Nobody mentioned it so far... with utf8mb4 which is 4-byte and can also store emoticons (we should never more use 3-byte utf8) and we can avoid errors like Incorrect string value: \xF0\x9F\x98\... we should not use typical VARCHAR(255) but rather VARCHAR(191) because in case utf8mb4 and VARCHAR(255) same part of data are stored off-page and you can not create index for column VARCHAR(255) but for VARCHAR(191) you can. It is because the maximum indexed column size is 767 bytes for ROW_FORMAT=COMPACT or ROW_FORMAT=REDUNDANT.

For newer row formats ROW_FORMAT=DYNAMIC or ROW_FORMAT=COMPRESSED (which requires newer file format innodb_file_format=Barracuda not older Antelope) maximum indexed column size is 3072. It is available since MySQL >= 5.6.3 when innodb_large_prefix=1 (disabled by default for MySQL <= 5.7.6 and enabled by default for MySQL >= 5.7.7). So in this case we can use VARCHAR(768) for utf8mb4 (or VARCHAR(1024) for old utf8) for indexed column. Option innodb_large_prefix is deprecated since 5.7.7 because its behavior is built-in MySQL 8 (in this version is option removed).

你可以在alter table请求中指定键的长度，就像这样:

alter table authors ADD UNIQUE(name_first(20), name_second(20));

我知道现在已经很晚了，但是删除唯一键约束解决了这个问题。我没有使用TEXT或LONGTEXT列作为PK，但我试图使它独特。我得到了1170的错误，但当我删除了英国，错误也被删除了。

我不完全明白为什么。

Nobody mentioned it so far... with utf8mb4 which is 4-byte and can also store emoticons (we should never more use 3-byte utf8) and we can avoid errors like Incorrect string value: \xF0\x9F\x98\... we should not use typical VARCHAR(255) but rather VARCHAR(191) because in case utf8mb4 and VARCHAR(255) same part of data are stored off-page and you can not create index for column VARCHAR(255) but for VARCHAR(191) you can. It is because the maximum indexed column size is 767 bytes for ROW_FORMAT=COMPACT or ROW_FORMAT=REDUNDANT.

For newer row formats ROW_FORMAT=DYNAMIC or ROW_FORMAT=COMPRESSED (which requires newer file format innodb_file_format=Barracuda not older Antelope) maximum indexed column size is 3072. It is available since MySQL >= 5.6.3 when innodb_large_prefix=1 (disabled by default for MySQL <= 5.7.6 and enabled by default for MySQL >= 5.7.7). So in this case we can use VARCHAR(768) for utf8mb4 (or VARCHAR(1024) for old utf8) for indexed column. Option innodb_large_prefix is deprecated since 5.7.7 because its behavior is built-in MySQL 8 (in this version is option removed).

DROP该表并再次运行Spring Project。 这可能会有所帮助。 有时你重写了foreignKey。

MySQL不允许索引BLOB, TEXT和长VARCHAR列的完整值，因为它们包含的数据可能很大，隐式DB索引也很大，这意味着索引没有好处。

MySQL要求定义要索引的前N个字符，诀窍是选择一个足够长的数字N，以提供良好的选择性，但又足够短，以节省空间。前缀应该足够长，以使索引几乎与索引整个列时一样有用。

在进一步讨论之前，让我们先定义一些重要的术语。索引选择性是总不同索引值与总行数之比。下面是一个测试表的例子:

+-----+-----------+
| id  | value     |
+-----+-----------+
| 1   | abc       |
| 2   | abd       |
| 3   | adg       |
+-----+-----------+

如果我们只索引第一个字符(N=1)，那么索引表将如下表所示:

+---------------+-----------+
| indexedValue  | rows      |
+---------------+-----------+
| a             | 1,2,3     |
+---------------+-----------+

