"""Extract sample.pdf's ruled table using Python 3 and pdfplumber.

Run: python3 extract.py
Requires pdfplumber (validated with 0.11.9). Paths are relative to this script.
"""

import csv
from pathlib import Path

import pdfplumber


def main():
    root = Path(__file__).resolve().parent
    with pdfplumber.open(root / "sample.pdf") as pdf:
        tables = [table for page in pdf.pages for table in page.extract_tables()]
    if len(tables) != 1:
        raise ValueError(f"Expected one ruled table, found {len(tables)}")
    rows = tables[0]
    if not rows or any(len(row) != 4 or any(cell is None for cell in row) for row in rows):
        raise ValueError("Expected a complete four-column table")
    # Keep cells as text to retain displayed precision and column order.
    # csv.writer quotes product names containing commas automatically.
    with (root / "table.csv").open("w", encoding="utf-8", newline="") as output:
        csv.writer(output).writerows(rows)


if __name__ == "__main__":
    main()