如何对包含数字和字母的字符串进行排序？

Question

我有 qt_no 值，例如

AM1,M3,M4,M14,M30,M40,MA01,A10,A13,A07,B01,B10,Z33,Z13

等（实际上是字母后的任何 int 2-3 位数字）。

我尝试过排序为

order by length(qt_no), qt_no

它没有达到我要求的输出。

我的预期输出是

A01,A07,A10,A13,B01,AM1,M3,M4,M14,M30,M40,MA01,Z13,Z33

记住这些 qt_no 值属于同一个字段和同一个表的不同行。

从这里开始我不知道该怎么做。

任何帮助将不胜感激。

编辑

这是play with 的示例数据库。

Answer 1

最好的方案是创建两列额外的列，一列用于字母部分，一列用于数字部分；那么它就像ORDER BY alpha_part ASC, num_part ASC 一样简单。如果你在这两列上有一个联合索引，它也会非常快。

如果您绝对必须在查询时解析列，这需要时间 - 并且还会使索引无用，这会使一切变得慢得多。但你可以这样做：

...
ORDER BY
  REGEXP_REPLACE(qt_no, '\d+', '') ASC, 
  CAST(REGEXP_REPLACE(qt_no, '\D+', '') AS INTEGER) ASC

编辑：我很抱歉，但我不知道如何在 5.7 上做到这一点，除了这样：

SELECT qt_no FROM t
ORDER BY
REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(qt_no, '0', ''), '1', ''), '2', ''), '3', ''), '4', ''), '5', ''), '6', ''), '7', ''), '8', ''), '9', '') ASC, 
CAST(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(REPLACE(qt_no, 'A', ''), 'B', ''), 'C', ''), 'D', ''), 'E', ''), 'F', ''), 'G', ''), 'H', ''), 'I', ''), 'J', ''), 'K', ''), 'L', ''), 'M', ''), 'N', ''), 'O', ''), 'P', ''), 'Q', ''), 'R', ''), 'S', ''), 'T', ''), 'U', ''), 'V', ''), 'W', ''), 'X', ''), 'Y', ''), 'Z', '') AS UNSIGNED) ASC;

Answer 2

由于 MySQL 版本

以下是此answer 的修改函数，它从输入字符串返回整数值。这里所做的修改是它返回字符串而不是 Int。因为你有像07 这样的数字字符串，需要按原样返回，而不是7。

DELIMITER $$

CREATE FUNCTION `ExtractNumber`(in_string VARCHAR(50)) 
RETURNS VARCHAR(50)
NO SQL
BEGIN
    DECLARE ctrNumber VARCHAR(50);
    DECLARE finNumber VARCHAR(50) DEFAULT '';
    DECLARE sChar VARCHAR(1);
    DECLARE inti INTEGER DEFAULT 1;

    IF LENGTH(in_string) > 0 THEN
        WHILE(inti <= LENGTH(in_string)) DO
            SET sChar = SUBSTRING(in_string, inti, 1);
            SET ctrNumber = FIND_IN_SET(sChar, '0,1,2,3,4,5,6,7,8,9'); 
            IF ctrNumber > 0 THEN
                SET finNumber = CONCAT(finNumber, sChar);
            END IF;
            SET inti = inti + 1;
        END WHILE;
        RETURN finNumber;
    ELSE
        RETURN '';
    END IF;    
END$$

DELIMITER ;

现在，您可以使用这个自定义函数，按字母部分排序，然后是数字部分（转换为unsigned）。

SELECT id, 
       name,
       REPLACE(name, ExtractNumber(name), '') as strpart, 
       CAST(ExtractNumber(name) AS UNSIGNED) as numpart 
FROM test
ORDER BY strpart, 
         numpart

DB Fiddle DEMO

Answer 3

不漂亮，因为我们正在解析字符串，所以它会很慢。 这假设您的格式是 Alpha，然后数字值从不混合。它找到第一个数值，然后根据该结果分成两列。

DEMO:

我只是想按 qt_no+0 排序以获得自然排序，但这不起作用。 所以我走了一条不同的路……

Field1 是你的 qt_no 字段...

