On Jun 2, 2006, at 1:25 PM, Yonik Seeley wrote:
On 6/2/06, Erik Hatcher <[EMAIL PROTECTED]> wrote:
I'll have to see what makes the most sense in terms of where this
will live in the directory structure and how we'll distribute it so
it can be used as a stand-alone API for a client.
Yeah, the fewer dependencies the better.
As for where it lives.... contrib/clients/java?
alongside contrib/clients/[ruby,python,cpp,etc]?
I think this kind of library is more "core" than contrib. This would
be THE way Java clients would interact with Solr, I suspect. But, of
course it doesn't really matter much to me as it'll just be a JAR
file I'll copy into my project once :) [I index with Java, query from
Rails]
Erik
I noticed it
leverages Solr's XML class
which makes this a little more tricky for
someone to easily use in a standalone Java project, but not a big
deal for us to arrange things so all someone would need is a solr-
client.jar or something like that.
Hmmm, yeah. For the Query side of clients, we might want some other
classes like NamedList too?
-Yonik
http://incubator.apache.org/solr Solr, the open-source Lucene
search server
