Otis Gospodnetic schrieb:
I just had a look at the demo and reeeally like it!
I didn't pay enough attention to this thread, though. Is the main concern that
by having a Solr search webapp that is really all in UI and uses your JS
library, the backend Solr server is directly exposed and thus somebody could
peek in the web page source, figure out Solr's address, and start issuing
delete and other damaging requests?
I think somebody mentioned a Servlet Filter. Couldn't we simply supply a
servlet filter that allows only some request URLs, possibly reading those URLs
from an external file, thus allowing easy customization?
Yep I think the main conclusion is that we may provide a "read only" url space
that only serves the json data needed by the js. We need at least SOME access to
solr as we want to get data from it :)
This dynamic stuff looks veeeery juicy.
Question about scalability:
How much is cached either client-side? With every new letter I type, is JS
hitting Solr, or is there some caching (planned) on the client?
In the demo there is no caching yet, so every typing causes a request. The
"real" implementation surely should include some caching etc.
matthias
Danke,
Otis
--
Sematext -- http://sematext.com/ -- Lucene - Solr - Nutch
----- Original Message ----
From: Matthew Runo <[EMAIL PROTECTED]>
To: solr-user@lucene.apache.org
Sent: Thursday, May 29, 2008 12:50:25 PM
Subject: Re: Announcement of Solr Javascript Client
Wow. This is really pretty cool. You're much further along than I
thought you were! I'd love to see this in as an 'official' Solr client.
Thanks!
Matthew Runo
Software Developer
Zappos.com
702.943.7833
On May 29, 2008, at 8:15 AM, Matthias Epheser wrote:
The server was rebooted yesterday without my knowledge, so the jetty
is restarted and should be reachable at
http://lovo.test.dev.indoqa.com/mepheser/moobrowser/
As you can see, this first demo uses widget classes and is built
with mootools.
