Postgresql ALTER TABLE ADD KEY equivalent - postgresql

What is the equivalent of this in postgresql ?
--
-- Index for the table `Artist`
--
ALTER TABLE `Artist`
ADD PRIMARY KEY (`idArtist`),
ADD UNIQUE KEY `name` (`name`,`firstName`);
ADD KEY `idFilm` (`idFilm`);


Adding a primary key is the same.
alter table artist
add primary key (idartist);
For the unique constraint you have two choices:
alter table artist
add constraint name unique (name, firstname);
or a unique index:
create unique index name on artist (name, firstname);
I think the key thing, simply adds a regular index:
create index idfilm on artist (idfilm);

Related

unique constraint to a foreign table

I have two tables, experiment and sample. Samples must be unique within experiments of the same type, but can be shared between experiments that have a different type. I understand how I can add a unique constraint to the samples to make them always unique, but is it possible to create a unique constraint based on the information in the foreign table?
create table experiment (
id integer primary key,
name text,
type text
);
create table sample (
id integer primary key,
name text,
experiment_id integer,
);
alter table sample add constraint exp_fkey foreign key (experiment_id) references experiment(id);

You can create a (redundant) unique constraint on experiment (id, type), add type to sample, create a foreign key constraint from (experiment_id, type) to (id, type) and a unique constraint on sample.type.

Alter the table by adding constraint and using index in postgresql getting an error syntax error at or near "("IDX_emp_PK

CREATE INDEX IDX_emp_PK ON
EMP(ID);
ALTER TABLE EMP ADD
CONSTRAINT PK_emp PRIMARY KEY (ID)
USING INDEX IDX_emp_PK;

There are two errors in your script:
First: you can't use a non-unique index for a primary key constraint, so you need
CREATE UNIQUE INDEX idx_emp_pk ON emp(id);
When you add a primary or unique constraint based on an index, you can't specify columns (as they are already defined in the index):
ALTER TABLE emp ADD
CONSTRAINT pk_emp PRIMARY KEY
USING INDEX idx_emp_pk;

Postgresql add existing column to composite primary key

I have a table in postgresql with a composite primary key. The primary key consists of two columns named:
DATETIME, UID
I have a another (non-null) column named ACTION already existing in this table. How do I add ACTION to the composite primary key? Ie: I'd like the resulting primary key of the table to be the triplet:
DATETIME, UID, ACTION

First drop the primary key constraint. You can get the name of the constraint by typing
\d my_table
and look under indexes for something like:
"my_table_pkey" PRIMARY KEY, btree (datetime, uid)
Drop it by doing:
alter table my_table drop constraint my_table_pkey;
Then create the new composite primary key by doing:
alter table my_table add constraint my_table_pkey primary key (datetime, uid, action);

Adding a NOT VALID foreign key in Postgres

Example schema: http://sqlfiddle.com/#!1/3d410
I've already got a table and I want to add a new, not valid foreign key to the table. What's the correct syntax for adding a NOT VALID foreign key?
CREATE TABLE junks (
id serial PRIMARY KEY,
name text
);
CREATE TABLE trunks (
id serial PRIMARY KEY,
name text
-- no fk
);
-- and the below does not work!
--ALTER TABLE trunks ADD junk serial REFERENCES junks(id) NOT VALID;

You first add the column:
alter table trunks add column junk serial;
Then add the constraint to the table:
alter table trunks add constraint the_constraint_name
FOREIGN KEY (junk)
REFERENCES junks (id)
not valid;

This works:
ALTER TABLE trunks ADD CONSTRAINT FK_junk_id
FOREIGN KEY (id)
REFERENCES junks(id)
NOT VALID
;
See, e.g.: http://www.postgresql.org/docs/devel/static/ddl-alter.html#AEN2758

T-SQL: foreign key that's not referencing a primary key

I have the following database:
CREATE TABLE ContentNodes
(
Id UNIQUEIDENTIFIER NOT NULL,
Revision INT IDENTITY(1,1) NOT NULL,
ParentId UNIQUEIDENTIFIER NULL
PRIMARY KEY (Id, Revision)
)
How do I limit ParentId to only contain values from the Id column. Trying to make ParentId a foreign key gives me:
PRINT 'FK_ContentNodes_ParentId_ContentNodes';
ALTER TABLE ContentNodes
ADD CONSTRAINT FK_ContentNodes_ParentId_ContentNodes FOREIGN KEY (ParentId) REFERENCES ContentNodes(Id);
GO
Error:
There are no primary or candidate keys in the referenced table
'ContentNodes' that match the
referencing column list in the foreign
key
'FK_ContentNodes_ParentId_ContentNodes'.

Since you have a compound primary key (Id, Revision) on your ContentNodes, you have to use both columns in a foreign key relation.
You cannot reference only parts of a primary key - simply cannot be done.
You have to either introduce a surrogate primary key into your table which is just a simple INT IDENTITY and then you can self-reference that single PK column, or you can (if it's possible in your data model) put a UNIQUE INDEX on that one column you want to reference:
CREATE UNIQUE NONCLUSTERED INDEX UIX_ID
ON ContentNodes(Id)
Once you have a UNIQUE INDEX on that column, then you can use it as a FK reference.