Mailing-List: contact issues-help@commons.apache.org; run by ezmlm
Precedence: bulk
Reply-To: issues@commons.apache.org
Message-ID: <2985905.126841288297644303.JavaMail.jira@thor>
Date: Thu, 28 Oct 2010 16:27:24 -0400 (EDT)
From: "Oliver Heger (JIRA)" <jira@apache.org>
To: issues@commons.apache.org
Subject: [jira] Resolved: (CONFIGURATION-424) HierachicalINIConfiguration
 has problems with configurations only containing a global section
In-Reply-To: <21897286.169801287176675700.JavaMail.jira@thor>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable


     [ https://issues.apache.org/jira/browse/CONFIGURATION-424?page=3Dcom.a=
tlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Oliver Heger resolved CONFIGURATION-424.
----------------------------------------

       Resolution: Fixed
    Fix Version/s: 1.7

A fix was applied to subversion in revision 1028460. Thanks for the report.

> HierachicalINIConfiguration has problems with configurations only contain=
ing a global section
> -------------------------------------------------------------------------=
--------------------
>
>                 Key: CONFIGURATION-424
>                 URL: https://issues.apache.org/jira/browse/CONFIGURATION-=
424
>             Project: Commons Configuration
>          Issue Type: Bug
>          Components: Format
>            Reporter: Andr=C3=A9 Weihe
>            Assignee: Oliver Heger
>            Priority: Critical
>             Fix For: 1.7
>
>
> HierachicalINIConfiguration#getSections() does not correctly deal with co=
nfigurations containing only a globalSection() leading for example to a con=
figuration containing only a global section to be saved as a zero length fi=
le.
> The current code only acknowledges a global section when at least one oth=
er named section is present, if none is present it just returns an empty se=
t.
> Since HierachicalINIConfiguration#save() saves only the contents of the s=
et returned by getSections() this leads to an ini file being loaded and sav=
ed again to be saved as a zero length file resulting in the loss of all dat=
a in the file (thus the critical priority).
> Workaround:
> Create at least one named section
> Proposed fix (sourcecode):=20
> Set sections =3D new ListOrderedSet();
> for (Iterator it =3D getRootNode().getChildren().iterator(); it.hasNext()=
;)
> {
>     ConfigurationNode node =3D (ConfigurationNode) it.next();
>     if (isSectionNode(node))
>     {
>         sections.add(node.getName());
>     }
>     else
>     {
>         sections.add(null);
>     }
> }
> return sections;
> Beware:
> I have NOT tested this (at all).
> It's just the way I would write this method instead of the way it was wri=
tten before. But since I don't know the big picture maybe there was a (good=
) reason to write the current version the way it is now so my seemingly eas=
y fix could not work correctly.

--=20
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.