Mailing-List: contact dev-help@hc.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "HttpComponents Project" <dev@hc.apache.org>
Received-SPF: error (nike.apache.org: local policy)
Message-ID: <1371486246.5115.13.camel@ubuntu>
Subject: Re: Writes to AbstractContentEncoder.completed field
From: Oleg Kalnichevski <olegk@apache.org>
To: HttpComponents Project <dev@hc.apache.org>
Date: Mon, 17 Jun 2013 18:24:06 +0200
In-Reply-To: 
 <CAOGo0Vb2ygT8Z-u7MShSgH_ebPkzse9KFToNSQxsXvFKYuc=xA@mail.gmail.com>
References: 
	<CAOGo0Vb2ygT8Z-u7MShSgH_ebPkzse9KFToNSQxsXvFKYuc=xA@mail.gmail.com>
Content-Type: text/plain; charset="UTF-8"
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit

On Mon, 2013-06-17 at 14:52 +0100, sebb wrote:
> At present, the classes ChunkEncoder and LengthDelimitedEncoder both
> use the code
> 
> this.completed = true;
> 
> This is equivalent to super.complete(); these are the only remaining
> external acceses to the field.
> 
> Any objections to making that change?
> 
> The completed field can then be deprecated so accidental external
> access will be flagged.
> 
> One day we can make it private, which will stop completed from ever
> being reset back to false (which I assume would cause some problems!)

This will probably help make FindBugs reports look prettier but will
hardly make a difference unless we make a decision to stop using direct
access to protected variables consistently across the code base.

Oleg


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