How to add data from table to biggest object query in PostgreSQL - sql

Postgres 9.1+ database contains different schema for every company named firma and company number, like firma1, firma5, firma99, firma12.
Every schema contains a table with company name:
-- this table contains always exactly one row:
create table firma5.company ( company char(50) not null );
Following query lists biggest objects:
select
(n.nspname||'.'||relname)::char(45) as tablename
, pg_size_pretty(pg_total_relation_size(c.oid))::char(10) as totalsize
, case
when c.relkind='i' then 'index'
when c.relkind='t' then 'toast'
when c.relkind='r' then 'table'
when c.relkind='v' then 'view'
when c.relkind='c' then 'composite type'
when c.relkind='S' then 'sequence'
else c.relkind::text
end ::char(14) as "type"
from
pg_class c
left join pg_namespace n on n.oid = c.relnamespace
left join pg_tablespace t on t.oid = c.reltablespace
where
(pg_total_relation_size(c.oid)>>20)>0 and c.relkind!='t'
order by
pg_total_relation_size(c.oid) desc
This query shows company schemas like firma1, firma5 etc.
How to show company names ( firman.company.company ) also in this query result? Query may return also tables from schemas other than firmaN. In this case company name column should be empty or null.

This is not possible with plain SQL because you cannot specify beforehand the table name to join on, so you need to run a dynamic query. You can, however, create a simple function that returns the company name from a dynamic query, if the schema has company tables:
CREATE FUNCTION company_name (sch text) RETURNS text AS $$
DECLARE
comp text := NULL;
BEGIN
IF strpos(sch, 'firma') = 1 THEN
EXECUTE 'SELECT company FROM ' || sch || '.company' INTO comp;
END IF;
RETURN comp;
END; $$ LANGUAGE plpgsql STRICT STABLE;
And then use that function in your query:
select
(n.nspname||'.'||c.relname)::char(45) as tablename
, pg_size_pretty(pg_total_relation_size(c.oid))::char(10) as totalsize
, case
when c.relkind='i' then 'index'
-- when c.relkind='t' then 'toast' FILTERED OUT IN WHERE CLAUSE
when c.relkind='r' then 'table'
when c.relkind='v' then 'view'
when c.relkind='c' then 'composite type'
when c.relkind='S' then 'sequence'
else c.relkind::text
end ::char(14) as "type"
, company_name(n.nspname) as company -- <<<<<<<<<<<<<<<<<<<<<<<
from pg_class c
left join pg_namespace n on n.oid = c.relnamespace
--left join pg_tablespace t on t.oid = c.reltablespace NOT USED
where (pg_total_relation_size(c.oid)>>20)>0 and c.relkind!='t'
order by pg_total_relation_size(c.oid) desc;

Related

How to create "statistics" for every table of a schema? (where statistics means 1 returned record per each table)

I have a schema myschema in which I have dozens of tables, each containing a specific column maincolumn. I would like to create a statistics of the whole schema, where I can see how many distinct values are in the maincolumn per each table.
I know I can list the tables in a schema as a result of this query:
SELECT tablename
FROM pg_tables
WHERE schemaname = 'myschema'
Results:
tablename
---------
table1
table2
table3
...
Also I can run the query for each table:
SELECT COUNT(DISTINCT(maincolumn))
FROM myschema.table1
Results:
count
-----
2972
But I wonder if there's a way to join these two tasks into one query, which would result this:
tablename | count
----------+------
table1 | 2972
table2 | 542
table3 | 1523
....
A not so elegant solution would be to list all my tables, put the result in an Excel table and use CONCATENATE function to create a SELECT query to every table, and then I could run it line by line, but I would prefer an easier way.
You would need dynamic SQL for that, probably in a PL/pgSQL function like this:
CREATE OR REPLACE FUNCTION table_counts() RETURNS TABLE (table_name text, count bigint)
LANGUAGE plpgsql AS
$$BEGIN
FOR table_counts.table_name IN
SELECT t.table_name
FROM information_schema.tables AS t
WHERE t.table_schema = 'myschema'
AND t.table_type = 'BASE TABLE'
LOOP
EXECUTE format(
'SELECT count(*) FROM myschema.%I',
table_counts.table_name
) INTO table_counts.count;
RETURN NEXT;
END LOOP;
END;$$;
Also, this is quite expensive.
If approximate counts are good enough for you, try
SELECT c.relname, c.reltuples
FROM pg_catalog.pg_class AS c
JOIN pg_catalog.pg_namespace AS n
ON c.relnamespace = n.oid
WHERE n.nspname = 'myschema'
AND c.relkind = 'r';
The result becomes more accurate if you run ANALYZE first.

