SQL SELECT to get the first N positive integers

Question

I need to get a result set containing the first N positive integers. Is it possible to use only standard SQL SELECT statement to get them (without any count table provided)?

Answer 1

A possible solution (admittedly not very elegant) is to use any table with a sufficiently large number of records.

For the first 10 integers (using the mysql.help_relation, but any table would do), you could use the following query:

SELECT  @N := @N +1 AS integers 
FROM mysql.help_relation , (SELECT @N:=0) dum LIMIT 10;

This could also be placed in a function taking Min and Max.

Answer 2

If you know that N is limited (and usually it is), you can use a construction such as:

select (a.digit + (10 * b.digit) + (100 * c.digit) + (1000 * d.digit) + (10000 * e.digit) + (100000 * f.digit)) as n
    from (select 0 as digit union all select 1 union all select 2 union all select 3 union all select 4 union all select 5 union all select 6 union all select 7 union all select 8 union all select 9) as a
    cross join (select 0 as digit union all select 1 union all select 2 union all select 3 union all select 4 union all select 5 union all select 6 union all select 7 union all select 8 union all select 9) as b
    cross join (select 0 as digit union all select 1 union all select 2 union all select 3 union all select 4 union all select 5 union all select 6 union all select 7 union all select 8 union all select 9) as c
    cross join (select 0 as digit union all select 1 union all select 2 union all select 3 union all select 4 union all select 5 union all select 6 union all select 7 union all select 8 union all select 9) as d
    cross join (select 0 as digit union all select 1 union all select 2 union all select 3 union all select 4 union all select 5 union all select 6 union all select 7 union all select 8 union all select 9) as e
    cross join (select 0 as digit union all select 1 union all select 2 union all select 3 union all select 4 union all select 5 union all select 6 union all select 7 union all select 8 union all select 9) as f;

which will generate the first million numbers. If you need only the positive numbers, simply add + 1 to the expression.

Note that in MySQL in particular, the results may not be sorted. You need to append order by n to the end if you need ordered numbers. This will increase the execution time dramatically, though (on my machine, it jumped up from 5 ms to 500 ms).

For simple queries, here's a query for just the first 10000 numbers:

select (a.digit + (10 * b.digit) + (100 * c.digit) + (1000 * d.digit)) as n
    from (select 0 as digit union all select 1 union all select 2 union all select 3 union all select 4 union all select 5 union all select 6 union all select 7 union all select 8 union all select 9) as a
    cross join (select 0 as digit union all select 1 union all select 2 union all select 3 union all select 4 union all select 5 union all select 6 union all select 7 union all select 8 union all select 9) as b
    cross join (select 0 as digit union all select 1 union all select 2 union all select 3 union all select 4 union all select 5 union all select 6 union all select 7 union all select 8 union all select 9) as c
    cross join (select 0 as digit union all select 1 union all select 2 union all select 3 union all select 4 union all select 5 union all select 6 union all select 7 union all select 8 union all select 9) as d;

This answer is adapted from the following query that returns a date range: https://stackoverflow.com/a/2157776/2948

Answer 3

Seems that what you want is a dummy rowset.

In MySQL, it's impossible without having a table.

Most major systems provide a way to do it:

• In Oracle:

  SELECT  level
  FROM    dual
  CONNECT BY
          level <= 10
  

• In SQL Server:

  WITH    q AS
          (
          SELECT  1 AS num
          UNION ALL
          SELECT  num + 1
          FROM    q
          WHERE   num < 10
          )
  SELECT  *
  FROM    q
  

• In PostgreSQL:

  SELECT  num
  FROM    generate_series(1, 10) num
  

MySQL lacks something like this and this is a serious drawback.

I wrote a simple script to generate test data for the sample tables in my blog posts, maybe it will be of use:

CREATE TABLE filler (
        id INT NOT NULL PRIMARY KEY AUTO_INCREMENT
) ENGINE=Memory;

CREATE PROCEDURE prc_filler(cnt INT)
BEGIN
        DECLARE _cnt INT;
        SET _cnt = 1;
        WHILE _cnt <= cnt DO
                INSERT
                INTO    filler
                SELECT  _cnt;
                SET _cnt = _cnt + 1;
        END WHILE;
END
$$

You call the procedure and the table gets filled with the numbers.

You can reuse it during the duration of the session.

Answer 4

The sequence I propose allows the programmer to execute the following query :

select value from sequence where value>=15 and value<100;

And to obtain the expected results : the sequence of integers between 15 (inclusive) and 100 (exclusive).

If that's what you want, you'll have to create the two following views, views that you'll declare only once :

create view digits as select 0 n union select 1 union select 2 union select 3 union select 4 union select 5 union select 6 union select 7 union select 8 union select 9;

create view sequence as select u.n+t.n*10+h.n*100 as value from digits as u cross join digits as t cross join digits as h;

This way you have your sequence with an intuitive SELECT...

Hope it helps.

Answer 5

If your database supports analytic windowing functions the following is simple works really well:

SELECT  row_number() over (partition by 1 order by 1) numbers
FROM SOME_TABLE
LIMIT 2700;

This statement returns a set of numbers from 1 to 2700.

Answer 6

I' pretty sure that you cannot do it, if I understand your question correctly.

As I understand your question, you want the list, from a single SQL statement, without having to reference a specific table?

I'm pretty sure that it is not possible in any SQL dialect. If you were to get a sequentially incrementing number along with the results of another query, then that would be possible (depending on SQL dialect, on mssql it would be rownumber(), but I don't know how in MySql, but it's probably there)

But that's not what I hear you ask?