Dynamic LIKE in Where Clause statement - sql

I've been trying my best to achieve dynamic query for the LIKE statement
From the below SQL query instead of doing ilike for every value and it may grow in large. I cannot rewrite the query again and again for new table_name value .. I can store those values in a separate table, But How do I achieve it dynamically
SELECT
t.table_name as table_name,
t.table_schema as table_schema
FROM
information_schema.tables T
WHERE (table_schema ilike 'stage' and table_name like 'ABC%') or (table_schema ilike 'stage' and table_name like 'EFG%');
I can have another table with a list of values like below
create or replace table tempdw.blk_table;
(
db_name varchar,
tbl_expr varchar
);
insert into tempdw.blk_table values ('stage','ABC%');
insert into tempdw.blk_table values ('stage','EFG%');
select * from tempdw.blk_table;

If you have a table with tables and schemas in them, you could always collate your join to take care of the case-sensitive part:
SELECT
t.table_name as table_name,
t.table_schema as table_schema
FROM
information_schema.tables T
JOIN
table_names tn
ON collate(t.table_name , 'en-ci') like collate(tn.table_name , 'en-ci')
AND collate(t.table_schema, 'en-ci') = collate(tn.schema_name, 'en-ci');

If you are just looking for a set of tables with prefixes, you could use regexp_like():
where table_schema ilike 'stage' and
regexp(lower(table_name), '^abc|efg')

Related

How to execute result of a query in PostgreSQL

I am trying to create a view by joining two tables. These two tables have a few columns with the same name. Which gives an error
SQL Error [42701]: ERROR: column "column_name" specified more than once
I cannot use column names while creating the view as there are 30+ columns and new columns will be added to both tables over the period of time. Hence, I have to use * to get all the columns.
Now, to eliminate columns which exist in both the table, I went ahead and did this:
SELECT 'SELECT ' || STRING_AGG('u2.' || column_name, ', ') || ' FROM schema_name.table_name u2'
FROM information_schema.columns
WHERE table_name = 'table_name'
AND table_schema = 'schema_name'
AND column_name NOT IN ('column_name');
This gives me the query to select data from schema_name.table_name without the column column_name. Great!!
The problem: How do I execute the result of the above query?
I tried PREPARE statement, it is just executing the above query and not the result of the above query.
Also, creating a temporary table with no column "column_name" isn't a viable solution.
You need to prepare a dynamic query and then EXECUTE it. It would be something like this:
DO
$do$
BEGIN
EXECUTE (
SELECT CONCAT('CREATE VIEW temp_view AS SELECT ',
-- SELECT table1 columns here
(SELECT STRING_AGG('u1.' || column_name, ', ')
FROM information_schema.columns
WHERE table_name = 'table1'
AND table_schema = 'schema_name'
-- AND column_name NOT IN ('column_name') -- not omitting for table1
),
', ',
-- SELECT table2 columns here
(SELECT STRING_AGG('u2.' || column_name, ', ')
FROM information_schema.columns
WHERE table_name = 'table2'
AND table_schema = 'schema_name'
AND column_name NOT IN ('column_name')),
-- Dynamically prepare the FROM and JOIN clauses
' FROM table1 u1 JOIN table2 u2 ON u1.id = u2.table1_id'));
END
$do$;
CHECK DEMO

How do I find table column in different databases?

I am trying to find table columns in different tables and different databases.
So far, I do have a query to do that, however it does not tell me in which database that column and table lays in.
Current code:
SELECT
sys.columns.name AS ColumnName,
tables.name AS TableName
FROM
sys.columns
JOIN
sys.tables ON sys.columns.object_id = tables.object_id
WHERE
sys.columns.name LIKE '%COLUMNNAME%'
Does anyone have an idea what I need to add to display the database name as well?
I'd suggest using INFORMATION_SCHEMA.tables or .columns like:
SELECT table_name, table_schema, table_catalog
from INFORMATION_SCHEMA.tables
where Table_name like '%<table>%'
Same with .columns. Just replace table_name with column_name
SELECT column_name, table_schema, table_catalog, *
from INFORMATION_SCHEMA.COLUMNS where column_name like '%<Column>%'

