POSTGRESQL：插入...选择自动生成的列

Question

我有两个表A(id，col1，col2)和B(col3，col4，col5，col6)

INSERT INTO A(col1, col2)
(SELECT col3, col4 from B)

ERROR:  null value in column "id" violates not-null constraint
DETAIL: Failing row contains (null, "abc", "def")

INSERT INTO A(id, col1, col2)
(SELECT DEFAULT, col3, col4 from B)

这引发错误

ERROR:  syntax error at or near "DEFAULT"

为什么postgres不自动为列“id”生成值？如果我从表B中提供“id”，或者如果我插入单行（没有select）并为自动生成的列提供“default”关键字，查询就可以工作。

编辑：表创建查询

CREATE TABLE A 
(
id bigint NOT NULL, 
col1 character varying(10), 
col2 character varying(10), 
CONSTRAINT A_pk PRIMARY KEY (id) 
); 
ALTER TABLE A OWNER TO user1; 

CREATE SEQUENCE A_id_seq 
START WITH 1 
INCREMENT BY 1 
NO MINVALUE 
NO MAXVALUE 
CACHE 1; 

ALTER TABLE A_id_seq OWNER TO user1; 

ALTER SEQUENCE A_id_seq OWNED BY A.id; 

Answer 1

INSERT INTO A(col1, col2)
(SELECT col3, col4 from B)

失败，因为没有为ID设置默认值。默认值应该是从序列中取nextval。

CREATE SEQUENCE A_id_seq 
  START WITH 1 
  INCREMENT BY 1 
  NO MINVALUE 
  NO MAXVALUE 
  CACHE 1; 

CREATE TABLE A 
(
  id bigint NOT NULL DEFAULT nextval('A_id_seq'::regclass), 
  col1 character varying(10), 
  col2 character varying(10), 
  CONSTRAINT A_pk PRIMARY KEY (id) 
); 

或者，您可以在创建表中使用bigserial而不是bigint，以便为您创建序列并自动设置默认设置。最终结果是一样的。