Mailing-List: contact issues-help@hbase.apache.org; run by ezmlm
Precedence: bulk
Date: Tue, 31 Oct 2017 20:55:01 +0000 (UTC)
From: "Jan Hentschel (JIRA)" <jira@apache.org>
To: issues@hbase.apache.org
Message-ID: <JIRA.13112480.1509085114000.115990.1509483301479@Atlassian.JIRA>
In-Reply-To: <JIRA.13112480.1509085114000@Atlassian.JIRA>
References: <JIRA.13112480.1509085114000@Atlassian.JIRA> <JIRA.13112480.1509085114799@jira-lw-us.apache.org>
Subject: [jira] [Commented] (HBASE-19103) Add BigDecimalComparator for
 filter
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Tue, 31 Oct 2017 20:55:07 -0000


    [ https://issues.apache.org/jira/browse/HBASE-19103?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16227534#comment-16227534 ] 

Jan Hentschel commented on HBASE-19103:
---------------------------------------

Your implementation of hashCode and equals look a little bit off. Why do you include hashCode and toString in the equals implementation?

I'm currently doing something similar in HBASE-19008 and the template looks similar to the following (taken from ColumnCountGetFilter)

{code:java}
@Override
  public boolean equals(Object o) {
    if (o == this) {
      return true;
    }

    if (!(o instanceof  ColumnCountGetFilter)) {
      return false;
    }

    final ColumnCountGetFilter that = (ColumnCountGetFilter) o;
    return Objects.equals(limit, that.limit);
  }

  @Override
  public int hashCode() {
    return Objects.hash(limit);
  }
{code}

A test case covering this looks like the following

{code:java}
@Test
  public void testEquals() {
    // Check that equals returns true for identical objects
    final ColumnCountGetFilter filter = new ColumnCountGetFilter(5);
    assertTrue(filter.equals(filter));
    assertEquals(filter.hashCode(), filter.hashCode());

    // Check that equals returns true for the same object
    final ColumnCountGetFilter filter1 = new ColumnCountGetFilter(5);
    final ColumnCountGetFilter filter2 = new ColumnCountGetFilter(5);
    assertTrue(filter1.equals(filter2));
    assertEquals(filter1.hashCode(), filter2.hashCode());

    // Check that equals returns false for different objects
    final ColumnCountGetFilter filter3 = new ColumnCountGetFilter(4);
    final ColumnCountGetFilter filter4 = new ColumnCountGetFilter(5);
    assertFalse(filter3.equals(filter4));
    assertNotEquals(filter3.hashCode(), filter4.hashCode());

    // Check that equals returns false for a different type
    final ColumnCountGetFilter filter5 = new ColumnCountGetFilter(5);
    assertFalse(filter5.equals(0));
  }
{code}

This also eliminates the need of a hashCode field in the class.

> Add BigDecimalComparator for filter
> -----------------------------------
>
>                 Key: HBASE-19103
>                 URL: https://issues.apache.org/jira/browse/HBASE-19103
>             Project: HBase
>          Issue Type: New Feature
>          Components: Client
>            Reporter: Qilin Cao
>            Assignee: Qilin Cao
>            Priority: Minor
>             Fix For: 1.2.0, 3.0.0
>
>         Attachments: HBASE-19103-1.2.0-v1.patch, HBASE-19103-1.2.0-v2.patch, HBASE-19103-trunk-v1.patch, HBASE-19103-trunk-v2.patch
>
>
> Should I add BigDecimalComparator for filter? Some scenarios need to calculate the data accurately may use it.


--
This message was sent by Atlassian JIRA
(v6.4.14#64029)