CSV Schema Definition and Validation

Install

pip install csv-schema

Usage

usage: csv-schema [-h] [--version] {validate-config,validate-csv,generate-config} ...

CSV Schema

optional arguments:
  -h, --help            show this help message and exit
  --version             show program's version number and exit

Commands:
  {validate-config,validate-csv,generate-config}
    validate-config     Validates the CSV schema JSON configuration file.
    validate-csv        Validates a CSV file against a schema.
    generate-config     Generate a CSV schema JSON configuration file.

Definition

The CSV file schema definition is defined in a JSON file in the following format.

File Definition:

{
  "name": null,
  "description": null,
  "filename": {
    "regex": null
  },
  "columns": []
}

Property	Description
name	The name of the schema.
description	The description of the schema.
filename	Properties for the name of the CSV filename to validate.
columns	List of column definitions.

filename

{
  "regex": null
}

Property	Description
regex	Regular expression to validate the name of the CSV file being validated.

Column Definitions:

string

{
  "type": "string",
  "name": null,
  "required": true,
  "null_or_empty": false,
  "regex": null,
  "min": null,
  "max": null
}

Property	Description
type	The column type.
name	The name of the column.
required	Whether or not the column is required in the file.
null_or_empty	Whether or not the value can be null (missing) or an empty string.
regex	Regular expression to validate the column value.
min	The minimum length of the string. null for no limit.
max	The maximum length of the string. null for no limit.

integer

{
  "type": "integer",
  "name": null,
  "required": true,
  "null_or_empty": false,
  "regex": null,
  "min": null,
  "max": null
}

Property	Description
type	The column type.
name	The name of the column.
required	Whether or not the column is required in the file.
null_or_empty	Whether or not the value can be null (missing) or an empty string.
regex	Regular expression to validate the column value.
min	The minimum value. null for no limit.
max	The maximum value. null for no limit.

decimal

{
  "type": "decimal",
  "name": null,
  "required": true,
  "null_or_empty": false,
  "regex": null,
  "min": null,
  "max": null,
  "precision": 2
}

Property	Description
type	The column type.
name	The name of the column.
required	Whether or not the column is required in the file.
null_or_empty	Whether or not the value can be null (missing) or an empty string.
regex	Regular expression to validate the column value.
min	The minimum value. null for no limit.
max	The maximum value. null for no limit.
precision	The decimal point precision.

enum

{
  "type": "enum",
  "name": null,
  "required": true,
  "null_or_empty": false,
  "values": []
}

Property	Description
type	The column type.
name	The name of the column.
required	Whether or not the column is required in the file.
null_or_empty	Whether or not the value can be null (missing) or an empty string.
values	Fixed set of constants.

Development

Setup

pipenv --three
pipenv shell
make pip_install
make build
make install_local

See Makefile for all commands.

Update README Definitions

The schema definition is auto generated by running scripts/generate_md_help.py. Copy/paste the output into the README.md file.

Testing

Create and activate a virtual environment:
Run the tests: make test

RomainTT/csv-schema-py