question: validate a list

Maybe I’m just approaching this all wrong, but I was wondering if the following is possible:

item = {
    'appid': {'type': 'integer', 'min': 1, 'required': True},
    'rank': {'type': 'integer', 'min': 1, 'required': True},
    'weight': {'type': 'float', 'required': True},
}

schema = {
    'type': 'list',
    'minlength': 1,
    'required': True,
    'schema': {
        'type': 'dict', 'schema': item
    }
}

I want to use the above to assert that a list I have, contains at least one element, and that each element of the list adheres to the item schema.

However, the above does not work since schema can only be applied to a dict.

So, as I understand it, I would have to wrap my list in a dict (e.g. {'list': list}), and then change my schema definition to:

schema = {
    'list': {
        'type': 'list',
        'minlength': 1,
        'required': True,
        'schema': {
            'type': 'dict', 'schema': item
        }
    }
}

I would really like to avoid this redundant nesting though but still use cerberus for the list validation, not “plain native” python (e.g. if isinstance(list, collections.Iterable) and len(list) > 0:).

[funkyfuture]

have you looked at http://docs.python-cerberus.org/en/stable/usage.html#schema-list ?

[dynnamitt]

Did you end up with wrapping in object or not?

  doc_ = dict(wrapper=doc) if isinstance(doc, list) else doc