"""
EV Camper Dashboard - FastAPI Backend
"""
import csv
import json
import os
from pathlib import Path
from typing import Optional
from fastapi import FastAPI, Query
from fastapi.middleware.cors import CORSMiddleware
from fastapi.staticfiles import StaticFiles
from fastapi.responses import FileResponse, JSONResponse

app = FastAPI(title="EV Camper Dashboard API", version="1.0")

app.add_middleware(
    CORSMiddleware,
    allow_origins=["*"],
    allow_methods=["*"],
    allow_headers=["*"],
)

DATA_DIR = Path("output")
CONFIG_PATH = Path("data.json")
POWER_RESOLUTIONS = {"1SEC", "1MIN", "15MIN", "1H", "1DAY"}
WATER_RESOLUTIONS = {"1MIN", "15MIN", "1H", "1DAY"}

# ── Helpers ──────────────────────────────────────────────────────────────

def read_csv(path: Path, limit: int = None, offset: int = 0) -> list[dict]:
    if not path.exists():
        return []
    with open(path) as f:
        reader = csv.DictReader(f)
        rows = []
        for i, row in enumerate(reader):
            if i < offset:
                continue
            parsed = {}
            for k, v in row.items():
                if k == "Time":
                    parsed[k] = v
                    continue
                try:
                    parsed[k] = float(v)
                except (ValueError, TypeError):
                    parsed[k] = v
            rows.append(parsed)
            if limit and len(rows) >= limit:
                break
    return rows


def load_config() -> dict:
    with open(CONFIG_PATH) as f:
        return json.load(f)


def get_csv_headers(path: Path) -> list[str]:
    """Return column names from first line of CSV."""
    if not path.exists():
        return []
    with open(path) as f:
        return next(csv.reader(f), [])


def infer_type(col: str) -> str:
    """Infer JSON schema type from column name."""
    if col == "Time":
        return "string (ISO8601)"
    if "Pct" in col or "Level" in col or "Ah" in col or "Voltage" in col or "Flow" in col or "Total" in col or "kWh" in col or "Lpm" in col or "_L" in col:
        return "number"
    return "string"


# ── Schema & Data Tables APIs ───────────────────────────────────────────

def _build_schema_and_mermaid() -> tuple[dict, str]:
    """Build schema dict and Mermaid ER diagram from actual CSV files."""
    tables = {}
    # Power: sample from 15MIN (flow) and 1H (totals) to cover both column sets
    for res in POWER_RESOLUTIONS:
        path = DATA_DIR / "power" / f"{res}.csv"
        if path.exists():
            headers = get_csv_headers(path)
            if headers and (res not in tables or len(headers) > len(tables.get(res, {}).get("columns", []))):
                tables[f"power_{res}"] = {
                    "stream": "power",
                    "resolution": res,
                    "columns": [{"name": h, "type": infer_type(h)} for h in headers],
                }
    # Water
    for res in WATER_RESOLUTIONS:
        path = DATA_DIR / "water" / f"{res}.csv"
        if path.exists():
            headers = get_csv_headers(path)
            if headers:
                tables[f"water_{res}"] = {
                    "stream": "water",
                    "resolution": res,
                    "columns": [{"name": h, "type": infer_type(h)} for h in headers],
                }

    # Mermaid ER diagram: one entity per logical table (power_series, water_series)
    power_cols = set()
    water_cols = set()
    for key, meta in tables.items():
        for c in meta["columns"]:
            if meta["stream"] == "power":
                power_cols.add((c["name"], c["type"]))
            else:
                water_cols.add((c["name"], c["type"]))
    power_cols = sorted(power_cols, key=lambda x: (0 if x[0] == "Time" else 1, x[0]))
    water_cols = sorted(water_cols, key=lambda x: (0 if x[0] == "Time" else 1, x[0]))

