Mailing-List: contact solr-user-help@lucene.apache.org; run by ezmlm
Precedence: bulk
Reply-To: solr-user@lucene.apache.org
Received-SPF: pass (athena.apache.org: domain of steve@b.abbies.us designates
 162.243.111.211 as permitted sender)
Content-Type: text/plain; charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 7.3 \(1878.2\))
Subject: Re: Indexing non-stored fields
From: Steve McKay <steve@b.abbies.us>
In-Reply-To: 
 <CAGKEjHbN-v1LEOAyAqc6ezH56sGmPE3POY1Prwo6mcqfWFLE_w@mail.gmail.com>
Date: Mon, 30 Jun 2014 20:54:14 -0400
Content-Transfer-Encoding: quoted-printable
Message-Id: <169BAC49-3052-4BA6-8C8B-529528FEA97A@b.abbies.us>
References: <1404172349190-4144893.post@n3.nabble.com>
 <77b12e4ade99a2d1d27e7892218c3948.squirrel@admin.elyograg.org>
 <CAGKEjHbN-v1LEOAyAqc6ezH56sGmPE3POY1Prwo6mcqfWFLE_w@mail.gmail.com>
To: solr-user@lucene.apache.org

Stored doesn't mean "stored to disk", more like "stored verbatim". When =
you index a field, Solr analyzes the field value and makes it part of =
the index. The index is persisted to disk when you commit, which is why =
it sticks around after a restart. Searching the index, mapping from =
search terms to doc ids, is very fast. However, the index is very very =
bad at going in reverse, from doc ids to terms. That's where stored =
fields come in. When you store a field, Solr takes the field value and =
stores the entire value separate from the index. This makes it trivial =
to get the value for a particular doc id, but it's terrible for =
searching.

So the stored attribute and the indexed attribute have different =
purposes. Indexed means you want to be able to search on the value, and =
stored means you want to be able to see the value in search results.

On Jun 30, 2014, at 8:15 PM, tomasv <dadkind@gmail.com> wrote:

> Thanks for the quick response.
>=20
> Follow-up newbie question:
> If the fields are not stored, how is the server able to search for =
them
> after a restart? Where does it get the data to be searched?
>=20
> Example:  "bob" (firstname) is indexed but not stored. After initial
> indexing, I query for "firstname:(bob)" and I get my document back. =
But if
> I restart the server, where does the server go to retrieve information =
that
> will allow me to query for "bob" once again? It would seem that "bob" =
got
> stored someplace if I can query on it after a restart.
>=20
> My untrained mind thinks that searching for "firstname:(bob)" (after a
> restart) will fail, but that searching for "recordid:(12345)" (in my
> original example) will succeed since it was indexed+stored.
>=20
> (stored + indexed makes total sense to me; it's the indexed but NOT =
stored
> that I can't get my head around).
>=20
> Thanks!
>=20
>=20
>=20
> On Mon, Jun 30, 2014 at 5:05 PM, Shawn Heisey-4 [via Lucene] <
> ml-node+s472066n4144894h24@n3.nabble.com> wrote:
>=20
>>> Hello All, (warning: newbie question)
>>>=20
>>> In our schema.xml we have defined many fields such as:
>>> <field name=3D"firstname" type=3D"string" indexed=3D"true" =
stored=3D"false" />
>>>=20
>>> Other fields are defined as this:
>>> <field name=3D"recordid" type=3D"long" indexed=3D"true" =
stored=3D"true" />
>>>=20
>>> Q: If my server is restarted/ rebooted, will I still be able to =
search
>> for
>>> documents using the "firstname" field? Or will my records need to be
>>> re-indexed before I can search by first name?
>>> It seems that after a re-boot, I can search for the "stored=3D'true'"
>> fields
>>> but not the "stored=3D'false'" fields.
>>>=20
>>> Am I interpreting this correctly? or am I missing something?
>>=20
>> Fields that are not stored simply mean that they will not be returned =
in
>> search results. If they are indexed, then you will be able to search =
on
>> those fields.
>>=20
>> This should be the case before or after a restart.
>>=20
>> Thanks,
>> Shawn
>>=20
>>=20
>>=20
>>=20
>>=20
>> ------------------------------
>> If you reply to this email, your message will be added to the =
discussion
>> below:
>>=20
>> =
http://lucene.472066.n3.nabble.com/Indexing-non-stored-fields-tp4144893p41=
44894.html
>> To unsubscribe from Indexing non-stored fields, click here
>> =
<http://lucene.472066.n3.nabble.com/template/NamlServlet.jtp?macro=3Dunsub=
scribe_by_code&node=3D4144893&code=3DZGFka2luZEBnbWFpbC5jb218NDE0NDg5M3wtM=
TcwMzcxNzM5Mg=3D=3D>
>> .
>> NAML
>> =
<http://lucene.472066.n3.nabble.com/template/NamlServlet.jtp?macro=3Dmacro=
_viewer&id=3Dinstant_html%21nabble%3Aemail.naml&base=3Dnabble.naml.namespa=
ces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.we=
b.template.NodeNamespace&breadcrumbs=3Dnotify_subscribers%21nabble%3Aemail=
.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Ae=
mail.naml>
>>=20
>=20
>=20
>=20
> --=20
> /*-----------------------
> * Tomas at Home
> * dadkind@gmail.com
> * ---------------------*/
>=20
>=20
>=20
>=20
> --
> View this message in context: =
http://lucene.472066.n3.nabble.com/Indexing-non-stored-fields-tp4144893p41=
44895.html
> Sent from the Solr - User mailing list archive at Nabble.com.