Hi Patrick et al. a trivial point and a javadoc question trivial item in SendCheck test, there's a comment referring to pkt3 & 4 and invalid port -1. This is no longer the case -- afaik it's now "can't send port 0 " in a SocketException ?
I note that there are some java doc changes and wonder is there an opportunity to make some further amendments: for getSocketAddress and setSocketAddress methods the wording "of the remote host" doesn't quite fit, and could be replaced with "of the application or service " in general remote host could be replaced with host, as localhost provides a valid IP address !! references to machine could be replaced with host for consistency, also ? regards Mark ________________________________ From: net-dev <net-dev-boun...@openjdk.java.net> on behalf of Patrick Concannon <patrick.concan...@oracle.com> Sent: Thursday 16 April 2020 13:39 To: Chris Hegarty <chris.hega...@oracle.com>; OpenJDK Network Dev list <net-dev@openjdk.java.net> Subject: Re: RFR[8237890]: 'DatagramPacket::getSocketAddress doesn't specify what happens if address or port are not set' Hi Chris, Thanks for that. I've added the new test cases as requested, and you can find them in the new webrev below. http://cr.openjdk.java.net/~pconcannon/8237890/webrevs/webrev.02/ Kind regards, Patrick On 15/04/2020 16:54, Chris Hegarty wrote: On 15 Apr 2020, at 16:40, Patrick Concannon <patrick.concan...@oracle.com><mailto:patrick.concan...@oracle.com> wrote: ... webrev: http://cr.openjdk.java.net/~pconcannon/8237890/webrevs/webrev.01/ I added myself as reviewer on the CSR. The changes mainly look good. Can you please add, or amend an existing, test for the newly specified default address and port values. -Chris.