    def mermaid_type(t: str) -> str:
        if "string" in t or "ISO" in t:
            return "string"
        return "decimal"
    def mermaid_row(name: str, t: str) -> str:
        attr = name.replace(" ", "_")
        typ = mermaid_type(t)
        return f"    {typ} {attr}"
    lines = [
        "erDiagram",
        "    power_timeseries {",
        *[mermaid_row(c[0], c[1]) for c in power_cols],
        "    }",
        "    water_timeseries {",
        *[mermaid_row(c[0], c[1]) for c in water_cols],
        "    }",
        "    power_timeseries ||--o| water_timeseries : aligned_by_Time",
    ]
    mermaid = "\n".join(lines)
    return {"tables": tables}, mermaid


@app.get("/api/schema")
def get_schema():
    """Return CSV data schema (tables, columns, types) and Mermaid ER diagram."""
    schema_dict, mermaid = _build_schema_and_mermaid()
    return {
        **schema_dict,
        "mermaid": mermaid,
    }


@app.get("/api/data/tables")
def get_data_tables_list():
    """Return list of available data tables (stream, resolution, columns, row_count)."""
    result = []
    for res in POWER_RESOLUTIONS:
        path = DATA_DIR / "power" / f"{res}.csv"
        if path.exists():
            headers = get_csv_headers(path)
            with open(path) as f:
                row_count = sum(1 for _ in f) - 1  # exclude header
            result.append({
                "id": f"power_{res}",
                "stream": "power",
                "resolution": res,
                "columns": headers,
                "row_count": max(0, row_count),
            })
    for res in WATER_RESOLUTIONS:
        path = DATA_DIR / "water" / f"{res}.csv"
        if path.exists():
            headers = get_csv_headers(path)
            with open(path) as f:
                row_count = sum(1 for _ in f) - 1
            result.append({
                "id": f"water_{res}",
                "stream": "water",
                "resolution": res,
                "columns": headers,
                "row_count": max(0, row_count),
            })
    return {"tables": result}


# ── Config & Specs ───────────────────────────────────────────────────────

@app.get("/api/config")
def get_config():
    """Return trip configuration and trailer specs."""
    cfg = load_config()
    return {
        "inputs": cfg["inputs"],
        "trailer_specs": cfg["trailer_specs"],
    }


@app.get("/api/components")
def get_components():
    """Return all trailer components by category."""
    cfg = load_config()
    comps = cfg["lookups"]["components"]
    schema = comps["schema"]
    result = {}
    for category in ["living", "cooking", "toilet", "shower", "laundry", "actuation", "dumping", "hvac"]:
        if category in comps:
            result[category] = [
                dict(zip(schema, item)) for item in comps[category]
            ]
    return result


# ── Budget Endpoints ─────────────────────────────────────────────────────

@app.get("/api/power/budget")
def get_power_budget():
    """Daily power budget breakdown in kWh."""
    cfg = load_config()
    from generate_mock_data import calc_power_budget
    params = cfg["inputs"]["params"]
    return calc_power_budget(
        cfg,
        params["user_type"]["value"],
        params["num_people"]["value"],
        params["temperature"]["value"],
        params["hvac_runtime_hrs"]["value"],
    )


@app.get("/api/water/budget")
def get_water_budget():
    """Daily water budget breakdown in litres."""
    cfg = load_config()
    from generate_mock_data import calc_water_budget
    params = cfg["inputs"]["params"]
    return calc_water_budget(
        cfg,
        params["user_type"]["value"],
        params["num_people"]["value"],
    )


