I use Hive version 3.0.0.3.1.
Create table with statement
CREATE TABLE IF NOT EXISTS ...
(
pk_... BIGINT DEFAULT SURROGATE_KEY(),
...
PRIMARY KEY (pk_...) DISABLE NOVALIDATE
)
After working for a while, today it has been duplicate key
Related
We have a table that is partitioned on a bigint field.
Here is an example definition :
CREATE TABLE IF NOT EXISTS public.demo_part
(
n bigint NOT NULL,
CONSTRAINT demo_part_pkey PRIMARY KEY (n)
) PARTITION BY RANGE (n);
If we attach a default partition with :
CREATE TABLE public.demo_part_default PARTITION OF public.demo_part
Later partitions cannot be attached without acquiring an AccessExclusiveLock because the default partition must be scanned to ensure no conflicting data is in the default partition.
I tried adding CHECK constraints to both the partition to be added and the default partition without success.
CREATE TABLE IF NOT EXISTS demo_part_default PARTITION OF demo_part DEFAULT;
ALTER TABLE demo_part_default DROP CONSTRAINT IF EXISTS def_check;
ALTER TABLE demo_part_default ADD CONSTRAINT def_check CHECK (n >= 1000 AND n < 500000);
CREATE TABLE IF NOT EXISTS demo_part_1_10 (LIKE demo_part);
ALTER TABLE demo_part_1_10 DROP CONSTRAINT IF EXISTS n_1_10;
ALTER TABLE demo_part_1_10 ADD CONSTRAINT n_1_10 CHECK (n >= 1 AND n < 10);
ALTER TABLE demo_part ATTACH PARTITION demo_part_1_10 FOR VALUES FROM (1) TO (10);
The later ATTACH always creates an AccessExclusiveLock even if the constraints are mutually exclusive.
Is it possible to attach a partition without AccessExclusiveLock when a default partition is attached ?
I am trying to remove a UNIQUE constraint on a column for sqlite but I do not have the name to remove the constraint. How can I find the name of the UNIQUE constraint name to remove it.
Below is the schema I see for the table I want to remove the constraint
UNIQUE (datasource_name)
sqlite> .schema datasources
CREATE TABLE "datasources" (
created_on DATETIME NOT NULL,
changed_on DATETIME NOT NULL,
id INTEGER NOT NULL,
datasource_name VARCHAR(255),
is_featured BOOLEAN,
is_hidden BOOLEAN,
description TEXT,
default_endpoint TEXT,
user_id INTEGER,
cluster_name VARCHAR(250),
created_by_fk INTEGER,
changed_by_fk INTEGER,
"offset" INTEGER,
cache_timeout INTEGER, perm VARCHAR(1000), filter_select_enabled BOOLEAN, params VARCHAR(1000),
PRIMARY KEY (id),
CHECK (is_featured IN (0, 1)),
CHECK (is_hidden IN (0, 1)),
FOREIGN KEY(created_by_fk) REFERENCES ab_user (id),
FOREIGN KEY(changed_by_fk) REFERENCES ab_user (id),
FOREIGN KEY(cluster_name) REFERENCES clusters (cluster_name),
UNIQUE (datasource_name),
FOREIGN KEY(user_id) REFERENCES ab_user (id)
);
SQLite only supports limited ALTER TABLE, so you can't remove the constaint using ALTER TABLE. What you can do to "drop" the column is to rename the table, create a new table with the same schema except for the UNIQUE constraint, and then insert all data into the new table. This procedure is documented in the Making Other Kinds Of Table Schema Changes section of ALTER TABLE documentation.
I just ran into this myself. An easy solution was using DB Browser for SQLite
It let me remove a unique constraint with just a checkbox in a gui.
PRAGMA foreign_keys=off;
BEGIN TRANSACTION;
ALTER TABLE table_name RENAME TO old_table;
CREATE TABLE table_name
(
column1 datatype [ NULL | NOT NULL ],
column2 datatype [ NULL | NOT NULL ],
...
);
INSERT INTO table_name SELECT * FROM old_table;
COMMIT;
PRAGMA foreign_keys=on;
Source: https://www.techonthenet.com/sqlite/unique.php
I was just working through this issue on a small database and found it easier to dump the data as SQL statements, it prints out your tables exactly as they are and also adds the INSERT INTO statements to rebuild the DB.
The .help terminal command shows:
.dump ?OBJECTS? Render database content as SQL
and prints the SQL to the terminal, you can update it in a TXT file. For once off changes and tidying this seems like a reasonable solution albeit a little inelegant
I'd like to add a NOT NULL column to a Redshift table that has records, an IDENTITY field, and that other tables have foreign keys to.
In PostgreSQL, you can add the column as NULL, fill it in, then ALTER it to be NOT NULL.
