Background

CPython rejects format-spec widths that exceed Py_ssize_t::MAX with ValueError: Too many decimal digits in format string. RustPython's FormatSpec::_parse only capped precision (via parse_precision); width digits were accepted up to usize::MAX, so values like sys.maxsize + 1 silently produced an effectively-ignored width.

Repro

import sys
fs = '{}f'.format(sys.maxsize + 1)  # '9223372036854775808f'

format(2.34, fs)
# CPython 3.14:    ValueError: Too many decimal digits in format string
# RustPython:      '2.340000'        (silently dropped huge width, before this PR)

Fix

After parse_number returns the width in FormatSpec::_parse, reject any value above i32::MAX with FormatSpecError::DecimalDigitsTooMany. This matches the cap precision already uses and produces the byte-identical CPython "Too many decimal digits in format string" ValueError.

Tests unmasked

• test_str.StrTest.test_format_huge_width

Verification

• CPython 3.14.4 byte-identical wording for the huge-width case
• Precision path unchanged (test_format_huge_precision still passes with its existing "Precision too big" message)
• Normal widths (e.g. format(2.34, '100.3f')) still pad correctly
• No regressions across 7 modules: test_str, test_format, test_unicode_identifiers, test_fstring, test_int, test_float, test_decimal

Summary by CodeRabbit

Release Notes

• Bug Fixes
  • Improved validation for format specifications to properly reject overly large width values with appropriate error handling during parsing.