OTOOH, it *is* a fairly convenient method. :-) Maybe it is best to keep
toString(Object) and toString(Object, boolean), but drop the rest.
On Sep 9, 2010, at 12:35 PM, Greg Brown wrote:
> OTOH, this is really just a convenience method, and the only place it is used
> in the codebase is in the unit tests. Also, there are a number of overloads
> whose primary purpose appears to be to prevent the caller from having to
> catch SerializationException. These methods are fairly old, and in
> retrospect, this seems misguided.
>
> I think it would probably be best to eliminate these methods.
>
> On Sep 9, 2010, at 12:25 PM, Greg Brown wrote:
>
>> That's a static method, so it can't see the "alwaysDelimitMapKeys" instance
>> variable. However, it probably makes sense to add an overloaded version of
>> this method that take an "alwaysDelimitMapKeys" boolean argument.
>>
>>
>> On Sep 9, 2010, at 11:27 AM, István Lakatos wrote:
>>
>>> Hi,
>>> it seems,that the
>>> org.apache.pivot.json.JSONSerializer.setAlwaysDelimitMapKeys() method has
>>> no effect to the .toString() method's result. If you vill look at the
>>> source code of the above class at apache-pivot-1.5.1 org.apache.pivot.json
>>> package, you find
>>>
>>> public static String toString(Object value) throws SerializationException {
>>> JSONSerializer jsonSerializer = new JSONSerializer();
>>> StringWriter writer = new StringWriter();
>>>
>>> meaning, that a new JSONSerializer object is created with the default
>>> alwaysDelimitMapKeys=false value, so the original setting has no meaning.
>>> I hope I'm mistaken.
>>
>