Bad attname value on pg_attribute

i have 2 postgres DB's with the same table X, and primary key PKEY (i'm not creator of this).
When im looking on it using my client (i try 2 different) or extracting ddl, i got identical source like this:
...
CONSTRAINT pkey PRIMARY KEY(column1, column2)
...
The problem is what i see in pg_attribute.attname - first DB have correct values (column1 and column2) but the second have column1 and id (?). The rest of data (attnum and other parameters) are identical...It's interesting that column id not exists on this table (X)...maybe one day it existed, but i'm not sure how to check it).
This is a production environment, so recreating index etc it's not easy...Have you met with a similar situation?
comparison method:
select cls.oid,
nsp.nspname as object_schema,
cls.relname as object_name,
a.attname,
case cls.relkind
when 'r' then 'TABLE'
when 'm' then 'MATERIALIZED_VIEW'
when 'i' then 'INDEX'
when 'S' then 'SEQUENCE'
when 'v' then 'VIEW'
when 'c' then 'TYPE'
else cls.relkind::text
end as type
from pg_class cls
join pg_roles rol on rol.oid = cls.relowner
join pg_namespace nsp on nsp.oid = cls.relnamespace
join pg_catalog.pg_attribute a on a.attrelid = cls.oid
left outer join pg_attrdef ad on (ad.adrelid = cls.oid and ad.adnum = a.attnum)
left outer join pg_constraint con on cls.oid = con.conrelid
where nsp.nspname not in ('information_schema', 'pg_catalog')
and nsp.nspname not like 'pg_toast%'
and a.attnum > 0
AND NOT a.attisdropped
and cls.relname like '%my_pkey_real_name%'
order by 1, 2
This query returns 'column1' and 'column2' in attname column on DB1 and 'column1' and 'id' on DB2.
As i wrote - the problem is that the column 'id' don't exists...when i'm extract ddl i'm getting somthing like:
ALTER TABLE X
ADD CONSTRAINT pkey
PRIMARY KEY (column1, column2) NOT DEFERRABLE;
on both DB's
So, i know the reason but not a solution :(
Let's create a test table:
CREATE TABLE test.test_x (
id VARCHAR(128) NOT NULL,
CONSTRAINT pkey PRIMARY KEY(id)
)
WITH (oids = false);
then query pg_class and pg_attribute:
select c.oid, c.* from pg_class c where relname = 'pkey'
select * from pg_attribute where attrelid = 854514857
the result of second query is:
next reneme the primary key column:
ALTER TABLE test.test_x RENAME COLUMN id TO id_x;
the name of this column have changed, ddl is ok, value in pg_class (queried for the table context) is ok (id_x, not id), but in pg_attribute is still the older one:
i tried reindex this table (or only index) but it didn't help

Building dymanic query with many laterail joins

I am building a dynamic query based on user-submitted conditions on user-specified fields.
I have a two tables
Visitors(id, name, email...)
and a
Trackings(id, visitor_id, field, string_value, integer_value, boolean_value, date_value)
The conditions comes in the form of an array of SQL fragments built earlier. Default filters for attributes that are hardcoded on the visitors table, and custom filters, for the values that are user-submitted and stored in an EAV schema (trackings)
Example:
Default:
{"name ILIKE 'Jack'", "(last_seen < (current_date - (7 || ' days')::interval))"}
Custom:
{"field = 'number_of_orders' > 10", "is_pro_user = true"}
A visitor can have many trackings, each one recording some custom, user-submitted data field for that visitor. But each visitor also has some default data that lies on the table itself, such as email, name or last_seen etc.
