Hello again,
Sorry, I took some time to process everything... I looked at some more
documentation and realized I am confusing documents with cores. Actually, I
was expecting to be able to have a USER core and a CITY core and be able to
relate them somehow. Thanks for the clarification, I will write another
email with my current doubts, but I think I understood the joins now, they
just were not what I was expecting. Maybe I am misunderstanding some
concepts...
Best regards,
Marcelo.
2012/10/24 Marcelo Elias Del Valle <mvall...@gmail.com>
> Shawn,
>
> First of all, thanks a lot for your answer, it was very useful.
> By the content of your email, it seemed to me the /browser is just
> something as a solr admin interface, so now I am confused. I am already
> using SOLRJ in my application and I am currently able to perform a query
> like follows, for example:
> http://localhost:8983/solr/user/select?q=*%3A*&wt=xml
>
> However, I want to use JOINs. I was trying to use the /browse handler
> because I was following the quick start on this page:
> http://wiki.apache.org/solr/Join
> In this wiki page, I couldn't see an example of query joining two
> documents.
>
> If I understood correctly, I don't need /browse, right? I could
> perform a query joining user document with city document, for instance,
> without relying on /browse?
> Do I need to configure anything to be able to use joins? Or the
> plugin comes already installed in solr 4?
> Sorry for the amount of questions. ;-)
>
> Thanks,
> Marcelo Valle.
>
>
> 2012/10/24 Shawn Heisey <s...@elyograg.org>
>
>> On 10/24/2012 8:05 AM, Marcelo Elias Del Valle wrote:
>>
>>> I saw there is some documentation in solr wiki for SearchHandler and
>>> VelocityResponseWriter, which I am trying to digest. However, I saw there
>>> are some configuration fields that aren't there, like this QF field. I am
>>> not sure on how to customize... Should I use only my custom fields there?
>>>
>>
>> Marcelo, the /browse handler that comes with the Solr example is just
>> that -- an example. It's not intended for production use without a lot of
>> customization, and definitely not intended to be directly available to
>> 'regular' people or the Internet. I'm not saying it can't be a useful
>> tool, but nothing in Solr is hardened against abuse, so it should not be
>> directly exposed to attack.
>>
>> Also, the /browse handler configuration is highly tied in with the
>> schema.xml in the example. If you change the schema, you'll probably have
>> to also perform surgery on the SolrItas config, which will likely require
>> an understanding of Velocity. For real help with Velocity, you'd need to
>> consult other resources. Here's some stuff that I was able to find:
>>
>> http://velocity.apache.org/**engine/releases/velocity-1.5/**overview.html<http://velocity.apache.org/engine/releases/velocity-1.5/overview.html>
>> http://velocity.apache.org/**engine/releases/velocity-1.5/**
>> user-guide.html<http://velocity.apache.org/engine/releases/velocity-1.5/user-guide.html>
>> http://velocity.apache.org/**engine/releases/velocity-1.5/**
>> developer-guide.html<http://velocity.apache.org/engine/releases/velocity-1.5/developer-guide.html>
>>
>> If you choose to customize the Velocity config and have questions about
>> the Solr pieces of the puzzle, then this list can probably give you the
>> answers you need.
>>
>> Generally speaking, rather than use /browse, you'll want to access Solr
>> directly from whatever application you have that faces your users, either
>> using constructed URLs and a standard http library, or a solr-specific
>> library that gives you the tools to tell it what you want and handles the
>> URL construction for you. For Java, that would be SolrJ. There are also
>> solr libraries for other languages.
>>
>> Thanks,
>> Shawn
>>
>>
>
>
> --
> Marcelo Elias Del Valle
> http://mvalle.com - @mvallebr
>
--
Marcelo Elias Del Valle
http://mvalle.com - @mvallebr
