0.15 - Tomorrow + Docs Updated

Heya,

Updated all the docs in the site to reflect 0.15 features, and it will be released tomorrow. Last chance to give it a go and see that all is well :slight_smile:

-shay.banon

great

thank you !

canal


From: Shay Banon shay.banon@elasticsearch.com
To: users users@elasticsearch.com
Sent: Fri, February 18, 2011 7:33:07 AM
Subject: 0.15 - Tomorrow + Docs Updated

Heya,

Updated all the docs in the site to reflect 0.15 features, and it will be
released tomorrow. Last chance to give it a go and see that all is well :slight_smile:

-shay.banon

Shay Banon skrev 2011-02-18 00:33:
<blockquote cite="mid:801066BA6AC348F09B2CCAA9AA6909B8@elasticsearch.com" type="cite">Heya, 

  Updated all the docs in the site to reflect 0.15
features, and it will be released tomorrow. Last chance to
give it a go and see that all is well :slight_smile:

-shay.banon
Hi!

Great news about the forthcoming release. I have some comments on
the docs about the Snowball Analyzer and Token Filter since I have
spent some time with them.




The <b>Analyzer </b>is limited to support <i>only </i>the
following languages with a <b>default </b>set of stopwords:


English, Dutch, German, German2 and French.


For the other  languages as mentioned in the docs for the Snowball
Token <b>Filter</b> that is supported you have to provide the
stopwords yourself.


A default set of stopwords for many of these is available from for
instance:

http://svn.apache.org/repos/asf/lucene/dev/branches/branch_3x/lucene/contrib/analyzers/common/src/resources/org/apache/lucene/analysis/

and

http://svn.apache.org/repos/asf/lucene/dev/branches/branch_3x/lucene/contrib/analyzers/common/src/resources/org/apache/lucene/analysis/snowball

And they can be used as thus (yaml format):
index :
    analysis :
        analyzer : 
           my_analyzer: 
                type: snowball
                name: Swedish
                stopwords: [och,det,att,i,en,jag,hon,som,han,pÄ,den,med,var,sig,för,sÄ,till,Àr,men,ett,om,hade,de,av,icke,mig,du,
                henne,dÄ,sin,nu,har,inte,hans,honom,skulle,hennes,dÀr,min,man,ej,vid,kunde,nÄgot,frÄn,ut,nÀr,efter,upp,vi,dem,vara,
                vad,över,Àn,dig,kan,sina,hÀr,ha,mot,alla,under,nÄgon,allt,mycket,sedan,ju,denna,sjÀlv,detta,Ät,utan,varit,hur,ingen,
                mitt,ni,bli,blev,oss,din,dessa,nÄgra,deras,blir,mina,samma,vilken,er,sÄdan,vÄr,blivit,dess,inom,mellan,sÄdant,varför,
                varje,vilka,ditt,vem,vilket,sitta,sÄdana,vart,dina,vars,vÄrt,vÄra,ert,era,vilkas]


In the Filter docs an example as this could be given. Stopwords will
have to be specified for the stop filter explicitly:
index :
    analysis :
        analyzer : 
           my_analyzer: 
                type: custom
                tokenizer: standard
                filter: [standard, lowercase, my_stop, snowball_stemming]
        filter :
         snowball_stemming:
           type: snowball
           language: Swedish
        my_stop:
                type: stop
                stopwords: [och,det,att,i,en,jag,hon,som,han,pÄ,den,med,var,sig,för,sÄ,till,Àr,men,ett,om,hade,de,av,icke,mig,du,
                henne,dÄ,sin,nu,har,inte,hans,honom,skulle,hennes,dÀr,min,man,ej,vid,kunde,nÄgot,frÄn,ut,nÀr,efter,upp,vi,dem,vara,
                vad,över,Àn,dig,kan,sina,hÀr,ha,mot,alla,under,nÄgon,allt,mycket,sedan,ju,denna,sjÀlv,detta,Ät,utan,varit,hur,ingen,
                mitt,ni,bli,blev,oss,din,dessa,nÄgra,deras,blir,mina,samma,vilken,er,sÄdan,vÄr,blivit,dess,inom,mellan,sÄdant,varför,
                varje,vilka,ditt,vem,vilket,sitta,sÄdana,vart,dina,vars,vÄrt,vÄra,ert,era,vilkas]
             