Now, users should be to ask queries such as:
Give me all the visitors for which no custom field named number_of_orders has been recored (is unknown)
Give me all visitors with the default attribute name set to Jack and for who the custom attribute total_purchase_value is greater than 1000
My attempt at solving it was using a stored procedure that dynamically concatenated a series of conditions using AND (for default data on the visitor table), and OR statements (for the custom data in the trackings table) inside a WHERE-caluse
CREATE OR REPLACE FUNCTION find_matching_visitors(app_id text, default_filters text[], custom_filters text[])
RETURNS TABLE (
id varchar
) AS
$body$
DECLARE
default_filterstring text;
custom_filterstring text;
default_filter_length integer;
custom_filter_length integer;
sql VARCHAR;
BEGIN
default_filter_length := COALESCE(array_length(default_filters, 1), 0);
custom_filter_length := COALESCE(array_length(custom_filters, 1), 0);
default_filterstring := array_to_string(default_filters, ' AND ');
custom_filterstring := array_to_string(custom_filters, ' OR ');
IF custom_filterstring = '' or custom_filterstring is null THEN
custom_filterstring := '1=1';
END IF;
IF default_filterstring = '' or default_filterstring is null THEN
default_filterstring := '1=1';
END IF;
sql := format('
SELECT v.id FROM visitors v
LEFT JOIN trackings t on v.id = t.visitor_id
WHERE v.app_id = app_id and (%s) and (%s)
group by v.id
having case when %s > 0 then count(v.id) = %s else true end
', custom_filterstring, default_filterstring, custom_filter_length, custom_filter_length);
RETURN QUERY EXECUTE sql;
END;
$body$
LANGUAGE 'plpgsql';
This works great, but AFAIK there is not way to express the unknown filter for a custom attribute. In that it would require using a left outer join, or a not exists subquery with access to the outer scope.
I am now looking at alternative ways to accomplish the same as above, but also support this kind of query. I am thinking something like the below, using a series of lateral joins for each condition, but this seems like it won't perform very well as soon as there is more than 1-2 conditions/joins.
select v.id, v.name from visitors v
inner join lateral ( <-- custom fields
select * from trackings t
where field = 'admin'
) as t1 on t1.visitor_id = v.id
inner join lateral (
select * from trackings t
where field = 'users_created'
) as t2 on t2.visitor_id = v.id
inner join lateral (
select * from trackings t
where field = 'teams_created' and integer_value > 0
) as t3 on t3.visitor_id = v.id
where v.app_id = 'ASnYW1-RgCl0I' and (v.type = 'lead' or v.type = 'user')
and name ILIKE 'mads' and email is not null // <-- default fields
Any suggestions?

Is there something like show create database in postgres? [duplicate]

How do you perform the equivalent of Oracle's DESCRIBE TABLE in PostgreSQL with psql command?
Try this (in the psql command-line tool):
\d+ tablename
See the manual for more info.
In addition to the PostgreSQL way (\d 'something' or \dt 'table' or \ds 'sequence' and so on)
The SQL standard way, as shown here:
select column_name, data_type, character_maximum_length, column_default, is_nullable
from INFORMATION_SCHEMA.COLUMNS where table_name = '<name of table>';
It's supported by many db engines.
If you want to obtain it from query instead of psql, you can query the catalog schema. Here's a complex query that does that:
SELECT
f.attnum AS number,
f.attname AS name,
f.attnum,
f.attnotnull AS notnull,
pg_catalog.format_type(f.atttypid,f.atttypmod) AS type,
CASE
WHEN p.contype = 'p' THEN 't'
ELSE 'f'
END AS primarykey,
CASE
WHEN p.contype = 'u' THEN 't'
ELSE 'f'
END AS uniquekey,
CASE
WHEN p.contype = 'f' THEN g.relname
END AS foreignkey,
CASE
WHEN p.contype = 'f' THEN p.confkey
END AS foreignkey_fieldnum,
CASE
WHEN p.contype = 'f' THEN g.relname
END AS foreignkey,
CASE
WHEN p.contype = 'f' THEN p.conkey
END AS foreignkey_connnum,
CASE
WHEN f.atthasdef = 't' THEN d.adsrc
END AS default
FROM pg_attribute f
JOIN pg_class c ON c.oid = f.attrelid
JOIN pg_type t ON t.oid = f.atttypid
LEFT JOIN pg_attrdef d ON d.adrelid = c.oid AND d.adnum = f.attnum
LEFT JOIN pg_namespace n ON n.oid = c.relnamespace
LEFT JOIN pg_constraint p ON p.conrelid = c.oid AND f.attnum = ANY (p.conkey)
LEFT JOIN pg_class AS g ON p.confrelid = g.oid
WHERE c.relkind = 'r'::char
AND n.nspname = '%s' -- Replace with Schema name
AND c.relname = '%s' -- Replace with table name
AND f.attnum > 0 ORDER BY number
;
It's pretty complex but it does show you the power and flexibility of the PostgreSQL system catalog and should get you on your way to pg_catalog mastery ;-). Be sure to change out the %s's in the query. The first is Schema and the second is the table name.
You can do that with a psql slash command:
\d myTable describe table
It also works for other objects:
\d myView describe view
\d myIndex describe index
\d mySequence describe sequence
Source: faqs.org
The psql equivalent of DESCRIBE TABLE is \d table.
