Dan Nicolici wrote:
Hi!
Please help me with Cocoon site searching. I posted another message, but
got no replies.
I read bits of documentation from where I could find any and formed an
overall image of the concept. At least please tell me if what I pictured
in my mind is correct. This is how I see it.
>>>>----------------------------------------------------------------
Index the content of the site:
- alternatives:
- crawl the site
- LuceneIndexTransformer
--------------------------------------------------------------------------
Search the previous formed index:
- using SearchGenerator
----------------------------------------------------------------<<<<<
I have a working webapp (almost all content is dynamic) that I would
like to enrich with searching capabilities. I am interested in
LuceneIndexTransformer for indexing.
The steps that I imagine I have to follow are:
- create a pipeline that aggregates all the other pipelines
(“master pipeline”);
- transform the resulting xml to a “lucene:index” document;
- transform the previous document with LuceneIndexTransformer
(index automatically gets built);
- use SearchGenerator for searching the index;
Is this a correct scenario? If it is, then it means I understood well so
far, otherwise please stand me corrected.
Yes, this is a scenario that should work. I used it for a project 1.5
years ago.
Assuming it is correct I have some questions about it:
1. The “master pipeline” is supposed to be called manually (like typing
it in the address bar) or can it be automated somehow (i. e. running
once a nigh)?
I suppose you could create a cron job for it. See Antonio's response.
2. What do I need the xml file that remains after LuceneIndexTransformer
created the index for?
It gives you some information on what was indexed. Other than that, you
can ignore it I think.
3. Can you please tell me where I can find a step by step SIMPLE example
of something like this?
Have you already read the Cocook Wiki page about the
LuceneIndexTransformer:
http://wiki.apache.org/cocoon/LuceneIndexTransformer. I followed the
steps on that page and got it working at the first run.
Niels
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
