BTW. the attached pom.xml patch changes the version of the plugin to 
1.2.1-tricode. This is purely so I can use the changed plugin directly in my 
project without conflicting with your version numbering. When you apply this 
patch, please update the version to 1.2.2 or something :) I'll then step back 
to the original versioning in my dependencies list.

Thank you for your contribution!

What other types of files are you trying to internationalize?

The maven gettext plugin handles only Java files so far and more changes would 
be needed to support other languages which might also be also outside the scope 
of this project.

When xgettext is invoked from the maven plugin the parameter --language=Java is 
passed to it which tells it to expect java files as input. Also the output of 
the maven plugin are either properties files or Java classes which only make 
sense for Java.

At the moment I'm using the gettext plugin to internationalize FreeMarker 
template files (it's based on JSP so at the end it's just Java :)). As long as 
the translation methods in the template engine are formed like normale Java 
method calls, it works. To get this working I wrote some viewHelper methods for 
FreeMarker, so now I just use ${_("text to be translated")} for singulars and 
${_p("got one", "got two or more", 3)} for plurals (with support for 
MessageFormat argument replacement). 

Hi,

I just released version 1.2.3 of the maven-gettext-plugin that supports files 
with different file endings.

I went with using Maven's FileSet type for specifying extra files that should 
be parsed by xgettext. The optional extra argument is:

<extraSourceFiles>
       <directory>${basedir}</directory>
        <includes>
           <include>**/*.jsp</include>
         </includes>
         <excludes>
           <exclude>**/*.txt</exclude>
         </excludes>
</extraSourceFiles>

I hope this is useful.

Perfect, thanks. I'll upgrade my dependencies to use your new version.