I have two functions- one to set a variable and one to view it. The idea is to use the variable in a View.
I'm really new to functions. Anyhoo, here's what they're supposed to do:
GetSiteId:
BEGIN
#Routine body goes here...
RETURN #siteContext;
END
SetSiteId:
BEGIN
#Routine body goes here...
set #siteContext := SiteId;
RETURN #siteContext;
END
This has an int parameter of SiteId.
But when I try to execute them, I get an error.
I execute them like so:
SetSiteId(1);
select GetSiteId();
If I perform a select on SetSiteId it works fine, but I don't want to see it, I only want to see the result of GetSiteId, which should be 1.
I thought maybe there was a specific keyword to execute the function, but I couldn't find any and Google was no help. I suspect I might be going about this all wrong, but I can't see how- it seems like it should be the simplest thing in the world to execute a function.
The oh-so-unhelpful error I get is this:
You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near 'SetSiteId(1)' at line 1
There are three basic kinds of stored code in MySQL:
Stored functions -- what you have
Stored procedures
Events (never mind these for now).
A function returns a value. Therefore it needs to be called from a context that can accept that value. Therefore, you can use
SET #something := YOUR_FUNCTION(param, param);
or
SELECT YOUR_FUNCTION(param, param)
A stored procedure returns a resultset (or, alternatively, no resultset). It's called with
CALL YOUR_PROCEDURE(param, param)
So, you can call your first stored function like this:
SET #junk := SetSiteId(1);
Or you can call it with SELECT and just ignore the result set.
Consider the following example function:
CREATE FUNCTION test()
RETURNS INTEGER
LANGUAGE SQL
AS $$ SELECT * FROM UNNEST(ARRAY[1,2,3,4,5]); $$
;
When we execute it like so:
SELECT test();
the result we get back is 1. In other words, the function just returned the first element of the result set.
I have a convenience function that retrieves an ID based on data in the rest of the row. It lets callers query on data that is often unique to a particular row but is not strictly unique. Primarily, it will be used to populate some mostly static data. Allowing the call to use the row data instead of hard coding the ID makes it more readable and easier to maintain when new data does come along. This depends on the caller having some familiarity with the data, but that's a reasonable assumption to make for the usage at hand.
The problem is if the user makes a mistake and gives arguments that are insufficient to filter the query's results to one row, I want my function to error out instead of returning the first result. How can I accomplish this? Do I have any options other than to switch my language to PL/PGSQL and check it manually?
(Using PostgreSQL 9.3).
You could put the entire query as a subquery in the SELECT list. When attempting to return more than one row, this will fail with this error:
ERROR: more than one row returned by a subquery used as an expression
Based on your example, this one would fail:
CREATE FUNCTION test()
RETURNS INTEGER
LANGUAGE SQL
AS $$ SELECT (SELECT * FROM UNNEST(ARRAY[1,2,3,4,5])); $$
and this one wouldn't fail:
CREATE FUNCTION test()
RETURNS INTEGER
LANGUAGE SQL
AS $$ SELECT (SELECT * FROM UNNEST(ARRAY[1])); $$
Situation: Having a SQL procedure which "returns" result via "SELECT x" statements. For some reasons it is not allowed to change it to a function or changing that procedure in any way. How can I obtain the result like:
set #result = 0;
#result = call SomeProcedure(#p1, #p2);
But since it is a procedure not a function above code won't compile/work. How can I achieve that in MySQL. In C++ it works but in MySQL I found no way ...
It is not possible.
Result sets returned from select ... will always be returned to the caller of the first procedure, even if you make several levels of sub calls.
Functions return a value (but not a result set) that you can use inside other procedures or functions.
Your only option is to either set session variables or to store the result in a temporary table that the calling procedure knows about.