See the psql portion of the PostgreSQL manual for more details.
This should be the solution:
SELECT * FROM information_schema.columns
WHERE table_schema = 'your_schema'
AND table_name = 'your_table'
You may do a \d *search pattern * with asterisks to find tables that match the search pattern you're interested in.
In addition to the command line \d+ <table_name> you already found, you could also use the information-schema to look up the column data, using info_schema.columns
SELECT *
FROM info_schema.columns
WHERE table_schema = 'your_schema'
AND table_name = 'your_table'
Use the following SQL statement
SELECT DATA_TYPE
FROM INFORMATION_SCHEMA.COLUMNS
WHERE table_name = 'tbl_name'
AND COLUMN_NAME = 'col_name'
If you replace tbl_name and col_name, it displays data type of the particular coloumn that you looking for.
You can use this :
SELECT attname
FROM pg_attribute,pg_class
WHERE attrelid=pg_class.oid
AND relname='TableName'
AND attstattarget <>0;
In MySQL , DESCRIBE table_name
In PostgreSQL , \d table_name
Or , you can use this long command:
SELECT
a.attname AS Field,
t.typname || '(' || a.atttypmod || ')' AS Type,
CASE WHEN a.attnotnull = 't' THEN 'YES' ELSE 'NO' END AS Null,
CASE WHEN r.contype = 'p' THEN 'PRI' ELSE '' END AS Key,
(SELECT substring(pg_catalog.pg_get_expr(d.adbin, d.adrelid), '\'(.*)\'')
FROM
pg_catalog.pg_attrdef d
WHERE
d.adrelid = a.attrelid
AND d.adnum = a.attnum
AND a.atthasdef) AS Default,
'' as Extras
FROM
pg_class c
JOIN pg_attribute a ON a.attrelid = c.oid
JOIN pg_type t ON a.atttypid = t.oid
LEFT JOIN pg_catalog.pg_constraint r ON c.oid = r.conrelid
AND r.conname = a.attname
WHERE
c.relname = 'tablename'
AND a.attnum > 0
ORDER BY a.attnum
This variation of the query (as explained in other answers) worked for me.
SELECT
COLUMN_NAME
FROM
information_schema.COLUMNS
WHERE
TABLE_NAME = 'city';
It's described here in details:
http://www.postgresqltutorial.com/postgresql-describe-table/
To improve on the other answer's SQL query (which is great!), here is a revised query. It also includes constraint names, inheritance information, and a data types broken into it's constituent parts (type, length, precision, scale). It also filters out columns that have been dropped (which still exist in the database).
SELECT
n.nspname as schema,
c.relname as table,
f.attname as column,
f.attnum as column_id,
f.attnotnull as not_null,
f.attislocal not_inherited,
f.attinhcount inheritance_count,
pg_catalog.format_type(f.atttypid,f.atttypmod) AS data_type_full,
t.typname AS data_type_name,
CASE
WHEN f.atttypmod >= 0 AND t.typname <> 'numeric'THEN (f.atttypmod - 4) --first 4 bytes are for storing actual length of data
END AS data_type_length,
CASE
WHEN t.typname = 'numeric' THEN (((f.atttypmod - 4) >> 16) & 65535)
END AS numeric_precision,
CASE
WHEN t.typname = 'numeric' THEN ((f.atttypmod - 4)& 65535 )
END AS numeric_scale,
CASE
WHEN p.contype = 'p' THEN 't'
ELSE 'f'
END AS is_primary_key,
CASE
WHEN p.contype = 'p' THEN p.conname
END AS primary_key_name,
CASE
WHEN p.contype = 'u' THEN 't'
ELSE 'f'
END AS is_unique_key,
CASE
WHEN p.contype = 'u' THEN p.conname
END AS unique_key_name,
CASE
WHEN p.contype = 'f' THEN 't'
ELSE 'f'
END AS is_foreign_key,
CASE
WHEN p.contype = 'f' THEN p.conname
END AS foreignkey_name,
CASE
WHEN p.contype = 'f' THEN p.confkey
END AS foreign_key_columnid,
CASE
WHEN p.contype = 'f' THEN g.relname
END AS foreign_key_table,
CASE
WHEN p.contype = 'f' THEN p.conkey
END AS foreign_key_local_column_id,
CASE
WHEN f.atthasdef = 't' THEN d.adsrc
END AS default_value
FROM pg_attribute f
JOIN pg_class c ON c.oid = f.attrelid
JOIN pg_type t ON t.oid = f.atttypid
LEFT JOIN pg_attrdef d ON d.adrelid = c.oid AND d.adnum = f.attnum
LEFT JOIN pg_namespace n ON n.oid = c.relnamespace
LEFT JOIN pg_constraint p ON p.conrelid = c.oid AND f.attnum = ANY (p.conkey)
LEFT JOIN pg_class AS g ON p.confrelid = g.oid
WHERE c.relkind = 'r'::char
AND f.attisdropped = false
AND n.nspname = '%s' -- Replace with Schema name
AND c.relname = '%s' -- Replace with table name
AND f.attnum > 0
ORDER BY f.attnum
;
You can also check using below query
Select * from schema_name.table_name limit 0;
Expmple : My table has 2 columns name and pwd. Giving screenshot below.