In Redshift, the best I've found so far is:
ALTER TABLE my_table ADD COLUMN new_column INTEGER;
-- Fill that column
CREATE TABLE my_table2 (
id INTEGER IDENTITY NOT NULL SORTKEY,
(... all the fields ... )
new_column INTEGER NOT NULL,
PRIMARY KEY(id)
) DISTSTYLE all;
UNLOAD ('select * from my_table')
to 's3://blah' credentials '<aws-auth-args>' ;
COPY my_table2
from 's3://blah' credentials '<aws-auth-args>'
EXPLICIT_IDS;
DROP table my_table;
ALTER TABLE my_table2 RENAME TO my_table;
-- For each table that had a foreign key to my_table:
ALTER TABLE another_table ADD FOREIGN KEY(my_table_id) REFERENCES my_table(id)
Is this the best way of achieving this?
You can achieve this w/o having to load to S3.
modify the existing table to create the desired column w/ a default value
update that column in some way (in my case it was copying from another column)
create a new table with the column w/o a default value
insert into the new table (you must list out the columns rather than using (*) since the order may be the same (say if you want the new column in position 2)
drop the old table
rename the table
alter table to give correct owner (if appropriate)
ex:
-- first add the column w/ a default value
alter table my_table_xyz
add visit_id bigint NOT NULL default 0; -- not null but default value
-- now populate the new column with whatever is appropriate (the key in my case)
update my_table_xyz
set visit_id = key;
-- now create the new table with the proper constraints
create table my_table_xzy_new
(
key bigint not null,
visit_id bigint NOT NULL, -- here it is not null and no default value
adt_id bigint not null
);
-- select all from old into new
insert into my_table_xyz_new
select key, visit_id, adt_id
from my_table_xyz;
-- remove the orig table
DROP table my_table_xzy_events;
-- rename the newly created table to the desired table
alter table my_table_xyz_new rename to my_table_xyz;
-- adjust any views, foreign keys or permissions as required
I am trying to create table having different indexes with single query but H2 gives Error for example:
create table tbl_Cust
(
id int primary key auto_increment not null,
fid int,
c_name varchar(50),
INDEX (fid)
);
but this gives error as
Unknown data type: "("; SQL statement:
[Error Code: 50004]
[SQL State: HY004]
Due to this I have to run 2 different queries to create table with Index. First query to create table and then second query to add index with
create INDEX c_fid on tbl_Cust(fid);
Is there something wrong in my query or H2 simply does not support this creation of table with index in single query?
Interesting question. The solution is even more interesting, as it involves MySQL compatibility mode.
It's actually possible to perform the exact same command you wrote without any modification, provided you just add to your jdbc url the MySQL mode.
Example URL like this: jdbc:h2:mem:;mode=mysql
SQL remains:
create table tbl_Cust
(
id int primary key auto_increment not null,
fid int,
c_name varchar(50),
INDEX (fid)
);
Update count: 0
(15 ms)
Too bad I did not see this question earlier... Hopefully the solution might become handy one day to someone :-)
I could resolve the problem. According to
http://www.h2database.com/html/grammar.html#create_index
I modified the query. It works fine with my H2 server.
CREATE TABLE subscription_validator (
application_id int(11) NOT NULL,
api_id int(11) NOT NULL,
validator_id int(11) NOT NULL,
PRIMARY KEY (application_id,api_id),
CONSTRAINT subscription_validator_ibfk_1 FOREIGN KEY (validator_id) REFERENCES validator (id) ON UPDATE CASCADE
);
CREATE INDEX validator_id ON subscription_validator(validator_id);
In PostgreSql, one can define a sequence and use it as the primary key of a table. In HsqlDB, one can still accomplish creating an auto-increment identity column which doesn't link to any user defined sequence. Is it possible to use a user defined sequence as the generator of an auto-increment identity column in HsqlDB?
Sample sql in PostgreSql:
CREATE SEQUENCE seq_company_id START WITH 1;
CREATE TABLE company (
id bigint PRIMARY KEY DEFAULT nextval('seq_company_id'),
name varchar(128) NOT NULL CHECK (name <> '')
);
What's the equivalent in HsqlDB?
Thanks.
In version 2.0, there is no direct feature for this. You can define a BEFORE INSERT trigger on the table to do this:
CREATE TABLE company ( id bigint PRIMARY KEY, name varchar(128) NOT NULL CHECK (name <> '') );
CREATE TRIGGER trigg BEFORE INSERT
ON company REFERENCING NEW ROW AS newrow
FOR EACH ROW
SET newrow.id = NEXT VALUE FOR seq_company_id;
and insert without using any vlue for id
INSERT INTO company VALUES null, 'test'
Update for HSQLDB 2.1 and later: A feature has been added to support this.
CREATE SEQUENCE SEQU
CREATE TABLE company ( id bigint GENERATED BY DEFAULT AS SEQUENCE SEQU PRIMARY KEY, name varchar(128) NOT NULL CHECK (name <> '') );
See the Guide under CREATE TABLE http://hsqldb.org/doc/2.0/guide/databaseobjects-chapt.html#dbc_table_creation
In addition, 2.1 and later has a PostgreSQL compatibility mode in which it accepts the PostgreSQL CREATE TABLE statement that references the sequence in the DEFAULT clause and translates it to HSQLDB syntax.