GUIProperties getPropertyValue() throws Exception

[dpulitano]

This shouldn't be done - we should throw the MissingResourceException and IllegalArgumentException instead, or a more specific GUIPropertiesException

[mchao47]

Seems like a reasonable thing to do.

[kood1]

I know we also had conversations around potentially returning null instead of exception. Thoughts?

[bryantrobbins]

I would recommend returning null in this case + adding an API call (or multiple calls) for "getPropertyValueOrDefault" to return a default instead.

Exception propagation for any tests which read a property always annoyed me. I would rather assertNotNull in the test than propagate; but maybe that is a preference.

Also (and this is a bigger impact) If trying to use a properties file in a widget class, you have to try/catch in every place and/or re-throw as WidgetException in order to comply with relevant interfaces.

[mike-glorioso]

I like the idea of avoiding unnecessary exception handling. My only concern would be that there may be existing automation that depends on the current behavior and won't function properly without it. I'd prefer to leave the current functionality with specific exceptions as suggested and add the getPropertyValueOrDefault.

[bryantrobbins]

CCB Feedback from 4/29:

1. "null" feature request is spec'ed out above in separate ticket.
2. The methods should not throw explicit exceptions, as these are RuntimeExceptions. However, the Javadocs for the methods should still annotate "ATthrow" for the possibly thrown RuntimeExceptions.

[kood1]

See #59

[dpulitano]

In which ticket is the null feature request spec'd out? Is the decision to have getPropertyValue() throw a certain RuntimeException if the property is not found, or to return null? If it's a RuntimeException, I will plan to create a new RuntimeException: PropertyNotFoundException, or use MissingResourceException.

[kood1]

I believe @bryantrobbins is referencing #45?