[
https://issues.apache.org/jira/browse/KYLIN-6007?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17928273#comment-17928273
]
Guoliang Sun commented on KYLIN-6007:
-------------------------------------
h3. RootCause
The `table_identity column in the metadata `table_info` table is not populated,
causing all tables to fail matching during fuzzy matching by table name,
resulting in an empty result.
h3. Dev Design
Some independent columns, such as `name` and `table_identity`, were extracted
from `table_info`. The principle is that only values already present in the
original JSON can be split into independent columns. However, `table_identity`
does not exist in the JSON, making it unsuitable for extraction.
* Modify the table structure: Remove the `table_identity` column from the
`table_info` table and add a `dbName` column.
* Update the filter logic: Split the fuzzy matching key into database name and
table name, then perform the filtering operation.
> On the task list page, snapshot tasks cannot be filtered by table name
> ----------------------------------------------------------------------
>
> Key: KYLIN-6007
> URL: https://issues.apache.org/jira/browse/KYLIN-6007
> Project: Kylin
> Issue Type: Bug
> Affects Versions: 5.0.0
> Reporter: Guoliang Sun
> Priority: Major
>
> The *table_identity* field in the table metadata is empty, causing incorrect
> job query condition concatenation.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