For the analyzer part, I think a default set of stopwords should be
programmed, so that the behaviour is similar regardless of language
chosen. This is not the case now.
-- 
Med vÀnlig hÀlsning
Kristian Jörg

Devo IT AB
Tel: 054 - 22 14 58, 0709 - 15 83 42
E-post: kristian.jorg@devo.se
Webb: http://www.devo.se

Thanks!, can you improve the docs and crete a pull request? :), here is some explanation on how to get the site going: Elasticsearch Platform — Find real-time answers at scale | Elastic.
On Friday, February 18, 2011 at 2:52 PM, Kristian Jörg wrote:

Shay Banon skrev 2011-02-18 00:33:

Heya,

Updated all the docs in the site to reflect 0.15 features, and it will be released tomorrow. Last chance to give it a go and see that all is well :slight_smile:

-shay.banon Hi!

Great news about the forthcoming release. I have some comments on the docs about the Snowball Analyzer and Token Filter since I have spent some time with them.

The Analyzer is limited to support only the following languages with a default set of stopwords:
English, Dutch, German, German2 and French.
For the other languages as mentioned in the docs for the Snowball Token Filter that is supported you have to provide the stopwords yourself.
A default set of stopwords for many of these is available from for instance:
http://svn.apache.org/repos/asf/lucene/dev/branches/branch_3x/lucene/contrib/analyzers/common/src/resources/org/apache/lucene/analysis/
and
http://svn.apache.org/repos/asf/lucene/dev/branches/branch_3x/lucene/contrib/analyzers/common/src/resources/org/apache/lucene/analysis/snowball

And they can be used as thus (yaml format):
index : analysis : analyzer : my_analyzer: type: snowball name: Swedish stopwords: [och,det,att,i,en,jag,hon,som,han,pÄ,den,med,var,sig,för,sÄ,till,Àr,men,ett,om,hade,de,av,icke,mig,du, henne,dÄ,sin,nu,har,inte,hans,honom,skulle,hennes,dÀr,min,man,ej,vid,kunde,nÄgot,frÄn,ut,nÀr,efter,upp,vi,dem,vara, vad,över,Àn,dig,kan,sina,hÀr,ha,mot,alla,under,nÄgon,allt,mycket,sedan,ju,denna,sjÀlv,detta,Ät,utan,varit,hur,ingen, mitt,ni,bli,blev,oss,din,dessa,nÄgra,deras,blir,mina,samma,vilken,er,sÄdan,vÄr,blivit,dess,inom,mellan,sÄdant,varför, varje,vilka,ditt,vem,vilket,sitta,sÄdana,vart,dina,vars,vÄrt,vÄra,ert,era,vilkas] In the Filter docs an example as this could be given. Stopwords will have to be specified for the stop filter explicitly:
index : analysis : analyzer : my_analyzer: type: custom tokenizer: standard filter: [standard, lowercase, my_stop, snowball_stemming] filter : snowball_stemming: type: snowball language: Swedish my_stop: type: stop stopwords: [och,det,att,i,en,jag,hon,som,han,pÄ,den,med,var,sig,för,sÄ,till,Àr,men,ett,om,hade,de,av,icke,mig,du, henne,dÄ,sin,nu,har,inte,hans,honom,skulle,hennes,dÀr,min,man,ej,vid,kunde,nÄgot,frÄn,ut,nÀr,efter,upp,vi,dem,vara, vad,över,Àn,dig,kan,sina,hÀr,ha,mot,alla,under,nÄgon,allt,mycket,sedan,ju,denna,sjÀlv,detta,Ät,utan,varit,hur,ingen, mitt,ni,bli,blev,oss,din,dessa,nÄgra,deras,blir,mina,samma,vilken,er,sÄdan,vÄr,blivit,dess,inom,mellan,sÄdant,varför, varje,vilka,ditt,vem,vilket,sitta,sÄdana,vart,dina,vars,vÄrt,vÄra,ert,era,vilkas] For the analyzer part, I think a default set of stopwords should be programmed, so that the behaviour is similar regardless of language chosen. This is not the case now.
-- Med vÀnlig hÀlsning Kristian Jörg Devo IT AB Tel: 054 - 22 14 58, 0709 - 15 83 42 E-post: kristian.jorg@devo.se Webb: http://www.devo.se