# ── Time Series Endpoints ────────────────────────────────────────────────


# ── Hourly Profile (MUST be before /{resolution} routes) ────────────────

@app.get("/api/power/hourly-profile")
def get_hourly_profile():
    """Average hourly power profile across all trip days."""
    hourly = read_csv(DATA_DIR / "power" / "1H.csv")
    circuits = ["HVAC", "Lighting", "Devices", "Fridge", "WaterPump", "Cooking", "Inverter"]

    from collections import defaultdict
    from datetime import datetime as dt
    hour_buckets = defaultdict(lambda: defaultdict(list))

    for row in hourly:
        ts = dt.fromisoformat(row["Time"].replace("Z", ""))
        h = ts.hour
        hour_buckets[h]["solar"].append(row.get("Solar_Total_kWh", 0))
        for c in circuits:
            hour_buckets[h][c].append(row.get(f"{c}_Total_kWh", 0))
        hour_buckets[h]["battery_pct"].append(row.get("Battery_Level_Pct", 0))

    profile = []
    for h in range(24):
        entry = {"hour": h}
        if h in hour_buckets:
            b = hour_buckets[h]
            entry["solar_kwh"] = round(sum(b["solar"]) / len(b["solar"]), 4)
            entry["battery_pct"] = round(sum(b["battery_pct"]) / len(b["battery_pct"]), 1)
            for c in circuits:
                entry[f"{c}_kwh"] = round(sum(b[c]) / len(b[c]), 4)
            entry["total_load_kwh"] = round(sum(entry.get(f"{c}_kwh", 0) for c in circuits), 4)
        profile.append(entry)
    return profile


@app.get("/api/water/hourly-profile")
def get_water_hourly_profile():
    """Average hourly water usage profile."""
    hourly = read_csv(DATA_DIR / "water" / "1H.csv")
    circuits = ["Shower", "Kitchen", "Toilet"]

    from collections import defaultdict
    from datetime import datetime as dt
    hour_buckets = defaultdict(lambda: defaultdict(list))

    for row in hourly:
        ts = dt.fromisoformat(row["Time"].replace("Z", ""))
        h = ts.hour
        for c in circuits:
            hour_buckets[h][c].append(row.get(f"{c}_Total_L", 0))
        hour_buckets[h]["fresh_pct"].append(
            row.get("FreshTank_Level_Pct", row.get("FreshTank_Level_L", 0) / 378.5 * 100)
        )

    profile = []
    for h in range(24):
        entry = {"hour": h}
        if h in hour_buckets:
            b = hour_buckets[h]
            for c in circuits:
                entry[f"{c}_L"] = round(sum(b[c]) / len(b[c]), 3)
            entry["total_L"] = round(sum(entry.get(f"{c}_L", 0) for c in circuits), 3)
            entry["fresh_pct"] = round(sum(b["fresh_pct"]) / len(b["fresh_pct"]), 1)
        profile.append(entry)
    return profile


@app.get("/api/power/peaks")
def get_power_peaks():
    """Identify peak power usage hours across the trip."""
    hourly = read_csv(DATA_DIR / "power" / "1H.csv")
    circuits = ["HVAC", "Lighting", "Devices", "Fridge", "WaterPump", "Cooking", "Inverter"]

    peaks = []
    for row in hourly:
        total = sum(row.get(f"{c}_Total_kWh", 0) for c in circuits)
        peaks.append({
            "time": row["Time"],
            "total_kwh": round(total, 4),
            "solar_kwh": row.get("Solar_Total_kWh", 0),
            "battery_pct": row.get("Battery_Level_Pct", 0),
        })

    peaks.sort(key=lambda x: x["total_kwh"], reverse=True)
    return {
        "top_10_peak_hours": peaks[:10],
        "bottom_10_hours": sorted(peaks, key=lambda x: x["total_kwh"])[:10],
    }


@app.get("/api/power/{resolution}")
def get_power_data(
    resolution: str,
    limit: Optional[int] = Query(None, ge=1, le=50000),
    offset: int = Query(0, ge=0),
    day: Optional[int] = Query(None, ge=1, le=30, description="Filter by day number"),
):
    """Power time-series data at given resolution."""
    res = resolution.upper()
    if res not in POWER_RESOLUTIONS:
        return JSONResponse({"error": f"Invalid resolution. Use: {POWER_RESOLUTIONS}"}, 400)

    path = DATA_DIR / "power" / f"{res}.csv"

