The change looks fine. Do you think it would be possible to make a
test case by using the loopback interface, when it's available and
when it doesn't have a broadcast address, which is probably a very common
situation?
Michael
On 30/05/13 12:53, Chris Hegarty wrote:
The Unix native implementation that determines the network interface's
broadcast address and network prefix/subnet mask, incorrectly ties the
mask to the existence of a broadcast address. This should be changed
so that the network prefix/subnet mask, if available, is exposed
through the Java API, getNetworkPrefixLength, regardless of the
broadcast address.
http://cr.openjdk.java.net/~chegar/7107883/webrev.00/webrev/
Creating a reliable automatic regression test for this be problematic.
Also, there is already an existing test that verify that
getNetworkPrefixLength returns values within specified bounds.
This issue can be easily seen with the loopback interface:
>: ifconfig lo
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
inet6 addr: ::1/128 Scope:Host
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:722 errors:0 dropped:0 overruns:0 frame:0
TX packets:722 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:102346 (102.3 KB) TX bytes:102346 (102.3 KB)
>: binaries/jdk1.7.0_07/bin/java Test lo
<0:0:0:0:0:0:0:1%1>
getBroadcast():null
getNetworkPrefixLength():128
<127.0.0.1>
getBroadcast():null
getNetworkPrefixLength():0
-Chris.
