jp-range 0.3.0

Utilities for parsing Japanese numeric ranges.

jp-range

jp-range is a small library for parsing Japanese numeric range expressions. It returns an Interval object implemented with Pydantic or None when parsing fails.

Installation

pip install jp-range
# or install from GitHub
pip install git+https://github.com/first-automation/jp-range.git

Python 3.12 or later is required.

Features

• Normalizes full width numbers and common punctuation
• Supports inclusive and exclusive bounds (以上, 未満, etc.)
• Parses connectors such as 〜, - and から
• Handles single-sided bounds, approximate expressions (90前後) and A±d notation
• Integrates with pandas via apply_parse for Series and DataFrame

Usage

Basic parsing

from jp_range import parse

interval = parse("40以上50未満")
print(interval)               # [40, 50)
print(interval.contains(45))  # True

Pandas integration

from pandas import Series
from jp_range import apply_parse

# Convert Series of textual ranges to pandas.Interval
s = Series(["20～30", "50超", "未満100"])
result = apply_parse(s)
# result is a Series of pandas.Interval objects

# Expand columns into numeric min/max
df = apply_parse(s.to_frame(name="range"), split_numeric=True)
# df has columns "range_max" and "range_min"

Supported expressions

• "20から30" – inclusive 20 to 30
• "20〜30" – inclusive 20 to 30 using a tilde connector
• "30以上40以下" – inclusive 30 to 40
• "40以上50未満" – 40 to under 50
• "70超90以下" – greater than 70 and up to 90
• "50より上" – greater than 50
• "60より下" – less than 60
• "90前後" – roughly around 90 (5% margin)
• "(2,3]" – standard interval notation

parse_jp_range returns None if the expression cannot be parsed.