演示示例。

必须导入的源 CSV 内容：

review_id,user,text 123,John,This is multiline 1, which contains a comma. 456,Jim,This is miltiline 2, which contains commas, 'quote' chars and "double quote" chars.

此数据必须导入到的表：

CREATE TABLE test (review_id INT, user VARCHAR(255), review_text TEXT);

将数据加载到表中的查询：

LOAD DATA INFILE 'C:/ProgramData/MySQL/MySQL Server 8.0/Uploads/test.csv' INTO TABLE test FIELDS TERMINATED BY '\0' ENCLOSED BY '' ESCAPED BY '' LINES TERMINATED BY '\r\n' IGNORE 1 LINES (@line) SET review_id = (@review_id := CASE WHEN @line REGEXP '^\\d' THEN SUBSTRING_INDEX(@line, ',', 1) ELSE @review_id END ), user = (@user := CASE WHEN @line REGEXP '^\\d' THEN SUBSTRING_INDEX(SUBSTRING_INDEX(@line, ',', 2), ',', -1) ELSE @user END ), review_text = (@review_text := CASE WHEN @line REGEXP '^\\d' THEN SUBSTRING(@line FROM 2 + LENGTH(SUBSTRING_INDEX(@line, ',', 2))) ELSE CONCAT_WS(' ', @review_text, @line) END );

加载后表格数据状态：

DELETE t1 FROM test t1 JOIN test t2 USING (user) WHERE t1.review_text < t2.review_text;