Ok, done.
Hope I did everything correctly. I am new to Git. Note that the docs
also had an error with the parameter "name" that should be "language".
Fixed in pull.

The pull request is here:
https://github.com/elasticsearch/elasticsearch.github.com/pull/12

/Kristian

Thanks!, can you improve the docs and crete a pull request? :), here
is some explanation on how to get the site going:
Elasticsearch Platform — Find real-time answers at scale | Elastic.

On Friday, February 18, 2011 at 2:52 PM, Kristian Jörg wrote:

Shay Banon skrev 2011-02-18 00:33:

Heya,

Updated all the docs in the site to reflect 0.15 features, and it
will be released tomorrow. Last chance to give it a go and see that
all is well :slight_smile:

-shay.banon
Hi!

Great news about the forthcoming release. I have some comments on the
docs about the Snowball Analyzer and Token Filter since I have spent
some time with them.

The *Analyzer *is limited to support /only /the following languages
with a *default *set of stopwords:
English, Dutch, German, German2 and French.
For the other languages as mentioned in the docs for the Snowball
Token Filter that is supported you have to provide the stopwords
yourself.
A default set of stopwords for many of these is available from for
instance:
http://svn.apache.org/repos/asf/lucene/dev/branches/branch_3x/lucene/contrib/analyzers/common/src/resources/org/apache/lucene/analysis/
and
http://svn.apache.org/repos/asf/lucene/dev/branches/branch_3x/lucene/contrib/analyzers/common/src/resources/org/apache/lucene/analysis/snowball

And they can be used as thus (yaml format):
index :
analysis :
analyzer :
my_analyzer:
type: snowball
name: Swedish
stopwords: [och,det,att,i,en,jag,hon,som,han,pÄ,den,med,var,sig,för,sÄ,till,Àr,men,ett,om,hade,de,av,icke,mig,du,
henne,dÄ,sin,nu,har,inte,hans,honom,skulle,hennes,dÀr,min,man,ej,vid,kunde,nÄgot,frÄn,ut,nÀr,efter,upp,vi,dem,vara,
vad,över,Àn,dig,kan,sina,hÀr,ha,mot,alla,under,nÄgon,allt,mycket,sedan,ju,denna,sjÀlv,detta,Ät,utan,varit,hur,ingen,
mitt,ni,bli,blev,oss,din,dessa,nÄgra,deras,blir,mina,samma,vilken,er,sÄdan,vÄr,blivit,dess,inom,mellan,sÄdant,varför,
varje,vilka,ditt,vem,vilket,sitta,sÄdana,vart,dina,vars,vÄrt,vÄra,ert,era,vilkas]

