float and int unsupported binary operator

[kocoten1992]

Describe the problem

Here is a simple reproduction

# postgres
select 8::float / 8::int;
 ?column? 
----------
        1
(1 row)

# cockroachdb
select 8::float / 8::int;
ERROR: unsupported binary operator: <float> / <int>
SQLSTATE: 22023

(I'm deleting all old content for brevity because it all boil down to the reproduction above, you can check back via github history log)

Jira issue: CRDB-16414

[blathers-crl]

Hello, I am Blathers. I am here to help you get the issue triaged.

Hoot - a bug! Though bugs are the bane of my existence, rest assured the wretched thing will get the best of care here.

I was unable to automatically find someone to ping.

If we have not gotten back to your issue within a few business days, you can try the following:

• Join our community slack channel and ask on #cockroachdb.
• Try find someone from here if you know they worked closely on the area and CC them.

_{🦉 Hoot! I am a Blathers, a bot for CockroachDB. My owner is otan.}

[kocoten1992]

Okay, so I'm missing one final cast float at:

WITH RECURSIVE per_investor_amount AS (
  SELECT 0::int AS investors_number, 0.00::float AS investment_amount, 0.00::float AS individual_amount
  UNION
  SELECT (investors_number + 1)::int, i.investment_amount, i.investment_amount / (investors_number + 1)::float /* HERE */
  FROM investment i, per_investor_amount pia WHERE investors_number < 3
)
SELECT * FROM per_investor_amount ORDER BY investment_amount, investors_number;

# cockroachdb
  investors_number | investment_amount |   individual_amount
-------------------+-------------------+-------------------------
                 0 |                 0 |                      0
                 1 | 5.322146e+06      | 5.322146e+06
                 2 | 5.322146e+06      | 2.661073e+06
                 3 | 5.322146e+06      | 1.7740486666666667e+06
                 1 | 5.6129486e+06     | 5.6129486e+06
                 2 | 5.6129486e+06     | 2.8064743e+06
                 3 | 5.6129486e+06     | 1.8709828666666665e+06
                 1 | 9.705321e+06      | 9.705321e+06
                 2 | 9.705321e+06      | 4.8526605e+06
                 3 | 9.705321e+06      | 3.235107e+06
(10 rows)

Still not sure if this is a bug (given postgres work?).

[kocoten1992]

Hi @yuzefovich, I see you triage this, I've an update in reproduction (also updated the first post):

# postgres
select 8::float / 8::int;
 ?column? 
----------
        1
(1 row)

# cockroachdb
select 8::float / 8::int;
ERROR: unsupported binary operator: <float> / <int>
SQLSTATE: 22023

[rafiss]

I think #38965 and #75101 would solve this, but in the meantime we should be able to explicitly add float/int overloads.

[surahman]

Hi @rafiss, could you please provide some context on where to look in the code base to get started with adding the explicit overloads? I have been out of code base for a while.