[jira] [Updated] (HBASE-11131) Improve essential column family handling when more than one SingleColumnValueFilter is involved

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

     [ https://issues.apache.org/jira/browse/HBASE-11131?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ted Yu updated HBASE-11131:
---------------------------

    Component/s: Filters

> Improve essential column family handling when more than one SingleColumnValueFilter is involved
> -----------------------------------------------------------------------------------------------
>
>                 Key: HBASE-11131
>                 URL: https://issues.apache.org/jira/browse/HBASE-11131
>             Project: HBase
>          Issue Type: Task
>          Components: Filters
>            Reporter: Ted Yu
>            Priority: Minor
>
> Currently, SingleColumnValueFilter determines essential column family based on:
> {code}
>   public boolean isFamilyEssential(byte[] name) {
>     return !this.filterIfMissing || Bytes.equals(name, this.columnFamily);
> {code}
> When more than one SingleColumnValueFilter is used through FilterList, some optimization can be done.
> e.g. when they are chained by Operator.MUST_PASS_ALL, the second and subsequent SingleColumnValueFilter can declare to be non-essential.
> This way their evaluation would depend on the outcome of the first SingleColumnValueFilter.



--
This message was sent by Atlassian JIRA
(v6.2#6252)