niagads-metadata-validator 0.2.0

JSON Schema based validation of dataset metadata developed to support submissions to the NIAGADS Data Sharing Service

NIAGADS JSON Schema based metadata validation tools

This tool allows the user to perform JSON Schema-based validation of a sample or file manifest metadata file arranged in tabular format (with a header row that has field names matching the validation schema).

The tool works for delimited text files (.tab, .csv., .txt) as well as excel (.xls, .xlsx) files.

This tool can be run as a script or can also be imported as a module. When run as a script, results are piped to STDOUT unless the --log option is specified.

Requirements

• Python: >3.12,<4.0

---

Usage

command-line

Run with the --help option to get full USAGE information

validate-metadata --help

module

Import package into your python script.

import niagads.metadata_validator_tool.core as mv_tool

Use mv_tool.initialize_validator to initialize and retrieve a validator object for further manipulation. Use mv_tool.run to initialize and run a validation with default configuration. See validator documentation for more information about validator properties and member functions.

Example code, with schema and metadata files are availble in the code repostory: examples/niagads-metadata-validator.

---

API Reference

---

function get_templated_schema_file

get_templated_schema_file(dir: str, template: str) → str

Verify that templated schema file {schemaDir}/{vType}.json exists.

Args:

• path (str): path to directory containing schema file
• template (str): template name

Raises:

• FileExistsError: if the schema file does not exist

Returns:

• str: schema file name

---

function get_templated_metadata_file

get_templated_metadata_file(
    prefix: str,
    template: str,
    extensions: List[str] = ['xlsx', 'xls', 'txt', 'csv', 'tab']
) → str

Find metadata file based on templated name {prefix}{validator_type}.{ext}.

Args:

• path (str): file path; may include prefix/file pattern to match (e.g. /files/study1/experiment1-)
• template (str): template name
• extensions (List[str], optional): allowable file extensions. Defaults to ["xlsx", "xls", "txt", "csv", "tab"].

Raises:

• FileNotFoundError: if metadata file does not exist

Returns:

• str: metadata file name

---

function initialize_validator

initialize_validator(
    file: str,
    schema: str,
    metadataType: MetadataValidatorType,
    idField: str = None
) → Union[BiosourcePropertiesValidator, FileManifestValidator]

Initialize and return a metadata validator.

Args:

• file (str): metadata file name
• schema (str): JSONschema file name
• metadataType (MetadataValidatorType): type of metadata to be validated
• idField (str, optional): biosource id field in the metadata file; required for BIOSOURCE_PROPERTIES validation. Defaults to None.

Raises:

• RuntimeError: if metadataType == 'BIOSOURCE_PROPERTIES' and no idField was provided
• ValueError: if invalid metadataType is specified

Returns:

• Union[BiosourcePropertiesValidator, FileManifestValidator]: the validator object

---

function run

run(
    file: str,
    schema: str,
    metadataType: str,
    idField: str = None,
    failOnError: bool = False
)

Run validation.

Validator initialization fully encapsulated. Returns validation result.

Args:

• file (str): metadata file name
• schema (str): JSONschema file name
• metadataType (MetadataValidatorType): type of metadata to be validated
• idField (str, optional): biosource id field in the metadata file; required for BIOSOURCE_PROPERTIES valdiatoin. Defaults to None.
• failOnError (bool, optional): raise an exception on validation error if true, otherwise returns list of validation errors. Defaults to False.

Returns:

• list: list of validation errors

---

class MetadataValidatorType

Enum defining types of supported tabular metadata files.

BIOSOURCE_PROPERTIES = '''biosource properties file;
a file that maps a sample or participant to descriptive properties
(e.g., phenotype or material) or a ISA-TAB-like sample file'''

FILE_MANIFEST = "file manifest or a sample-data-relationship (SDRF) file"

---

This file was automatically generated via lazydocs.