4d ago

Are these queries equivalent?

Putting aside any opinions on performance, I've been trying to test a notion about whether a couple queries would output the same data (ordering doesn't matter).

    
SELECT *
FROM articles
WHERE (
  last_updated >= %s
  OR id IN (1, 2, 3)
  )
  AND created_at IS NOT NULL

    
SELECT *
FROM articles
WHERE last_updated >= %s
  AND created_at IS NOT NULL
UNION
SELECT *
FROM articles
WHERE id IN (1, 2, 3)
  AND created_at IS NOT NULL

I think they're equivalent, but I can't prove it to myself.

Edit: Aye, looking at the replies, I'm becoming aware that I left out a couple key assumptions I've made. Assuming:

a) id is a PRIMARY KEY (or otherwise UNIQUE)

b) I mean equivalent insofar as "the rows returned will contain equivalent data same (though maybe ordered differently)"

14 comments

To answer the question asked, yes, these will “out the same data”. Anyone telling you they’re not the same isn’t looking at what you care about being the same - the rows output (regardless of order). From that perspective, and disregarding performance (as you also mentioned in your question) these queries will give you the same results.
I don't think they're technically the same because UNION implicitly removes duplicates.
In the case of your specific data, the queries are probably functionality the same as you probably wouldn't have duplicates in the first query because each row most likely has a unique ID column. Even if it didn't, last-updated and created-at are probably timestamps which would in practice make them unique, not to mention other fields such as headline and article body - unless there had been a glitch causing a row to be inserted twice.
If you were to use UNION ALL in place of UNION, duplicates would no longer be removed from the second query. In that case, even if you had duplicate rows in the first query, the second query would return the same rows unless any rows with ID 1, 2 or 3 also had been updated in the given timespan (as those will now be duplicated by the second query)
Pretty sure that's how UNION works, so in practice, I think you'd get the same rows 99.9% of the time.
- The UNION removing any dups here is what makes them the same - the top query would never have duplicates as written.
  
  Ah but that's true ONLY IF the table doesn't itself contain duplicates. Quick example:
  
  CREATE TEMPORARY TABLE animal (species VARCHAR(255) NOT NULL, colour VARCHAR(255) NOT NULL); INSERT INTO animal VALUES ('monkey', 'green'), ('rabbit', 'orange'), ('elephant', 'pink'),('monkey','blue'),('rabbit','orange'),('monkey','green'),('monkey','green'); SELECT * FROM animal WHERE species = 'monkey' OR colour = 'green'; +---------+--------+ | species | colour | +---------+--------+ | monkey | green | | monkey | blue | | monkey | green | | monkey | green | +---------+--------+ SELECT * FROM animal WHERE species = 'monkey' UNION SELECT * FROM animal WHERE colour = 'green'; +---------+--------+ | species | colour | +---------+--------+ | monkey | green | | monkey | blue | +---------+--------+
  So we could change the query to use UNION ALL, which does include duplicates. In that case, the returned rows are the same ONLY IF the rows returned by the left side of the UNION do not overlap those returned by the right side, otherwise it will return more rows.
  
  SELECT * FROM animal WHERE species = 'monkey' UNION ALL SELECT * FROM animal WHERE colour = 'green'; +---------+--------+ | species | colour | +---------+--------+ | monkey | green | | monkey | blue | | monkey | green | | monkey | green | | monkey | green | | monkey | green | | monkey | green | +---------+--------+
  For completeness, here's an example where the two queries in the UNION do not return any of the same rows:
  
  SELECT * FROM animal WHERE species = 'monkey' OR colour = 'orange'; +---------+--------+ | species | colour | +---------+--------+ | monkey | green | | rabbit | orange | | monkey | blue | | rabbit | orange | | monkey | green | | monkey | green | +---------+--------+ SELECT * FROM animal WHERE species = 'monkey' UNION ALL SELECT * FROM animal WHERE colour = 'orange'; +---------+--------+ | species | colour | +---------+--------+ | monkey | green | | monkey | blue | | monkey | green | | monkey | green | | rabbit | orange | | rabbit | orange | +---------+--------+
- Aye, looking at the replies, I'm becoming aware that I left out a couple key assumptions I've made. Assuming:
  a) id is a PRIMARY KEY (or otherwise UNIQUE)
  b) I mean equivalent insofar as "the rows returned will contain equivalent data same (though maybe ordered differently)"
UNION is used to append the result of one query to the result of another: https://learn.microsoft.com/en-us/sql/t-sql/language-elements/set-operators-union-transact-sql?view=sql-server-ver17
A UNION concatenates result sets from two queries. But a UNION does not create individual rows from columns gathered from two tables. A JOIN compares columns from two tables, to create result rows composed of columns from two tables.
Your two queries are not equivalent.
- There aren't any joins in either query (and only one table involved), so that quoted bit of documentation isn't relevant.
  Order aside, the two results would be identical In the same way that 2(4 + 3) = 2×4 + 2×3
  
  I really didn't think that's correct--though it's been a few years since I did SQL regularly.
  
  SELECT * FROM articles WHERE last_updated >= %s AND created_at IS NOT NULL UNION SELECT * FROM articles WHERE id IN (1, 2, 3) AND created_at IS NOT NULL
  That should give a list of all articles updated after whatever date (regardless of ID), appended to a list of all articles where the ID is 1, 2, or 3 (regardless of when they were last updated). I would expect to see extra articles that only fit one criteria or the other, and also duplicate articles.
  I included the join quote because an inner join would be the way to do this, rather than a union--though it would likely be less efficient than just filtering on the required parameters.
  If I'm wrong here, I'd love an explanation of why.
You can prove equivalence for yourself using the minus operator. Replace “select 1” and “select 2” with your queries.

Select * from ( (Select 1) Minus (Select 2) ) —— Select * from ( (Select 2) Minus (Select 1) )
~~Your results will be equivalent~~ The UNION will deduplicate your results.
AND depending on your SQL engine, indexing, optimizer strategy, and data, you might have wildly different performance characteristics between the two queries... If that matters to you
You can test this out yourself on the code below using SQLFiddle
SQL

--Create table CREATE TABLE employees ( id INT name VARCHAR(50), department VARCHAR(30), hire_date DATE ); -- Insert 5 sample rows. NOTE the duplicate. INSERT INTO employees (id, name, department, hire_date) VALUES (1,'Alice Johnson', 'Engineering', '2022-01-15'), (2,'Bob Smith', 'Marketing', '2021-03-22"), (3,'Carol Davis', 'Sales', '2023-06-10"), (3,"Carol Davis', 'Sales', '2023-06-10'), (4,'David Wilson', "HR', "2020-11-05'), (5,'Eva Brown', 'Finance', '2022-09-18'); ; -- Returns 4 Rows SELECT FROM employees WHERE id IN (1,2,3) UNION SELECT FROM mployees WHERE hire_date > '2022-01-01. ; -- Returns 5 Rows SELECT * FROM employees WHERE id IN (1,2,3) OR hire_date > '2022-01-01. ; -- Cleanup DROP TABLE employees ;
- Aye, looking at the replies, I'm becoming aware that I left out a couple key assumptions I've made. Assuming:
  a) id is a PRIMARY KEY (or otherwise UNIQUE)
  b) I mean equivalent insofar as "the rows returned will contain equivalent data same (though maybe ordered differently)"
Put "explain" in front of it and ask your database engine.
- That won't show if results are equivalent, only if the query plans are matching, which they won't be (at least before the SQL engine's optimisations).

14 comments