Maybe I'm burnout, but I don't understand this. I have two tables in Oracle: TBL_a and TBL_x. I'm trying to create a foreign key between thos two tables as follows and get error
ORA-02270: no matching unique or primary key.
CREATE TABLE tbl_a (
cod_op integer,
cod_dni char(8),
cod_correl integer,
varchar2(50)
);
CREATE TABLE tbl_x (
cod_op integer,
cod_dni char(8),
blabla varchar2(50)
);
CREATE UNIQUE INDEX TBL_A_PK ON TBL_A (COD_OP);
CREATE UNIQUE INDEX TBL_x_PK ON TBL_x (COD_OP);
ALTER TABLE TBL_a ADD CONSTRAINT TBL_a_R01
FOREIGN KEY (COD_OP) REFERENCES TBL_x (COD_OP);
The problem is that you've created unique INDEXES on your tables, but you didn't create a unique or primary key CONSTRAINT. Oracle requires that the constraints exist in order to establish a foreign key relationship.
If you drop your existing indexes and add the appropriate constraints you can establish your foreign key relationship:
DROP INDEX TBL_A_PK;
DROP INDEX TBL_x_PK;
ALTER TABLE TBL_A
ADD CONSTRAINT UQ_A
UNIQUE(COD_OP)
USING INDEX;
ALTER TABLE TBL_X
ADD CONSTRAINT UQ_X
UNIQUE(COD_OP)
USING INDEX;
dbfiddle here
The table that is referred by the foreign key (here, tbl_x) must have a primary key or a unique constraint.
In your use case, as you are declaring a unique index on cod_op, you could simply make cod_op the primary key of tbl_x instead: that would make the error disappear.
Demo on DB Fiddle
In general, it is a good practice to have a primary key on any table. Extending the principe of turning your unique indexes to primary keys, your DDL statements could be simplified as follows:
CREATE TABLE tbl_x (
cod_op INTEGER PRIMARY KEY,
cod_dni CHAR(8),
blabla VARCHAR2(50)
);
CREATE TABLE tbl_a (
cod_op INTEGER PRIMARY KEY,
cod_dni CHAR(8),
cod_correl INTEGER,
blabla VARCHAR2(50),
CONSTRAINT TBL_a_R01 FOREIGN KEY (COD_OP) REFERENCES TBL_x (COD_OP)
);
Demo on DB Fiddle
Related
I have a DB2 (for IBMi) table created as below. I would like to drop the forign key constraint while running in an SQLRPGLE program. Is this possible?
create table grid_action_details(id integer not null
generated always as identity
(start with 1 increment by 1)
PRIMARY KEY,grid_details_id integer,foreign key(grid_details_id)
references grid_details(id),action_code_details_id integer,
foreign key(action_code_details_id)
references action_code_details(id),action_code_status varchar(2),
created_date date default
current_date,created_by varchar(30),
last_updated_date date default current_date,updated_by
varchar(30),required_parameter clob);
I tried the below syntax but it just doesn't seem to work for me:
ALTER TABLE table-name
DROP FOREIGN KEY foreign_key_name
alter table iesqafile.grid_action_details
drop foreign key action_code_details_id
ACTION_CODE_DETAILS_ID in IESQAFILE type *N not found.
You drop the FK constraint via name, not via column.
Since you didn't specify one during create, you'll need to look to see what name the system generated.
Always a best practice to name things yourself.
CONSTAINT <name> FOREIGN KEY (<columns>)
create table grid_action_details (
id integer not null generated always as identity (
start with 1 increment by 1
)
,constraint grid_action_details_pk
primary key
,grid_details_id integer
,constraint grid_details_fk
foreign key (grid_details_id)
references grid_details (id)
,action_code_details_id integer
,constraint action_code_details_fk
foreign key (action_code_details_id)
references action_code_details (id)
,action_code_status varchar(2)
,created_date date default current_date
,created_by varchar(30)
,last_updated_date date default current_date
,updated_by varchar(30)
,required_parameter clob
);
Now you can drop by the known name
alter table iesqafile.grid_action_details
drop foreign key action_code_details_fk
EDIT
To find the generated name use:
the ACS Schema component
DSPFD
SQL against one of the catalog views (QSYS2.SYSCST, SYSIBM.SQLFOREIGNKEYS, SYSIBM.REFERENTIAL_CONSTRAINTS )
I am new to here, please help me to find an answer to the below query.
"How to add a primary key constraint on two columns to make a composite key column in a table?"
Instead of writing so many lines, kindly provide the answer as short as possible with code.
(I am creating a table in which I can have only one primary key on the values of two-columns as a composite key.)
JacknJill
You can create such constraint in scope of CREATE TABLE operator
CREATE TABLE foo (
id INTEGER,
name VARCHAR(25),
bar VARCHAR(255),
CONSTRAINT foo_pk PRIMARY KEY (id,name)
);
or ALTER TABLE operator
CREATE TABLE foo (
id INTEGER,
name VARCHAR(25),
bar VARCHAR(255)
);
ALTER TABLE foo ADD CONSTRAINT foo_pk PRIMARY KEY (id,name);
How can I inherit from two(or more) parent classes in ORACLE SQL? I've tried something like this:
ID(13),
OTHER_ID(9),
CONSTRAINT FK_ID FOREIGN
KEY(ID) REFERENCES TABLE_ONE(ID),
CONSTRAINT FK_OTHER_GROUP FOREIGN
KEY(OTHER_ID) REFERENCES TABLE_TWO(OTHER_ID)
I've read the documentation and the code I've found is this one:
INDEX (product_category, product_id),
INDEX (customer_id),
FOREIGN KEY (product_category, product_id)
REFERENCES product(category, id)
ON UPDATE CASCADE ON DELETE RESTRICT,
FOREIGN KEY (customer_id)
REFERENCES customer(id)
It doesn't quite work for me. Is there any special way to define these indexes? I haven't met them before. The first example had a problem that data types weren't specified. But I'd like someone to explain me how could I do it the second(oracle documentation) way.
