I am using solr in cloud mode with schemaless mode. I don’t want to
update/touch managed schema. Is there any way I can send the copy of those
fields using solj ? So the text value will be copied to string value.
Thanks
Subhasis Patra
240-755-2601
subhasis.pa...@e2open.com
-----Original Message-----
From: ufuk yılmaz <uyil...@vivaldi.net.INVALID>
Sent: Thursday, May 11, 2023 5:58 AM
To: users@solr.apache.org
Subject: Re: Streaming of Documents with text columns (_txt)
PHISH ALERT! CHECK VALIDITY IF CLICKING, SHARING, RESPONDING
My solution to this kind of situation is to have a docValues enabled copyField
for each text field in the schema, so I can export all of the fields when
necessary
-ufuk yilmaz
—
> On 11 May 2023, at 05:08, Subhasis Patra <subhasis.pa...@e2open.com> wrote:
>
> Hi All,
>
> I am using CloudSolrStream to get stream data for documents in Solr. I am
> using /export When documents have columns of type STRING, DATE, DOUBLE, LONG.
> It does not allow /export when documents have _txt column(DocValues=false).
> So I use as below. I use _txt to support case insensitive search.
>
> StreamFactory factory = new
> StreamFactory().withCollectionZkHost(collection, zkHost);
> StreamExpression streamExpression =
> StreamExpressionParser.parse("search(" + collection + ",
> q=\""+filter+"\", fl=\""+filedsCommaSeparated+"\",rows=\""+count+"\",
> sort=\"id asc\")");
>
> This works, but it does not support memory management like /export. Limiting
> rows by using start parameters slows down the process.
> Can anyone help me how to achieve this ?
>
>
> Thanks
> Subhasis Patra
> 240-755-2601
> subhasis.pa...@e2open.com<mailto:subhasis.pa...@e2open.com>
>
