[
https://issues.apache.org/jira/browse/NET-437?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13182864#comment-13182864
]
Sebb commented on NET-437:
--------------------------
I'm not quite sure why it's important to check __bytesAvailable?
Perhaps the condition should be coded as follows:
{code}
if (__threaded) {
return __bytesAvailable;
} else {
return __bytesAvailable + super.available();
}
{code}
This should return a value closer to the actual number available.
> TelnetInputStream doesn't support non-blocking IO when reader thread is not
> enabled
> -----------------------------------------------------------------------------------
>
> Key: NET-437
> URL: https://issues.apache.org/jira/browse/NET-437
> Project: Commons Net
> Issue Type: Bug
> Components: Telnet
> Affects Versions: 3.0.1
> Environment: Java 6 + commons-net-3.0.1
> Reporter: Gavin Camp
> Original Estimate: 1h
> Remaining Estimate: 1h
>
> When the telnet client is used without allowing it to create it's own reader
> thread (i.e. setReaderThread(false)) then the TelnetInputStream.available()
> method will always return 0 bytes available. This makes non-blocking IO
> impossible as you need to actualy call read to get the data without knowing
> if it will block or not.
> This fix to the available method in
> org.apache.commons.net.telnet.TelnetInputStream, seems to fix the issue, and
> should work for reader threads as well:
> {noformat}
> @Override
> public int available() throws IOException
> {
> // Critical section because run() may change __bytesAvailable
> synchronized (__queue)
> {
> if (__bytesAvailable == 0 && !__threaded) {
> return super.available();
> } else {
> return __bytesAvailable;
> }
> }
> }
> {noformat}
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