I want to pass a table name as a parameter in a Postgres function. I tried this code:
CREATE OR REPLACE FUNCTION some_f(param character varying) RETURNS integer
AS $$
BEGIN
IF EXISTS (select * from quote_ident($1) where quote_ident($1).id=1) THEN
return 1;
END IF;
return 0;
END;
$$ LANGUAGE plpgsql;
select some_f('table_name');
And I got this:
ERROR: syntax error at or near "."
LINE 4: ...elect * from quote_ident($1) where quote_ident($1).id=1)...
^
********** Error **********
ERROR: syntax error at or near "."
And here is the error I got when changed to this select * from quote_ident($1) tab where tab.id=1:
ERROR: column tab.id does not exist
LINE 1: ...T EXISTS (select * from quote_ident($1) tab where tab.id...
Probably, quote_ident($1) works, because without the where quote_ident($1).id=1 part I get 1, which means something is selected. Why may the first quote_ident($1) work and the second one not at the same time? And how could this be solved?
Before you go there: for only few, known tables names, it's typically simpler to avoid dynamic SQL and spell out the few code variants in separate functions or in a CASE construct.
That said, what you are trying to achieve can be simplified and improved:
CREATE OR REPLACE FUNCTION some_f(_tbl regclass, OUT result integer)
LANGUAGE plpgsql AS
$func$
BEGIN
EXECUTE format('SELECT (EXISTS (SELECT FROM %s WHERE id = 1))::int', _tbl)
INTO result;
END
$func$;
Call with schema-qualified name (see below):
SELECT some_f('myschema.mytable'); -- would fail with quote_ident()
Or:
SELECT some_f('"my very uncommon table name"');
Major points
Use an OUT parameter to simplify the function. You can directly select the result of the dynamic SQL into it and be done. No need for additional variables and code.
EXISTS does exactly what you want. You get true if the row exists or false otherwise. There are various ways to do this, EXISTS is typically most efficient.
You seem to want an integer back, so I cast the boolean result from EXISTS to integer, which yields exactly what you had. I would return boolean instead.
I use the object identifier type regclass as input type for _tbl. That does everything quote_ident(_tbl) or format('%I', _tbl) would do, but better, because:
.. it prevents SQL injection just as well.
.. it fails immediately and more gracefully if the table name is invalid / does not exist / is invisible to the current user. (A regclass parameter is only applicable for existing tables.)
.. it works with schema-qualified table names, where a plain quote_ident(_tbl) or format(%I) would fail because they cannot resolve the ambiguity. You would have to pass and escape schema and table names separately.
It only works for existing tables, obviously.
I still use format(), because it simplifies the syntax (and to demonstrate how it's used), but with %s instead of %I. Typically, queries are more complex so format() helps more. For the simple example we could as well just concatenate:
EXECUTE 'SELECT (EXISTS (SELECT FROM ' || _tbl || ' WHERE id = 1))::int'
No need to table-qualify the id column while there is only a single table in the FROM list. No ambiguity possible in this example. (Dynamic) SQL commands inside EXECUTE have a separate scope, function variables or parameters are not visible there - as opposed to plain SQL commands in the function body.
Here's why you always escape user input for dynamic SQL properly:
db<>fiddle here demonstrating SQL injection
Old sqlfiddle
If at all possible, don't do this.
That's the answer—it's an anti-pattern. If the client knows the table it wants data from, then SELECT FROM ThatTable. If a database is designed in a way that this is required, it seems to be designed sub-optimally. If a data access layer needs to know whether a value exists in a table, it is easy to compose SQL in that code, and pushing this code into the database is not good.
To me this seems like installing a device inside an elevator where one can type in the number of the desired floor. After the Go button is pressed, it moves a mechanical hand over to the correct button for the desired floor and presses it. This introduces many potential issues.
Please note: there is no intention of mockery, here. My silly elevator example was *the very best device I could imagine* for succinctly pointing out issues with this technique. It adds a useless layer of indirection, moving table name choice from a caller space (using a robust and well-understood DSL, SQL) into a hybrid using obscure/bizarre server-side SQL code.
