You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@ignite.apache.org by "Roman Puchkovskiy (Jira)" <ji...@apache.org> on 2023/12/19 11:46:00 UTC

[jira] [Updated] (IGNITE-20133) Compute hashes for integral/decimal columns in a stable way

     [ https://issues.apache.org/jira/browse/IGNITE-20133?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Roman Puchkovskiy updated IGNITE-20133:
---------------------------------------
    Epic Link: IGNITE-17767  (was: IGNITE-20782)

> Compute hashes for integral/decimal columns in a stable way
> -----------------------------------------------------------
>
>                 Key: IGNITE-20133
>                 URL: https://issues.apache.org/jira/browse/IGNITE-20133
>             Project: Ignite
>          Issue Type: Improvement
>            Reporter: Roman Puchkovskiy
>            Priority: Minor
>              Labels: ignite-3
>             Fix For: 3.0.0-beta2
>
>
> The idea is to make hash computation for integral and decimal types satisfy the following property: if a column type is changed from an integral to a decimal type, the hashes for values that are already stored remain the same.
> This will allow us to permit chaning type (integral -> decimal and decimal -> longer decimal) of a column that is included in a HASH index.
> A hash that has this property is the following function: hash(val.toString(TRIM_TRAILING_ZEROS)). For instance, for 1 it will be hash("1"), for 1.000 it will also be hash("1"), but for 1.23 it will give hash("1.23").



--
This message was sent by Atlassian Jira
(v8.20.10#820010)