Mailing-List: contact issues-help@jmeter.apache.org; run by ezmlm
Precedence: bulk
Reply-To: issues@jmeter.apache.org
From: bugzilla@apache.org
To: issues@jmeter.apache.org
Subject: [Bug 61050] Crash in JMeterUtils due to null ResourceBundle in
 method getResStringDefault while trying to use DurationAssertion
Date: Fri, 28 Apr 2017 05:57:21 +0000
Message-ID: <bug-61050-58895-l5OUhJ22bK@https.bz.apache.org/bugzilla/>
In-Reply-To: <bug-61050-58895@https.bz.apache.org/bugzilla/>
References: <bug-61050-58895@https.bz.apache.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Auto-Submitted: auto-generated
MIME-Version: 1.0
archived-at: Fri, 28 Apr 2017 05:57:27 -0000

https://bz.apache.org/bugzilla/show_bug.cgi?id=3D61050

--- Comment #2 from chuha.billi@gmail.com ---
Hi Phillippe,

I understand the sentiment, but while digging up in the JMeter code I found=
 out
that in in Class JMeterUtils and method getResStringDefault the bundle is
assigned the reference of resources object as ResourceBundle bundle =3D
resources;, now the resources object is initialized only if the locale is s=
et,
if not set then the bundle is null and that then before accessing the bundle
the check of it being not null is not checked. which causes the crash.

now if this is designed like this then setting locale is mandatory but if n=
ot
then I think won't it be a good option to do a null check before using the
object and if found null then setting locale using system default values.??
This was the query in my mind with which I raised the bug.

--=20
You are receiving this mail because:
You are the assignee for the bug.=