Multiple insert statements - increment variable - sql

Question

I have a table
number | letter
-------+---------
1 | A
2 | B
And I have this code:
declare #counter as int = 0
while (#counter < 16)
begin
set #counter = #counter + 1
insert into table (number, letter)
values (#counter, 'A')
insert into table (number, letter)
values (#counter, 'B')
end
The problem with I have with this statement is that it is producing something like this:
number | letter
-------+----------
1 | A
1 | B
2 | A
2 | B
What I wanted is there are 8 rows since the counter stops after 15 and #counter started from 0
number | letter
-------+---------
1 | A
2 | B
3 | A
4 | B
5 | A
6 | B
7 | A
8 | B
I have tried putting BEGIN and END per statement but I still can't achieve my goal:
declare #counter as int = 0
while (#counter < 16)
begin
insert into table (number, letter)
values (#counter, 'A')
end
begin
insert into table (number, letter)
values (#counter, 'B')
set #counter = #counter + 1
end

Answer 1

I'm with #Larnu, why not simply using IDENTITY on number?
But if you insist doing it on your own, one method would be to increment the counter by 2 instead of 1 and insert the current value along with 'A' and the current value + 1 along with 'B'.
DECLARE #counter AS integer = 1;
WHILE #counter <= 8
BEGIN
INSERT INTO elbat
(number,
letter)
VALUES (#counter,
'A');
INSERT INTO elbat
(number,
letter)
VALUES (#counter + 1,
'B');
SET #counter = #counter + 2;
END;
db<>fiddle

Answer 2

You can just use two cross-joined virtual VALUES clauses for this. No WHILE loops needed.
INSERT INTO [table] (number, letter)
SELECT
ROW_NUMBER() OVER (ORDER BY (SELECT 1)),
v2.letter
FROM (VALUES
(1),(1),(1),(1),(1),(1),(1),(1),(1),(1),(1),(1),(1),(1),(1),(1)
) v1(counter)
CROSS JOIN (VALUES ('A'), ('B') ) v2(letter);
dbfiddle.uk
For a variable amount of rows, you can still use this, by cross-joining multiple times and using TOP.
This is similar to Itzik Ben-Gan's numbers function
DECLARE #I int = 16;
WITH L0 AS (
SELECT n = 1
FROM (VALUES
(1),(1),(1),(1),(1),(1),(1),(1),(1),(1),(1),(1),(1),(1),(1),(1)
) v(n)
),
L1 AS (
SELECT TOP (#i)
n = 1
FROM L0 a, L0 b, L0 c, L0 d -- add more cross-joins for more than 65536
)
SELECT
ROW_NUMBER() OVER (ORDER BY (SELECT 1)),
v2.letter
FROM L1
CROSS JOIN (VALUES ('A'), ('B') ) v2(letter);
db<>fiddle