    # When day filter is active, read all rows first, then filter
    if day is not None:
        rows = read_csv(path)
        if rows:
            from datetime import datetime
            start_date = datetime.fromisoformat(rows[0]["Time"].replace("Z", "")).date()
            rows = [r for r in rows
                    if (datetime.fromisoformat(r["Time"].replace("Z", "")).date() - start_date).days + 1 == day]
        if limit:
            rows = rows[offset:offset + limit]
    else:
        rows = read_csv(path, limit=limit, offset=offset)

    return {"resolution": res, "count": len(rows), "data": rows}


@app.get("/api/water/{resolution}")
def get_water_data(
    resolution: str,
    limit: Optional[int] = Query(None, ge=1, le=50000),
    offset: int = Query(0, ge=0),
    day: Optional[int] = Query(None, ge=1, le=30),
):
    """Water time-series data at given resolution."""
    res = resolution.upper()
    if res not in WATER_RESOLUTIONS:
        return JSONResponse({"error": f"Invalid resolution. Use: {WATER_RESOLUTIONS}"}, 400)

    path = DATA_DIR / "water" / f"{res}.csv"

    if day is not None:
        rows = read_csv(path)
        if rows:
            from datetime import datetime
            start_date = datetime.fromisoformat(rows[0]["Time"].replace("Z", "")).date()
            rows = [r for r in rows
                    if (datetime.fromisoformat(r["Time"].replace("Z", "")).date() - start_date).days + 1 == day]
        if limit:
            rows = rows[offset:offset + limit]
    else:
        rows = read_csv(path, limit=limit, offset=offset)

    return {"resolution": res, "count": len(rows), "data": rows}


# ── Summary / Aggregated Stats ───────────────────────────────────────────

@app.get("/api/summary")
def get_summary():
    """Trip-wide summary statistics."""
    daily_power = read_csv(DATA_DIR / "power" / "1DAY.csv")
    daily_water = read_csv(DATA_DIR / "water" / "1DAY.csv")
    hourly_power = read_csv(DATA_DIR / "power" / "1H.csv")

    if not daily_power:
        return {"error": "No data"}

    circuits = ["HVAC", "Lighting", "Devices", "Fridge", "WaterPump", "Cooking", "Inverter", "Unmetered"]

    total_solar = sum(d.get("Solar_Total_kWh", 0) for d in daily_power)
    total_shore = sum(d.get("Shore_Total_kWh", 0) for d in daily_power)
    total_consumption = sum(
        sum(d.get(f"{c}_Total_kWh", 0) for c in circuits)
        for d in daily_power
    )

    # Per-circuit totals
    circuit_totals = {}
    for c in circuits:
        circuit_totals[c] = round(sum(d.get(f"{c}_Total_kWh", 0) for d in daily_power), 3)

    # Battery stats
    min_battery = min(d.get("Battery_Level_Pct", 100) for d in hourly_power) if hourly_power else 0
    max_battery = max(d.get("Battery_Level_Pct", 0) for d in hourly_power) if hourly_power else 100
    avg_battery = sum(d.get("Battery_Level_Pct", 0) for d in hourly_power) / len(hourly_power) if hourly_power else 0

    # Water totals
    total_fresh_used = 0
    if daily_water:
        total_fresh_used = sum(d.get("Pump_Total_L", 0) for d in daily_water)

    # Final tank levels
    last_water = daily_water[-1] if daily_water else {}
    last_power = hourly_power[-1] if hourly_power else {}