In the Filter docs an example as this could be given. Stopwords will
have to be specified for the stop filter explicitly:
index :
analysis :
analyzer :
my_analyzer:
type: custom
tokenizer: standard
filter: [standard, lowercase, my_stop, snowball_stemming]
filter :
snowball_stemming:
type: snowball
language: Swedish
my_stop:
type: stop
stopwords: [och,det,att,i,en,jag,hon,som,han,pÄ,den,med,var,sig,för,sÄ,till,Àr,men,ett,om,hade,de,av,icke,mig,du,
henne,dÄ,sin,nu,har,inte,hans,honom,skulle,hennes,dÀr,min,man,ej,vid,kunde,nÄgot,frÄn,ut,nÀr,efter,upp,vi,dem,vara,
vad,över,Àn,dig,kan,sina,hÀr,ha,mot,alla,under,nÄgon,allt,mycket,sedan,ju,denna,sjÀlv,detta,Ät,utan,varit,hur,ingen,
mitt,ni,bli,blev,oss,din,dessa,nÄgra,deras,blir,mina,samma,vilken,er,sÄdan,vÄr,blivit,dess,inom,mellan,sÄdant,varför,
varje,vilka,ditt,vem,vilket,sitta,sÄdana,vart,dina,vars,vÄrt,vÄra,ert,era,vilkas]

For the analyzer part, I think a default set of stopwords should be
programmed, so that the behaviour is similar regardless of language
chosen. This is not the case now.

Med vÀnlig hÀlsning
Kristian Jörg

Devo IT AB
Tel: 054 - 22 14 58, 0709 - 15 83 42
E-post:kristian.jorg@devo.se
Webb:http://www.devo.se

Cool, thanks!. Applied :slight_smile:
On Monday, February 21, 2011 at 2:13 PM, Kristian Jörg wrote:

Ok, done.
Hope I did everything correctly. I am new to Git. Note that the docs also had an error with the parameter "name" that should be "language". Fixed in pull.

The pull request is here: https://github.com/elasticsearch/elasticsearch.github.com/pull/12

/Kristian

Thanks!, can you improve the docs and crete a pull request? :), here is some explanation on how to get the site going: Elasticsearch Platform — Find real-time answers at scale | Elastic.
On Friday, February 18, 2011 at 2:52 PM, Kristian Jörg wrote:

Shay Banon skrev 2011-02-18 00:33:

Heya,

Updated all the docs in the site to reflect 0.15 features, and it will be released tomorrow. Last chance to give it a go and see that all is well :slight_smile:

-shay.banon Hi!

Great news about the forthcoming release. I have some comments on the docs about the Snowball Analyzer and Token Filter since I have spent some time with them.

The Analyzer is limited to support only the following languages with a default set of stopwords:
English, Dutch, German, German2 and French.
For the other languages as mentioned in the docs for the Snowball Token Filter that is supported you have to provide the stopwords yourself.
A default set of stopwords for many of these is available from for instance:
http://svn.apache.org/repos/asf/lucene/dev/branches/branch_3x/lucene/contrib/analyzers/common/src/resources/org/apache/lucene/analysis/
and
http://svn.apache.org/repos/asf/lucene/dev/branches/branch_3x/lucene/contrib/analyzers/common/src/resources/org/apache/lucene/analysis/snowball

