You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by Uwe Schindler <uw...@thetaphi.de> on 2009/01/14 15:43:10 UTC
Question on small optimization in TrieUtils
Hi,
I was just looking through some more optimizations of my trie encoded
fields. They are currently indexed using omitTf. Additionally I could also
add omitNorms to the index options for such type of fields (because they are
only used for sorting and the special TrieRangeFilter). Does this has any
effect on other fields and boosting when used together with normal
TermQueries? If no, I commit shortly!
-----
Uwe Schindler
H.-H.-Meier-Allee 63, D-28213 Bremen
http://www.thetaphi.de
eMail: uwe@thetaphi.de
---------------------------------------------------------------------
To unsubscribe, e-mail: java-dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-dev-help@lucene.apache.org
Re: Question on small optimization in TrieUtils
Posted by Mark Miller <ma...@gmail.com>.
Right. Keyword and other such fields (likely not analyzed) should def be
omitTf and OmitNorms, unless you have some special situation.
Uwe Schindler wrote:
> Yes, that was may opinion when reading the java docs. I wanted to be sure.
>
> In principle we could write down somewhere, that disabling norms is good for
> not-analyzed fields like document identifiers, numbers,.... Such a field do
> not need norms, as e.g. the document identifier is only used for searching
> for only the document or sorting or something like that. So in my opinion,
> the best default for non-analyzed fields is to use no norms, right?
>
> -----
> Uwe Schindler
> H.-H.-Meier-Allee 63, D-28213 Bremen
> http://www.thetaphi.de
> eMail: uwe@thetaphi.de
>
>
>
>> -----Original Message-----
>> From: Mark Miller [mailto:markrmiller@gmail.com]
>> Sent: Wednesday, January 14, 2009 4:01 PM
>> To: java-dev@lucene.apache.org
>> Subject: Re: Question on small optimization in TrieUtils
>>
>> It shouldn't have an affect, I'd make the change. The norm setting is
>> only respective to the field you set it on, and a sort field won't
>> benefit IMO. Norms are for index time boost and field length
>> normalization right? Turn them off I say.
>>
>> Uwe Schindler wrote:
>>
>>> Hi,
>>>
>>> I was just looking through some more optimizations of my trie encoded
>>> fields. They are currently indexed using omitTf. Additionally I could
>>>
>> also
>>
>>> add omitNorms to the index options for such type of fields (because they
>>>
>> are
>>
>>> only used for sorting and the special TrieRangeFilter). Does this has
>>>
>> any
>>
>>> effect on other fields and boosting when used together with normal
>>> TermQueries? If no, I commit shortly!
>>>
>>> -----
>>> Uwe Schindler
>>> H.-H.-Meier-Allee 63, D-28213 Bremen
>>> http://www.thetaphi.de
>>> eMail: uwe@thetaphi.de
>>>
>>>
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: java-dev-unsubscribe@lucene.apache.org
>>> For additional commands, e-mail: java-dev-help@lucene.apache.org
>>>
>>>
>>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: java-dev-unsubscribe@lucene.apache.org
>> For additional commands, e-mail: java-dev-help@lucene.apache.org
>>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: java-dev-unsubscribe@lucene.apache.org
> For additional commands, e-mail: java-dev-help@lucene.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: java-dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-dev-help@lucene.apache.org
RE: Question on small optimization in TrieUtils
Posted by Uwe Schindler <uw...@thetaphi.de>.
Yes, that was may opinion when reading the java docs. I wanted to be sure.
In principle we could write down somewhere, that disabling norms is good for
not-analyzed fields like document identifiers, numbers,.... Such a field do
not need norms, as e.g. the document identifier is only used for searching
for only the document or sorting or something like that. So in my opinion,
the best default for non-analyzed fields is to use no norms, right?
-----
Uwe Schindler
H.-H.-Meier-Allee 63, D-28213 Bremen
http://www.thetaphi.de
eMail: uwe@thetaphi.de
> -----Original Message-----
> From: Mark Miller [mailto:markrmiller@gmail.com]
> Sent: Wednesday, January 14, 2009 4:01 PM
> To: java-dev@lucene.apache.org
> Subject: Re: Question on small optimization in TrieUtils
>
> It shouldn't have an affect, I'd make the change. The norm setting is
> only respective to the field you set it on, and a sort field won't
> benefit IMO. Norms are for index time boost and field length
> normalization right? Turn them off I say.
>
> Uwe Schindler wrote:
> > Hi,
> >
> > I was just looking through some more optimizations of my trie encoded
> > fields. They are currently indexed using omitTf. Additionally I could
> also
> > add omitNorms to the index options for such type of fields (because they
> are
> > only used for sorting and the special TrieRangeFilter). Does this has
> any
> > effect on other fields and boosting when used together with normal
> > TermQueries? If no, I commit shortly!
> >
> > -----
> > Uwe Schindler
> > H.-H.-Meier-Allee 63, D-28213 Bremen
> > http://www.thetaphi.de
> > eMail: uwe@thetaphi.de
> >
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: java-dev-unsubscribe@lucene.apache.org
> > For additional commands, e-mail: java-dev-help@lucene.apache.org
> >
> >
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: java-dev-unsubscribe@lucene.apache.org
> For additional commands, e-mail: java-dev-help@lucene.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: java-dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-dev-help@lucene.apache.org
Re: Question on small optimization in TrieUtils
Posted by Mark Miller <ma...@gmail.com>.
It shouldn't have an affect, I'd make the change. The norm setting is
only respective to the field you set it on, and a sort field won't
benefit IMO. Norms are for index time boost and field length
normalization right? Turn them off I say.
Uwe Schindler wrote:
> Hi,
>
> I was just looking through some more optimizations of my trie encoded
> fields. They are currently indexed using omitTf. Additionally I could also
> add omitNorms to the index options for such type of fields (because they are
> only used for sorting and the special TrieRangeFilter). Does this has any
> effect on other fields and boosting when used together with normal
> TermQueries? If no, I commit shortly!
>
> -----
> Uwe Schindler
> H.-H.-Meier-Allee 63, D-28213 Bremen
> http://www.thetaphi.de
> eMail: uwe@thetaphi.de
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: java-dev-unsubscribe@lucene.apache.org
> For additional commands, e-mail: java-dev-help@lucene.apache.org
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: java-dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: java-dev-help@lucene.apache.org