On 4/2/13 6:27 PM, chris...@zoulas.com wrote:
On Apr 2,  5:26pm, kurchi.subhra.ha...@oracle.com (Kurchi Hazra) wrote:
-- Subject: Re: DefaultProxySelector socks override

| Also, I should have clarified why I am changing the test. Since now we
| are defining the socks non-proxy property -
| localhost gets added to the list of non-proxy hosts by default in our
| implementation (as is the case with other protocols
| too). So "localhost" no more acts as the socksProxyHost.

Ah, ok. Makes sense; still the change in the test does not appear to
do anything different?!? Am I missing something?

Hi Christos,

It is ok to push the test changes alongwith the code changes - we do not want
to break our tests as a result of new changes..
I am making an attempt at reading the actual IP address of the machine that
the test is running on, rather than the string "localhost". It is a workaround, the other option is to remove the test - I do not see any other way of automating
what the changed test is trying to check.

- Kurchi

Reply via email to