pydantic
from datamodel_code_generator.imports import Import
import os
import re
datamodel_code_generator.model.pydantic.types.IMPORT_SECRET_STR = Import.from_full_path(
"metadata.ingestion.models.custom_pydantic.CustomSecretStr"
)
from datamodel_code_generator.__main__ import main
current_directory = os.getcwd()
ingestion_path = "./" if current_directory.endswith("/ingestion") else "ingestion/"
directory_root = "../" if current_directory.endswith("/ingestion") else "./"
UTF_8 = "UTF-8"
UNICODE_REGEX_REPLACEMENT_FILE_PATHS = [
f"{ingestion_path}src/metadata/generated/schema/entity/classification/tag.py",
f"{ingestion_path}src/metadata/generated/schema/entity/events/webhook.py",
f"{ingestion_path}src/metadata/generated/schema/entity/teams/user.py",
f"{ingestion_path}src/metadata/generated/schema/entity/type.py",
f"{ingestion_path}src/metadata/generated/schema/type/basic.py",
]
args = f"--custom-template-dir {directory_root}/ingestion/codegen_custom_templates --input {directory_root}openmetadata-spec/src/main/resources/json/schema --output-model-type pydantic_v2.BaseModel --use-annotated --base-class metadata.ingestion.models.custom_pydantic.BaseModel --input-file-type jsonschema --output {ingestion_path}src/metadata/generated/schema --set-default-enum-member".split(" ")
main(args)
for file_path in UNICODE_REGEX_REPLACEMENT_FILE_PATHS:
with open(file_path, "r", encoding=UTF_8) as file_:
content = file_.read()
# Python now requires to move the global flags at the very start of the expression
content = content.replace("(?U)", "(?u)")
with open(file_path, "w", encoding=UTF_8) as file_:
file_.write(content)
# Until https://github.com/koxudaxi/datamodel-code-generator/issues/1895
# TODO: This has been merged but `Union` is still not there. We'll need to validate
MISSING_IMPORTS = [f"{ingestion_path}src/metadata/generated/schema/entity/applications/app.py",]
WRITE_AFTER = "from __future__ import annotations"
for file_path in MISSING_IMPORTS:
with open(file_path, "r", encoding=UTF_8) as file_:
lines = file_.readlines()
with open(file_path, "w", encoding=UTF_8) as file_:
for line in lines:
file_.write(line)
if line.strip() == WRITE_AFTER:
file_.write("from typing import Union # custom generate import\n\n")
# unsupported rust regex pattern for pydantic v2
# https://docs.pydantic.dev/2.7/api/config/#pydantic.config.ConfigDict.regex_engine
# We'll remove validation from the client and let it fail on the server, rather than on the model generation
UNSUPPORTED_REGEX_PATTERN_FILE_PATHS = [
f"{ingestion_path}src/metadata/generated/schema/type/basic.py",
f"{ingestion_path}src/metadata/generated/schema/entity/data/searchIndex.py",
f"{ingestion_path}src/metadata/generated/schema/entity/data/table.py",
]
for file_path in UNSUPPORTED_REGEX_PATTERN_FILE_PATHS:
with open(file_path, "r", encoding=UTF_8) as file_:
content = file_.read()
content = content.replace("pattern='^((?!::).)*$',", "")
with open(file_path, "w", encoding=UTF_8) as file_:
file_.write(content)