*Using PG admin3
In postgres \d is used to describe the table structure.
e.g. \d schema_name.table_name
this command will provide you the basic info of table such as, columns, type and modifiers.
If you want more info about table use
\d+ schema_name.table_name
this will give you extra info such as, storage, stats target and description
The best way to describe a table such as a column, type, modifiers of columns, etc.
\d+ tablename or \d tablename
When your table is not part of the default schema, you should write:
\d+ schema_name.table_name
Otherwise, you would get the error saying that "the relation doesn not exist."
When your table name starts with a capital letter you should put your table name in the quotation.
Example: \d "Users"
Use this command
\d table name
like
\d queuerecords
Table "public.queuerecords"
Column | Type | Modifiers
-----------+-----------------------------+-----------
id | uuid | not null
endtime | timestamp without time zone |
payload | text |
queueid | text |
starttime | timestamp without time zone |
status | text |
1) PostgreSQL DESCRIBE TABLE using psql
In psql command line tool, \d table_name or \d+ table_name to find the information on columns of a table
2) PostgreSQL DESCRIBE TABLE using information_schema
SELECT statement to query the column_names,datatype,character maximum length of the columns table in the information_schema database;
SELECT
COLUMN_NAME, DATA_TYPE, CHARACTER_MAXIMUM_LENGTH
from INFORMATION_SCHEMA.COLUMNS where table_name = 'tablename';
For more information https://www.postgresqltutorial.com/postgresql-describe-table/
I'll add the pg_dump command even thou the psql command was requested. because it generate an output more common to previous MySQl users.
# sudo -u postgres pg_dump --table=my_table_name --schema-only mydb
/dt is the commad which lists you all the tables present in a database. using
/d command and /d+ we can get the details of a table. The sysntax will be like
* /d table_name (or) \d+ table_name
The command below can describe multiple tables simply
\dt <table> <table>
The command below can describe multiple tables in detail:
\d <table> <table>
The command below can describe multiple tables in more detail:
\d+ <table> <table>
I worked out the following script for get table schema.
'CREATE TABLE ' || 'yourschema.yourtable' || E'\n(\n' ||
array_to_string(
array_agg(
' ' || column_expr
)
, E',\n'
) || E'\n);\n'
from
(
SELECT ' ' || column_name || ' ' || data_type ||
coalesce('(' || character_maximum_length || ')', '') ||
case when is_nullable = 'YES' then ' NULL' else ' NOT NULL' end as column_expr
FROM information_schema.columns
WHERE table_schema || '.' || table_name = 'yourschema.yourtable'
ORDER BY ordinal_position
) column_list;

List all sequences in a Postgres db 8.1 with SQL

I'm converting a db from postgres to mysql.
Since i cannot find a tool that does the trick itself, i'm going to convert all postgres sequences to autoincrement ids in mysql with autoincrement value.
So, how can i list all sequences in a Postgres DB (8.1 version) with information about the table in which it's used, the next value etc with a SQL query?
Be aware that i can't use the information_schema.sequences view in the 8.4 release.
The following query gives names of all sequences.
SELECT c.relname FROM pg_class c WHERE c.relkind = 'S';
Typically a sequence is named as ${table}_id_seq. Simple regex pattern matching will give you the table name.
To get last value of a sequence use the following query:
SELECT last_value FROM test_id_seq;
Note, that starting from PostgreSQL 8.4 you can get all information about sequences used in the database via:
SELECT * FROM information_schema.sequences;
Since I'm using a higher version of PostgreSQL (9.1), and was searching for same answer high and low, I added this answer for posterity's sake and for future searchers.