SELECT Field1,

#Use this to just get the number values but since we lose trailing zeros...Step 2 we reverse the value so numbers are first allowing the convert to drop the letters.  unfortunately this also drops the trailing (leading since we reversed) zeros.
       @NumStep1 := reverse(CONVERT(reverse(Field1), SIGNED)) NumStep1,

#We got the postiion of the first number... so get the  whole number now.
       @NumStep2 :=substring(Field1,locate(@numStep1,Field1),length(Field1)) NumStep2,
       @Alpha:= substring(Field1,1,Locate(@numStep2,Field1)-1) Alpha

FROM (
SELECT 'AM1' as Field1 UNION ALL 
SELECT 'M3' as Field1 UNION ALL
SELECT 'M4' as Field1 UNION ALL
SELECT 'M14' as Field1 UNION ALL
SELECT 'M30' as Field1 UNION ALL
SELECT 'M40' as Field1 UNION ALL
SELECT 'MA01' as Field1 UNION ALL
SELECT 'A10' as Field1 UNION ALL
SELECT 'A13' as Field1 UNION ALL
SELECT 'A07' as Field1 UNION ALL
SELECT 'B01' as Field1 UNION ALL
SELECT 'B10' as Field1 UNION ALL
SELECT 'Z33' as Field1 UNION ALL
SELECT 'Z13' as Field1) Z
ORDER BY Alpha, NumStep2*1

给我们：

+----+--------+----------+----------+-------+
|    | Field1 | NumStep1 | NumStep2 | Alpha |
+----+--------+----------+----------+-------+
|  1 | A10    |        1 |       10 | A     |
|  2 | A13    |       13 |       13 | A     |
|  3 | A07    |       07 |       07 | A     |
|  4 | AM1    |        1 |        1 | AM    |
|  5 | B10    |        1 |       10 | B     |
|  6 | B01    |       01 |       01 | B     |
|  7 | M3     |        3 |        3 | M     |
|  8 | M4     |        4 |        4 | M     |
|  9 | M14    |       14 |       14 | M     |
| 10 | M30    |        3 |       30 | M     |
| 11 | M40    |        4 |       40 | M     |
| 12 | MA01   |       01 |       01 | MA    |
| 13 | Z33    |       33 |       33 | Z     |
| 14 | Z13    |       13 |       13 | Z     |
+----+--------+----------+----------+-------+

没有用户变量，但将数据拆分为字母和数字。

SELECT Field1,
       substring(Field1,locate(reverse(CONVERT(reverse(Field1), SIGNED)),Field1),length(Field1)) NumStep2,
       substring(Field1,1,Locate(substring(Field1,locate(reverse(CONVERT(reverse(Field1), SIGNED)),Field1),length(Field1)),Field1)-1) Alpha

FROM (
SELECT 'AM1' as Field1 UNION ALL 
SELECT 'M3' as Field1 UNION ALL
SELECT 'M4' as Field1 UNION ALL
SELECT 'M14' as Field1 UNION ALL
SELECT 'M30' as Field1 UNION ALL
SELECT 'M40' as Field1 UNION ALL
SELECT 'MA01' as Field1 UNION ALL
SELECT 'A10' as Field1 UNION ALL
SELECT 'A13' as Field1 UNION ALL
SELECT 'A07' as Field1 UNION ALL
SELECT 'B01' as Field1 UNION ALL
SELECT 'B10' as Field1 UNION ALL
SELECT 'Z33' as Field1 UNION ALL
SELECT 'Z13' as Field1) Z
ORDER BY Alpha, NumStep2*1

Answer 4

您可以将顺序分为三个阶段：首先是字母，然后是字符串的长度，最后是字母顺序：

select * from test order by substring( qt_no, 1, 1 ), length(qt_no), qt_no;