commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jörg Schaible <joerg.schai...@gmx.de>
Subject Re: [VOTE] Release JXPath 1.3 based on RC4
Date Sat, 05 Jul 2008 13:00:57 GMT
Forgot to add the path for the POM, just in case you will do another RC.

Jörg Schaible wrote:

> Hi Matt,
> 
> Builds from source and works fine on my compiler zoo except for all
> IBM-JDKs, but that's a different story
> (http://jira.codehaus.org/browse/MNG-3580).
> 
> Some minor issues though in the docs:
> 
> - download page goes nowhere, I suspect this is caused by the current
> location and automatically fixed for the release
> - the links to the older APIs does not work, again I suspect this is fixed
> for the release
> - the user's guide contains a table of content at its top. However, quite
> all of the link anchors do not work
> 
> Issue in the POM:
> 
> - it defines a property "commons.release.version" with value 1.2. This is
> used by the download-page-template.xml to generate the download page in
> xdocs. Therefore currently all liks are generated for JXPath 1.2. Maybe
> the property's value in the POM should be defined as ${project.version} in
> general and its definition can move up to the parent-pom.
> - it declares an optional provided dep to commons-logging, which is simply
> not true, since jxpath does not make usage of commons-logging. However
> commons-beanutils has it as transitive dep in an old version, therefore
> commons-logging should be declared in a dependencyManagement with version
> 1.1.1 to overwrite the old transitive version
> - it declares an optional runtime dep to commons-collection, but it is not
> used by any other dependency and can therefore removed completely (all
> tests run fine without it)
> - I don't under stand why deps are declared with scope provided *and*
> optional. IMHO provided is enough, but it may have also effect on the
> generated dependencies page
> 
> Wiki:
> 
> - talks about release plan for 1.2 ;-)
> 
> 
> Summary:
> -1, simply because of the wrong download links, anything else is not
> critical
> 
> - Jörg


Mime
View raw message