Launch psql with the -E flag ("echo the actual queries generated by \d and other backslash commands"), then enter the \ds command to list all sequences. You should see something like this:
# \ds
********* QUERY **********
SELECT n.nspname as "Schema",
c.relname as "Name",
CASE c.relkind WHEN 'r' THEN 'table' WHEN 'v' THEN 'view' WHEN 'm' THEN 'materialized view' WHEN 'i' THEN 'index' WHEN 'S' THEN 'sequence' WHEN 's' THEN 'special' WHEN 'f' THEN 'foreign table' WHEN 'p' THEN 'partitioned table' WHEN 'I' THEN 'partitioned index' END as "Type",
pg_catalog.pg_get_userbyid(c.relowner) as "Owner"
FROM pg_catalog.pg_class c
LEFT JOIN pg_catalog.pg_namespace n ON n.oid = c.relnamespace
WHERE c.relkind IN ('S','')
AND n.nspname <> 'pg_catalog'
AND n.nspname <> 'information_schema'
AND n.nspname !~ '^pg_toast'
AND pg_catalog.pg_table_is_visible(c.oid)
ORDER BY 1,2;
**************************
List of relations
Schema | Name | Type | Owner
--------+--------------------------------+----------+-------
public | assignments_id_seq | sequence | root
public | framework_users_id_seq | sequence | root
public | lending_items_id_seq | sequence | root
public | proxy_borrower_requests_id_seq | sequence | root
public | roles_id_seq | sequence | root
public | stack_requests_id_seq | sequence | root
(6 rows)
To examine a particular sequence, you can then run \d <sequence name>:
# \d lending_items_id_seq
********* QUERY **********
(...about four queries...)
**************************
Sequence "public.lending_items_id_seq"
Type | Start | Minimum | Maximum | Increment | Cycles? | Cache
--------+-------+---------+---------------------+-----------+---------+-------
bigint | 1 | 1 | 9223372036854775807 | 1 | no | 1
Owned by: public.lending_items.id
after a little bit of pain, i got it.
the best way to achieve this is to list all tables
select * from pg_tables where schemaname = '<schema_name>'
and then, for each table, list all columns with attributes
select * from information_schema.columns where table_name = '<table_name>'
then, for each column, test if it has a sequence
select pg_get_serial_sequence('<table_name>', '<column_name>')
and then, get the information about this sequence
select * from <sequence_name>
sequence info : max value
SELECT * FROM information_schema.sequences;
sequence info : last value
SELECT * FROM <sequence_name>
The relationship between automatically generated sequences ( such as those created for SERIAL columns ) and the parent table is modelled by the sequence owner attribute.
You can modify this relationship using the OWNED BY clause of the ALTER SEQUENCE commmand
e.g.
ALTER SEQUENCE foo_id OWNED by foo_schema.foo_table
to set it to be linked to the table foo_table
or
ALTER SEQUENCE foo_id OWNED by NONE
to break the connection between the sequence and any table
The information about this relationship is stored in the pg_depend catalogue table.
the joining relationship is the link between pg_depend.objid -> pg_class.oid WHERE relkind = 'S' - which links the sequence to the join record and then pg_depend.refobjid -> pg_class.oid WHERE relkind = 'r' , which links the join record to the owning relation ( table )
This query returns all the sequence -> table dependencies in a database. The where clause filters it to only include auto generated relationships, which restricts it to only display sequences created by SERIAL typed columns.
WITH fq_objects AS (SELECT c.oid,n.nspname || '.' ||c.relname AS fqname ,
c.relkind, c.relname AS relation
FROM pg_class c JOIN pg_namespace n ON n.oid = c.relnamespace ),
sequences AS (SELECT oid,fqname FROM fq_objects WHERE relkind = 'S'),
tables AS (SELECT oid, fqname FROM fq_objects WHERE relkind = 'r' )
SELECT
s.fqname AS sequence,
'->' as depends,
t.fqname AS table
FROM
pg_depend d JOIN sequences s ON s.oid = d.objid
JOIN tables t ON t.oid = d.refobjid
WHERE
d.deptype = 'a' ;
Get all sequence:
select * from pg_sequences;
PSQL:
\ds
\ds+
\ds *actor*
\ds *actor* will get all the sequence that sequence name contains actor characters.
