from __future__ import annotations

import json
from datetime import date, timedelta

import httpx
import pandas as pd
from loguru import logger

from ..config import API_BASE_URL

# Keep chunks small to avoid truncated responses from the API.
_MAX_DAYS_PER_REQUEST = 90

_NDJSON_HEADERS = {"Accept": "application/x-ndjson"}


def _build_url(start: date, end: date, base: str) -> str:
    return f"{API_BASE_URL}/rates?from={start.isoformat()}&to={end.isoformat()}&base={base}"


def _parse_ndjson(text: str) -> list[dict]:
    """Parse NDJSON (one JSON object per line) into our column schema."""
    rows: list[dict] = []
    for line in text.splitlines():
        line = line.strip()
        if not line:
            continue
        row = json.loads(line)
        rows.append(
            {
                "rate_date": row["date"],
                "base_currency": row["base"],
                "quote_currency": row["quote"],
                "rate": row["rate"],
            }
        )
    return rows


def fetch_rates(start_date: date, end_date: date, base: str = "USD") -> pd.DataFrame:
    """Fetch daily spot rates from Frankfurter API for a date range.

    Large ranges are automatically chunked into 90-day requests.
    Uses NDJSON streaming to avoid JSON truncation on large responses.
    Returns a DataFrame with columns: rate_date, base_currency, quote_currency, rate.
    """
    all_rows: list[dict] = []
    chunk_start = start_date

    with httpx.Client(timeout=120.0) as client:
        while chunk_start <= end_date:
            chunk_end = min(chunk_start + timedelta(days=_MAX_DAYS_PER_REQUEST), end_date)
            url = _build_url(chunk_start, chunk_end, base)

            logger.info(f"Fetching rates {chunk_start} -> {chunk_end}")
            response = client.get(url, headers=_NDJSON_HEADERS)
            response.raise_for_status()

            rows = _parse_ndjson(response.text)
            all_rows.extend(rows)
            logger.debug(f"  Got {len(rows)} rate rows")

            chunk_start = chunk_end + timedelta(days=1)

    df = pd.DataFrame(all_rows)
    if not df.empty:
        df["rate_date"] = pd.to_datetime(df["rate_date"]).dt.date
    logger.info(f"Total: {len(df)} rate rows fetched")
    return df