Mailing-List: contact dev-help@jackrabbit.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@jackrabbit.apache.org
Received-SPF: pass (nike.apache.org: domain of gcaj-jackrabbit-dev@m.gmane.org
 designates 80.91.229.2 as permitted sender)
To: dev@jackrabbit.apache.org
From: Christoph Kiehl <christoph@sulu3000.de>
Subject: Re: improving  the scalability in searching
Date: Fri, 17 Aug 2007 18:34:34 +0200
Lines: 31
Message-ID: <fa4io7$r6f$1@sea.gmane.org>
References: <A955EA1F8FE31749AEC8C998082F6C7C01AC1341@hai01.hippo.local>
 <46C5A4AA.7030907@gmx.net> <fa4evo$dor$1@sea.gmane.org>
 <46C5C578.7080003@gmx.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
In-Reply-To: <46C5C578.7080003@gmx.net>
Sender: news <news@sea.gmane.org>

Marcel Reutegger wrote:

> 1) New QueryHandler class
> 2) Introduce parameter in configuration
> 3) Auto-detect in SearchIndex
> 
> I prefer 1) because it makes it explicit. I have reservations regarding 
> 3) because it introduces some magic. I don't like 2) because we probably 
> cannot come up with a sensible name ;)

I'm a bit indifferent about 1) because I think the change is not fundamentally 
enough to justify a new QueryHandler class. Do you have any other plans with the 
new QueryHandler implementation? If I were to implement a SQL based QueryHandler 
solution I would create a new QueryHandler implementation, but not for a small 
change like that. Why don't you like my suggestion to replace the old behaviour 
using the strategy pattern? Just because you have to specify or auto-detect 
which strategy to use? Or is there any other reason?
I don't like 2) as well because I prefer 3) ;). I would like to reduce the 
number of choices for the user (see i.e. 
http://www.joelonsoftware.com/items/2006/11/21.html). The system should just 
work optimally by itself. Which problem do you see in not explicitly stating in 
the configuration which index format is used? The only reason I could come up 
with is for people not knowing which format they use. But I would just log a 
warning if the index is in the old format and that query performance will 
improve if they rebuild their indexes.
I don't want to be academically, I just want the best solution for the user. I 
wouldn't object if you decide to go for 1). It's just that I would like to 
understand your reasoning.

Cheers & a happy weekend ;)
Chris