[
https://issues.apache.org/jira/browse/SOLR-6343?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14091296#comment-14091296
]
Steve Rowe commented on SOLR-6343:
----------------------------------
{quote}
bq. Do you plan on ignoring all keys in all nested arrays/objects?
yes
{quote}
I suspect this may be problematic for some structures, e.g.:
{code:javascript}
{ "John Q. Public": { "phone": "XXX-XXX-XXXX", "location": "...", ... },
"Jane S. Private": { "phone": ... },
...
}
{code}
Those names are keys, and would not be indexed according to your plan.
Maybe you want to ignore keys because they can be repeated many times, like
"phone" above? (if that's not the reason why, then: why?) If so, an
alternative to ignoring keys might be to only index each unique key once.
> add a new end point /json/raw to index json for full text search
> -----------------------------------------------------------------
>
> Key: SOLR-6343
> URL: https://issues.apache.org/jira/browse/SOLR-6343
> Project: Solr
> Issue Type: New Feature
> Reporter: Noble Paul
> Assignee: Noble Paul
>
> I should be able to just put in any random json to an index to the end point
> {{/json/raw}} and all the terms can just get indexed to the default search
> field. There should be a way to store the entire JSON as well by adding an
> extra param {{target=fieldName}} and it can store the whole payload to that
> field
--
This message was sent by Atlassian JIRA
(v6.2#6252)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@lucene.apache.org
For additional commands, e-mail: dev-h...@lucene.apache.org
