Code Monkey home page Code Monkey logo

Comments (8)

comphead avatar comphead commented on September 7, 2024 2

This is a good point @jonahgao so this gives me even more confidence to introduce a new DF param which controls should DF go with IEEE 754

from arrow-datafusion.

parthchandra avatar parthchandra commented on September 7, 2024 1
ERROR:  division by zero

I wasn't doubting it. :) But I am surprised.

@comphead I'll take back my 'NULL is not the right thing to return here' comment. It's better than failing for most users.
Just one thing to keep in mind - Spark sql has a isnan function which should be consistent with this.

from arrow-datafusion.

parthchandra avatar parthchandra commented on September 7, 2024

Surprising that Postgres fails. According to the docs, Infinity, -Infinity, and NaN are legitimate floating point values. This is also the expected behavior per the ANSI SQL standard, afaik.
NULL is not the right thing to return here.

from arrow-datafusion.

alamb avatar alamb commented on September 7, 2024

Postgres fails:

andrewlamb@Andrews-MacBook-Pro-2:~/Software/arrow-rs$ psql -h localhost -U postgres
psql (14.12 (Homebrew), server 16.1 (Debian 16.1-1.pgdg120+1))
WARNING: psql major version 14, server major version 16.
         Some psql features might not work.
Type "help" for help.

postgres=# select 1 % -0.0;
ERROR:  division by zero

from arrow-datafusion.

alamb avatar alamb commented on September 7, 2024

I don't have a strong preference one way or the other, personally

from arrow-datafusion.

comphead avatar comphead commented on September 7, 2024

I don't have a strong preference one way or the other, personally

I would go with Spark/DuckDB way :) if there is no other strong opinions, I'll make a fix soon

from arrow-datafusion.

comphead avatar comphead commented on September 7, 2024

Thanks @parthchandra for the suggestion, I just checked we should be fine.

scala> spark.sql("select 1 % -0.0").show(false)
+---------+
|(1 % 0.0)|
+---------+
|NULL     |
+---------+

scala> spark.sql("select isnan(1 % -0.0)").show(false)
+----------------+
|isnan((1 % 0.0))|
+----------------+
|false           |
+----------------+

from arrow-datafusion.

jonahgao avatar jonahgao commented on September 7, 2024

arrow-rs follows the rules of IEEE 754. If we intend to be compatible with other engines, many cases will also need to be modified, such as division by zero.

DataFusion CLI v39.0.0
> select 1.0/0.0;
+-------------------------+
| Float64(1) / Float64(0) |
+-------------------------+
| inf                     |
+-------------------------+


> select 0.0/0.0;
+-------------------------+
| Float64(0) / Float64(0) |
+-------------------------+
| NaN                     |
+-------------------------+

In PostgreSQL:

postgres=# select 1.0/0.0;
ERROR:  division by zero
postgres=# select 0.0/0.0;
ERROR:  division by zero

from arrow-datafusion.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.