如何在MySQL中的单个列中删除重复的逗号分隔值

Question

SELECT id, country FROM my_records

我从MySQL查询得到了上述结果，我想从结果中删除重复的ID。不是在PHP代码的帮助下，而是使用MySQL查询。是否有任何功能或查询来做同样的事情。

谢谢

Answer 1

我陷入了类似的情况，发现MySql没有提供任何预定义的功能来克服这个问题。

为了克服我创建了一个UDF，请看下面的定义和用法。

DROP FUNCTION IF EXISTS `get_unique_items`;
DELIMITER //
CREATE FUNCTION `get_unique_items`(str varchar(1000)) RETURNS varchar(1000) CHARSET utf8
BEGIN

        SET @String      = str;
        SET @Occurrences = LENGTH(@String) - LENGTH(REPLACE(@String, ',', ''));
        SET @ret='';
        myloop: WHILE (@Occurrences > 0)
        DO 
            SET @myValue = SUBSTRING_INDEX(@String, ',', 1);
            IF (TRIM(@myValue) != '') THEN
                IF((LENGTH(@ret) - LENGTH(REPLACE(@ret, @myValue, '')))=0) THEN
                        SELECT CONCAT(@ret,@myValue,',') INTO @ret;
                END if;
            END IF;
            SET @Occurrences = LENGTH(@String) - LENGTH(REPLACE(@String, ',', ''));
            IF (@occurrences = 0) THEN 
                LEAVE myloop; 
            END IF;
            SET @String = SUBSTRING(@String,LENGTH(SUBSTRING_INDEX(@String, ',', 1))+2);
        END WHILE;    
SET @ret=concat(substring(@ret,1,length(@ret)-1), '');
return @ret;

END //
DELIMITER ;

样品用法：

SELECT get_unique_items('2,2,2,22,2,3,3,3,34,34,,54,5,45,,65,6,5,,67,6,,34,34,2,3,23,2,32,,3,2,,323') AS 'Items';

结果：

2,22,3,34,54,45,65,67,23,32,323

希望这有帮助！

Answer 2

这可能对你有所帮助。

DELIMITER //

DROP FUNCTION IF EXISTS `find_duplicate_using_comma` //
CREATE FUNCTION `find_duplicate_using_comma` (in_str LONGTEXT) RETURNS LONGTEXT
DETERMINISTIC
NO SQL
BEGIN


DECLARE out_str LONGTEXT DEFAULT NULL; -- pending output
DECLARE next_str TEXT DEFAULT NULL;    -- next element under consideration

dedup:
LOOP

  IF CHAR_LENGTH(TRIM(in_str)) = 0 OR in_str IS NULL THEN
    LEAVE dedup; -- no more data to consider
  END IF;

  SET next_str = SUBSTRING_INDEX(in_str,',',1);                   -- find the next element
  SET in_str = SUBSTRING(in_str FROM (CHAR_LENGTH(next_str) + 1 + 1)); -- remove that element

  SET in_str = TRIM(in_str), next_str = TRIM(next_str); -- trim the new and the rest

  IF FIND_IN_SET(next_str,out_str) OR CHAR_LENGTH(next_str) = 0 THEN -- if empty or already found
    ITERATE dedup;
  END IF;

  SET out_str = CONCAT_WS(',',out_str,next_str); -- append the new to pending output 

END LOOP;

RETURN out_str;

END //

DELIMITER ;

示例：

SELECT find_duplicate_using_comma('6675,8661,8661,8661,8661,8661,8661,8661,8661,8661,8661,8661,8661,8661,8661,8661,8661,8661') AS data;

Result : 6675,8661

Answer 3

给出的其他解决方案不区分大小写，如果您希望结果区分大小写并且不删除重音匹配（如'o'和'ö'），并查找完全匹配和严格匹配，则此处为代码

如果utf8mb4_bin不起作用，请尝试utf8_bin或其他二进制类型。

DELIMITER //

DROP FUNCTION IF EXISTS `find_duplicate_using_comma` //
CREATE FUNCTION `find_duplicate_using_comma` (in_str LONGTEXT) RETURNS LONGTEXT
DETERMINISTIC
NO SQL
BEGIN


DECLARE out_str LONGTEXT DEFAULT NULL; -- pending output
DECLARE next_str TEXT DEFAULT NULL;    -- next element under consideration

dedup:
LOOP

  IF CHAR_LENGTH(TRIM(in_str)) = 0 OR in_str IS NULL THEN
    LEAVE dedup; -- no more data to consider
  END IF;

  SET next_str = SUBSTRING_INDEX(in_str,',',1);                   -- find the next element
  SET in_str = SUBSTRING(in_str FROM (CHAR_LENGTH(next_str) + 1 + 1)); -- remove that element

  SET in_str = TRIM(in_str), next_str = TRIM(next_str); -- trim the new and the rest

  IF FIND_IN_SET(next_str collate utf8mb4_bin,out_str collate utf8mb4_bin) OR CHAR_LENGTH(next_str) = 0 THEN -- if empty or already found
    ITERATE dedup;
  END IF;

  SET out_str = CONCAT_WS(',',out_str,next_str); -- append the new to pending output 

END LOOP;

RETURN out_str;

END //

DELIMITER ;