Path: csiph.com!x330-a1.tempe.blueboxinc.net!usenet.pasdenom.info!aioe.org!eternal-september.org!feeder.eternal-september.org!.POSTED!not-for-mail
From: Stanimir Stamenkov <s7an10@netscape.net>
Newsgroups: comp.lang.java.help
Subject: Re: TeeOutputStream?
Date: Tue, 07 Jun 2011 00:45:17 +0300
Organization: A noiseless patient Spider
Lines: 26
Message-ID: <isjhpb$5bu$1@dont-email.me>
References: <YVzGp.501$_I7.309@newsfe08.iad> <ihpmu61t1pdfja3rsvh1f8rqaoafqof838@4ax.com> <isgvlo$8e1$1@dont-email.me> <jcSdnTfDIfrGinHQnZ2dnUVZ_hmdnZ2d@earthlink.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Injection-Date: Mon, 6 Jun 2011 21:45:15 +0000 (UTC)
Injection-Info: mx04.eternal-september.org; posting-host="VdcnnljfizG2LUdZ5iOA6g"; logging-data="5502"; mail-complaints-to="abuse@eternal-september.org";	posting-account="U2FsdGVkX1+C0pFp+BOCH9z/Mq2ttPHK"
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:2.0.1) Gecko/20110531 SeaMonkey/2.1pre
In-Reply-To: <jcSdnTfDIfrGinHQnZ2dnUVZ_hmdnZ2d@earthlink.com>
X-Face: )>>ChyF_H<b)u~Zjgo/=wa~;=qyW%.F\L.d^fKL[;y\=tY\]M}2t(a^;PKS}9g|k@\vkA<P Q|4?kcJ52334f:CaCrQZ=]D~txPPh6[y{xHkZ+4/KPKZ~|*K#?EqeP0W]iU*Ldy-hyjh0)N4c.I<m) K}GsUUe0)~24Xp`Jt
Cancel-Lock: sha1:EEseAPmzhlu57FfhR4nxDZfA71c=
Xref: x330-a1.tempe.blueboxinc.net comp.lang.java.help:752

Sun, 05 Jun 2011 16:55:09 -0700, /Patricia Shanahan/:
> On 6/5/2011 3:23 PM, Stanimir Stamenkov wrote:
>> Sun, 05 Jun 2011 12:23:09 +0100, /rossum/:
>>
>>> public void close() throws IOException {
>>>
>>>    IOException firstFail = null;
>>>    IOException secondFail = null;
>>>
>>>    try {
>>>      try {
>>>        os1.close();
>>>      } catch (IOException ioe1) {
>>>        firstFail = ioe1;
>>
>> Wouldn't this approach fail in face of RuntimeException or Error thrown?
>
> There may be a case for saying that the operation has failed as a whole,
> and should be abandoned with the appropriate throw, if there is a
> problem other than an IOException from one of the streams.

But then, shouldn't an effort to close all resources be made before 
letting the first throw pop up the stack, in most cases?

-- 
Stanimir