When using rounding numbers that are negative and round to 0 show a negative sign

[jrgleason]

We ran into this while rounding to one decimal

http://plnkr.co/edit/cUqjvRuBnSfcrAC7Visz?p=preview

Basically if I were to have a value like -.01 and I would use number:1 then it shows -0.0 which would seem to be invalid.

[jrgleason]

Now that I am looking at it seems there is a problem with Math.round and fractions in general with JS.

[jrgleason]

I changed my code to this and it seems to be working maybe we can use that in the code?

[petebacondarwin]

Do you think you could provide a pull request for this?

[smilli]

I created a patch for this, but it seems like this is actually expected behavior? Or at least according to this test

[jrgleason]

Huh that makes little sense to me
On Aug 23, 2014 1:24 AM, "Smitha Milli" [email protected] wrote:

I created a patch for this, but it seems like this is actually expected
behavior? Or at least according to this test

angular.js/test/ng/filter/filtersSpec.js

Line 179 in c94329a

expect(number(-1e-50, 0)).toEqual('-0');

—
Reply to this email directly or view it on GitHub
#8489 (comment).

[jrgleason]

I mean I get it is supposed to signify the original number was negative but
I thought when rounding it became that number and there is no negative 0.
Unless I am missing some complicated math rule
On Aug 23, 2014 9:24 AM, [email protected] wrote:

Huh that makes little sense to me
On Aug 23, 2014 1:24 AM, "Smitha Milli" [email protected] wrote:

I created a patch for this, but it seems like this is actually expected
behavior? Or at least according to this test

angular.js/test/ng/filter/filtersSpec.js

Line 179 in c94329a

expect(number(-1e-50, 0)).toEqual('-0');

—
Reply to this email directly or view it on GitHub
#8489 (comment)
.

[caitp]

in regular math, 0 has no sign --- but with IEEE 754 math, 0 does have a sign because that's just how the format works. But yeah we generally don't want to format it that way =P It should be a pretty small patch to write, why not have a go at it? -0 === +0 should evaluate to true, so you could instead say if (number === 0) number = 0; or something similar.

[jrgleason]

Will do give me a bit to get tests and all setup. I will work on it tonight
On Aug 23, 2014 10:18 AM, "Caitlin Potter" [email protected] wrote:

in regular math, 0 has no sign --- but with IEEE 754 math, 0 does have a
sign because that's just how the format works. But yeah we generally don't
want to format it that way =P It should be a pretty small patch to write,
why not have a go at it? -0 === +0 should evaluate to true, so you could
instead say if (number === 0) number = 0; or something similar.

—
Reply to this email directly or view it on GitHub
#8489 (comment).

[smilli]

Actually, I already created a fix earlier, just updated that test case from above in #8745.