MySQL Bugs: #95205: Support of U+32FF, a ligature of Japanese new regnal year "Reiwa"

Bug #95205	Support of U+32FF, a ligature of Japanese new regnal year "Reiwa"
Submitted:	30 Apr 2019 15:01	Modified:	1 May 2019 6:42
Reporter:	Ryusuke Kajiyama	Email Updates:
Status:	Verified	Impact on me:	None
Category:	MySQL Server: Charsets	Severity:	S4 (Feature request)
Version:	8.0	OS:	Any
Assigned to:		CPU Architecture:	Any

Description:
Collation of MySQL is not collate "U+32FF" (a ligature of Japanese new regnal year "Reiwa") and Decomposition Mapping "U+4EE4 U+548C" as expected.
U+32FF is defined as of Unicode 12.1;
 http://unicode.org/versions/Unicode12.1.0/

The new Japanese regnal name was officially effective as of May 1, 2019.

Code points of ligatures of previous eras U+337E (㍾), U+337D (㍽), U+337C (㍼), and U+337B (㍻) are handled properly.

How to repeat:
MySQL > SELECT _utf8mb4 0xe5b9b3e68890 = _utf8mb4 0xE38DBB AS "平成 = U+337B";
+---------------+
| 平成 = U+337B |
+---------------+
|             1 |
+---------------+
1 row in set (0.0005 sec)

MySQL > SELECT _utf8mb4 0xE4BBA4E5928C = _utf8mb4 0xE38BBF AS "令和 = U+32FF";
+------------------------+
| 令和 = ligature U+32FF |
+------------------------+
|                      0 |
+------------------------+
1 row in set (0.0090 sec)

Please refer 5th code in the following blog entity by Masahiro Tomita;
 https://tmtms.hatenablog.com/entry/201904/mysql-reiwa

Suggested fix:
Support Unicode 12.1 in collations or update existing utf8mb4_is_0900_* collations.
The latter should be also 1.

Thank you for the feature request.

May 1 2019 is in past. Maybe time to update ICU code, shipped with the server?

ICU is only used for regular expressions in MySQL 8.0.

In order to support the Reiwa ligature, we will have to add new collations based on Unicode 12.1.0 or later. We will need new collations in any case, since old collations are never modified. I don't have a timeline for this work.

(The ICU version is scheduled for upgrade in 8.0.21.)