在这种情况下，索引选择性等于is =1/3 = 0.33。

现在让我们看看如果我们将索引字符的数量增加到两个(N=2)会发生什么。

+---------------+-----------+
| indexedValue  | rows      |
+---------------+-----------+
| ab             | 1,2      |
| ad             | 3        |
+---------------+-----------+

在这个场景中，IS=2/3=0.66，这意味着我们增加了索引的选择性，但我们也增加了索引的大小。诀窍是找到最小的数字N，这将导致最大的索引选择性。

有两种方法可以对数据库表进行计算。我将对此数据库转储进行演示。

假设我们想要将表employees中的列last_name添加到索引中，并且想要定义能产生最佳索引选择性的最小数字N。

首先让我们来看看最常见的姓氏:

select count(*) as cnt, last_name 
from employees 
group by employees.last_name 
order by cnt

+-----+-------------+
| cnt | last_name   |
+-----+-------------+
| 226 | Baba        |
| 223 | Coorg       |
| 223 | Gelosh      |
| 222 | Farris      |
| 222 | Sudbeck     |
| 221 | Adachi      |
| 220 | Osgood      |
| 218 | Neiman      |
| 218 | Mandell     |
| 218 | Masada      |
| 217 | Boudaillier |
| 217 | Wendorf     |
| 216 | Pettis      |
| 216 | Solares     |
| 216 | Mahnke      |
+-----+-------------+
15 rows in set (0.64 sec)

如你所见，姓巴巴的人出现频率最高。现在我们要找出最常出现的last_name前缀，从五个字母的前缀开始。

+-----+--------+
| cnt | prefix |
+-----+--------+
| 794 | Schaa  |
| 758 | Mande  |
| 711 | Schwa  |
| 562 | Angel  |
| 561 | Gecse  |
| 555 | Delgr  |
| 550 | Berna  |
| 547 | Peter  |
| 543 | Cappe  |
| 539 | Stran  |
| 534 | Canna  |
| 485 | Georg  |
| 417 | Neima  |
| 398 | Petti  |
| 398 | Duclo  |
+-----+--------+
15 rows in set (0.55 sec)

每个前缀出现的次数都要多得多，这意味着我们必须增加数字N，直到值几乎与前一个例子中的值相同。

下面是N=9时的结果

select count(*) as cnt, left(last_name,9) as prefix 
from employees 
group by prefix 
order by cnt desc 
limit 0,15;

+-----+-----------+
| cnt | prefix    |
+-----+-----------+
| 336 | Schwartzb |
| 226 | Baba      |
| 223 | Coorg     |
| 223 | Gelosh    |
| 222 | Sudbeck   |
| 222 | Farris    |
| 221 | Adachi    |
| 220 | Osgood    |
| 218 | Mandell   |
| 218 | Neiman    |
| 218 | Masada    |
| 217 | Wendorf   |
| 217 | Boudailli |
| 216 | Cummings  |
| 216 | Pettis    |
+-----+-----------+

下面是N=10时的结果。

+-----+------------+
| cnt | prefix     |
+-----+------------+
| 226 | Baba       |
| 223 | Coorg      |
| 223 | Gelosh     |
| 222 | Sudbeck    |
| 222 | Farris     |
| 221 | Adachi     |
| 220 | Osgood     |
| 218 | Mandell    |
| 218 | Neiman     |
| 218 | Masada     |
| 217 | Wendorf    |
| 217 | Boudaillie |
| 216 | Cummings   |
| 216 | Pettis     |
| 216 | Solares    |
+-----+------------+
15 rows in set (0.56 sec)

这是非常好的结果。这意味着我们可以在列last_name上创建索引，只索引前10个字符。在表定义中，列last_name被定义为VARCHAR(16)，这意味着每个条目节省了6个字节(如果姓氏中有UTF8个字符，则节省了更多字节)。在这个表中，有1637个不同的值乘以6个字节大约是9KB，想象一下，如果我们的表包含数百万行，这个数字将如何增长。

你可以在我的文章中阅读其他计算N的方法。