rhinoerror v1.0.3

RhinoError

The RhinoError constructor takes the following arguments:

1. code: A string / number that can be used to identify the error.
2. message: A string that can be used to provide an explanation to the user
3. cause: An Error object that caused the error, or null (optional).
4. debugInfo: debug information for developers (optional)
5. userInfo: debug information for users (optional)

import RhinoError from 'rhinoerror';
try {
  JSON.parse(input)
} catch(error) {
  throw new RhinoError('UserInputError', 'you provided an invalid JSON file', error, null, {input});
}

The default value of the name property of a RhinoError is 'RhinoError'.

If the type of the code argument is a number, we'll try to set the name property of the error to the name of the appropriate HttpCode if it exists. For example, passing code as 403 will set the name of the error to 'HttpErrorForbidden'.

If the type of the code argument is a string, we'll set the name property of the error to the value of the code argument.

The stack trace of the cause error is preserved when it is wrapped in a RhinoError.

The debugInfo and userInfo properties are used to separate information meant for internal use by developers to debug the error and information meant for display to end users.

Typically, these are both arbitrary JSON objects. In the context of a web application, debugInfo is logged for debugging later by the developers; and userInfo is returned in the API's response to explain to the user what just happened.

You wouldn't need to add any debugInfo for 400-class (user induced) errors, whereas 500-class (application fault) errors usually need debugInfo to be logged.