I have two Sql Server 2008 R2 databases, called MyDbOne and MyDbTwo.
I'd like to merge them (all tables and procedures) into a new database called MyDb.
Assume there are no conflicts in table and procedure names between them.
The problem is: there are LOTS of code and procedures that execute queries using the database name, including procedures declared in one of the databases referecing tables from the other. There are queries like:
select * from MyDbOne..SomeTable;
select * from MyDbTwo..AnotherTable;
The tables SomeTable and AnotherTable would then exist in the MyDb database. But I need to support querying them using the leagcy names MyDbOne and MyDbTwo. If I would run the queries above, I'd like them to be effectively the same as:
select * from MyDb..SomeTable;
select * from MyDb..AnotherTable;
Is there some way to do it? Maybe create some sort of global aliases on the new database, resolving MyDbOne and MyDbTwo to MyDb? That would be perfect, but I don't know how to do it.
Thank you ver much!
There's no easy way around it. You'll have to update all those references manually. You can search the SQL server for all objects containing those database names by using one of these queries from this answer;
SELECT ROUTINE_NAME, ROUTINE_DEFINITION
FROM INFORMATION_SCHEMA.ROUTINES
WHERE ROUTINE_DEFINITION LIKE '%Foo%'
AND ROUTINE_TYPE='PROCEDURE'
SELECT OBJECT_NAME(id)
FROM SYSCOMMENTS
WHERE [text] LIKE '%Foo%'
AND OBJECTPROPERTY(id, 'IsProcedure') = 1
GROUP BY OBJECT_NAME(id)
SELECT OBJECT_NAME(object_id)
FROM sys.sql_modules
WHERE OBJECTPROPERTY(object_id, 'IsProcedure') = 1
AND definition LIKE '%Foo%'
I don't think there is an issue solution to this. If I were you, I would keep existing MyDbOne and MyDbTwo databases intact and create a brand new one that references these two tables via views. Basically, you will create a Data Warehouse that seats on top of these existing databases. Good thing about this setup is, you don't have to rewrite any legacy code, which is huge.
Related
How do I drop a few tables (e.g. 1 - 3) using the output of a SELECT statement for the table names? This is probably standard SQL, but specifically I'm using Apache Impala SQL accessed via Apache Zeppelin.
So I have a table called tables_to_drop with a single column called "table_name". This will have one to a few entries in it, each with the name of another temporary table that was generated as the result of other processes. As part of my cleanup I need to drop these temporary tables whose names are listed in the "tables_to_drop" table.
Conceptually I was thinking of an SQL command like:
DROP TABLE (SELECT table_name FROM tables_to_drop);
or:
WITH subquery1 AS (SELECT table_name FROM tables_to_drop) DROP TABLE * FROM subquery1;
Neither of these work (syntax errors). Any ideas please?
even in standard sql this is not possible to do it the way you showed.
in standard sql usually you can use dynamic sql which impala doesn't support.
however you can write an impala script and run it in impala shell but it's going to be complicated for such task, I would prepare the drop statement using select and run it manually if this is one-time thing:
select concat('DROP TABLE IF EXISTS ',table_name) dropstatements
from tables_to_drop
I have to add a few columns to a table and I also need to add these columns to all the views that use this table.
Is it possible to get a list of all the views in a database that use a certain table?
This should do it:
SELECT *
FROM INFORMATION_SCHEMA.VIEWS
WHERE VIEW_DEFINITION like '%YourTableName%'
To find table dependencies you can use the sys.sql_expression_dependencies catalog view:
SELECT
referencing_object_name = o.name,
referencing_object_type_desc = o.type_desc,
referenced_object_name = referenced_entity_name,
referenced_object_type_desc =so1.type_desc
FROM sys.sql_expression_dependencies sed
INNER JOIN sys.views o ON sed.referencing_id = o.object_id
LEFT OUTER JOIN sys.views so1 ON sed.referenced_id =so1.object_id
WHERE referenced_entity_name = 'Person'
You can also try out ApexSQL Search a free SSMS and VS add-in that also has the View Dependencies feature. The View Dependencies feature has the ability to visualize all SQL database objects’ relationships, including those between encrypted and system objects, SQL server 2012 specific objects, and objects stored in databases encrypted with Transparent Data Encryption (TDE)
Disclaimer: I work for ApexSQL as a Support Engineer
If you need to find database objects (e.g. tables, columns, triggers) by name - have a look at the FREE Red-Gate tool called SQL Search which does this - it searches your entire database for any kind of string(s).
It's a great must-have tool for any DBA or database developer - did I already mention it's absolutely FREE to use for any kind of use??
I find this works better:
SELECT type, *
FROM sys.objects
WHERE OBJECT_DEFINITION(object_id) LIKE '%' + #ObjectName + '%'
AND type IN ('V')
ORDER BY name
Filtering VIEW_DEFINTION inside INFORMATION_SCHEMA.VIEWS is giving me quite a few false positives.
