[mayur-dhavan]

Fixes #431.

Problem

intcomma formats floats with f"{value:,}", which uses repr-style general formatting and switches to an exponent at 1e16 (and below 1e-4), so no commas are added:

>>> humanize.intcomma(1e16)
'1e+16'            # before
'10,000,000,000,000,000.0'   # after
>>> humanize.intcomma(0.00001)
'1e-05'            # before
'0.00001'          # after

Fix

When the grouped result contains an exponent marker, reformat through Decimal(repr(value)), which never uses scientific notation. Integral floats keep their trailing .0, matching the existing output for values below 1e16 (e.g. intcomma(1e15) stays '1,000,000,000,000,000.0'). The ndigits path and string inputs are unchanged.

Tests

Seven new parametrized cases in test_intcomma (1e16, 1.5e20, -1e16, 1e16 with ndigits=0, 0.00001, 1.23e-5, 1234.5e15). tests/test_number.py passes locally (248 passed) and the full suite passes except the pre-existing test_benchmarks.py collection errors, which require the optional pytest-benchmark plugin and fail identically on unmodified master.

AI disclosure: this contribution was prepared with AI assistance (an automated coding agent acting on behalf of the author); the change and tests were validated locally against the project's test suite as described above.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota.

[mayur-dhavan]

Closing this as a duplicate of #432 (Oct 2), which fixes #431 with the same approach (formatting exponent-form floats through Decimal(str(value)) so they expand to grouped digits, preserving the trailing .0). I missed it when checking for existing work. Sorry for the noise.