incubator-jspwiki-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ethan Larson <>
Subject Re: Embedding JSPWiki 2.6
Date Fri, 04 Jan 2008 06:21:46 GMT
Ok, I created a dummy page provider and a dummy authorizer and I got a 
lot farther. I don't even need a MemoryPageProvider since I all I need 
is the output (thanks just the same Florian - it was instructive).  I 
actually got translated output.  The problem is that I had to modify the 
source code to do it.  I had to comment out line 532 of WikiEngine:

//m_authorizationManager.initialize( this, props );

As near as I can tell, there's no way to create an authorization manager 
that doesn't involve a jspwiki.policy under WEB-INF.  However, since I'm 
running it as a standalone app, I don't have a web container and 
therefore no WEB-INF.  I could create this under the working directory, 
but I really don't want to put blank, unused metadata in my app.  Is 
there any way to configure this such that I can start the authorization 
manager without a jspwiki.policy?

On a broader note, I'd be over the moon if this were an easier process.  
JSPWiki seems to be the most actively developed and feature-rich Java 
wiki there is, and has support for plugins and filters which I will 
eventually need.  If there were an easy way to run the wiki translation, 
complete with plugins and filters, that didn't involve a web container 
and any extra memory/disk usage, it could broaden the usage quite a 
bit.  I've looked at other java wiki translators out there, and none of 
them are doing a good job of the features/active development/ease of 
standalone combo (Radeox, Bliki, VQWiki to name a few).  Other 
forum/mailing list posts confirm there is a demand.

Thanks for all your help,

P.S. -- Here's my current code for anyone reading this in the future:

Properties props = new Properties();

WikiEngine engine = new WikiEngine(props);

WikiContext context = new WikiContext(engine, new WikiPage(engine, "test"));

System.out.println("output: \n" + engine.textToHTML(context, "this is a 
test\n\n* more stuff"));

MyPageProvider and MyAuthorizer are both empty implentations of the 
interfaces.  Just return an empty List for MyPageProvider.getAllPages.

Janne Jalkanen wrote:
> Yup, the problem is that there needs to be *some* kind of a page 
> provider, because the system needs to check if e.g. a page exists or 
> not when it encounters a link.  The generated HTML differs in each case.
> A dummy provider will do just fine, e.g. the MemoryPageProvider.
> /Janne
> On Jan 2, 2008, at 00:23 , Florian Holeczek wrote:
>> Hi Ethan,
>> maybe this can help you:
>> Regards,
>>  Florian
>>> The problem with the page directory is that I don't want one. I will
>>> be managing the input/output of the text myself. I really just want
>>> to give some wiki markup to the parser and get back html. Is there
>>> currently a way to do this?

View raw message