#!/usr/bin/env python3
"""
Normalize a schema document that contains multiple CREATE TABLE blocks but may
miss semicolons between statements.

Usage:
  python fix_schema_sql.py input.sql > fixed.sql
  python fix_schema_sql.py input.sql output.sql
"""

from __future__ import annotations

import re
import sys
from pathlib import Path


CREATE_TABLE_RE = re.compile(r"(^|\n)(\s*CREATE\s+TABLE\b)", re.IGNORECASE)


def normalize_schema(text: str) -> str:
    text = text.replace("\r\n", "\n")

    pieces = []
    last = 0
    first = True

    for match in CREATE_TABLE_RE.finditer(text):
        start = match.start(2)
        if first:
            first = False
            last = start
            continue

        chunk = text[last:start]
        stripped = chunk.rstrip()
        if stripped and not stripped.endswith(";"):
            chunk = stripped + ";\n\n"
        pieces.append(chunk)
        last = start

    tail = text[last:]
    pieces.append(tail)
    return "".join(pieces)


def main() -> int:
    if len(sys.argv) not in (2, 3):
        print("usage: fix_schema_sql.py <input.sql> [output.sql]", file=sys.stderr)
        return 2

    src = Path(sys.argv[1])
    fixed = normalize_schema(src.read_text(encoding="utf-8"))

    if len(sys.argv) == 3:
        Path(sys.argv[2]).write_text(fixed, encoding="utf-8")
    else:
        sys.stdout.write(fixed)
    return 0


if __name__ == "__main__":
    raise SystemExit(main())