Question

我有一个select语句，它返回表中的重复项。我想现在能够删除这些重复项。如何使用ORACLE SQL执行此操作？

SELECT
  c1.name                     AS c1name,
  c2.name                     AS c2name,
  c1.cy_code                  AS c1country,
  c2.cy_code                  AS c2country,
  c1.latitude                 AS c1lat,
  c1.longitude                AS c1long,
  c2.latitude                 AS c2lat,
  c2.longitude                AS c2long,
  calc_distance(c1.latitude,
                c1.longitude,
                c2.latitude,
                c2.longitude) AS distance
FROM city_temp c1,
  city_temp c2
WHERE c2.name = c1.name
  AND c2.cy_code = c1.cy_code
  AND calc_distance(c1.latitude, c1.longitude, c2.latitude, c2.longitude) = 0
  AND c1.latitude = c2.latitude
  AND c1.longitude = c2.longitude;

Answer 1

如果你可以改变功能calc_distance

，你可以尝试这个

DELETE FROM city_temp
WHERE rowid not in
(SELECT MIN(rowid)
FROM city_temp
WHERE calc_distance(latitude, longitude)=0
GROUP BY name, cy_code, latitude,longitude
)

Answer 2

您将使用exists子句删除所有与较高rowid相等的行：

delete from city_temp c1
where exists
(
  select *
  from city_temp c2
  where c2.name = c1.name 
    and c2.cy_code = c1.cy_code 
    and c2.latitude = c1.latitude 
    and c2.longitude = c1.longitude
    and calc_distance(c1.latitude, c1.longitude, c2.latitude, c2.longitude) = 0 
    and c2.rowid > c1.rowid
);

Answer 3

使用rowid，您可以执行此操作

DELETE FROM city_temp where rowid in (
            SELECT c1.rowid
            FROM city_temp c1, city_temp c2
            WHERE c2.name = c1.name  and c2.cy_code = c1.cy_code and calc_distance(c1.latitude, c1.longitude, c2.latitude, c2.longitude)=0 
                        and c1.latitude = c2.latitude and c1.longitude = c2.longitude)

Answer 4

删除重复行的经典方法是

DELETE FROM table_name A WHERE ROWID > (
SELECT min(rowid) FROM table_name B
WHERE A.key_values = B.key_values)

只需将SELECT语句插入其中即可。

或者对于大型数据集的最高速度 - 您可以将不同的数据复制到临时表中，截断原始数据，复制数据并删除临时表。

如何在ORACLE中将其转换为DELETE语句？

4 个答案: