How to enable a new service

Question

I'd really like to create a new search listing page for the wiki, without simply replacing the existing one.
So, I'd like to duplicate the existing search.xsl , rename it search2.xsl and then having modified this one, call either as the fancy takes me. e.g. /groups/foobar/search2/?.....
Can anyone advise as to how new services, I think that's what the wiki calls them, can be 'registered'
ta.

Answer 1

You are correct that we create a virtual XML document structure in memory and >pass it through the XSLT template, which is also turned into a virtual document >structure in memory at startup. We put everything inside a context node, so you >could theoretically pass this through as-is using a template and see what's >available to you.
Forgive me, atom nodes are at the edge of my knowledgebase, but I'm prepared to go away and expand my frontiers. Could you offer any hints as to how I would "pass this through as-is using a template and see what's available"? I feel that this could be very useful to a lot of people.
If I stripped out the bulk of the contents of a default.xsl template, what would I replace it with to create an HTML listing of the elements/attributes that were available?
Another avenue by which you can add functionality without changing the Python >files is via JavaScript. Remember that any JavaScript files you include in your >custom theme will get included in all of your pages. You can use these files to >manipulate the DOM and even look up additional information from the server.
Yes I can see this though my instinct at the moment is to try and leave the functionality on the server. I can see situations where one wouldn't but..
I'm working on a couple of code samples currently (and will put them online >eventually) but if you have something specific in mind that you're trying to >accomplish, I may be able to help you.
We'd love to see some code samples, sooner rather than later, but eventually would be good enough.
I've got a number of ideas and a general feeling that tailoring search, for example, will prove very useful
Specific example.
I want to create a listing of pages that relate to a number of books I've written. A page for each book has been created in the wiki and each has a animage of the front cover of the book. I'd also like to create a 'summary' paragraph for each book. Each book is tagged as a 'book'
So, I want to search for 'book' and create a listing, similar to a current search, but with the image for each book in place of the generic 'star' image followed by the book title (page title) and the summary paragraph.
Search at the moment uses the snippet - but is there any way to control its length or 'end char' that it 'snips' at ?
Could I create a named 'div' in a page that search, or the search.xsl, could recognise? Or is there a more elegant method.
I can get the image by ensuring that each book image is simply called booktitle.jpg which is equivalent to pageTitle.jpg but this is not particularly flexible or in keeping with the spirit of the wiki. Unfortunately the page.plist DOESN'T contain any image/imageName key pairs.
Is there any legitimate way that we can write into the page.plist and then be able to access entries ?