cocoon-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Unico Hommes" <Un...@hippo.nl>
Subject RE: [PATCH] CocoonCrawler enhancements
Date Mon, 28 Jul 2003 19:31:39 GMT
Hi Upayavira,

I missed your email somehow. Yep this one is used by the Lucene block.
It depends on a link view being available in the sitemap and crawls
cocoon over http. I had started implementing a publisher block in the
same vein as the Lucene block because I couldn't really see how to
achieve it the way we discussed a few weeks back, i.e. within the VM,
without first changing a lot internals. This patch is a step toward
that.

I haven't had time to continue with it unfortunately. Also, the boss
decided on our current publisher - that depends on incompatible usage of
cocoon internals - as we already put some effort into that. But I can't
see how I could change the code easily so that it integrates with the
current cocoon. I hope to continue with the cocoon publisher block in
the fall. Sorry, I should have let you know what was happening before
especially because you showed so much interest before.

In a nutshell the idea I had for the Publisher interface:

interface Publisher extends SingleThreaded {
  
  /**
   * parametrize the publisher with the location where
   * from where to start publishing and the depth to publish
   * to.
   */
  void publish(URL url, int depth);
  
  /**
   * after parametrizing the publisher as above, obtain a 
   * an iterator of Publications on the publish path.
   */
  Iterator iterator();

}

This interface is inspired by the CocoonCrawler interface that I found
very elegantly allows client code control over it through the iterator.
The Iterator could hold Publication objects:

Interface Publication {
   
  Long lastModifiedDate();
  
  Long lastPublicationDate();
  
  void publish();

}

The publisher could eventually also have an implementation that
processes directly on the Cocoon object instead of over http. That were
my initial ideas about that I haven't been able to continue yet. Off
course it could be implemented in a hundred other ways equally elegant
:)

Regards,
Unico

> -----Original Message-----
> From: Vadim Gritsenko [mailto:vadim.gritsenko@verizon.net]
> Sent: maandag 28 juli 2003 20:37
> To: dev@cocoon.apache.org
> Subject: Re: [PATCH] CocoonCrawler enhancements
> 
> Upayavira wrote:
> 
> >Unico,
> >
> >I've seen this crawler before, and wondered what it is used for. Can
you
> >tell me?
> >
> >
> 
> It is used to create a lucene search index. See lucene block,
> SimpleLuceneCocoonIndexerImpl, and samples
> 
> Vadim
> 


Mime
View raw message