Mailing-List: contact dev-help@hc.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "HttpComponents Project" <dev@hc.apache.org>
Received-SPF: pass (nike.apache.org: local policy includes SPF record at
 spf.trusted-forwarder.org)
Message-ID: <1395752725.10953.18.camel@ubuntu>
Subject: Re: Java 1.5 -> 1.6 or 1.7? was Re: java 1.5?
From: Oleg Kalnichevski <olegk@apache.org>
To: HttpComponents Project <dev@hc.apache.org>
In-Reply-To: <20140325114800.GA3447@localhost.yandex.net>
References: 
	  <CAF=Pa5_38Oh2iW2RrPoVotr9niYuGcwFPAZ90S9p-S6FzZuRQw@mail.gmail.com>
	 <CACZkXPxGYV2PGkGAk=eqNgotGpDif2bf-pTdKrzY4ArcyWPoTw@mail.gmail.com>
	 <1379157734.21618.11.camel@ubuntu> <-1205251862486749141@unknownmsgid>
	 <1379163630.21618.12.camel@ubuntu>
	 <20140325114800.GA3447@localhost.yandex.net>
Content-Type: text/plain; charset="UTF-8"
Date: Tue, 25 Mar 2014 14:05:25 +0100
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit

On Tue, 2014-03-25 at 15:48 +0400, Dmitry Potapov wrote:
> Gentlemen,
> 
> Have you came to any conclusion concerning this?
> Currently AbstractMultiworkerIOReactor.execute(...) has finally block which
> calls doShutdown() which can throw exception itself (see HTTPCORE-376), so
> all information about current exception will be lost.

I/O reactors are complex beasts. If things go sideways, there can be
multiple exceptions in multiple places. I am not sure suppressed
exceptions introduced in Java 1.7 would help us much here. For that
reason the standard I/O reactors maintain an audit log. Suppressed
exceptions should be logged there.

Oleg 

> One of the ways to fix this and have information concerning both exception is
> to use suppressed exceptions idiom implemented in Java 7 (probably with
> conjunction with try-with-resources), so I wish to know is it allowed to use
> Java 7 features in future patches.
> 


---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org