Hi Ray, It seems that we are fixing the same problem :)
On Thu, Jul 15, 2010 at 3:49 PM, Ray Chen <clrayc...@gmail.com> wrote: > On 07/15/2010 02:38 PM, Shi Jun Zhang (JIRA) wrote: > >> [ >> https://issues.apache.org/jira/browse/HARMONY-6584?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel] >> >> Shi Jun Zhang updated HARMONY-6584: >> ----------------------------------- >> >> Attachment: HARMONY-6584.diff >> >> The patch contains the fix and the test case. Because >> Java_org_apache_harmony_luni_platform_OSNetworkSystem_readDirect() method is >> invoked in lots of places, we cannot simply throw the >> AsynchronousCloseException. I throw a SocketException with "Asyn Close" >> message then catch this in SocketChannelImpl.java and throw an >> AsynchronousCloseException. >> The test case is just closing the socket channel in another thread while >> it is still reading in the main thread. >> >> >> >>> [classlib][nio]SocketChannel should throw AsynchronousCloseException if >>> the channel is closed in another thread >>> >>> --------------------------------------------------------------------------------------------------------------- >>> >>> Key: HARMONY-6584 >>> URL: https://issues.apache.org/jira/browse/HARMONY-6584 >>> Project: Harmony >>> Issue Type: Bug >>> Components: Classlib >>> Affects Versions: 5.0M14 >>> Environment: all >>> Reporter: Shi Jun Zhang >>> Priority: Minor >>> Fix For: 5.0M15 >>> >>> Attachments: HARMONY-6584.diff >>> >>> >>> According to Java spec, if a SocketChannel is closed in another thread >>> while it is reading, it should throw an AsynchronousCloseException. But >>> harmony doesn't implement this. >>> >>> >> >> > Hi Chance, > Would you please take a look at JIRA-6473, are they same one? > -- 张时骏
