Vlang ClassTransformer/Validator

Accepts any map[string]string and transforms into the struct you pass to it.

Example:

import validator

struct BasicObject {
    str string [req;min_length: 10; max_length: 30]
    number int [min: 5; max; 10]
}
...

data := validator.transform_and_validate[BasicObject]({
    "str": "Hello world!"
    "number": "42"
})
your_basic_object := data.value


if data.has_errors() {
    // you can get the errors from data.errors
    // where the key of the map is the field with error, 
    // internally you will have a list of errors
}

Example of usage for a configuration

import dracks.validator {transform_and_validate}

pub struct Config {
	debug           bool = true
	default_locale  string = 'en_US'
	login_demo_user bool
	port            int = 8081
}

pub fn load_config() !Config {
	env := os.environ()
	transform := transform_and_validate[Config](env)
	if transform.has_errors(){
		errors_str := transform.errors.keys().map(fn [transform] (field string) string {
			return '\t${field}: ${transform.errors[field]}\n'
		})

		return error('Invalid configuration\n${errors_str}')
	}
	return transform.value
}


pub fn (self Config) to_json() string {
	return json.encode(self)
}

Notes

• SubStructs are not possible to work yet, for limitations of Vlang
• You cannot flag the attribute required from v, as the tool needs to create an empty object you can use req to force
• floats not implemented yet

Accepted attributes

• max_length: validates the string has not more than the specified length
• min_length: validates the string has not less than the specified length
• max: Checks the number is not greather than specified
• min: checks the number is not lower than specified
• req: make the field required, it will fail if there is no default

Valid types for the structure

• bool
• int
• string