Levi Siuzdak (sile)
Levi Siuzdak (sile)
@tde-banana-odoo of course, added one, even helped me find a bug in my bugfix, so good call
Script used to find `1795630070.0` ```python from random import random, randint from decimal import Decimal to_invert = [10 ** -i for i in range(1, 30)] expected = [[float(Decimal(str(10 ** i))...
Made some small adjustment to satisfy the new style checker, and opened a PR in enterprise to adjust tests there: https://github.com/odoo/enterprise/pull/60903 > I will bring more eyes to this, as...
> I'll see if I can find any other places that relied on this quirk in order to skip rounding intermediate values. Okay, so I found a couple of examples,...
# Updated tests The ones with the `...0000000003` stripped are self-explanatory, but here I'll list other changes. ## base test_rounding_03 https://github.com/odoo/odoo/blob/b0980a9bbe65d2c237c4f095e1ad900995d14000/odoo/addons/base/tests/test_float.py#L110-L111 The nearest 3 decimal even to ±357.4555 should be...
> By the way, do you know that your example numbers actually don't exist? I don't see how that's relevant. Fact is that in the IEEE 754 standard, there's only...
Removed the [FIX] commits now that they've been backported & merged into master: - https://github.com/odoo/odoo/pull/163786 - https://github.com/odoo/odoo/pull/164673 - https://github.com/odoo/odoo/pull/162977 - https://github.com/odoo/odoo/pull/163154 - https://github.com/odoo/odoo/pull/164665 - https://github.com/odoo/odoo/pull/163182
>That is factually incorrect: no float is strictly equal to the decimal number 99.99. I thought it was obvious I was referring to the `==` operator when I said "strictly...
> because their is testcase failing in upgrade due to these changes in 17.4. Can you be more specific about the nature of the failing test? This PR shouldn't have...