And they can be used as thus (yaml format):
index : analysis : analyzer : my_analyzer: type: snowball name: Swedish stopwords: [och,det,att,i,en,jag,hon,som,han,pÄ,den,med,var,sig,för,sÄ,till,Àr,men,ett,om,hade,de,av,icke,mig,du, henne,dÄ,sin,nu,har,inte,hans,honom,skulle,hennes,dÀr,min,man,ej,vid,kunde,nÄgot,frÄn,ut,nÀr,efter,upp,vi,dem,vara, vad,över,Àn,dig,kan,sina,hÀr,ha,mot,alla,under,nÄgon,allt,mycket,sedan,ju,denna,sjÀlv,detta,Ät,utan,varit,hur,ingen, mitt,ni,bli,blev,oss,din,dessa,nÄgra,deras,blir,mina,samma,vilken,er,sÄdan,vÄr,blivit,dess,inom,mellan,sÄdant,varför, varje,vilka,ditt,vem,vilket,sitta,sÄdana,vart,dina,vars,vÄrt,vÄra,ert,era,vilkas] In the Filter docs an example as this could be given. Stopwords will have to be specified for the stop filter explicitly:
index : analysis : analyzer : my_analyzer: type: custom tokenizer: standard filter: [standard, lowercase, my_stop, snowball_stemming] filter : snowball_stemming: type: snowball language: Swedish my_stop: type: stop stopwords: [och,det,att,i,en,jag,hon,som,han,pÄ,den,med,var,sig,för,sÄ,till,Àr,men,ett,om,hade,de,av,icke,mig,du, henne,dÄ,sin,nu,har,inte,hans,honom,skulle,hennes,dÀr,min,man,ej,vid,kunde,nÄgot,frÄn,ut,nÀr,efter,upp,vi,dem,vara, vad,över,Àn,dig,kan,sina,hÀr,ha,mot,alla,under,nÄgon,allt,mycket,sedan,ju,denna,sjÀlv,detta,Ät,utan,varit,hur,ingen, mitt,ni,bli,blev,oss,din,dessa,nÄgra,deras,blir,mina,samma,vilken,er,sÄdan,vÄr,blivit,dess,inom,mellan,sÄdant,varför, varje,vilka,ditt,vem,vilket,sitta,sÄdana,vart,dina,vars,vÄrt,vÄra,ert,era,vilkas] For the analyzer part, I think a default set of stopwords should be programmed, so that the behaviour is similar regardless of language chosen. This is not the case now.
-- Med vÀnlig hÀlsning Kristian Jörg Devo IT AB Tel: 054 - 22 14 58, 0709 - 15 83 42 E-post: kristian.jorg@devo.se Webb: http://www.devo.se

Oops!




I hade missed one instance of the faulty name param (in the filter
text). It is corrected in this pull now. Can you apply again?




/Kristian




Shay Banon skrev 2011-02-21 18:38:
<blockquote cite="mid:9B9D5BB678E045F1880C1BC8224E77D3@elasticsearch.com" type="cite">Cool, thanks!. Applied :)

On Monday, February 21, 2011 at 2:13 PM, Kristian Jörg wrote:

Ok, done.
              Hope I did everything correctly. I am new to Git. Note
              that the docs also had an error with the parameter
              "name" that should be "language". Â Fixed in pull.




              The pull request is here: <a moz-do-not-send="true" href="https://github.com/elasticsearch/elasticsearch.github.com/pull/12">https://github.com/elasticsearch/elasticsearch.github.com/pull/12</a>




              /Kristian
Thanks!, can you improve the docs and crete a pull request? :), here is some explanation on how to get the site going:Â http://www.elasticsearch.org/community/contributing-site.html.

On Friday, February 18, 2011 at 2:52 PM, Kristian Jörg wrote:

Shay Banon skrev 2011-02-18 00:33:
Heya,

  Updated all
the docs in the site to
reflect 0.15 features,
and it will be released
tomorrow. Last chance to
give it a go and see
that all is well :slight_smile:

-shay.banon


Hi!

                                Great news about the forthcoming
                                release. I have some comments on the
                                docs about the Snowball Analyzer and
                                Token Filter since I have spent some
                                time with them.




                                The <b>Analyzer </b>is limited to
                                support <i>only </i>the following
                                languages with a <b>default </b>set
                                of stopwords:


                                English, Dutch, German, German2 and
                                French.


                                For the other  languages as
                                mentioned in the docs for the
                                Snowball Token <b>Filter</b> that
                                is supported you have to provide the
                                stopwords yourself.


                                A default set of stopwords for many
                                of these is available from for
                                instance:

http://svn.apache.org/repos/asf/lucene/dev/branches/branch_3x/lucene/contrib/analyzers/common/src/resources/org/apache/lucene/analysis/

                                and

http://svn.apache.org/repos/asf/lucene/dev/branches/branch_3x/lucene/contrib/analyzers/common/src/resources/org/apache/lucene/analysis/snowball

