processor.py

"""
processor.py — main CSV log processing pipeline
Module D: Files + Error Handling

Usage:
    python processor.py data/access.log.csv
    python processor.py data/access.log.csv --out-dir results/
"""

import argparse
import csv
import logging
from pathlib import Path

from parser import ParseError, parse_row, validate_row
from reporter import compute_stats, write_error_log, write_summary

# ── Logging setup ─────────────────────────────────────────────────────────────
logging.basicConfig(
    level=logging.INFO,
    format="%(levelname)s  %(message)s",
)
log = logging.getLogger(__name__)


# ── Pipeline ──────────────────────────────────────────────────────────────────

def read_log(path: Path) -> list[dict]:
    """
    Read a CSV log file and return raw rows as dicts.

    Args:
        path: Path to the CSV file.

    Returns:
        List of raw row dicts.

    Raises:
        FileNotFoundError: If the file doesn't exist.
        ValueError: If the file is empty or has no headers.
    """
    if not path.exists():
        raise FileNotFoundError(f"Log file not found: {path}")

    rows = []
    with path.open("r", encoding="utf-8-sig") as f:
        reader = csv.DictReader(f)
        if reader.fieldnames is None:
            raise ValueError(f"CSV file is empty or missing headers: {path}")
        for row in reader:
            rows.append(row)

    if not rows:
        raise ValueError(f"CSV file has headers but no data rows: {path}")

    return rows


def process(log_path: Path, out_dir: Path) -> None:
    """
    Full pipeline: read → parse → validate → report.

    Args:
        log_path: Path to the input CSV.
        out_dir:  Directory for output files.
    """
    out_dir.mkdir(parents=True, exist_ok=True)

    # ── Read ──────────────────────────────────────────────────────────────────
    log.info(f"Reading {log_path} ...")
    raw_rows = read_log(log_path)
    log.info(f"  {len(raw_rows)} rows found.")

    # ── Parse + validate ──────────────────────────────────────────────────────
    entries = []
    errors  = []

    for raw in raw_rows:
        try:
            parsed = parse_row(raw)
            validate_row(parsed)
            entries.append(parsed)
        except ParseError as e:
            errors.append({"row": e.row, "reason": e.reason})
            log.warning(f"  Skipping malformed row: {e.reason}")

    log.info(f"  {len(entries)} rows parsed OK, {len(errors)} errors.")

    # ── Stats + reports ───────────────────────────────────────────────────────
    stats        = compute_stats(entries)
    summary_path = out_dir / "summary.txt"
    errors_path  = out_dir / "parse_errors.txt"

    write_summary(stats, len(entries), len(errors), summary_path)
    log.info(f"  Summary written → {summary_path}")

    if errors:
        write_error_log(errors, errors_path)
        log.info(f"  Error log written → {errors_path}")

    # ── Print summary to terminal ─────────────────────────────────────────────
    print()
    print(summary_path.read_text(encoding="utf-8"))


# ── CLI ───────────────────────────────────────────────────────────────────────

def main() -> None:
    parser = argparse.ArgumentParser(
        description="Parse a server access log CSV and generate a summary report.",
        formatter_class=argparse.RawDescriptionHelpFormatter,
        epilog="""
Examples:
  python processor.py data/access.log.csv
  python processor.py data/access.log.csv --out-dir results/
        """,
    )
    parser.add_argument("log_file", type=Path, help="Path to the input CSV log file.")
    parser.add_argument(
        "--out-dir", type=Path, default=Path("output"),
        help="Directory for output files (default: output/)",
    )
    args = parser.parse_args()

    try:
        process(args.log_file, args.out_dir)
    except (FileNotFoundError, ValueError) as e:
        log.error(str(e))
        raise SystemExit(1)


if __name__ == "__main__":
    main()