Mailing-List: contact dev-help@commons.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Commons Developers List" <dev@commons.apache.org>
Received-SPF: neutral (athena.apache.org: local policy)
MIME-Version: 1.0
In-Reply-To: <31cc37360904090143x43975f41qecf40d4f6bbdeef5@mail.gmail.com>
References: <31cc37360904090143x43975f41qecf40d4f6bbdeef5@mail.gmail.com>
From: James Carman <james@carmanconsulting.com>
Date: Thu, 9 Apr 2009 06:43:11 -0400
Message-ID: <f2e8eedf0904090343t11be1662ie26ac61457ec317d@mail.gmail.com>
Subject: Re: [LANG] Null-protection methods
To: Commons Developers List <dev@commons.apache.org>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

On Thu, Apr 9, 2009 at 4:43 AM, Henri Yandell <flamefew@gmail.com> wrote:
> Is the following worth keeping in Lang?
>
> public Foo someJdkMethod(Object input) {
> =A0 =A0if(input =3D=3D null) {
> =A0 =A0 =A0 =A0handleAccordingly();
> =A0 =A0}
> =A0 =A0return input.jdkMethod();
> }
>
> I feel that we've bloated up with these methods, and is it really
> worth it? I still have lots of if(x !=3D null && x.callFoo()) type
> invocations, having a small handful of methods available in Lang to
> let me say if(FooUtils.callFoo(x)) isn't enough to hide all of the
> other times when I might have to worry about it.
>
> Personally I'd like to see these methods removed from Lang [especially
> given Stephen's blogging about null protected setting in some future
> JVM].
>
> An example:
>
> =A0 =A0public static String trim(String str) {
> =A0 =A0 =A0 =A0return str =3D=3D null ? null : str.trim();
> =A0 =A0}
>
> Given that I'm likely to do N things to the string, chances are I'll
> check for null up front and then invoke methods rather than repeatedly
> writing ugly code by using StringUtils.
>
> Any thoughts in favour of these methods?

What do bug catchers such as FindBugs do with situations like these?
Would they still report it as a potential bug?  Or, do they dig into
the FooUtils.callFoo() methods to see if they're doing the null check?
 I'm just curious, I guess.  If FindBugs is still going to report it
as a potential bug, then it would seem like folks would probably be
doing the null check on their own to quiet down FindBugs (not that
everyone uses it of course).

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