SQL Server - Search for a column across all database tables

I would like to execute a SQL query to find all tables within a database that does not have a particular column?
E.g. I would like a list of tables that do not have a column called 'BranchID'.
Thank you for your attention.
Information Schema Views contain metadata about the database. For more information, go here
Using these views you can do something like this...
SELECT
t.TABLE_NAME
FROM
INFORMATION_SCHEMA.TABLES AS t
WHERE NOT EXISTS
(SELECT TABLE_NAME FROM INFORMATION_SCHEMA.COLUMNS AS c WHERE c.TABLE_NAME = t.TABLE_NAME AND c.COLUMN_NAME = 'BranchID')
select table_name
from INFORMATION_SCHEMA.COLUMNS
except
select table_name
from INFORMATION_SCHEMA.COLUMNS
where column_name = 'BranchID'
(from memory, I might have the column names wrong but should get you most of the way there :) )
This should work.
SELECT table_name
FROM INFORMATION_SCHEMA.columns c
WHERE NOT EXISTS (
SELECT table_name
FROM INFORMATION_SCHEMA.columns cc
WHERE cc.table_name = c.table_name
AND cc.column_name = 'Branch_id'
)
First of all, we select the table_name from Information_Schema.Columns where the table_name is NOT in a query that selects tables WITH the Branch_ID Column
You can use the INFORMATION_SCHEMA.COLUMNS table and INFORMATION_SCHEMA.TABLES table to get the information you want using normal SQL Queries.
In your case you would use a subquery:
select TABLE_NAME from INFORMATION_SCHEMA.TABLES where TABLE_NAME NOT IN (select TABLE_NAME from INFORMATION_SCHEMA.COLUMNS where COLUMN_NAME='BranchID')
I have created a re-usable procedure that could help you.
CREATE PROC SP_UTIL_findcolumns
#SearchString As varchar(250) = NULL
AS
BEGIN
SELECT
t.TABLE_NAME
FROM INFORMATION_SCHEMA.TABLES AS t
WHERE NOT EXISTS
(SELECT TABLE_NAME
FROM INFORMATION_SCHEMA.COLUMNS AS c
WHERE c.TABLE_NAME = t.TABLE_NAME
AND c.COLUMN_NAME = #SearchString )
END

Finding the specific column (field) name from multiple tables in a database

I have large numbers of tables in my database. I am searching for a column named Country, but don't know which table contains that column. Is there a specific query that will help me to find the name of the table containing this column?
Yes, you can use INFORMATION_SCHEMA.COLUMNS
SELECT DISTINCT
TABLE_NAME
FROM
INFORMATION_SCHEMA.COLUMNS WHERE COLUMN_NAME = 'Country'
SELECT t.name AS table_name,
SCHEMA_NAME(schema_id) AS schema_name,
c.name AS column_name
FROM sys.tables AS t
INNER JOIN sys.columns c ON t.OBJECT_ID = c.OBJECT_ID
WHERE c.name LIKE '%ColumnName%'
ORDER BY schema_name, table_name;
Replace ColumnName to your actual column name
select distinct table_schema, table_name from information_schema.columns where table_name = 'Country';
You can find such of information in the INFORMATION_SCHEMA.COLUMNS
USE YourDBName;
SELECT DISTINCT Table_Name
FROM INFORMATION_SCHEMA.COLUMNS
WHERE Column_Name = 'Country';
If we want to find a column name with like operator in more than one tables (ex: 3 tables) then go for below query:
select COLUMN_NAME, TABLE_NAME
from ALL_TAB_COLUMNS
where TABLE_NAME in ('SIL_RLS_UPLOAD_TMP','SIL_CUSTOMER_MST','SIL_PERSONAL_CUSTOMER_MST')
and upper(column_name) like upper('%loan%');
If we want to find a column name with like operator in all tables then go for below query:
select *
from all_tab_columns
where upper(column_name) like upper('%card%');

How can I get column names from a table in Oracle?

