Skip to content

Round, Rounded, Correct, Distance(x, y, z, w) and operator< addition to Vector 4#146

Merged
mjcarroll merged 5 commits intogazebosim:ign-math6from
luccosta:round-min-max-abs
Aug 7, 2020
Merged

Round, Rounded, Correct, Distance(x, y, z, w) and operator< addition to Vector 4#146
mjcarroll merged 5 commits intogazebosim:ign-math6from
luccosta:round-min-max-abs

Conversation

@luccosta
Copy link
Copy Markdown
Contributor

@luccosta luccosta commented Jul 31, 2020

Resolves #71 .

This implements the functions Round(), Rounded(), Correct(), Distance(T x, T y, T z, T w) and operator overload operator<(Vector4) in Vector4 API.

It enables:

vec4.Round();
Vector4 roundedVector = vec4.Rounded();
vec4.Correct();
vec4.Distance(1, 2, 3, 4);
if (vec4_a < vec4_b) {
  ...
}

@luccosta
Round, Rounded, Correct, Distance(x, y, z, w) and operator< addition …
439f05e 
…to Vector4

Signed-off-by: Lucas Fernando <lucas.costa@ee.ufcg.edu.br>
@luccosta luccosta requested a review from scpeters as a code owner July 31, 2020 23:01
@luccosta
Copy link
Copy Markdown
Contributor Author

luccosta commented Jul 31, 2020 •
edited
Loading

I tried to test the Correct function generating an NaN first with 0/0 but this generate an Exception and them with math::sqrt(-1) that works fine, but generate the following message in code checker:

(warning) Passing value -1 to sqrt() leads to implementation-defined result.

Should I change/remove the test?

@chapulina
Copy link
Copy Markdown
Contributor

chapulina commented Aug 1, 2020

test the Correct function generating an NaN

While working on #139 I noticed that the code checker would complain about 0.0 / 0.0 but not 1.0 / 0.0, maybe this is an option?

Also another option may be to use std::nan().

@luccosta
Copy link
Copy Markdown
Contributor Author

luccosta commented Aug 1, 2020

but not 1.0 / 0.0, maybe this is an option?

Is a good option, I will check the result.

Also another option may be to use std::nan().

Looks like a more consistent test, I will let with this.

@luccosta
Correct test
b37bed6 
Signed-off-by: Lucas Fernando <lucas.costa@ee.ufcg.edu.br>
@luccosta
Copy link
Copy Markdown
Contributor Author

luccosta commented Aug 1, 2020

I will check the result.

With 1/0 I get Floating-point exception running the test as with 0/0.

@mjcarroll
Copy link
Copy Markdown

mjcarroll commented Aug 3, 2020

I don't think that you'll be able to get around the divide-by-zero issue either triggering the code check, compiler warning, or runtime error (since it's a signaling nan).

I agree that using std::nan here is the best approach, since that generates a quiet nan and won't trigger any floating point exceptions.

@luccosta
Copy link
Copy Markdown
Contributor Author

luccosta commented Aug 3, 2020

I agree that using std::nan here is the best approach

Great, the current PR is using just std::nan.

@mjcarroll mjcarroll merged commit 1a7ae9f into gazebosim:ign-math6 Aug 7, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@mjcarroll mjcarroll mjcarroll approved these changes

@scpeters scpeters Awaiting requested review from scpeters scpeters is a code owner

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@luccosta @chapulina @mjcarroll