Put the data in two different fields. You can also include
a type field and use the fq=type:type1 or fq=type:type2
if you need to distinguish them separately.
Best,
Erick
On Sun, Sep 15, 2013 at 11:00 AM, Baskar Sikkayan <baskar....@gmail.com>wrote:
> Hi Erick,
> Thanks a lot for your reply. I am still not clear.
> I will have 2 different searches.
> So, there will be 2 different kind of documents with different fields.
>
> But the example below, gives me a impression that
>
> SolrInputDocument doc = new SolrInputDocument();even there are 2 diff
> searches, the data is getting stored in the same
>
> doc.addField("cat", "book");
> doc.addField("id", "book-" + i);
> doc.addField("name", "The Legend of the Hobbit part " + i);
> server.add(doc);
>
> If i use the same doc, how to diff data for 2 different searches.
> Could you explain me clear on this??????//
>
>
> Thanks,
> Baskar.S
>
>
>
> On Sun, Sep 15, 2013 at 7:57 PM, Erick Erickson <erickerick...@gmail.com
> >wrote:
>
> > Listen to Wunder. You really have to switch
> > gears and think in terms of searching rather
> > than database tables.
> >
> > First, there's no requirement that all documents
> > have the same fields. There's very little penalty
> > for this.
> >
> > Second, you have to get over normalized data.
> > It's unclear how many employees/records you have,
> > but unless it's a really big set of records (i.e multiple
> > tens of millions) you don't have to worry much
> > about saving space.
> >
> > BTW, your java example works, but you do NOT
> > want to commit after every document. Configure
> > your auto soft and hard commits to do this on a
> > time-based basis.
> >
> > So I'd recommend you approach it differently.
> > Don't get stuck on your current data model. Instead
> > define the searches you want to make and _then_
> > decide the appropriate fields in the docs.
> >
> > But to your question, you can certainly create
> > two cores. That's a good strategy if there is
> > little data shared between the two types of users.
> > It's probably conceptually cleaner as well.
> >
> > Best,
> > Erick
> >
> >
> > On Sun, Sep 15, 2013 at 8:18 AM, Baskar Sikkayan <baskar....@gmail.com
> > >wrote:
> >
> > > Hi,
> > > Thank you very much for your reply.
> > > Let me clearly explain my requirement.
> > >
> > > Its a kind of job site.
> > >
> > > I have 2 tables that need to be added in solr.
> > >
> > > 1) employee ( id, name, skills, location, .....)
> > > 2) job_postings ( id, name, job title, description, salary,
> date_posted,
> > > ....)
> > >
> > >
> > > Here, there are 2 different kind of searches in my application.
> > >
> > > 1) employee search by employer
> > > 2) job postings search by employee
> > >
> > > New row can be added at any time. Also the existing row can be altered
> at
> > > any time. For example, employee mobile number can be changed at any
> time
> > > and the solr data has to be updated.
> > >
> > > Here there are 2 different searches. So i need to maintain two
> different
> > > kind of indexes in solr.
> > >
> > > The java example says, to add a document, below is the sample code. My
> > > doubt is, how can i add two different kind of search data here.
> > > Should i create a 2 different core for this 2 different searches?
> > >
> > > SolrInputDocument doc = new SolrInputDocument();
> > >
> > > doc.addField("cat", "book");
> > > doc.addField("id", "book-" + i);
> > > doc.addField("name", "The Legend of the Hobbit part " +
> i);
> > > server.add(doc);
> > > server.commit();
> > >
> > > Thanks,
> > > Baskar.S
> > >
> > >
> > > On Sun, Sep 15, 2013 at 9:14 AM, Walter Underwood <
> wun...@wunderwood.org
> > > >wrote:
> > >
> > > > Solr does not have tables, and you can't add an index.
> > > >
> > > > Solr's data model is flat, like a single table with lots of columns.
> > > Think
> > > > about creating a view where each row has all the information for one
> > > search
> > > > result. It includes everything that is searched (indexed=true in the
> > > > schema) and returned (stored=true in the schema).
> > > >
> > > > wunder
> > > >
> > > > On Sep 14, 2013, at 7:32 PM, Amit Jha wrote:
> > > >
> > > > > Question is not clear to me. Please be more elaborative in your
> > query.
> > > > Why do u want to store index to DB tables?
> > > > >
> > > > > Rgds
> > > > > AJ
> > > > >
> > > > > On 15-Sep-2013, at 7:20, Baskar Sikkayan <baskar....@gmail.com>
> > wrote:
> > > > >
> > > > >> How to add index to 3 diff tables from java ...
> > > > >>
> > > > >>
> > > > >> On Sun, Sep 15, 2013 at 6:49 AM, Amit Jha <shanuu....@gmail.com>
> > > wrote:
> > > > >>
> > > > >>> Add a field called "source" in schema.xml and value would be your
> > > table
> > > > >>> names.
> > > > >>>
> > > > >>>
> > > > >>>
> > > > >>> Rgds
> > > > >>> AJ
> > > > >>>
> > > > >>> On 15-Sep-2013, at 5:38, Baskar Sikkayan <baskar....@gmail.com>
> > > wrote:
> > > > >>>
> > > > >>>> Hi,
> > > > >>>> I am new to Solr and trying to use Solr java client instead of
> > using
> > > > the
> > > > >>>> Data handler.
> > > > >>>> Is there any configuration i need to do for this?
> > > > >>>>
> > > > >>>> I got the following sample code.
> > > > >>>>
> > > > >>>> SolrInputDocument doc = new SolrInputDocument();
> > > > >>>>
> > > > >>>> doc.addField("cat", "book");
> > > > >>>> doc.addField("id", "book-" + i);
> > > > >>>> doc.addField("name", "The Legend of the Hobbit part
> " +
> > > i);
> > > > >>>> server.add(doc);
> > > > >>>> server.commit(); // periodically flush
> > > > >>>>
> > > > >>>> I am confused here. I am going to index 3 different tables for 3
> > > > >>> different
> > > > >>>> kind of searches. Here i dont have any option to differentiate 3
> > > kind
> > > > of
> > > > >>>> indexes.
> > > > >>>> Am i missing anything here. Could anyone please shed some light
> > > here?
> > > > >>>>
> > > > >>>> Thanks,
> > > > >>>> Baskar.S
> > > > >>>
> > > >
> > > >
> > > >
> > > >
> > > >
> > > >
> > >
> >
>
