Skip to content

Keep three digits when metric rounds into quetta #424

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
SashaMIT wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Keep three digits when metric rounds into quetta #424

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 5 additions & 1 deletion src/humanize/number.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -573,6 +573,7 @@ def metric(value: float, unit: str = "", precision: int = 3) -> str:

if not math.isfinite(value):
return _format_not_finite(value)
original = value
exponent = math.floor(math.log10(abs(value))) if value != 0 else 0

if exponent >= 33 or exponent < -30:
Expand All @@ -582,8 +583,11 @@ def metric(value: float, unit: str = "", precision: int = 3) -> str:
value /= 10**old_bucket
digits = int(max(0, precision - exponent % 3 - 1))
# Absorb a rounding carry (9.999 -> 10.0) by bumping the exponent.
if exponent < 30 and round(abs(value), digits) >= 10 ** (exponent % 3 + 1):
# Quetta still has a prefix for exponents 30 and 31. Exponent 33 does not.
if round(abs(value), digits) >= 10 ** (exponent % 3 + 1):
exponent += 1
if exponent >= 33:
return scientific(original, precision - 1) + unit

@ub3dqy ub3dqy Oct 30, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

metric() already accepts precision=0 (including cases in the existing test table), but this new branch passes -1 to scientific():

humanize.metric(9.996e32, precision=0)
# ValueError: Format specifier missing precision

The base commit returns 1000 Q for this input. Negative values and calls with a unit are affected too. Clamping the decimal precision here preserves zero precision when switching to scientific notation:

Suggested change
return scientific(original, precision - 1) + unit
return scientific(original, max(0, precision - 1)) + unit

I checked positive/negative values, unit/no unit, and precisions 0, 1, 2, 3 and 6: all 30 additional cases pass with this guard, alongside the 248 existing number tests. Four of those cases raise on the current PR head.

Dima Mak / ub3dqy

new_bucket = exponent // 3 * 3
value /= 10 ** (new_bucket - old_bucket)
digits = int(max(0, precision - exponent % 3 - 1))
Expand Down
7 changes: 7 additions & 0 deletions tests/test_number.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -332,6 +332,13 @@ def test_clamp(test_args: list[typing.Any], expected: str) -> None:
([1.235e29, "A"], "124 RA"),
([2.56e-30, "V"], "2.56 qV"),
([2.596e32, "F"], "260 QF"),
([9.996e30], "10.0 Q"),
([9.996e31], "100 Q"),
([9.996e32], "1.00 x 10³³"),
([-9.996e32], "-1.00 x 10³³"),
([-9.996e30], "-10.0 Q"),
([9.996e30, "V"], "10.0 QV"),
([9.996e32, "F"], "1.00 x 10³³F"),
([1e50], "1.00 x 10⁵⁰"),
([1e-50], "1.00 x 10⁻⁵⁰"),
([-1500, "V"], "-1.50 kV"),
Expand Down