Support superscript decimal point and fraction slash in exponents (#2250)

[apoorvdarshan]

Closes #2250.

Problem

Unit strings using a superscript decimal point (⋅, U+22C5) or a superscript fraction slash (⸍, U+2E0D) inside an exponent raise UndefinedUnitError, even though ordinary superscript exponents already work:

>>> ureg.Unit("gr²")        # grain ** 2        (works)
>>> ureg.Unit("gr⁰⋅³³³")    # UndefinedUnitError: '⋅' is not defined
>>> ureg.Unit("gr¹⸍³")      # UndefinedUnitError: '⸍' is not defined

Fix

As suggested by @andrewgsavage in the issue, this extends the existing superscript handling in string_preprocessor (pint/util.py):

• _pretty_table now maps ⋅ → . (decimal point) and ⸍ → / (fraction slash).
• _pretty_exp_re now accepts ⋅ as a decimal separator and ⸍ as a fraction separator within a superscript exponent.

Because the exponent is already wrapped as **(...), the fraction stays correctly parenthesized:

>>> ureg.Unit("gr⁰⋅³³³")       # grain ** 0.333
>>> ureg.Unit("gr¹⸍³")         # grain ** 0.3333333333333333   (== 1/3)
>>> ureg.Unit("ft²·gr¹⸍³/lb")  # foot ** 2 * grain ** (1/3) / pound

· (U+00B7, the multiplication interpunct) is a distinct character from ⋅ (U+22C5) and is unaffected — kg·m/s² still parses as kilogram * meter / second ** 2.

Testing

• Added TestStringProcessor::test_superscript covering the existing forms (gr², gr⁻³), the interpunct-is-multiplication case (kg·m), and the two new separators. It fails on main and passes with this change.
• Full test suite (excluding the benchmark dir, which needs pytest-benchmark): 1636 passed, no regressions.
• ruff check / ruff format --check clean; CHANGES updated.

---

Disclosure: this change was prepared with the assistance of an AI tool (Claude Code). I reproduced the issue, implemented and verified the fix and tests, ran the suite and linters, and take responsibility for the contribution and will respond to review feedback personally.

[codspeed-hq]

Merging this PR will improve performance by 23.22%

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 2 improved benchmarks
✅ 446 untouched benchmarks
⏩ 448 skipped benchmarks¹

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	WallTime	test_load_definitions_stage_1[True]	184.1 µs	146.1 µs	+26.01%
⚡	WallTime	test_import	323.9 ms	268.8 ms	+20.49%

Tip

Curious why this is faster? Comment @codspeedbot explain why this is faster on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing apoorvdarshan:superscript-decimal-fraction-exponents-2250 (61643f5) with master (5d9c534)}

Footnotes

1. 448 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