Integer max/min errors are reports as floats

[jmcnamara]

Hi,

Integer maximum and minimum exceptions are reported as floats.

Here is an example based on one of the examples in the docs:

import json
import validictory

data = json.loads(''' {"result": 10, "resultTwo": 12}''')

schema = {
    "properties": {
        "result": { # passes
            "minimum": 9,
            "maximum": 10
        },
        "resultTwo": { # fails
            "type": "integer",
            "minimum": 13
        }
    }
}

try:
    validictory.validate(data, schema)
except validictory.validator.FieldValidationError as e:
    print e

Running this give:

Value 12 for field 'resultTwo' is less than minimum value: 13.000000

While this isn't strictly incorrect it looks a little odd.

Replacing the format string %f with %g in the error handler would probably be a sufficient fix rather than varying the format string based on the validation type.

Regards,

John

[jmcnamara]

It looks like I had a slightly older version of valedictory on the machine I was testing on. This issues doesn't occur on the latest PyPi version or the GitHub master.

Sorry for the noise. Closing.

[jmcnamara]

Brief followup:

I had installed the module as follows:

$ pip install validictory

And got the 0.9.3 release. Installing as follows picked up the latest release:

$ pip install --pre validictory