    # Daily breakdown
    daily_breakdown = []
    for i, dp in enumerate(daily_power):
        dw = daily_water[i] if i < len(daily_water) else {}
        daily_breakdown.append({
            "day": i + 1,
            "solar_kwh": round(dp.get("Solar_Total_kWh", 0), 2),
            "consumption_kwh": round(sum(dp.get(f"{c}_Total_kWh", 0) for c in circuits), 2),
            "shore_kwh": round(dp.get("Shore_Total_kWh", 0), 2),
            "battery_end_pct": dp.get("Battery_Level_Pct", 0),
            "fresh_used_L": round(dw.get("Pump_Total_L", 0), 1),
            "fresh_remaining_pct": round(dw.get("FreshTank_Level_Pct", 0), 1),
        })

    return {
        "trip_days": len(daily_power),
        "total_solar_kwh": round(total_solar, 2),
        "total_shore_kwh": round(total_shore, 2),
        "total_consumption_kwh": round(total_consumption, 2),
        "self_sufficiency_pct": round(total_solar / max(total_consumption, 0.01) * 100, 1),
        "circuit_totals_kwh": circuit_totals,
        "battery": {
            "min_pct": round(min_battery, 1),
            "max_pct": round(max_battery, 1),
            "avg_pct": round(avg_battery, 1),
            "current_pct": last_power.get("Battery_Level_Pct", 0),
        },
        "water": {
            "total_fresh_used_L": round(total_fresh_used, 1),
            "fresh_remaining_L": last_water.get("FreshTank_Level_L", 0),
            "fresh_remaining_pct": last_water.get("FreshTank_Level_Pct", 0),
            "grey_level_L": last_water.get("GreyTank_Level_L", 0),
            "grey_level_pct": last_water.get("GreyTank_Level_Pct", 0),
            "black_level_L": last_water.get("BlackTank_Level_L", 0),
            "black_level_pct": last_water.get("BlackTank_Level_Pct", 0),
        },
        "daily_breakdown": daily_breakdown,
    }


# ── Serve Frontend ───────────────────────────────────────────────────────

# ── Data Regeneration ────────────────────────────────────────────────────

from pydantic import BaseModel

class GenerateRequest(BaseModel):
    user_type: str = "Typical"
    num_people: int = 2
    trip_duration_days: int = 5
    temperature: str = "Hot"
    sunlight: str = "Hi- Sunny"
    humidity: str = "Comfortable"
    seed: int = 42

@app.post("/api/generate")
def regenerate_data(req: GenerateRequest):
    """Regenerate mock data with new parameters."""
    import subprocess, sys

    # Update config in memory for budget endpoints
    cfg = load_config()
    cfg["inputs"]["params"]["user_type"]["value"] = req.user_type
    cfg["inputs"]["params"]["num_people"]["value"] = req.num_people
    cfg["inputs"]["params"]["trip_duration_days"]["value"] = req.trip_duration_days
    cfg["inputs"]["params"]["temperature"]["value"] = req.temperature
    cfg["inputs"]["params"]["sunlight"]["value"] = req.sunlight
    cfg["inputs"]["params"]["humidity"]["value"] = req.humidity

    # Write updated config
    with open(CONFIG_PATH, "w") as f:
        json.dump(cfg, f, indent=2)

    # Run the generator
    result = subprocess.run(
        [sys.executable, "generate_mock_data.py",
         "--config", str(CONFIG_PATH),
         "--out", str(DATA_DIR),
         "--seed", str(req.seed),
         "--user", req.user_type,
         "--people", str(req.num_people),
         "--days", str(req.trip_duration_days),
         "--temp", req.temperature],
        capture_output=True, text=True, timeout=120,
    )

    if result.returncode != 0:
        return JSONResponse({"error": result.stderr}, 500)

    return {"status": "ok", "message": f"Generated {req.trip_duration_days}-day trip for {req.user_type} profile"}


# ── Serve Frontend (catch-all, must be LAST) ─────────────────────────────

STATIC_DIR = Path("static")

@app.get("/")
def serve_index():
    return FileResponse(STATIC_DIR / "index.html")

@app.get("/{path:path}")
def serve_static(path: str):
    file_path = STATIC_DIR / path
    if file_path.exists() and file_path.is_file():
        return FileResponse(file_path)
    return FileResponse(STATIC_DIR / "index.html")


if __name__ == "__main__":
    import uvicorn
    uvicorn.run(app, host="0.0.0.0", port=8000)