I know this post is pretty old, but I found the solution by CMS to be very useful as I was looking for an automated way to link a sequence to the table AND column, and wanted to share. The use of pg_depend catalog table was the key. I expanded what was done to:
WITH fq_objects AS (SELECT c.oid,n.nspname || '.' ||c.relname AS fqname ,
c.relkind, c.relname AS relation
FROM pg_class c JOIN pg_namespace n ON n.oid = c.relnamespace ),
sequences AS (SELECT oid,fqname FROM fq_objects WHERE relkind = 'S'),
tables AS (SELECT oid, fqname FROM fq_objects WHERE relkind = 'r' )
SELECT
s.fqname AS sequence,
'->' as depends,
t.fqname AS table,
a.attname AS column
FROM
pg_depend d JOIN sequences s ON s.oid = d.objid
JOIN tables t ON t.oid = d.refobjid
JOIN pg_attribute a ON a.attrelid = d.refobjid and a.attnum = d.refobjsubid
WHERE
d.deptype = 'a' ;
This version adds column to the list of fields returned. With both the table name and the column name in hand, a call to pg_set_serial_sequence makes it easy to ensure that all sequences in the database are set correctly. For example:
CREATE OR REPLACE FUNCTION public.reset_sequence(tablename text, columnname text)
RETURNS void
LANGUAGE plpgsql
AS $function$
DECLARE
_sql VARCHAR := '';
BEGIN
_sql := $$SELECT setval( pg_get_serial_sequence('$$ || tablename || $$', '$$ || columnname || $$'), (SELECT COALESCE(MAX($$ || columnname || $$),1) FROM $$ || tablename || $$), true)$$;
EXECUTE _sql;
END;
$function$;
Hope this helps someone with resetting sequences!
I know the question was about postgresql version 8 but I wrote this simple way here for people who want to get sequences in version 10 and upper
you can use the bellow query
select * from pg_sequences
view-pg-sequences
This statement lists the table and column that is associated with each sequence:
Code:
SELECT t.relname as related_table,
a.attname as related_column,
s.relname as sequence_name
FROM pg_class s
JOIN pg_depend d ON d.objid = s.oid
JOIN pg_class t ON d.objid = s.oid AND d.refobjid = t.oid
JOIN pg_attribute a ON (d.refobjid, d.refobjsubid) = (a.attrelid, a.attnum)
JOIN pg_namespace n ON n.oid = s.relnamespace
WHERE s.relkind = 'S'
AND n.nspname = 'public'
more see here link to answer
select sequence_name, (xpath('/row/last_value/text()', xml_count))[1]::text::int as last_value
from (
select sequence_schema,
sequence_name,
query_to_xml(format('select last_value from %I.%I', sequence_schema, sequence_name), false, true, '') as xml_count
from information_schema.sequences
where sequence_schema = 'public'
) new_table order by last_value desc;
Partially tested but looks mostly complete.
select *
from (select n.nspname,c.relname,
(select substring(pg_catalog.pg_get_expr(d.adbin, d.adrelid) for 128)
from pg_catalog.pg_attrdef d
where d.adrelid=a.attrelid
and d.adnum=a.attnum
and a.atthasdef) as def
from pg_class c, pg_attribute a, pg_namespace n
where c.relkind='r'
and c.oid=a.attrelid
and n.oid=c.relnamespace
and a.atthasdef
and a.atttypid=20) x
where x.def ~ '^nextval'
order by nspname,relname;
Credit where credit is due... it's partly reverse engineered from the SQL logged from a \d on a known table that had a sequence. I'm sure it could be cleaner too, but hey, performance wasn't a concern.
Improvement of the previous answer:
select string_agg('select sequence_name, last_value from ' || relname, chr(13) || 'union' || chr(13) order by relname)
from pg_class where relkind ='S'
Thanks for your help.
Here is the pl/pgsql function which update each sequence of a database.
---------------------------------------------------------------------------------------------------------
--- Nom : reset_sequence
--- Description : Générique - met à jour les séquences au max de l'identifiant
---------------------------------------------------------------------------------------------------------
CREATE OR REPLACE FUNCTION reset_sequence() RETURNS void AS
$BODY$
DECLARE _sql VARCHAR := '';
DECLARE result threecol%rowtype;
BEGIN
FOR result IN
WITH fq_objects AS (SELECT c.oid,n.nspname || '.' ||c.relname AS fqname ,c.relkind, c.relname AS relation FROM pg_class c JOIN pg_namespace n ON n.oid = c.relnamespace ),
sequences AS (SELECT oid,fqname FROM fq_objects WHERE relkind = 'S'),
tables AS (SELECT oid, fqname FROM fq_objects WHERE relkind = 'r' )
SELECT
s.fqname AS sequence,
t.fqname AS table,
a.attname AS column
FROM
pg_depend d JOIN sequences s ON s.oid = d.objid
JOIN tables t ON t.oid = d.refobjid
JOIN pg_attribute a ON a.attrelid = d.refobjid and a.attnum = d.refobjsubid
WHERE
d.deptype = 'a'
LOOP
EXECUTE 'SELECT setval('''||result.col1||''', COALESCE((SELECT MAX('||result.col3||')+1 FROM '||result.col2||'), 1), false);';
END LOOP;
END;$BODY$ LANGUAGE plpgsql;
SELECT * FROM reset_sequence();
Here is another one which has the schema name beside the sequence name
select nspname,relname from pg_class c join pg_namespace n on c.relnamespace=n.oid where relkind = 'S' order by nspname
Get sequences by each column of each table via parsing of DEFAULT clause. This method provides info about to which column sequences are linked and does not use dependencies which may not exist for some sequences. Even pg_get_serial_sequence(sch.nspname||'.'||tbl.relname, col.attname) function found not all sequences for me!