This is how I understood the problem:
there are two tables, each having primary key constraint
in my example, those are t_emp and t_dept
there's also the third table whose columns are supposed to reference primary keys from previous two tables
those are fk_th_emp and fk_th_dep foreign key constraints
If that's so, here's how to do that:
SQL> create table t_emp
2 (empno number constraint pk_temp primary key,
3 ename varchar2(20)
4 );
Table created.
SQL> create table t_dept
2 (deptno number constraint pk_dept primary key,
3 dname varchar2(20)
4 );
Table created.
SQL> create table third
2 (id number constraint pk_third primary key,
3 other_id number,
4 --
5 constraint fk_th_emp foreign key (id) references t_emp (empno),
6 constraint fk_th_dep foreign key (other_id) references t_dept (deptno)
7 );
Table created.
SQL>
I use oracle and I try to have a recursive relation
CREATE TABLE "EVENT"
(
"EVENT_ID" NUMBER(18) NOT NULL, //primary key
"NAME" VARCHAR(20) NULL,
"RELATED_EVENT_ID" NUMBER(18) NULL //foreign key
);
Event 1 parent is Event 2....
When I try to create this table, I get this error.
ALTER TABLE "EVENT"
ADD CONSTRAINT "FK_RELATED_EVENT_ID"
FOREIGN KEY ("RELATED_EVENT_ID") REFERENCES "EVENT" ("RELATED_EVENT_ID")
Error report -
SQL Error: ORA-02270: no matching unique or primary key for this column-list
02270. 00000 - "no matching unique or primary key for this column-list"
*Cause: A REFERENCES clause in a CREATE/ALTER TABLE statement
gives a column-list for which there is no matching unique or primary
key constraint in the referenced table.
*Action: Find the correct column names using the ALL_CONS_COLUMNS
catalog view
You have two problems:
There is no primary key constraint on this table.
The foreign key constraint you defined has RELATED_EVENT_ID referencing RELATED_EVENT_ID. I suspect that was just a typo.
Change your table definition to:
CREATE TABLE EVENT
(EVENT_ID NUMBER
NOT NULL
CONSTRAINT PK_EVENT
PRIMARY KEY
USING INDEX,
NAME VARCHAR2(20),
RELATED_EVENT_ID NUMBER);
Then add the foreign key constraint as
ALTER TABLE EVENT
ADD CONSTRAINT EVENT_FK1
FOREIGN KEY (RELATED_EVENT_ID) REFERENCES EVENT(EVENT_ID);
db<>fiddle here
EDIT
Note that the better way to handle this is to use a junction table, such as:
CREATE TABLE EVENT_EVENT
(EVENT_ID1 NUMBER
CONSTRAINT EVENT_EVENT_FK1
REFERENCES EVENT(EVENT_ID),
EVENT_ID2 NUMBER
CONSTRAINT EVENT_EVENT_FK2
REFERENCES EVENT(EVENT_ID),
CONSTRAINT PK_EVENT_EVENT
PRIMARY KEY (EVENT_ID1, EVENT_ID2)
USING INDEX);
Then you can drop the RELATED_EVENT_ID column from EVENT as you no longer need it.
According to oracle document :
Foreign key specifies that the values in the column must correspond to values in a
referenced primary key or unique key column or that they are NULL.
In your case, create primary key on column (EVENT_ID) and use it in reference clause as following:
ALTER TABLE "EVENT"
ADD CONSTRAINT "FK_RELATED_EVENT_ID"
FOREIGN KEY ("RELATED_EVENT_ID")
REFERENCES "EVENT" ("EVENT_ID") -- this
Now, use EVENT2's EVENT_ID as RELATED_EVENT_ID in EVENT1 record to make EVENT2 as parent of EVENT1.
Cheers!!
I have one simple table:
-- Create table
create table FAVOURITE_RULES
( rule_id NUMBER(9) not null,
user_id NUMBER(9) not null);
-- Create/Recreate primary, unique and foreign key constraints
alter table FAVOURITE_RULES
add constraint FAV_RULES_PK primary key (RULE_ID, USER_ID)
alter table FAVOURITE_RULES
add constraint FAV_RULES_RULE_ID_FK foreign key (RULE_ID)
references RULES (RULE_ID) on delete cascade;
alter table FAVOURITE_RULES
add constraint FAV_RULES_USER_ID foreign key (USER_ID)
references USER_AUTHENTICATION (USER_ID) on delete cascade;
I have a rule (from .Net code) that doesn't exist in the original table RULES. It has the Id=-999.
When I try to insert into the FAVOURITE_RULES I get an error about integrity constraint violation (as expected) (FAV_RULES_RULE_ID_FK) violated - parent key not found.
Can I keep the foreign key (FAV_RULES_RULE_ID_FK ) and allow extra only this value (-999) to be inserted?
May be this can help.
Step 1: drop fk constraint
Step 2: insert your violating row
Step 3: again create fk constraint with ENABLE NOVALIDATE