Such responsibility-splitting through movement of query construction logic into dynamic SQL makes the code harder to understand. It violates a standard and reliable convention (how a SQL query chooses what to select) in the name of custom code fraught with potential for error.
Here are detailed points on some of the potential problems with this approach:
Dynamic SQL offers the possibility of SQL injection that is hard to recognize in the front end code or the back end code alone (one must inspect them together to see this).
Stored procedures and functions can access resources that the SP/function owner has rights to but the caller doesn't. As far as I understand, without special care, then by default when you use code that produces dynamic SQL and runs it, the database executes the dynamic SQL under the rights of the caller. This means you either won't be able to use privileged objects at all, or you have to open them up to all clients, increasing the surface area of potential attack to privileged data. Setting the SP/function at creation time to always run as a particular user (in SQL Server, EXECUTE AS) may solve that problem, but makes things more complicated. This exacerbates the risk of SQL injection mentioned in the previous point, by making the dynamic SQL a very enticing attack vector.
When a developer must understand what the application code is doing in order to modify it or fix a bug, he'll find it very difficult to get the exact SQL query being executed. SQL profiler can be used, but this takes special privileges and can have negative performance effects on production systems. The executed query can be logged by the SP but this increases complexity for questionable benefit (requiring accommodating new tables, purging old data, etc.) and is quite non-obvious. In fact, some applications are architected such that the developer does not have database credentials, so it becomes almost impossible for him to actually see the query being submitted.
When an error occurs, such as when you try to select a table that doesn't exist, you'll get a message along the lines of "invalid object name" from the database. That will happen exactly the same whether you're composing the SQL in the back end or the database, but the difference is, some poor developer who's trying to troubleshoot the system has to spelunk one level deeper into yet another cave below the one where the problem exists, to dig into the wonder-procedure that Does It All to try to figure out what the problem is. Logs won't show "Error in GetWidget", it will show "Error in OneProcedureToRuleThemAllRunner". This abstraction will generally make a system worse.
An example in pseudo-C# of switching table names based on a parameter:
string sql = $"SELECT * FROM {EscapeSqlIdentifier(tableName)};"
results = connection.Execute(sql);
While this does not eliminate every possible issue imaginable, the flaws I outlined with the other technique are absent from this example.
Inside plpgsql code, The EXECUTE statement must be used for queries in which table names or columns come from variables. Also the IF EXISTS (<query>) construct is not allowed when query is dynamically generated.
Here's your function with both problems fixed:
CREATE OR REPLACE FUNCTION some_f(param character varying) RETURNS integer
AS $$
DECLARE
v int;
BEGIN
EXECUTE 'select 1 FROM ' || quote_ident(param) || ' WHERE '
|| quote_ident(param) || '.id = 1' INTO v;
IF v THEN return 1; ELSE return 0; END IF;
END;
$$ LANGUAGE plpgsql;
I know this is an old thread, but I ran across it recently when trying to solve the same problem - in my case, for some fairly complex scripts.
Turning the entire script into dynamic SQL is not ideal. It's tedious and error-prone work, and you lose the ability to parameterize: parameters must be interpolated into constants in the SQL, with bad consequences for performance and security.
Here's a simple trick that lets you keep the SQL intact if you only need to select from your table - use dynamic SQL to create a temporary view:
CREATE OR REPLACE FUNCTION some_f(_tbl varchar) returns integer
AS $$
BEGIN
drop view if exists myview;
execute format('create temporary view myview as select * from %s', _tbl);
-- now you can reference myview in the SQL
IF EXISTS (select * from myview where myview.id=1) THEN
return 1;
END IF;
return 0;
END;
$$ language plpgsql;
The first doesn't actually "work" in the sense that you mean, it works only in so far as it does not generate an error.
Try SELECT * FROM quote_ident('table_that_does_not_exist');, and you will see why your function returns 1: the select is returning a table with one column (named quote_ident) with one row (the variable $1 or in this particular case table_that_does_not_exist).
