Hmmm, I'd expect you to have an error in your log file if you haven't
removed the default field type named "string". If you have removed it
from your schema, this should work...
But I'd change the name anyway, it'll cause endless confusion.
And don't forget to re-index *everything*, preferably af
A note: in the first schema, you had the stopwords after the stemmer.
This would not work, since the stopwords are not stemmed.
On Wed, Aug 24, 2011 at 12:59 AM, _snake_ wrote:
> I forgot to say that my stopwords file is in the same location as the schema
> file and the solrconfig file.
>
> --
>
I forgot to say that my stopwords file is in the same location as the schema
file and the solrconfig file.
--
View this message in context:
http://lucene.472066.n3.nabble.com/Problem-using-stop-words-tp3274598p3280319.html
Sent from the Solr - User mailing list archive at Nabble.com.
Thanks everybody for your help!!
I change the stopwords file, and I only use one word per line, without start
/ ending spaces, and without comments.
I change it to UTF-8.
I am using the TermsComponent to suggest words to the user (JQuery UI
Autocomplete). So, the stopwords are still showed here...
No, I think you're right, i've never seen pipes as comments before...
2011/8/22 Erick Erickson
> Ahh, you're right. I was wy off base there
>
> So I guess the question is how you know the words aren't being removed? A
> common
> problem is to look at *stored* fields rather than what's ac
Ahh, you're right. I was wy off base there
So I guess the question is how you know the words aren't being removed? A common
problem is to look at *stored* fields rather than what's actually in
the inverted index.
The TermsComponent can help here:
http://wiki.apache.org/solr/TermsComponent
That very txt said "A Spanish stop word list. Comments begin with vertical
bar. Each stop word is at the start of a line."
Solr's comments are #s not pipes.
Brazilian stopwords file is kinda raw...
http://svn.apache.org/repos/asf/lucene/dev/trunk/modules/analysis/common/src/resources/org/apache/l
Funny thing is that stopwords files in the examples shown in
http://wiki.apache.org/solr/LanguageAnalysis#Spanish are actually using pipe
and other terms. See the spanish one in
http://svn.apache.org/repos/asf/lucene/dev/trunk/modules/analysis/common/src/resources/org/apache/lucene/analysis/snowbal
What does the admin/analysis page show? And if you're really
putting the pipe symbol (|) in you stopwords file, I have no clue what
Solr will make of it. The stopwords file format is usually just one
word per line.
I'm assuming your name of "string" for the field type is just a placeholder
or