                                And they can be used as thus (yaml
                                format):
index :
    analysis :
        analyzer : 
           my_analyzer: 
                type: snowball
                name: Swedish
                stopwords: [och,det,att,i,en,jag,hon,som,han,pÄ,den,med,var,sig,för,sÄ,till,Àr,men,ett,om,hade,de,av,icke,mig,du,
                henne,dÄ,sin,nu,har,inte,hans,honom,skulle,hennes,dÀr,min,man,ej,vid,kunde,nÄgot,frÄn,ut,nÀr,efter,upp,vi,dem,vara,
                vad,över,Àn,dig,kan,sina,hÀr,ha,mot,alla,under,nÄgon,allt,mycket,sedan,ju,denna,sjÀlv,detta,Ät,utan,varit,hur,ingen,
                mitt,ni,bli,blev,oss,din,dessa,nÄgra,deras,blir,mina,samma,vilken,er,sÄdan,vÄr,blivit,dess,inom,mellan,sÄdant,varför,
                varje,vilka,ditt,vem,vilket,sitta,sÄdana,vart,dina,vars,vÄrt,vÄra,ert,era,vilkas]


                                In the Filter docs an example as
                                this could be given. Stopwords will
                                have to be specified for the stop
                                filter explicitly:
index :
    analysis :
        analyzer : 
           my_analyzer: 
                type: custom
                tokenizer: standard
                filter: [standard, lowercase, my_stop, snowball_stemming]
        filter :
         snowball_stemming:
           type: snowball
           language: Swedish
        my_stop:
                type: stop
                stopwords: [och,det,att,i,en,jag,hon,som,han,pÄ,den,med,var,sig,för,sÄ,till,Àr,men,ett,om,hade,de,av,icke,mig,du,
                henne,dÄ,sin,nu,har,inte,hans,honom,skulle,hennes,dÀr,min,man,ej,vid,kunde,nÄgot,frÄn,ut,nÀr,efter,upp,vi,dem,vara,
                vad,över,Àn,dig,kan,sina,hÀr,ha,mot,alla,under,nÄgon,allt,mycket,sedan,ju,denna,sjÀlv,detta,Ät,utan,varit,hur,ingen,
                mitt,ni,bli,blev,oss,din,dessa,nÄgra,deras,blir,mina,samma,vilken,er,sÄdan,vÄr,blivit,dess,inom,mellan,sÄdant,varför,
                varje,vilka,ditt,vem,vilket,sitta,sÄdana,vart,dina,vars,vÄrt,vÄra,ert,era,vilkas]
             
                                For the analyzer part, I think a
                                default set of stopwords should be
                                programmed, so that the behaviour is
                                similar regardless of language
                                chosen. This is not the case now.
-- 
Med vÀnlig hÀlsning
Kristian Jörg

Devo IT AB
Tel: 054 - 22 14 58, 0709 - 15 83 42
E-post: kristian.jorg@devo.se
Webb: http://www.devo.se
-- 
Med vÀnlig hÀlsning
Kristian Jörg

Devo IT AB
Tel: 054 - 22 14 58, 0709 - 15 83 42
E-post: kristian.jorg@devo.se
Webb: http://www.devo.se

Go ahead Kristian, pull requests are welcome.

Lukas

On Tue, Feb 22, 2011 at 10:01 AM, Kristian Jörg krjg@devo.se wrote:

Oops!

I hade missed one instance of the faulty name param (in the filter text).
It is corrected in this pull now. Can you apply again?

/Kristian

Shay Banon skrev 2011-02-21 18:38:

Cool, thanks!. Applied :slight_smile:

On Monday, February 21, 2011 at 2:13 PM, Kristian Jörg wrote:

Ok, done.
Hope I did everything correctly. I am new to Git. Note that the docs also
had an error with the parameter "name" that should be "language". Fixed in
pull.

The pull request is here:
https://github.com/elasticsearch/elasticsearch.github.com/pull/12

/Kristian

Thanks!, can you improve the docs and crete a pull request? :), here is
some explanation on how to get the site going:
Elasticsearch Platform — Find real-time answers at scale | Elastic.

