[
https://issues.apache.org/jira/browse/LUCENE-7863?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16054803#comment-16054803
]
Mikhail Khludnev commented on LUCENE-7863:
------------------------------------------
Got it. Nice decision!
So, instead of searching for name:\*ar, it flips query to name_rev:ra*, then
for every doc (if we need phrase logic or highlighting): it seeks original
term's postings to the same doc, and read positions and offsets.
Thinking about EdgeNGramms (searching for name:\*a\*), derivative field should
go like this: {{ar_bar}}, {{r_bar}} to be able to switch to original term's
posting. So, I still think that even with this approach (second DOCS_ONLY
field) blowing postings by these derivative terms still might not be
affordable.
And coming back to your question:
bq. have you thought of using another field that only has the reversed terms?
No. I haven't thought about it. It's a great idea! Thanks for contributing it.
> Don't repeat postings and positions on ReverseWF, EdgeNGram, etc
> ------------------------------------------------------------------
>
> Key: LUCENE-7863
> URL: https://issues.apache.org/jira/browse/LUCENE-7863
> Project: Lucene - Core
> Issue Type: Improvement
> Components: core/index
> Reporter: Mikhail Khludnev
> Attachments: LUCENE-7863.hazard
>
>
> h2. Context
> \*suffix and \*infix\* searches on large indexes.
> h2. Problem
> Obviously applying {{ReversedWildcardFilter}} doubles an index size, and I'm
> shuddering to think about EdgeNGrams...
> h2. Proposal
> _DRY_
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
