hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ted Yu (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-19550) Wrap the Mutation in cp layer to make sure all passed cells are of ExtendedCell
Date Sun, 24 Dec 2017 02:23:00 GMT

    [ https://issues.apache.org/jira/browse/HBASE-19550?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16302641#comment-16302641

Ted Yu commented on HBASE-19550:

Looks good overall.
Minor comments below.
+    } catch (IOException e) {
+      // we eat the exception of wrong row for BC..
Should the IOE be logged ?
In Mutation :
+  Mutation add(Cell cell) throws IOException {
+    //Checking that the row of the kv is the same as the put
'put' is not accurate, right ?
+  private static final class CellWrap implements ExtendedCell {
CellWrap -> CellWrapper
+    public Optional<Tag> getTag(byte type) {
The implementation is similar to PrivateCellUtil#getTag(). Is it possible to refactor the
code ?

> Wrap the Mutation in cp layer to make sure all passed cells are of ExtendedCell
> -------------------------------------------------------------------------------
>                 Key: HBASE-19550
>                 URL: https://issues.apache.org/jira/browse/HBASE-19550
>             Project: HBase
>          Issue Type: Task
>            Reporter: Chia-Ping Tsai
>            Assignee: Chia-Ping Tsai
>             Fix For: 2.0.0-beta-2
>         Attachments: HBASE-19550.v0.patch
> We assume all cells in server are of ExtendedCell. However, cp user can add their cell
impl via Put#add(Cell) in observer. That will cause UnsupportedOperationException when rs
try to update the cell's timestamp and seq Id. We should do something for cp user...For example,
wrap the passed cells to be a extendcell type.

This message was sent by Atlassian JIRA

View raw message