> convertedinput and rawinput are 2 completely seperate things.
> rawinput really caches the raw string data. So that forms can render
> themselfs again without doing any validation or conversion.
why not use getInput()? why not have getInput() use rawInput internally, for caching?
Thats what getValue() is for.
getInput and getInputArray is just an abstraction for the request params.
rawinput sits between the model object and the input
It is not really input it is more output, it overrides the model value (see getValue())
Maybe getInput() could use that value if it is set, i don't think that would be a big problem
i can't think of an situation that rawinput would override new input then when calling getInput()
only problem is that rawinput != exactly getInput because rawinput == getInputAsArray() concatted.
So i gues we then first need to make rawinput a String[] (as i said before)
and let getInputArray() return the rawinput array when set.
>
> ConvertedInput is only there when required check is ok and the
> conversion did work. After that the validators will be runned.
my suggestion is to not access convertedInput anywhere but in getConvertedInput. the validators will run on the result by getConvertedInput. you can return null, or throw exception, or some special String reference to show errors in conversion.
that is the case.
convertedInput will only be set in the right places
and it will only be read in the getConvertedInput()
johan