SELECT VIEW_NAME
FROM INFORMATION_SCHEMA.VIEW_TABLE_USAGE
WHERE TABLE_NAME = 'Your Table'
select your table -> view dependencies -> Objects that depend on
Simplest way to find used view or stored procedure for the tableName using below query -
exec dbo.dbsearch 'Your_Table_Name'
Platform: SQL Server 2008
Language: TSQL
I have a number of queries that currently take the general form of (for simplicity sake)
-- Sample begin results
SELECT * from DB01.dbo.table UNION ALL
SELECT * from DB02.dbo.table UNION ALL --many other databases follow with same syntax
How can I modify these queries such that, when a new database comes into existence (named, say DB39C), I ensure that my queries already includes those new records?
--Sample end results
SELECT * from DB01.dbo.table UNION ALL
SELECT * from DB02.dbo.table UNION ALL
SELECT * from DB39C.dbo.table -- this was created as soon as a new database came into existence
I am looking to make sure programmatically, that this happens without my awareness as new databases are added quite regularly and I need the queries I rely on to keep pace.
You might want to have a look at using something like
SELECT name AS DATABASENAME
FROM master.dbo.sysdatabases
and creating dynamic queries
sys.databases (Transact-SQL)
I have created a new database with copies of existing tables but changed the names of these tables, is there a SQL script that I can run (maybe using SysObjects) to change all references to these tables in all stored procedures?
DO NOT RELY ON INFORMATION_SCHEMA.ROUTINES because ROUTINE_DEFINITION is only nvarchar(4000). You need to sys.sql_modules where definition is nvarchar(max)
try any of these to find the procedure that you need to modify:
SELECT DISTINCT
LEFT(s.name+'.'+o.name, 100) AS Object_Name,o.type_desc --, m.definition
FROM sys.sql_modules m
INNER JOIN sys.objects o ON m.object_id=o.object_id
INNER JOIN sys.schemas s ON o.schema_id=s.schema_id
WHERE m.definition Like '%'+#SearchValue+'%'
ORDER BY 1
SELECT
OBJECT_SCHEMA_NAME(m.object_id)+'.'+OBJECT_NAME(m.object_id) --, m.definition
FROM sys.sql_modules m
WHERE m.definition like '%whatever%'
SELECT
OBJECT_SCHEMA_NAME(m.object_id)+'.'+OBJECT_NAME(m.object_id), o.type_desc
--,m.definition
FROM sys.sql_modules m
INNER JOIN sys.objects o ON m.object_id=o.object_id
WHERE m.definition like '%whatever%'
you can uncomment m.definition to list out the content, but I find it better to just ID all the procedures and then review them manually, because you don't want to run UPDATE commands on the system tables. Script out the necessary procedures, make the changes (search/replace or manually), and then run the scripts!!!
No.
I believe SQL Refactor from Redgate has this functionality. Otherwise you could script out all objects and either manually or via code do a search and replace.
SQL Server 2005 also has support for synonyms that might be of some help.
Here is a chunk of SQL that you could use to retrieve the definition of stored procedures that match a certain search criteria. You could simply change it to do a search and replace like Martin had suggested.
Simply change '%TABLE_NAME%' to your search criteria or the table name you would like to change.
SELECT ROUTINE_NAME, ROUTINE_DEFINITION
FROM INFORMATION_SCHEMA.ROUTINES
WHERE ROUTINE_DEFINITION LIKE '%TABLE_NAME%' AND ROUTINE_TYPE='PROCEDURE'
If it's short term (like a testing database), the easier solution may be to make a VIEW for each table that you changed, using the old name. For instance, if you changed the table tests to tests_new you can do:
CREATE VIEW dbo.tests
AS
SELECT * FROM dbo.tests_new
All your procedures will that reference dbo.tests will actually look at the data in dbo.tests_new.
This is a very very bad idea if this will be a permanent/production DB, as it just adds a layer of obfuscation to your structure and will make it a nightmare to maintain.
You can't change the table references in sprocs through the system data dictionary - you will have to get the script that creates the stored procedure and change the table names in the script. If you have the scripts this is a simple search and replace for the most part.
If you don't have the scripts you can get the text of the stored procedure scripts from sys.sql_modules, or retrieve it through SSMS.
Is there a way to identify column-level dependencies within and between databases? I'd like to generate a report of all columns in a database that are unused by anything (views, procs, UDFs).
e.g.
In database 'DB1', there is a table with a column called 'col1'. How do I determine if 'col1' is being used by procs, views or UDFs in either database 'DB1' or a second database 'DB2'?
Thank you
You can try using:
SELECT OBJECT_NAME(m.object_id), m.*
FROM SYS.SQL_MODULES m
WHERE m.definition LIKE N'%my_column_name%'
SYSCOMMENTS and INFORMATION_SCHEMA.routines have nvarchar(4000) columns. So if "my_column_name" is used at position 3998, it won't be found. SYSCOMMENTS does have multiple lines, but ROUTINES truncates.
But that won't be any help for SELECT * situations, because the column name won't be in text.
If it's only col1, db1, and db2, you can script out the objects in db1 and db2 and search for references to col1.