像@Venkatramanan和其他人一样，我找到了INFORMATION_SCHEMA。TABLES不可靠(使用InnoDB, MySQL 5.1.44)，每次运行时给出不同的行数，即使是在静态表上。这里有一种生成大型SQL语句的相对hack(但是灵活/适应性强)的方法，您可以将其粘贴到新的查询中，而不需要安装Ruby宝石之类的东西。

SELECT CONCAT(
    'SELECT "', 
    table_name, 
    '" AS table_name, COUNT(*) AS exact_row_count FROM `', 
    table_schema,
    '`.`',
    table_name, 
    '` UNION '
) 
FROM INFORMATION_SCHEMA.TABLES 
WHERE table_schema = '**my_schema**';

它产生如下输出:

SELECT "func" AS table_name, COUNT(*) AS exact_row_count FROM my_schema.func UNION                         
SELECT "general_log" AS table_name, COUNT(*) AS exact_row_count FROM my_schema.general_log UNION           
SELECT "help_category" AS table_name, COUNT(*) AS exact_row_count FROM my_schema.help_category UNION       
SELECT "help_keyword" AS table_name, COUNT(*) AS exact_row_count FROM my_schema.help_keyword UNION         
SELECT "help_relation" AS table_name, COUNT(*) AS exact_row_count FROM my_schema.help_relation UNION       
SELECT "help_topic" AS table_name, COUNT(*) AS exact_row_count FROM my_schema.help_topic UNION             
SELECT "host" AS table_name, COUNT(*) AS exact_row_count FROM my_schema.host UNION                         
SELECT "ndb_binlog_index" AS table_name, COUNT(*) AS exact_row_count FROM my_schema.ndb_binlog_index UNION 

复制粘贴，除了最后一个UNION，可以得到漂亮的输出，

+------------------+-----------------+
| table_name       | exact_row_count |
+------------------+-----------------+
| func             |               0 |
| general_log      |               0 |
| help_category    |              37 |
| help_keyword     |             450 |
| help_relation    |             990 |
| help_topic       |             504 |
| host             |               0 |
| ndb_binlog_index |               0 |
+------------------+-----------------+
8 rows in set (0.01 sec)

像许多其他人一样，我很难用InnoDB在INFORMATION_SCHEMA表上获得准确的值，并且能够通过count()进行查询将无限受益，并且希望在一次查询中完成它。

首先，确保启用大规模group_concats:

SET SESSION group_concat_max_len = 1000000;

然后运行此查询以获得将为数据库运行的结果查询。

SELECT CONCAT('SELECT ', GROUP_CONCAT(table1.count SEPARATOR ',\n')) FROM (
    SELECT concat('(SELECT count(id) AS \'',table_name,' Count\' ','FROM ',table_name,') AS ',table_name,'_Count') AS 'count'
    FROM information_schema.tables 
    WHERE table_schema = '**YOUR_DATABASE_HERE**'
) AS table1

这将生成诸如…

SELECT (SELECT count(id) AS 'table1 Count' FROM table1) AS table1_Count,
   (SELECT count(id) AS 'table2 Count' FROM table2) AS table2_Count,
   (SELECT count(id) AS 'table3 Count' FROM table3) AS table3_Count;

这反过来又产生了以下结果:

*************************** 1. row ***************************
table1_Count: 1
table2_Count: 1
table3_Count: 0

对于这个估算问题，有一点hack/workaround。

Auto_Increment -由于某些原因，如果您在表上设置了自动增量，则此函数将为数据库返回更准确的行数。

在探索为什么显示表信息与实际数据不匹配时发现了这一点。

SELECT
table_schema 'Database',
SUM(data_length + index_length) AS 'DBSize',
SUM(TABLE_ROWS) AS DBRows,
SUM(AUTO_INCREMENT) AS DBAutoIncCount
FROM information_schema.tables
GROUP BY table_schema;


+--------------------+-----------+---------+----------------+
| Database           | DBSize    | DBRows  | DBAutoIncCount |
+--------------------+-----------+---------+----------------+
| Core               |  35241984 |   76057 |           8341 |
| information_schema |    163840 |    NULL |           NULL |
| jspServ            |     49152 |      11 |            856 |
| mysql              |   7069265 |   30023 |              1 |
| net_snmp           |  47415296 |   95123 |            324 |
| performance_schema |         0 | 1395326 |           NULL |
| sys                |     16384 |       6 |           NULL |
| WebCal             |    655360 |    2809 |           NULL |
| WxObs              | 494256128 |  530533 |        3066752 |
+--------------------+-----------+---------+----------------+
9 rows in set (0.40 sec)

然后，您可以轻松地使用PHP或其他工具返回2个数据列的最大值，以给出行数的“最佳估计”。

即。

SELECT
table_schema 'Database',
SUM(data_length + index_length) AS 'DBSize',
GREATEST(SUM(TABLE_ROWS), SUM(AUTO_INCREMENT)) AS DBRows
FROM information_schema.tables
GROUP BY table_schema;

Auto Increment将始终是+1 *(表数)行，但即使有4000个表和300万行，这也是99.9%的准确性。比估计的行数好多了。

这样做的好处是，performance_schema中返回的行计数也会被擦除，因为greatest对null无效。但是，如果没有带有自动递增功能的表，这可能是个问题。

你可以用表格把一些东西组合在一起。我从来没有这样做过，但它看起来有一个列用于TABLE_ROWS和一个列用于TABLE NAME。

要获取每个表的行，你可以使用这样的查询:

SELECT table_name, table_rows
FROM INFORMATION_SCHEMA.TABLES
WHERE TABLE_SCHEMA = '**YOUR SCHEMA**';

基于上面@Nathan的回答，但不需要“删除最终的联合”，并带有对输出进行排序的选项，我使用以下SQL。它生成另一个SQL语句，然后运行:

select CONCAT( 'select * from (\n', group_concat( single_select SEPARATOR ' UNION\n'), '\n ) Q order by Q.exact_row_count desc') as sql_query
from (
    SELECT CONCAT(
        'SELECT "', 
        table_name, 
        '" AS table_name, COUNT(1) AS exact_row_count
        FROM `', 
        table_schema,
        '`.`',
        table_name, 
        '`'
    ) as single_select
    FROM INFORMATION_SCHEMA.TABLES 
    WHERE table_schema = 'YOUR_SCHEMA_NAME'
      and table_type = 'BASE TABLE'
) Q 

您确实需要一个足够大的group_concat_max_len服务器变量的值，但是从MariaDb 10.2.4开始，它应该默认为1M。

我不知道为什么这么难，但这就是生活。 下面是执行实际计数的bash脚本。只需将其保存为(例如count_rows.sh)，使其可执行(例如chmod 755 count_rows.sh)，并运行它(例如。/count_rows.sh)

#!/bin/bash

readarray -t TABLES < <(mysql --skip-column-names -u myuser -pmypassword mydbname -e "show tables")

# now we have an array like:
# TABLES='([0]="customer" [1]="order" [2]="product")'
# You can print out the array with:
#declare -p TABLES


for i in "${TABLES[@]}"
do
    #echo $i
    COUNT=$(mysql --skip-column-names -u username -pmypassword mydbname -e  "select count(*) from $i")
    echo $i : $COUNT
done