Hello Guido,
A new RFE has been entered in order to handle your welcomed suggestion :
http://restlet.tigris.org/issues/show_bug.cgi?id=1236.
Thansk for your help.
Best regards,
Thierry Boileau
Hi,
> the methods mentioned above in 2.1snapshot-jse all use a similar approach:
> An extra thread is started that writes a representation to a pipe's sink
> and
> the pipe's source is returned.
>
> In a server environment where representations are not only served
> statically, but processed in one way or another (cached to disk,
> reformatted, analysed, accumulated etc.) while being sent to the client
> these methods are used heavily.
>
> BTW, we try to avoid the BioUtils methods as the NioUtils is using the >10x
> faster PipeChannel.
>
> My suggestion is to modify the extra thread's run() method to
>
>
> SomeWriter writer = null
> try {
> writer = pipesink
> representation.write(writer)
> } catch(IOException e) {
> log e
> } finally {
> try { representiaton.exhaust } catch (Exception e) {}
> try { representiaton.release} catch (Exception e) {}
> try { if (null != writer) writer.close } catch (Exception e) {}
> }
>
> I think this could reduce the risk of resource leaks (threads, sockets).
>
> Guido Schmidt.
>
>
> --
> View this message in context:
> http://restlet-discuss.1400322.n2.nabble.com/Suggestion-modify-B-N-ioUtils-getChannel-Stream-Reader-tp5981336p5981336.html
> Sent from the Restlet Discuss mailing list archive at Nabble.com.
>
> ------------------------------------------------------
>
> http://restlet.tigris.org/ds/viewMessage.do?dsForumId=4447&dsMessageId=2701834
>
------------------------------------------------------
http://restlet.tigris.org/ds/viewMessage.do?dsForumId=4447&dsMessageId=2702994