On Friday, February 18, 2011 at 2:52 PM, Kristian Jörg wrote:

Shay Banon skrev 2011-02-18 00:33:

Heya,

Updated all the docs in the site to reflect 0.15 features, and it will

be released tomorrow. Last chance to give it a go and see that all is well
:slight_smile:

-shay.banon

Hi!

Great news about the forthcoming release. I have some comments on the docs
about the Snowball Analyzer and Token Filter since I have spent some time
with them.

The *Analyzer *is limited to support *only *the following languages with a
*default set of stopwords:
English, Dutch, German, German2 and French.
For the other languages as mentioned in the docs for the Snowball Token *
Filter
that is supported you have to provide the stopwords yourself.
A default set of stopwords for many of these is available from for
instance:

http://svn.apache.org/repos/asf/lucene/dev/branches/branch_3x/lucene/contrib/analyzers/common/src/resources/org/apache/lucene/analysis/
and

http://svn.apache.org/repos/asf/lucene/dev/branches/branch_3x/lucene/contrib/analyzers/common/src/resources/org/apache/lucene/analysis/snowball

And they can be used as thus (yaml format):

index :
analysis :
analyzer :
my_analyzer:
type: snowball
name: Swedish
stopwords: [och,det,att,i,en,jag,hon,som,han,pÄ,den,med,var,sig,för,sÄ,till,Àr,men,ett,om,hade,de,av,icke,mig,du,
henne,dÄ,sin,nu,har,inte,hans,honom,skulle,hennes,dÀr,min,man,ej,vid,kunde,nÄgot,frÄn,ut,nÀr,efter,upp,vi,dem,vara,
vad,över,Àn,dig,kan,sina,hÀr,ha,mot,alla,under,nÄgon,allt,mycket,sedan,ju,denna,sjÀlv,detta,Ät,utan,varit,hur,ingen,
mitt,ni,bli,blev,oss,din,dessa,nÄgra,deras,blir,mina,samma,vilken,er,sÄdan,vÄr,blivit,dess,inom,mellan,sÄdant,varför,
varje,vilka,ditt,vem,vilket,sitta,sÄdana,vart,dina,vars,vÄrt,vÄra,ert,era,vilkas]

In the Filter docs an example as this could be given. Stopwords will have
to be specified for the stop filter explicitly:

index :
analysis :
analyzer :
my_analyzer:
type: custom
tokenizer: standard
filter: [standard, lowercase, my_stop, snowball_stemming]
filter :
snowball_stemming:
type: snowball
language: Swedish
my_stop:
type: stop
stopwords: [och,det,att,i,en,jag,hon,som,han,pÄ,den,med,var,sig,för,sÄ,till,Àr,men,ett,om,hade,de,av,icke,mig,du,
henne,dÄ,sin,nu,har,inte,hans,honom,skulle,hennes,dÀr,min,man,ej,vid,kunde,nÄgot,frÄn,ut,nÀr,efter,upp,vi,dem,vara,
vad,över,Àn,dig,kan,sina,hÀr,ha,mot,alla,under,nÄgon,allt,mycket,sedan,ju,denna,sjÀlv,detta,Ät,utan,varit,hur,ingen,
mitt,ni,bli,blev,oss,din,dessa,nÄgra,deras,blir,mina,samma,vilken,er,sÄdan,vÄr,blivit,dess,inom,mellan,sÄdant,varför,
varje,vilka,ditt,vem,vilket,sitta,sÄdana,vart,dina,vars,vÄrt,vÄra,ert,era,vilkas]

For the analyzer part, I think a default set of stopwords should be
programmed, so that the behaviour is similar regardless of language chosen.
This is not the case now.

--
Med vÀnlig hÀlsning
Kristian Jörg

Devo IT AB
Tel: 054 - 22 14 58, 0709 - 15 83 42
E-post: kristian.jorg@devo.se
Webb: http://www.devo.se

--
Med vÀnlig hÀlsning
Kristian Jörg

Devo IT AB
Tel: 054 - 22 14 58, 0709 - 15 83 42
E-post: kristian.jorg@devo.se
Webb: http://www.devo.se