Thanks Erick for this last confirmation. I've at the end I've used the
standard "text_ws":
And the field
On Fri, Sep 6, 2019 at 2:54 AM Erick Erickson
wrote:
> bq. What I do not understand is what happens to the Analyzers, Tokenizers,
> and
> Filters in the i
bq. What I do not understand is what happens to the Analyzers, Tokenizers, and
Filters in the indexing chain
They are irrelevant. The analysis chain is only executed when indexed=true.
Best,
Erick
> On Sep 5, 2019, at 9:03 AM, Vincenzo D'Amore wrote:
>
> What I do not understand is what happe
I agree, stored=true and indexed =false should resolve this size issue.
On Thu, 5 Sep 2019 at 21:54, Erick Erickson wrote:
> Use a text field with stored=true and indexed=false? That'll allow you to
> return it...
>
> On Thu, Sep 5, 2019, 07:04 Vincenzo D'Amore wrote:
>
> > Hi all,
> >
> > sorr
Thanks Erick for the prompt answer.
What I do not understand is what happens to the Analyzers, Tokenizers, and
Filters in the indexing chain.
Are they executed or not? Well, answering to my own question I think no,
but so what's the difference between string and text when they are not
indexed?
Just
Use a text field with stored=true and indexed=false? That'll allow you to
return it...
On Thu, Sep 5, 2019, 07:04 Vincenzo D'Amore wrote:
> Hi all,
>
> sorry for the silly question, I need to store in Solr a string field larger
> than 32k (index="false").
>
> Given that storing field larger than
fieldType "string" is not tokenized, so your observation is correct. You need
to use a fieldType with analysis and tokenization to get the behavior you want.
--
Jan Høydahl, search solution architect
Cominvent AS - www.cominvent.com
25. juni 2013 kl. 02:35 skrev "Mugoma Joseph O." :
>
> It loo
It looks like partial search works only with copied to field. This works:
$ curl
"http://localhost:8282/solr/links/select?q=text_ngrams:yengas&wt=json&indent=on&fl=id,domain,score";
On Tue, June 25, 2013 12:39 am, Mugoma Joseph O. wrote:
> Hello,
>
> I am newbie to solr.
>
> I am trying out part
Hi,
You can try to increase the "pf" boost for your string field, I don't think
you'll have success in having it boosted with pf since it's a string? Check
explain output with &debugQuery=true and see whether you get a phrase boost.
--
Jan Høydahl, search solution architect
Cominvent AS - www.c
Hi Jan
my question is when I tweak pf and qf parameter and the results change
slightly and I do not think for exact match you need to implement the
solution that you mentioned in your reply. you can always have string field
and in your pf parameter you can boost that field to get the exact match
r
Hi,
The "pf" feature will only kick in for phrases, i.e. multiple tokens. Per
definition a "string" is one single token, so it will never kick in for strings.
A workaround can be found here: https://github.com/cominvent/exactmatch
--
Jan Høydahl, search solution architect
Cominvent AS - www.com
The full import wasn't spitting out any errors on the web page but in
looking at the logs, there were errors. Correcting those errors solved that
issue.
Thanks,
Brian Lamb
On Tue, Mar 29, 2011 at 2:44 PM, Erick Erickson wrote:
> try the schema browser from the admin page to be sure the fields
>
try the schema browser from the admin page to be sure the fields
you *think* are in the index really are. Did you do a commit
after indexing? Did you re-index after the schema changes? Are
you 100% sure that, if you did re-index, the new fields were in the
docs submitted?
Best
Erick
On Tue, Mar 2
First, make sure your request handler is set to spit out everything. I take
it you did, but I hate to assume.
Second, I suggest indexing your data twice. One as tokenized-text, the
other as a string. It'll save you from howling at the moon in anguish...
Unless you really only do care about pure
Thank you,It is perfectly working
--
View this message in context:
http://lucene.472066.n3.nabble.com/String-field-with-lower-case-filter-tp1930941p1935283.html
Sent from the Solr - User mailing list archive at Nabble.com.
> But with above configuration i am not getting any
> results.Can anybody has idea.
class="solr.StrField" should be replaced with class="solr.TextField"
15 matches
Mail list logo
