Support utf8mb3 charset naming for MySQL 8 and MariaDB 10.6 (#1323)

MySQL and MariaDB have long aliased the `utf8` charset to the
underlying `utf8mb3`. They both switched to report the underlying
charset name instead of the alias name, allowing `utf8` to target
`utf8mb4` in the future.

That means we need to explicitly map `utf8mb3` for folks running
newer MySQL/MariaDB with older `utf8` dbs.

Author: jeremy

ext/mysql2/mysql_enc_name_to_ruby.h

@@ -51,7 +51,7 @@ mysql2_mysql_enc_name_to_rb_hash (str, len)
       74, 74, 74, 74, 74, 74, 74, 74, 74, 74,
       74, 74, 74, 74, 74, 74, 74, 74, 74, 74,
       74, 74, 74, 74, 74, 74, 74, 74, 15,  5,
-       0, 74,  5, 25, 40, 10, 20, 50, 74, 74,
+       0, 30,  5, 25, 40, 10, 20, 50, 74, 74,
       74, 74, 74, 74, 74, 74, 74, 74, 74, 74,
       74, 74, 74, 74, 74, 74, 74, 74, 74, 74,
       74, 74, 74, 74, 74, 74, 74, 74, 74, 74,
@@ -89,7 +89,7 @@ mysql2_mysql_enc_name_to_rb (str, len)
 {
   enum
     {
-      TOTAL_KEYWORDS = 41,
+      TOTAL_KEYWORDS = 42,
       MIN_WORD_LENGTH = 3,
       MAX_WORD_LENGTH = 8,
       MIN_HASH_VALUE = 3,
@@ -133,7 +133,8 @@ mysql2_mysql_enc_name_to_rb (str, len)
       {"big5", "Big5"},
       {"euckr", "EUC-KR"},
       {"latin2", "ISO-8859-2"},
-      {""}, {""},
+      {"utf8mb3", "UTF-8"},
+      {""},
       {"dec8", NULL},
       {"cp850", "CP850"},
       {"latin1", "ISO-8859-1"},

ext/mysql2/mysql_enc_to_ruby.h

@@ -306,5 +306,20 @@ static const char *mysql2_mysql_enc_to_rb[] = {
   "UTF-8",
   "UTF-8",
   "UTF-8",
+  "UTF-8",
+  "UTF-8",
+  "UTF-8",
+  "UTF-8",
+  "UTF-8",
+  "UTF-8",
+  "UTF-8",
+  "UTF-8",
+  "UTF-8",
+  "UTF-8",
+  "UTF-8",
+  "UTF-8",
+  "UTF-8",
+  "UTF-8",
+  "UTF-8",
   "UTF-8"
 };

support/mysql_enc_to_ruby.rb

@@ -33,6 +33,7 @@
   "macroman" => "macRoman",
   "cp852"    => "CP852",
   "latin7"   => "ISO-8859-13",
+  "utf8mb3"  => "UTF-8",
   "utf8mb4"  => "UTF-8",
   "cp1251"   => "Windows-1251",
   "utf16"    => "UTF-16",

support/ruby_enc_to_mysql.rb

@@ -28,6 +28,7 @@
   "macroman" => "macRoman",
   "cp852"    => "CP852",
   "latin7"   => "ISO-8859-13",
+  "utf8mb3"  => "UTF-8",
   "utf8mb4"  => "UTF-8",
   "cp1251"   => "Windows-1251",
   "utf16"    => "UTF-16",