I need to query the database to get the column names, not to be confused with data in the table. For example, if I have a table named EVENT_LOG that contains eventID, eventType, eventDesc, and eventTime, then I would want to retrieve those field names from the query and nothing else.
I found how to do this in:
Microsoft SQL Server
MySQL
PostgreSQL
But I need to know: how can this be done in Oracle?
You can query the USER_TAB_COLUMNS table for table column metadata.
SELECT table_name, column_name, data_type, data_length
FROM USER_TAB_COLUMNS
WHERE table_name = 'MYTABLE'
In SQL Server...
SELECT [name] AS [Column Name]
FROM syscolumns
WHERE id = (SELECT id FROM sysobjects WHERE type = 'V' AND [Name] = 'Your table name')
Type = 'V' for views
Type = 'U' for tables
You can do this:
describe EVENT_LOG
or
desc EVENT_LOG
Note: only applicable if you know the table name and specifically for Oracle.
For SQL Server 2008, we can use information_schema.columns for getting column information
SELECT *
FROM information_schema.columns
WHERE table_name = 'Table_Name'
ORDER BY ordinal_position
For SQLite I believe you can use something like the following:
PRAGMA table_info(table-name);
Explanation from sqlite.org:
This pragma returns one row for each column in the named table. Columns in the result set include the column name, data type, whether or not the column can be NULL, and the default value for the column. The "pk" column in the result set is zero for columns that are not part of the primary key, and is the index of the column in the primary key for columns that are part of the primary key.
See also: Sqlite.org Pragma Table Info
That information is stored in the ALL_TAB_COLUMNS system table:
SQL> select column_name from all_tab_columns where table_name = 'DUAL';
DUMMY
Or you could DESCRIBE the table if you are using SQL*PLUS:
SQL> desc dual
Name Null? Type
----------------------------------------------------- -------- ---------------------- -------------
DUMMY VARCHAR2(1)
The other answers sufficiently answer the question, but I thought I would share some additional information. Others describe the "DESCRIBE table" syntax in order to get the table information. If you want to get the information in the same format, but without using DESCRIBE, you could do:
SELECT column_name as COLUMN_NAME, nullable || ' ' as BE_NULL,
SUBSTR(data_type || '(' || data_length || ')', 0, 10) as TYPE
FROM all_tab_columns WHERE table_name = 'TABLENAME';
Probably doesn't matter much, but I wrote it up earlier and it seems to fit.
For Oracle
SELECT column_name FROM user_tab_cols WHERE table_name=UPPER('tableName');
describe YOUR_TABLE;
In your case :
describe EVENT_LOG;
Even this is also one of the way we can use it
select * from product where 1 != 1
select column_name,* from information_schema.columns
where table_name = 'YourTableName'
order by ordinal_position
For MySQL, use
SELECT column_name
FROM information_schema.columns
WHERE
table_schema = 'Schema' AND table_name = 'Table_Name'
For SQL Server:
SELECT [name] AS [Column Name]
FROM syscolumns
WHERE id = object_id('TABLE_NAME')
SELECT COLUMN_NAME 'all_columns'
FROM INFORMATION_SCHEMA.COLUMNS
WHERE TABLE_NAME='user';
You could also try this, but it might be more information than you need:
sp_columns TABLE_NAME
SELECT A.COLUMN_NAME, A.* FROM all_tab_columns a
WHERE table_name = 'Your Table Name'
AND A.COLUMN_NAME = 'COLUMN NAME' AND a.owner = 'Schema'
Mysql
SHOW COLUMNS FROM a_table_named_users WHERE Field REGEXP 'user_id|user_name|user_pass'
This will return a result something like this:
Field | Type | Null | Key | Default | Extra
user_id int(8) NO PRI NULL auto_increment
user_name varchar(64) NO MUL NULL
user_pass varchar(64) NO NULL
Then to pull out the values you can simply
fetch row[0]
This is also great for passing input dynamically since the REGEXP needs the '|' for multiple inputs, but is also a way to keeps data separated and easy to store/pass to classes/functions.
Try throwing in dummy data as well for security when sending it out and compare what was returned when receiving any errors.
In Oracle, there is two views that describe columns:
DBA_TAB_COLUMNS describes the columns of all tables, views, and
clusters in the database.
USER_TAB_COLUMNS describes the columns of the tables, views, and
clusters owned by the current user. This view does not display the
OWNER column.
The answer is here: http://php.net/manual/en/function.mysql-list-fields.php
I'd use the following code in your case:
$result = mysql_query("SHOW COLUMNS FROM sometable");
if (!$result) {
echo 'Could not run query: ' . mysql_error();
exit;
}
$fields = array();
if (mysql_num_rows($result) > 0) {
while ($row = mysql_fetch_assoc($result)) {
$fields[] = $row['Field'];
}
}
you can run this query
SELECT t.name AS table_name,
SCHEMA_NAME(schema_id) AS schema_name,
c.name AS column_name
FROM sys.tables AS t
INNER JOIN sys.columns c ON t.OBJECT_ID = c.OBJECT_ID
WHERE c.name LIKE '%%' --if you want to find specific column write here
ORDER BY schema_name, table_name;
Try this
select * from sys.all_columns c join sys.objects o on c.object_id=o.object_id where o.name = 'TABLENAME' and c.name like '%COLUMN NAME%'
Just select first row from the table , for oracle : select * from <table name> where rownum = 1;
Came across this question looking for access to column names on Teradata, so I'll add the answer for their 'flavour' of SQL:
SELECT ColumnName
FROM DBC.Columns
WHERE DatabaseName='DBASE_NAME'
AND TableName='TABLE_NAME';
The info is stored in the DBC dbase.
Getting data types is a little bit more involved:
Get column type using teradata system tables
I did it like this
SELECT
TOP 0
*
FROM
Posts
It works even in http://data.stackexchange.com whose service tables I am not aware of!
SELECT COLUMN_NAME
FROM YourDatabase.INFORMATION_SCHEMA.COLUMNS
WHERE TABLE_NAME = 'YourTableName'