[jira] [Commented] (LUCENE-8041) All Fields.terms(fld) impls should be O(1) not O(log(N))

["Robert Muir (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/LUCENE-8041?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16508692#comment-16508692 ] 

Robert Muir commented on LUCENE-8041:
-------------------------------------

This has the downside that it sorts all fields on every call to iterator(). My concern is mainly that it will introduce performance problems down the line, ones that are difficult to find/debug because of java's syntactic sugar around iterator(). Especially if someone is using MultiFields (slow-wrapper crap), they will be doing a bunch of sorts on each segment, then merging those, and all hidden behind a single call to iterator().

I still feel the best would be to remove this map entirely: then you can be sure there aren't traps. The only thing blocking this is the fact that term-vector options are configurable per-doc, which doesnt make sense anyway.

> All Fields.terms(fld) impls should be O(1) not O(log(N))
> --------------------------------------------------------
>
>                 Key: LUCENE-8041
>                 URL: https://issues.apache.org/jira/browse/LUCENE-8041
>             Project: Lucene - Core
>          Issue Type: Improvement
>            Reporter: David Smiley
>            Priority: Major
>         Attachments: LUCENE-8041.patch
>
>
> I've seen apps that have a good number of fields -- hundreds.  The O(log(N)) of TreeMap definitely shows up in a profiler; sometimes 20% of search time, if I recall.  There are many Field implementations that are impacted... in part because Fields is the base class of FieldsProducer.  
> As an aside, I hope Fields to go away some day; FieldsProducer should be TermsProducer and not have an iterator of fields. If DocValuesProducer doesn't have this then why should the terms index part of our API have it?  If we did this then the issue here would be a simple transition to a HashMap.
> Or maybe we can switch to HashMap and relax the definition of Fields.iterator to not necessarily be sorted?
> Perhaps the fix can be a relatively simple conversion over to LinkedHashMap in many cases if we can assume when we initialize these internal maps that we consume them in sorted order to begin with.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org