Question

我有桌城

CREATE TABLE city
(
  id serial NOT NULL,
  name character varying(255) NOT NULL,
  country character varying(30) DEFAULT NULL::character varying,
  CONSTRAINT city_pkey PRIMARY KEY (id)
)
WITH (
  OIDS=FALSE
);
ALTER TABLE city
  OWNER TO uks;

CREATE UNIQUE INDEX uniq_2d5b02345373c966
  ON city
  USING btree
  (country COLLATE pg_catalog."default");

CREATE UNIQUE INDEX uniq_2d5b02345e237e06
  ON city
  USING btree
  (name COLLATE pg_catalog."default");

现在，当我插入表格时，一切正常，序列也增加了。当我尝试添加具有现有名称的行时因为uniqe键名称而出现错误，这是我所期望的，但在这种情况下序列也增加了：/

uks=> SELECT * FROM city;
 id | name | country 
----+------+---------
(0 rows)

uks=> INSERT INTO city (name) VALUES('ONE');
INSERT 0 1
uks=> INSERT INTO city (name) VALUES('TWO');
INSERT 0 1
uks=> SELECT * FROM city;
 id | name | country 
----+------+---------
  1 | ONE  | 
  2 | TWO  | 
(2 rows)

uks=> INSERT INTO city (name) VALUES('TWO');
ERROR:  duplicate key value violates unique constraint "uniq_2d5b02345e237e06"
DETAIL:  Key (name)=(TWO) already exists.
uks=> INSERT INTO city (name) VALUES('TWO');
ERROR:  duplicate key value violates unique constraint "uniq_2d5b02345e237e06"
DETAIL:  Key (name)=(TWO) already exists.
uks=> INSERT INTO city (name) VALUES('TWO');
ERROR:  duplicate key value violates unique constraint "uniq_2d5b02345e237e06"
DETAIL:  Key (name)=(TWO) already exists.
uks=> INSERT INTO city (name) VALUES('THREE');
INSERT 0 1
uks=> SELECT * FROM city;
 id | name  | country 
----+-------+---------
  1 | ONE   | 
  2 | TWO   | 
  6 | THREE | 
(3 rows)

这是postgres中的正常行为吗？我认为，当错误发生时，序列不应该增加。

即使由于唯一引起的错误，PostgreSQL也会增加序列 - 这是正常的行为吗？

0 个答案: