[
https://issues.apache.org/jira/browse/HBASE-8388?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ted Yu updated HBASE-8388:
--------------------------
Attachment: 8388-trunk-v1.txt
In StoreFile#passesGeneralBloomFilter(), there was this check for ROW bloom :
{code}
- if (rowOffset != 0 || rowLen != row.length) {
- throw new AssertionError("For row-only Bloom filters the row "
- + "must occupy the whole array");
- }
{code}
Trunk patch v1 replaces this check and remembers key offset and key length.
> Use ROW bloom filter for scanning
> ---------------------------------
>
> Key: HBASE-8388
> URL: https://issues.apache.org/jira/browse/HBASE-8388
> Project: HBase
> Issue Type: Bug
> Reporter: Lars Hofhansl
> Fix For: 0.98.0, 0.94.8, 0.95.1
>
> Attachments: 8388-0.94-sketch.txt, 8388-trunk-v1.txt
>
>
> Currently only ROWCOL BFs are used during the scanning process. It seems we
> could make use of ROW BFs in StoreFileScanner.requestSeek.
> See HBASE-8362 for background.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