Solution:
SELECT
seq_sch.nspname AS sequence_schema
, seq.relname AS sequence_name
, seq_use."schema" AS used_in_schema
, seq_use."table" AS used_in_table
, seq_use."column" AS used_in_column
FROM pg_class seq
INNER JOIN pg_namespace seq_sch ON seq_sch.oid = seq.relnamespace
LEFT JOIN (
SELECT
sch.nspname AS "schema"
, tbl.relname AS "table"
, col.attname AS "column"
, regexp_split_to_array(
TRIM(LEADING 'nextval(''' FROM
TRIM(TRAILING '''::regclass)' FROM
pg_get_expr(def.adbin, tbl.oid, TRUE)
)
)
, '\.'
) AS column_sequence
FROM pg_class tbl --the table
INNER JOIN pg_namespace sch ON sch.oid = tbl.relnamespace
--schema
INNER JOIN pg_attribute col ON col.attrelid = tbl.oid
--columns
INNER JOIN pg_attrdef def ON (def.adrelid = tbl.oid AND def.adnum = col.attnum) --default values for columns
WHERE tbl.relkind = 'r' --regular relations (tables) only
AND col.attnum > 0 --regular columns only
AND def.adsrc LIKE 'nextval(%)' --sequences only
) seq_use ON (seq_use.column_sequence [1] = seq_sch.nspname AND seq_use.column_sequence [2] = seq.relname)
WHERE seq.relkind = 'S' --sequences only
ORDER BY sequence_schema, sequence_name;
Note that 1 sequence can be used in multiple tables, so it can be listed in multiple rows here.
This function shows the last_value of each sequence.
It outputs a 2 columns table that says the sequence name plus it's last generated value.
drop function if exists public.show_sequence_stats();
CREATE OR REPLACE FUNCTION public.show_sequence_stats()
RETURNS TABLE(tablename text, last_value bigint)
LANGUAGE 'plpgsql'
COST 100
VOLATILE
ROWS 1000
AS $BODY$
declare r refcursor; rec record; dynamic_query varchar;
BEGIN
dynamic_query='select tablename,last_value from (';
open r for execute 'select nspname,relname from pg_class c join pg_namespace n on c.relnamespace=n.oid where relkind = ''S'' order by nspname';
fetch next from r into rec;
while found
loop
dynamic_query=dynamic_query || 'select '''|| rec.nspname || '.' || rec.relname ||''' "tablename",last_value from ' || rec.nspname || '.' || rec.relname || ' union all ';
fetch next from r into rec;
end loop;
close r;
dynamic_query=rtrim(dynamic_query,'union all') || ') x order by last_value desc;';
return query execute dynamic_query;
END;
$BODY$;
select * from show_sequence_stats();
Kind of a hack, but try this:
select 'select ''' || relname || ''' as sequence, last_value from ' || relname || ' union'
FROM pg_catalog.pg_class c
WHERE c.relkind IN ('S','');
Remove the last UNION and execute the result
Assuming exec() function declared in this post https://stackoverflow.com/a/46721603/653539 , sequences together with their last values can be fetched using single query:
select s.sequence_schema, s.sequence_name,
(select * from exec('select last_value from ' || s.sequence_schema || '.' || s.sequence_name) as e(lv bigint)) last_value
from information_schema.sequences s
Here is an example how to use psql to get a list of all sequences with their last_value:
psql -U <username> -d <database> -t -c "SELECT 'SELECT ''' || c.relname || ''' as sequence_name, last_value FROM ' || c.relname || ';' FROM pg_class c WHERE (c.relkind = 'S')" | psql -U <username> -d <database> -t
SELECT nextval(sequence_schema || '.' || sequence_name)
FROM information_schema.sequences