Question

当我使用HSQLDB 2.2.5运行以下脚本（q.sql）时，results.csv的内容将变为以下内容：

results.csv ：

C1,C2
[null],e
a,b

q.sql：

DROP TABLE "PUBLIC"."TABLE1" IF EXISTS;

CREATE TABLE "PUBLIC"."TABLE1" (
  "C1" VARCHAR(10),
  "C2" VARCHAR(10),
  "C3" VARCHAR(10)
);

INSERT INTO "PUBLIC"."TABLE1" ("C1", "C2", "C3") VALUES ('a', 'b', 'c');
INSERT INTO "PUBLIC"."TABLE1" ("C1", "C2", "C3") VALUES ('d', 'e', 'f');

DROP TABLE "PUBLIC"."TABLE2" IF EXISTS;

CREATE TABLE "PUBLIC"."TABLE2" (
  "C1" VARCHAR(10),
  "C2" VARCHAR(10),
);

INSERT INTO "PUBLIC"."TABLE2" ("C1", "C2") VALUES ('a', 'b');
INSERT INTO "PUBLIC"."TABLE2" ("C1", "C2") VALUES ('u', 'v');

DROP TABLE "PUBLIC"."TABLE3" IF EXISTS;

CREATE TABLE "PUBLIC"."TABLE3" (
  "C1" VARCHAR(10),
  "C2" VARCHAR(10)
);

INSERT INTO "PUBLIC"."TABLE3" (
  "C1",
  "C2"
)
SELECT
(SELECT
"T2"."C1" AS "T2C1"
FROM "PUBLIC"."TABLE2" "T2"
WHERE "T2"."C1" = "T1"."C1") AS "T2C1",
"T1"."C2" AS "T1C2"
FROM "PUBLIC"."TABLE1" "T1"
ORDER BY "T2C1", "T1C2";

* *DSV_COL_DELIM=,
* *DSV_ROW_DELIM=\n
* *DSV_TARGET_FILE=results.csv

\x "PUBLIC"."TABLE3"

如何更改我的脚本，使“[null]”替换为“？”在生成的CSV文件中？

Answer 1

取决于dbengine：

for hSQL ... 3个作品中的任何一个根据：hSQLDB.org

合并（MOST）

NVL（oracle）

IFNULL（MySQL的）编辑添加（...（哎呀我忘了！）

SELECT
coalesce((SELECT
"T2"."C1" AS "T2C1"
FROM "PUBLIC"."TABLE2" "T2"
WHERE "T2"."C1" = "T1"."C1"),'?') AS "T2C1", 
"T1"."C2" AS "T1C2"
FROM "PUBLIC"."TABLE1" "T1"
ORDER BY "T2C1", "T1C2";

Answer 2

不是从表中插入值，而是在插入上使用case语法来检查值是否为空。

这样的事情：

case when value = [null] then ? else variable end

在HSQLDB中的JOIN操作期间用其他值替换NULL

2 个答案: