Mailing-List: contact lucene-dev-help@jakarta.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Lucene Developers List" <lucene-dev@jakarta.apache.org>
Message-ID: <3FBD0FC8.9050000@detego-software.de>
Date: Thu, 20 Nov 2003 20:02:32 +0100
From: Christoph Goller <goller@detego-software.de>
User-Agent: Mozilla/5.0 (X11; U; Linux i686; de-AT; rv:1.4) Gecko/20030624
MIME-Version: 1.0
To: Lucene Developers List <lucene-dev@jakarta.apache.org>
Subject: Re: CompoundFileReader
References: <20031119201857.83953.qmail@web12704.mail.yahoo.com>
In-Reply-To: <20031119201857.83953.qmail@web12704.mail.yahoo.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit


Otis Gospodnetic schrieb:
> Hello Christoph,
> 
> I don't know enough to judge all changes you made with confidence, and
> don't have the code on this machine to apply the patch and look at it
> properly.  However, it looks like seekInternal(long) should now be left
> with only the conditionals that can cause IOException, and nothing
> else.  Why do we need to keep the body of the method then?

I discussed that already with Dmitry and he was not very passionate
regarding these conditionals. Finally I kept them since they were required
by Dmitry's test. But throwing an exception on seek is
not a behavior implemented by the other subclasses of InputStream.
I decided to remove the conditionals and slightly changed the unit test.

If more index checks are desirable, I would put them into InputStream.
Despite that I added a "read past EOF" check to CSInputStream.readInternal
to guarantee the same behavior as FSInputStream.

> 
> I also see getFilePointer() call, but can't tell where that method is. 
> I'll assume it's from the parent InputStream.

yes.


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