[jira] [Commented] (ASTERIXDB-1907) Unify the two methods in IBinaryComparatorFactoryProvider

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

    [ https://issues.apache.org/jira/browse/ASTERIXDB-1907?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16009566#comment-16009566 ] 

Till commented on ASTERIXDB-1907:
---------------------------------

One possible fix would be to extend the type system with a string-type that ignores case during comparison.

But as this comparison seems to be only used for the comparison of tokens for inverted indexes it could be more appropriate to create an instance of {{IBinaryComparatorFactoryProvider}} solely for secondary indexes.  This provider could provide BinaryComparatorFactories that ignore case when comparing (string) tokens.

[~wangsaeu], what do you think about these options?

> Unify the two methods in IBinaryComparatorFactoryProvider
> ---------------------------------------------------------
>
>                 Key: ASTERIXDB-1907
>                 URL: https://issues.apache.org/jira/browse/ASTERIXDB-1907
>             Project: Apache AsterixDB
>          Issue Type: Bug
>          Components: Hyracks Core
>            Reporter: Abdullah Alamoudi
>            Assignee: Abdullah Alamoudi
>
> The interface {{IBinaryComparatorFactoryProvider}} has 2 methods {{getBinaryComparatorFactory}}, one with the parameters
> {noformat}
> (Object type, boolean ascending)
> {noformat}
> and one with the parameters
> {noformat}
> (Object type, boolean ascending, boolean ignoreCase)
> {noformat}
> The 2nd signature seems problematic, as it is not clear what {{ignoreCase}} should mean for a non-string datatype.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)