Details
-
Type:
Task
-
Status: Open
-
Priority:
Major
-
Resolution: Unresolved
-
Fix Version/s: None
-
Component/s: None
-
Labels:None
Description
Can you implement a function (or any other syntax, if a function is not appropriated) returning the name of a partition/subpartition containing a PK entry? For example:
SELECT PARTITION(id) FROM user WHERE id = 98765;
Why am I asking this? Consider RANGE [COLUMNS] and LIST [COKUMNS] partitioning. The name of the (sub)partition is often a useful information. Of course, if we select the columns involved in the partitioning expression, we can calculate this information; but having the information pre-calculated by MariaDB could be more efficient, and more convenient. It could even allow to hide the DB complexity to developers.
(Apache Cassandra has even Static Columns, that is, a single value can be assigned to all rows in a given partition. But it seems to me unnecessarily complicated)
Gliffy Diagrams
Attachments
Activity
- All
- Comments
- Work Log
- History
- Activity
- Transitions
i think this could be developed with a UDF, but i don't know how could more than one column be used, something like:
partition_name( schema name, table name, column1, value1, column2, value2, ...... , column<n>, value<n>)