When I try to use it, I get the following error:

File "/home/luke/Documents/code/git/notebook/venv/lib/python3.6/site-packages/flask_inputs/inputs.py", line 34, in __init__
for field, validators in input.iteritems():
AttributeError: 'dict' object has no attribute 'iteritems'

AFAIK dict.iteritems isn't available in Python 3

Numeric values fail to validate with, for example, NumberRange built-in validator.
Maybe the structure should be kind of like:

rules = {
    'age': {type:'integer',
            validators:[DataRequired(u'age is required'), NumberRange(1,120)]}
}

and then the internal input representation is a Wtforms IntegerField. This way input is properly coerced.

Another option would be to just directly set type equal to any of the field types that Wtforms provides, but I don't know if the "Field" suffix of all the field types would introduce improper naming, since Flask-inputs validates more than form fields.
Example:

from wtforms.fields import IntegerField

rules = {
    'age': {type:IntegerField,
            validators:[DataRequired(u'age is required'), NumberRange(1,120)]}
}

When there is an error in the during the validation process only the error messages are returned the fields that generated the error is not return along with the validation message

Hello,
Thanks for "flask-inputs" library. It is working on my python2 but on python3, it gives an error 'dict'`` object has no attribute 'iteritems'''

I also search that iteritems is no longer supported in python3, is there other way to use flask-inputs on python3? Or if none, what other ways I can use to validate the json input for my [POST] method request.

Thank you!

You apparently missed adding the actual files to the package, so doing
$ pip install flask-inputs
installs an empty package

Python 3 use dict.items() instead of dict.iteritems()
Would you plan to support python3?
Thanks

pass any script with other than string parameter the library returns the message as same script which can be used as XSS injection.

How to define nested JSON?

i want to know if Cross-Site Request Forgery (CSRF) is prevent when using flask_input cos i know it depends on wtf forms which prevent it , will like to know if the case is the same for flask_input

In order to validate url rules, flask-inputs uses "rule". In the Flask documentation (http://flask.pocoo.org/docs/0.12/api/#url-route-registrations) these are called "url_rule".

Please consider offering also url_rule in flask-inputs for consistency.

jsonschema isn't listed in the requirements.txt which means you can't use the example in the documentation.

Weirdly it is listed in the requirements-dev.txt. It's also only listed in the tests_require section in the setup.py

How can I compare multiple inputs inside a single validator?
e.g. group of coordinates' spanned area must be less than a value

schema = { 'type': 'object', 'properties': { 'size': {'type': 'integer', 'minLength': 1, 'multipleOf': 10 } } }

here is my schema, but in my input, it can accept . and if I input a string the error says that "str" is not a number

Did anyone experience the same case? May I know how did your solve this issue?
Thank you!

I have a form which contains a variable number of fields notated as p1_value, p2_value, p3_value, etc. (e.g. sometimes there may only be p1 and p2, yet other times they may be a p4). How can I represent this when validating the form? Thank you.

form = {
        '_value': [
            Regexp(SOME_PATTERN, message=u'MESSAGE')
        ]
    }

Hey, guys,

Thanks for your effort! I found this package exactly is what I wanted. Here, I want to discuss several improvable functions or missing features.

How do you think about let Flask-Inputs also return the HTTP status code for error handling.

For now, Flask-Inputs only raise error message when validation failed. eg:

class CustomInputs(Inputs):
    headers = {
        'Content-Type': [
            AnyOf(
                ('application/json', ),
                message="Content-Type of Headers must be 'application/json'",
            ),
        ]
    }

@app.route('/api', methods=('GET', 'POST')
def api_inference():
    inputs = CustomInputs(request)

    if not inputs.validate():
        return jsonify(success=False, errors=inputs.errors)

Flask-Inputs works well and return a error message to client, if the request has a wrong Content-Type. But the client still would receive status code 200 which indicates that the client's request was successfully received, understood, and accepted (according to RFC 2616). Actually, this request was just received but accepted. The correct status code should be 415 indicates that the origin server is refusing to service the request because the payload is in a format not supported.

Hence, what I expected is:

class CustomInputs(Inputs):
    headers = {
        'Content-Type': [
            AnyOf(
                ('application/json', ),
                message="Content-Type of Headers must be 'application/json'",
            ),
        ]
    }

@app.route('/api', methods=('GET', 'POST')
def api_inference():
    inputs = CustomInputs(request)

    if not inputs.validate():
        return jsonify(success=False, errors=inputs.errors), inputs.status_code

The major different is in second return variable of last line inputs.status_code. This will make both client/server much easier to debug, check and log information.

I think this is really a valuable and necessary function.

Of course, there are also problems would be troubles when implementing this function:

(1) Priority of status code. If there are multiple wrong validations, which one should be returned. Two choices: a. Protocol sequence. Check headers first, then validate json or form or else things defined in HTTP protocol. Return first error status codes. b. Depending on implementation of codes, like follow valid_attrs = ['args', 'form', 'values', 'cookies', 'headers', 'json', 'rule'] attributes.

(2) The way to define or custom status code for specific validation. Where to define error status code. Here, I provide one solution. eg:

class CustomInputs(Inputs):
    headers = {
        'Content-Type': [
            AnyOf(
                ('application/json', ),
                message="Content-Type of Headers must be 'application/json'",
                errorcode=415,
            ),
        ]
    }

Now, validaors of Flask-Inputs are depending on wtforms packages. If necessary, could we rewrite or reimplement ourself validators based on wtforms.validators. (Yeah, sometimes, I ported several validators directly from source codes, becuase I didn't use package wtforms in my codes.)

This probably isn't the best way. We can discuss to get a better solution to define status code for each validation.

Thanks for reading. Looking forwarding to knowing your opinions. Or whether developer accept pull request in this new feature.

Regards.