InetAddress.getByName/getAllByName for empty host string

[Chris Hegarty]

Jaikiran,

On 13/04/18 16:29, Jaikiran Pai wrote:
> Hi Chris,
> 
> Thank you creating that JIRA.
> 
> If the fix involves just updating the javadoc, is this something that 
> youwould like me to contribute as a patch? I have a signed and approved 
> OCA, but I will need a sponsor if I do come up with the patch.

A patch containing the specification clarification and an
update to an existing test to cover said specification
clarification would be helpful.

The change will require a CSR, but I can do that on your
behalf.

-Chris.

> -Jaikiran
> 
> 
> On 13/04/18 8:41 PM, Chris Hegarty wrote:
>> Hi Jaikiran,
>>
>> On 13/04/18 15:29, Jaikiran Pai wrote:
>>> The javadoc of InetAddress.getByName(host) and getAllByName(host) 
>>> states that:
>>>
>>> If the host is null then an InetAddress representing an address of 
>>> the loopback interface is returned.
>>>
>>> For non-null values the javadoc explains what the implementation 
>>> does. However, there seems to be no mention of what it does with an 
>>> empty string (length == 0) value for the host param. Right now, the 
>>> implementation seems to treat an empty value the same way it treats 
>>> the host == null case and returns an InetAddress representing the 
>>> loopback address.
>>>
>>> Consider the following example:
>>>
>>> public class InetAddressTest {
>>>      public static void main(final String[] args) throws Exception {
>>>          System.out.println("InetAddress.getByName() for empty string 
>>> returns " + java.net.InetAddress.getByName(""));
>>>          System.out.println("InetAddress.getAllByName() for empty 
>>> string returns "
>>>              + 
>>> java.util.Arrays.toString(java.net.InetAddress.getAllByName("")));
>>>
>>>      }
>>> }
>>>
>>> This outputs:
>>>
>>> InetAddress.getByName() for empty string returns localhost/127.0.0.1
>>> InetAddress.getAllByName() for empty string returns 
>>> [localhost/127.0.0.1]
>>>
>>>
>>> Is it intentional for these APIs to behave this way for empty string?
>>
>> Yes.
>>
>>> If so, should the javadoc be updated to explicitly state this behaviour?
>>
>> Yeah, probably.
>>
>> The following JIRA issue has been filed to track this:
>>   https://bugs.openjdk.java.net/browse/JDK-8201545
>>
>> -Chris.
>