What you want to do will require dynamic SQL, which is actually the place that the quote_* functions are meant to be used.
If the question was to test if the table is empty or not (id=1), here is a simplified version of Erwin's stored proc :
CREATE OR REPLACE FUNCTION isEmpty(tableName text, OUT zeroIfEmpty integer) AS
$func$
BEGIN
EXECUTE format('SELECT COALESCE ((SELECT 1 FROM %s LIMIT 1),0)', tableName)
INTO zeroIfEmpty;
END
$func$ LANGUAGE plpgsql;
If you want table name, column name and value to be dynamically passed to function as parameter
use this code
create or replace function total_rows(tbl_name text, column_name text, value int)
returns integer as $total$
declare
total integer;
begin
EXECUTE format('select count(*) from %s WHERE %s = %s', tbl_name, column_name, value) INTO total;
return total;
end;
$total$ language plpgsql;
postgres=# select total_rows('tbl_name','column_name',2); --2 is the value
I have 9.4 version of PostgreSQL and I always use this code:
CREATE FUNCTION add_new_table(text) RETURNS void AS
$BODY$
begin
execute
'CREATE TABLE ' || $1 || '(
item_1 type,
item_2 type
)';
end;
$BODY$
LANGUAGE plpgsql
And then:
SELECT add_new_table('my_table_name');
It works good for me.
Attention! Above example is one of those which shows "How do not if we want to keep safety during querying the database" :P
Here's my configuration:
I have a re-runnable batch script that I use to update my database.
Inside of that batch script, I have code that says the following:
If Table 'A' doesn't exist, then create Table 'A' and insert rows into it.
Later on in that batch script, I create an schemabound indexed view on that table.
And if you didn't already know, indexed views require specific client settings.
Sometimes, when I re-run the script, that is after the table has been created, SQL Server Management Studio evaluates the "insert rows" code, which is protected by the 'If this table doesn't exist' code, and yields the following error:
Msg 1934, Level 16, State 1, Line 15
INSERT failed because the following SET options have incorrect settings: 'CONCAT_NULL_YIELDS_NULL, ANSI_WARNINGS, ANSI_PADDING, ARITHABORT'. Verify that SET options are correct for use with indexed views and/or indexes on computed columns and/or filtered indexes and/or query notifications and/or XML data type methods and/or spatial index operations.
Please note: If someone were to try this INSERT statement in a vacuum, I would fully expect SSMS to generate this error.
But not when it's protected by a conditional block.
My Question:
Does the SSMS compiler evaluate all expressions, regardless of whether they will actually be executed?
Yes, it evaluates all of them,take a look at this
declare #i int
select #i =1
if #i = 1
begin
declare #i2 int
set #i2 = 5
end
else
begin
declare #i2 int
set #i2 = 5
end
Msg 134, Level 15, State 1, Line 12
The variable name '#i2' has already been declared. Variable names must be unique within a query batch or stored procedure.
Another example with temp tables is here: What is deferred name resolution and why do you need to care?
your only way out would be to wrap it inside dynamic SQL
Note that most of the settings you mention are connection-level, i.e. in case you set/change them they stay in effect unless you close the connection or explicitly change their value.
Returning to your question. The error you mention looks like runtime error, i.e. the INSERT is actually being executed. It would be better if you could show your script (omitting details, but keeping batches).
Edit: it is not SSMS compiler that evaluates SQL you try to execute - it is SQL Server. What do you meant by 'evaluate'? Is it 'execute'? When you run a batch (which is what actually is being executed by a server), SQL Server first does syntactic analysis and throws error in case it finds any syntactic error, nothing is being executed at this point of time. In case syntax is ok, the server starts executing you batch.
Again, the error you show seems to be runtime - so I guess you'd carefully watch for the conditions and track what happens (or provide us more details about 'sometimes').