charset: support collation utf8mb4_unicode_ci and utf8_unicode_ci #17596
Closed
Description
Feature Request
Is your feature request related to a problem? Please describe:
Currently, TiDB doesn't support utf8mb4_unicode_ci and utf8_unicode_ci when new collation is enabled.
tidb> set names utf8 collate utf8_unicode_ci;
ERROR 1273 (HY000): Unsupported collation when new collation is enabled: 'utf8_unicode_ci'.
unicode_ci is a widely used collation in MySQL, it would be better if TiDB can support it.
Describe the feature you'd like:
Support collation utf8mb4_unicode_ci and utf8_unicode_ci when new collation is enabled.
Besides implementing the algorithm for unicode_ci, we need to think over how to incorporate it into the current new collation frame.
For example, what if concat(general_ci_str, unicode_ci_str)? How constant propagation work with unicode_ci?
Mentor(s)
Contact the mentors: #ddl-sig channel in TiDB Community Slack Workspace
Recommended Skills
- Golang
- Rust
Learning Materials
- Chinese: 字符集和排序规则
- English: Character Set and Collation
Schedule
- GanttStart: 2020-07-01
- GanttDue: 2020-10-15
- GanttProgress: 100%
Activity