If you want to reuse Wikistream/Filter then starting from an existing XML does not make much sense. Indenting an XML is a pretty standard thing using only XML APIs. You can look at what the XAR plugin is doing at https://github.com/xwiki/xwiki-commons/blob/master/xwiki-commons-tools/xwiki.... I would not recommend using WikiStream/Filter for an extension anyway unless you want your extension to work only starting with 6.2 or trigger a lot of dependencies before 6.2 (all the API moved from "wikistream" to "filter/filterstream" naming in 6.2 and also moved to xwiki-commons). Anyway for the record here is a solution with Filter in 6.2: from Filter point of view it would be better to directly serialize the document to XAR XML format. There is no real public API that takes directly a XWikiDocument in input yet (not that it's really hard to do, just did not had the use case yet) but there is one that gets a document reference and then you can give a Writer for example to the XAR module to get your XML: In Java that would be something like:

...

...

...

...

...

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

@Inject @Named(FilterStreamType.XWIKI_INSTANCE) InputFilterStreamFactory inputFilterStreamFactory; @Inject @Named(FilterStreamType.XWIKI_XAR_11) OutputFilterStreamFactory xarFilterStreamFactory; [...] ///////////// // Instance DocumentInstanceInputProperties inputProperties = new DocumentInstanceInputProperties(); // I gess you don't want any history inputProperties.setWithJRCSRevisions(false); inputProperties.setWithRevisions(false); // Indicate which document you want in input EntityReferenceSet entities = new EntityReferenceSet(); entities.includes(myDocumentReference); inputProperties.setEntities(entities); InputFilterStream inputFilterStream = inputFilterStreamFactory.createInputFilterStream(inputProperties); ///////////// // XAR // It format by default but it's possible to disable it with setEncoding(false) XAROutputProperties xarProperties = new XAROutputProperties(); // Set a writer as output StringWriterOutputTarget target = new StringWriterOutputTarget(); xarProperties.setTarget(target); OutputFilterStream outputFilterStream = xarFilterStreamFactory.createOutputFilterStream(xarProperties); ///////////// // Serialization // Do the actual serialization inputFilterStream.read(outputFilterStream.getFilter()); inputFilterStream.close(); outputFilterStream.close(); // You can use StringWriterOutputTarget#getBuffer() to access the written String System.out.println(target.getBuffer()) <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< Note that it's not very hard to do all that in Velocity using the "filter" script service (see https://github.com/xwiki/xwiki-platform/blob/master/xwiki-platform-core/xwik...). On Sat, Sep 6, 2014 at 5:17 PM, Ludovic Dubost <[email protected]> wrote:

Hi,

I'd like to improve the GitHub app to perform the equivalent of mvn xar:format, and I'm looking for the easiest way to do this using groovy in an XWiki instance.

I've seen that the wikistream module is also indenting XML I think using this code:

https://github.com/xwiki/xwiki-commons/blob/master/xwiki-commons-core/xwiki-...

However I did not really understand how to use it. What I have initially is the result of

doc.toXML()

What's the best way to do this ? Knowing that I will want to fallback to no indenting on older XWiki versions.

Ludovic

-- <http://www.xwiki.com/>*Ludovic Dubost* *Founder and CEO* [email protected] skype: ldubost Blog: http://blog.ludovic.orgXWiki SAS organise un petit déjeuner <http://www.xwiki.com/lang/fr/BlogFr/PetitDejeunerXWikiOctobre2014> "retour d'expérience client" le mardi 7 octobre 2014. Inscription gratuite <http://info.xwiki.com/petit-dejeuner-XWiki-octobre-2014.html>. _______________________________________________ devs mailing list [email protected] http://lists.xwiki.org/mailman/listinfo/devs

-- Thomas Mortagne