Skip to main content

LinkML Validator

Project description

LinkML Validator

Run tests PyPI

The LinkML Validator is a library for performing validation on data objects that conform to a given LinkML schema.

The Validator is initialized using a LinkML schema YAML, and is designed to allow for flexible validation where each type of validation is done by a plugin.

For example, JSONSchema validation is performed by JsonSchemaValidationPlugin.

Motivation

The LinkML Validator is built with the following goals in mind:

  • the Validator should respond with parseable validation messages
  • the Validator should not break the validation process even if one object from a list of objects fail validation
  • the Validator should provide the ability to perform more than one type of validation on an object

Installation

python setup.py install

To install development dependencies (like pytest, mkdocs, etc.):

pip install -e ".[dev]"

Running the LinkML Validator via CLI

To run the LinkML Validator,

linkml-validator --inputs <INPUT JSON> \
    --schema <SCHEMA YAML> \
    --output <OUTPUT>

You can pass filepath or a URL that points to the LinkML schema YAML.

Input data as a dictionary of objects

The input JSON can be a dictionary of objects keyed by the object type.

{
    "<OBJECT_TYPE>": [
        {

        }
    ]
}

Where the <OBJECT_TYPE> is the pythonic representation of a class defined in the schema YAML.

For example, consider examples/example_data1.json:

{
    "NamedThing": [
        {
            "id": "obj1",
            "name": "Object 1",
            "type": "X"
        },
        {
            "id": "obj2",
            "name": "Object 2",
            "type": "Y"
        }
    ]
}

In the above example, the NamedThing is the target_class, which is the pythonic representation of the class named thing as defined in the examples/example_schema.yaml.

You can run the validator on the above data as follows:

linkml-validator --inputs examples/example_data1.json \
    --schema examples/example_schema.yaml \
    --output examples/example_data1_validation_report.json

Input data as an array of objects

The input JSON can also be an array of objects:

[
    {},
    {}
]

In this case, one must also specify the object type via --target-class argument in the CLI.

For example, consider examples/example_data2.json:

[
    {
        "id": "obj1",
        "name": "Object 1",
        "type": "X"
    },
    {
        "id": "obj2",
        "name": "Object 2",
        "type": "Y"
    }
]

You can run the validator on the above data as follows:

linkml-validator --inputs examples/example_data2.json \
    --schema examples/example_schema.yaml \
    --output examples/example_data2_validation_report.json \
    --target-class NamedThing

Running selected plugins

To run only certain plugins as part of the validation,

linkml-validator --inputs data.json \
    --schema schema.yaml \
    --output validation_results.json \
    --plugins JsonSchemaValidationPlugin

To perform strict validation,

linkml-validator --inputs data.json \
    --schema schema.yaml \
    --output validation_results.json \
    --plugins JsonSchemaValidationPlugin \
    --strict

Under normal (default) mode, the validator will run all the checks defined in all referenced plugins on a given object.

When in strict mode, the validator will stop the validation for an object if even one of the plugins report a failed validation.

Running your own plugins with the Validator (via CLI)

To run your custom plugin as part of the validation,

linkml-validator --inputs data.json \
    --schema schema.yaml \
    --output validation_results.json \
    --plugins JsonSchemaValidationPlugin \
    --plugins <CUSTOM_PLUGIN_CLASS>

where <CUSTOM_PLUGIN_CLASS> the reference to a custom plugin class.

Note: The custom plugin class must be a subclass of linkml_validator.plugins.base.BasePlugin and must implement all the methods defined in BasePlugin class.

Using LinkML Validator as a module

You can use the linkml_validator.validator.Validator class directly in your codebase to perform validation on objects that you are working with.

The following code snippet provides a quick way of instantiating the Validator class and performing validation on an object:

from linkml_validator.validator import Validator

data_obj = {
    "id": "obj1",
    "name": "Object 1",
    "type": "X"
}
validator = Validator(schema="examples/example_schema.yaml")
validator.validate(obj=data_obj, target_class="NamedThing")

Note: The above code makes the assumption that there is a class named thing defined in the examples/example_schema.yaml and that NamedThing is its Pythonic representation.

You can also provide your own custom plugin class to run with the Validator,

from linkml_validator.validator import Validator
from linkml_validator.plugins.base import BasePlugin
from linkml_validator.models import ValidationResult

class MyCustomPlugin(BasePlugin):
    NAME = "MyCustomPlugin"

    def __init__(self, schema: str, **kwargs) -> None:
        super().__init__(schema)

    def process(self, obj: dict, **kwargs) -> ValidationResult:
        # Add your custom logic for processing and validating the incoming object
        valid = False
        print("In MyCustomPlugin.process method")
        result = ValidationResult(
            plugin_name=self.NAME,
            valid=valid,
            validation_messages=[]
        )
        return result

data_obj = {
    "id": "obj1",
    "name": "Object 1",
    "type": "X"
}
validator = Validator(schema="examples/example_schema.yaml", plugins=[{"plugin_class": "MyCustomPlugin", "args": {}])
validator.validate(obj=data_obj, target_class="NamedThing")

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

linkml_validator-0.4.5.tar.gz (12.1 kB view details)

Uploaded Source

Built Distribution

linkml_validator-0.4.5-py3-none-any.whl (14.3 kB view details)

Uploaded Python 3

File details

Details for the file linkml_validator-0.4.5.tar.gz.

File metadata

  • Download URL: linkml_validator-0.4.5.tar.gz
  • Upload date:
  • Size: 12.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.8.15

File hashes

Hashes for linkml_validator-0.4.5.tar.gz
Algorithm Hash digest
SHA256 bce72053c1593a678ccefd28ed5cf4b39a878e78da884b5b41fef65610a23867
MD5 5a07b109dd32dc34e1dd8f048e49fbe2
BLAKE2b-256 d6d72b67879fdd0b77e2451e073ac62d1284b6e88c85ceb919dbad2bf487d6b9

See more details on using hashes here.

Provenance

File details

Details for the file linkml_validator-0.4.5-py3-none-any.whl.

File metadata

File hashes

Hashes for linkml_validator-0.4.5-py3-none-any.whl
Algorithm Hash digest
SHA256 7a9799830cba3e1df203cde86263d8ea85f9ed9f6a45a218e3f752483abac030
MD5 7c14e8848bf595b56a4bd4b3bec4ae73
BLAKE2b-256 87e14a9473997ae94d899008b9ae0c5da0702c7dfdc9ee1dd53c0a00c86873cf

See more details on using hashes here.

Provenance

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page