LedgerLogic — Optimal Change Calculator (ll_change_maker)

CLI and library code that breaks a US cash amount into the minimum number of bills and coins using the standard greedy method (valid for US denominations). Parsing uses integer cents internally so totals stay exact.

This folder is a standalone slice of the larger LedgerLogic idea: the interactive change maker, shared parsing (CSV dates/amounts), storage (atomic CSV/JSON writes, format_money), and schemas (TypedDict shapes for typing).

Layout

Module	Role
change_maker.py	Parse amounts, greedy breakdown, CLI menu, printing
parsing.py	parse_date, parse_amount for bank/CSV-style strings
storage.py	format_money, categorized CSV + JSON helpers, data dir
schemas.py	TypedDicts for change results and categorized rows
tests/	pytest suite

Setup

cd ll_change_maker
python -m venv .venv
.venv\Scripts\activate
pip install -r requirements.txt

Run the interactive app

From the project directory (so import schemas / import storage resolve):

python change_maker.py

Tests

python -m pytest tests -q

Two parsers (on purpose)

Function	Use
change_maker.parse_amount_to_cents	Interactive change input: e.g. bare 1473 can mean 14.73 (total cents) when there is no $.
parsing.parse_amount	CSV / bank strings: debits in parentheses, grouping spaces; always yields a non‑negative float in dollars, no “integer cents” shortcut.

Both round half-up to cents before use; scientific notation is rejected in both.

Data directory (storage)

If you use storage.get_data_dir() without a base_dir, it checks the environment variable LEDGERLOGIC_DATA_DIR, otherwise creates ./ledgerlogic_data under the current working directory (legacy naming from the parent app).

Features

• Denominations: $100, $50, $20, $10, $5, $1 bills; quarter, dime, nickel, penny
• Edge cases: zero (message, no breakdown), negatives rejected, non-numeric errors, optional note when sub-cent rounding happened
• Verbose mode: step-by-step greedy trace
• Output: bills vs coins, counts, verification line, unused denominations

Code style

Functions and dicts only (no classes in the change flow). schemas types are for mypy/readability; runtime values are plain dicts.