#!/usr/bin/env bash

set -euo pipefail

ROOT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
INPUT_FILE="${1:-$ROOT_DIR/public_ipv4_from_master.txt}"
STAMP="$(date +%Y%m%d-%H%M%S)"
OUTPUT_DIR="${2:-$ROOT_DIR/reports/inventory-$STAMP}"
TARGETS_FILE="$OUTPUT_DIR/targets.txt"
RAW_BASENAME="$OUTPUT_DIR/nmap_inventory"
SUMMARY_CSV="$OUTPUT_DIR/summary.csv"
SUMMARY_MD="$OUTPUT_DIR/summary.md"

mkdir -p "$OUTPUT_DIR"

if ! command -v nmap >/dev/null 2>&1; then
  echo "nmap is not installed. Install it first, then rerun this script." >&2
  echo "Expected input: $INPUT_FILE" >&2
  echo "Planned output directory: $OUTPUT_DIR" >&2
  exit 1
fi

if [[ ! -f "$INPUT_FILE" ]]; then
  echo "Input file not found: $INPUT_FILE" >&2
  exit 1
fi

python3 - "$INPUT_FILE" "$TARGETS_FILE" <<'PY'
import ipaddress
import sys

src, dst = sys.argv[1], sys.argv[2]
valid = []
seen = set()

with open(src, "r", encoding="utf-8") as fh:
    for raw in fh:
        line = raw.strip()
        if not line or line.startswith("#"):
            continue
        try:
            ip = ipaddress.IPv4Address(line)
        except ipaddress.AddressValueError:
            print(f"Skipping invalid IPv4: {line}", file=sys.stderr)
            continue
        text = str(ip)
        if text not in seen:
            seen.add(text)
            valid.append(text)

with open(dst, "w", encoding="utf-8") as fh:
    for ip in valid:
        fh.write(f"{ip}\n")

print(len(valid))
PY

TARGET_COUNT="$(wc -l < "$TARGETS_FILE" | tr -d ' ')"

if [[ "$TARGET_COUNT" -eq 0 ]]; then
  echo "No valid IPv4 targets found in $INPUT_FILE" >&2
  exit 1
fi

echo "Validated $TARGET_COUNT targets"
echo "Running a conservative external inventory scan"

nmap \
  -Pn \
  -n \
  -T3 \
  --top-ports 20 \
  -sV \
  --version-light \
  --open \
  --max-retries 2 \
  --host-timeout 2m \
  -iL "$TARGETS_FILE" \
  -oA "$RAW_BASENAME"

python3 - "$RAW_BASENAME.gnmap" "$SUMMARY_CSV" "$SUMMARY_MD" "$TARGET_COUNT" <<'PY'
import csv
import sys
from collections import defaultdict

gnmap_path, csv_path, md_path, total_targets = sys.argv[1:5]
rows = []
ports_by_ip = defaultdict(list)
seen_hosts = set()

with open(gnmap_path, "r", encoding="utf-8", errors="replace") as fh:
    for raw in fh:
        line = raw.strip()
        if not line.startswith("Host: "):
            continue
        if "Ports: " not in line:
            continue
        host = line.split()[1]
        seen_hosts.add(host)
        ports_blob = line.split("Ports: ", 1)[1]
        for item in ports_blob.split(", "):
            parts = item.split("/")
            if len(parts) < 7:
                continue
            port, state, proto, _, service, product, extra = parts[:7]
            if state != "open":
                continue
            product_info = " ".join(x for x in (product, extra) if x).strip()
            rows.append(
                {
                    "ip": host,
                    "port": port,
                    "protocol": proto,
                    "state": state,
                    "service": service or "unknown",
                    "product": product_info or "-",
                }
            )
            ports_by_ip[host].append(f"{port}/{proto} {service or 'unknown'}".strip())

with open(csv_path, "w", newline="", encoding="utf-8") as fh:
    writer = csv.DictWriter(
        fh,
        fieldnames=["ip", "port", "protocol", "state", "service", "product"],
    )
    writer.writeheader()
    writer.writerows(rows)

sorted_hosts = sorted(ports_by_ip.items(), key=lambda item: (-len(item[1]), item[0]))

with open(md_path, "w", encoding="utf-8") as fh:
    fh.write("# External Inventory Summary\n\n")
    fh.write(f"- Total validated targets: {total_targets}\n")
    fh.write(f"- Hosts with at least one open top port: {len(sorted_hosts)}\n")
    fh.write(f"- CSV details: `{csv_path}`\n")
    fh.write(f"- Raw Nmap files: `{gnmap_path[:-6]}` (`.nmap`, `.gnmap`, `.xml`)\n\n")
    fh.write("## Prioritized review queue\n\n")
    fh.write("| IP | Open ports found | Services |\n")
    fh.write("| --- | ---: | --- |\n")
    for ip, entries in sorted_hosts:
        fh.write(f"| {ip} | {len(entries)} | {', '.join(entries)} |\n")

print(f"Wrote {len(rows)} rows to {csv_path}")
print(f"Wrote Markdown summary to {md_path}")
PY

echo
echo "Inventory complete"
echo "Output directory: $OUTPUT_DIR"
echo "Summary CSV:      $SUMMARY_CSV"
echo "Summary report:   $SUMMARY_MD"