#!/usr/bin/env bash
set -euo pipefail

SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
# shellcheck source=00_env.sh
. "$SCRIPT_DIR/00_env.sh"

SEARCH_PATH="/"
TOP_N=20

usage() {
  printf 'Usage: %s [--path <path>] [--top <N>]\n' "$(basename "$0")"
}

while [[ "$#" -gt 0 ]]; do
  case "$1" in
    --path) SEARCH_PATH="${2:-}"; shift 2 ;;
    --top) TOP_N="${2:-}"; shift 2 ;;
    -h|--help) usage; exit 0 ;;
    *) critical "Unknown argument: $1"; usage; exit 2 ;;
  esac
done

if ! [[ "$TOP_N" =~ ^[0-9]+$ ]] || (( TOP_N < 1 )); then
  critical "--top must be a positive integer"
  exit 2
fi

validate_path "$SEARCH_PATH" || exit 2
require_cmd find || exit 1
require_cmd sort || exit 1
require_cmd head || exit 1

section "Largest Files Under $SEARCH_PATH"
warning "Read-only scan. Permission errors can be normal without root access."

find "$SEARCH_PATH" -xdev \
  \( -path /proc -o -path /sys -o -path /dev -o -path /run \) -prune -o \
  -type f -printf '%s\t%p\n' 2>/dev/null |
  sort -rn |
  head -n "$TOP_N" |
  awk '
    function human(bytes) {
      split("B KB MB GB TB PB", unit)
      size = bytes
      idx = 1
      while (size >= 1024 && idx < 6) {
        size = size / 1024
        idx++
      }
      return sprintf("%.1f%s", size, unit[idx])
    }
    {
      size = $1
      $1 = ""
      sub(/^\t/, "")
      printf "%10s  %s\n", human(size), $0
    }
  ' | tee -a "$LOG_FILE"

ok "No files were modified."