RFR: 8237967: No proper error message when --runtime-image points to non-existent path

[Andy Herrick]

I sent this in reply to Alexey, needed to send it to alias:

On 2/6/2020 4:54 PM, Andy Herrick wrote:
> I had fixed that and failed to push the revised webrev [4]
>
> [4] http://cr.openjdk.java.net/~herrick/8237967/webrev.03/
>
> /Andy
>
> On 2/6/2020 11:46 AM, Alexey Semenyuk wrote:
>> Andy,
>>
>> defaultSupported() is defined, but not called form the test. What is 
>> the point of this method?
>>
>> - Alexey
>>
>> On 2/6/2020 8:42 AM, Andy Herrick wrote:
>>> revised fir at [3] to check for non-existent resource-dir as well as 
>>> runtime-image dir.
>>>
>>> [3] http://cr.openjdk.java.net/~herrick/8237967/webrev.02/
>>>
>>> /Andy
>>>
>>> On 2/5/2020 7:46 PM, Andy Herrick wrote:
>>>> OK - this makes sense - will update tomorrow.
>>>>
>>>> /Andy
>>>>
>>>> On 2/5/2020 4:24 PM, Alexander Matveev wrote:
>>>>> Hi Andy,
>>>>>
>>>>> Fix looks good, but I think we should do same for --resource-dir. 
>>>>> I do not think that non-existing directory should be same as 
>>>>> empty. User might misspell directory name and it will not be clear 
>>>>> for user why we not using information from --resource-dir.
>>>>>
>>>>> Thanks,
>>>>> Alexander
>>>>>
>>>>> On 2/5/2020 10:23 AM, Andy Herrick wrote:
>>>>>> This simple fix [1] to jpackage bug [2] simply adds a proper 
>>>>>> error message when --runtime-image points to a non-existent 
>>>>>> directory path.
>>>>>>
>>>>>> [1] http://cr.openjdk.java.net/~herrick/8237967/webrev.01/index.html
>>>>>>
>>>>>> [2] https://bugs.openjdk.java.net/browse/JDK-8237967
>>>>>>
>>>>>> /Andy
>>>>>>
>>>>>
>>