[Python] JSON validation using Voluptuous

How to perform JSON value validation using the Voluptuous library in Python.

What is Voluptuous?

Validation library such as JSON of Python.

Official repository: https://github.com/alecthomas/voluptuous

Voluptuous, despite the name, is a Python data validation library. It is primarily intended for validating data coming into Python as JSON, YAML, etc.

As mentioned in the description in the repository, it's a great name.

voluptuous

1 Indulge in lust, drown in liquor, sensual.
2 Sensual, sexy, adorable, sexy.
3 <Feeling, etc.> Comfortable.

install

Python 2.6 and above

Can be installed with pip.

# pip install voluptuous

Python 2.5 Although it is not officially supported, it can be installed and operated at this article level with Install Voluptuous with Python2.5.

Sample program A sample program that shows the operation. Where the Schema declaration in get_schema is at the core. This is the definition of JSON schema in Voluptuous.

from __future__ import with_statement
import re
import sys
import json  # Python 2.5 is simplejson
import six
from voluptuous import Schema, Required, Any, Range, Invalid, ALLOW_EXTRA


def validate_io_size(v):
	if re.search("^[0-9]+[bkm]", v) is None:
		raise Invalid("not a valid value (%s)" % str(v))


def get_schema():
	schema = Schema({
		"comment"					: six.text_type,
		Required("operation")		: Any("read", "write"),
		"thread"					: Range(min=1),
		Required("io_size")			: validate_io_size,
		Required("access_percentage")	: Range(min=1, max=100),
	}, extra=ALLOW_EXTRA)
	
	return schema


def main():
	schema	= get_schema()
	
	with open(sys.argv[1], "r") as fp:
		dict_sample	= json.load(fp)
	
	schema(dict_sample)
	six.print_(dict_sample)
	
	return 0


if __name__ == '__main__':
	sys.exit(main())

Details

Voluptuous term

Validation method

One of the following

1. Direct specification of possible values. It is also possible to have a width with Any, Range, etc.
2. Class (Int, float, etc., six.text_type can be specified) --Any value is OK as long as the classes match
3. Specify the validation function --This is the validate_io_size function mentioned above --The validation function throws an Invalid exception if the condition is not met

Operation example

environment

• Python 2.5/2.7
• voluptuous 0.8.8

Normal operation example

valid1.json

{
	"comment"			: "comment",
	"operation"			: "write",
	"thread"			: 8,
	"io_size"			: "8k",
	"access_percentage"	: 100
}

# python voluptuous_sample.py valid1.json
{u'comment': u'comment', u'operation': u'write', u'access_percentage': 100, u'thread': 8, u'io_size': u'8k'}
#

Example of normal operation (required parameters omitted)

valid2.json

{
	"operation"			: "write",
	"io_size"			: "8k",
	"access_percentage"	: 100
}

# python voluptuous_sample.py valid2.json
{u'operation': u'write', u'access_percentage': 100, u'io_size': u'8k'}

No error occurs even if there are no required parameters.

Error operation example

invalid.json

{
	"comment"			: "comment",
	"operation"			: "<invalid value>",
	"thread"			: 8,
	"io_size"			: "8k",
	"access_percentage"	: 100
}

# python voluptuous_sample.py invalid.json
Traceback (most recent call last):
  File "voluptuous_sample.py", line 36, in <module>
    sys.exit(main())
  File "voluptuous_sample.py", line 29, in main
    schema(dict_sample)
  File "/usr/local/lib/python2.7/dist-packages/voluptuous.py", line 337, in __call__
    return self._compiled([], data)
  File "/usr/local/lib/python2.7/dist-packages/voluptuous.py", line 635, in validate_dict
    return base_validate(path, iteritems(data), out)
  File "/usr/local/lib/python2.7/dist-packages/voluptuous.py", line 471, in validate_mapping
    raise MultipleInvalid(errors)
voluptuous.MultipleInvalid: not a valid value for dictionary value @ data[u'operation']
#

If there is a parameter that does not match the schema, voluptuous.MultipleInvalid will be sent and I get a message that the value of the parameter (operation in the example) is incorrect.

Error operation example (when validation function is specified)

invalid2.json

{
	"comment"			: "comment",
	"operation"			: "write",
	"thread"			: 8,
	"io_size"			: "a8k",
	"access_percentage"	: 100
}

# python voluptuous_sample.py invalid2.json
Traceback (most recent call last):
  File "voluptuous_sample.py", line 38, in <module>
    sys.exit(main())
  File "voluptuous_sample.py", line 31, in main
    schema(dict_sample)
  File "/usr/local/lib/python2.7/dist-packages/voluptuous.py", line 337, in __call__
    return self._compiled([], data)
  File "/usr/local/lib/python2.7/dist-packages/voluptuous.py", line 635, in validate_dict
    return base_validate(path, iteritems(data), out)
  File "/usr/local/lib/python2.7/dist-packages/voluptuous.py", line 471, in validate_mapping
    raise MultipleInvalid(errors)
voluptuous.MultipleInvalid: not a valid value (a8k) for dictionary value @ data[u'io_size']

This is also the case when the validation function is specified.