svr/usr/bin/expresso_parser

#!/usr/bin/env python
#!/usr/bin/env python3
import sys
import subprocess
import re
from datetime import datetime

# === AUTO INSTALL ===
def ensure(pkg):
    try:
        __import__(pkg)
    except ImportError:
        subprocess.check_call([sys.executable, "-m", "pip", "install", pkg])

ensure("pandas")
ensure("matplotlib")

import pandas as pd
import matplotlib.pyplot as plt

# === ARGS ===
if len(sys.argv) < 2:
    print("Usage: python script.py input.txt [output.csv]")
    sys.exit(1)

INPUT = sys.argv[1]
OUTPUT = sys.argv[2] if len(sys.argv) > 2 else "poker_clean.csv"

# === HELPERS ===
def money(x):
    if not x or x.strip() in ["-", ""]:
        return 0.0
    return float(x.replace("€", "").replace(",", ".").strip())

def category(name):
    n = name.lower()
    if "freeroll" in n: return "freeroll"
    if "nitro" in n: return "nitro"
    return "expresso"

def stake(b):
    return "micro" if b <= 1 else "low" if b <= 5 else "mid" if b <= 20 else "high"

def session(h):
    if 6 <= h < 12: return "morning"
    if 12 <= h < 18: return "afternoon"
    if 18 <= h < 24: return "evening"
    return "night"

# === PARSE ===
rows = []
errors = 0

with open(INPUT, encoding="utf-8") as f:
    for line in f:
        raw = line.strip()
        if not raw:
            continue

        parts = re.split(r"\s{2,}", raw)

        try:
            # DEBUG FIRST RUN ONLY (comment after)
            # print(parts)

            dt = datetime.strptime(parts[0], "%d/%m/%Y %H:%M:%S")
            name = parts[1]
            buy = float(parts[2]) if len(parts) > 2 else 0.0
            players = int(parts[5]) if len(parts) > 5 and parts[5].isdigit() else None

            pos = parts[6] if len(parts) > 6 and parts[6] != "-" else None
            prize = money(parts[7]) if len(parts) > 7 else 0.0
            net = money(parts[8]) if len(parts) > 8 else 0.0

            profit = net - buy
            h = dt.hour

            rows.append({
                "datetime": dt,
                "game": name,
                "buy_in": buy,
                "players": players,
                "position": pos,
                "prize": prize,
                "net": net,
                "profit": profit,
                "category": category(name),
                "stake": stake(buy),
                "hour": h,
                "day": dt.strftime("%A"),
                "session": session(h),
                "result": "win" if pos == "1" else "itm" if prize > 0 else "loss"
            })

        except Exception as e:
            errors += 1
            if errors < 5:
                print("Skip:", raw)
                print("Reason:", e)

# === VALIDATION ===
if not rows:
    print("❌ No valid rows parsed. Your format is different. Enable debug print.")
    sys.exit(1)

df = pd.DataFrame(rows)

if "datetime" not in df.columns:
    print("❌ Broken parse. Columns:", df.columns)
    sys.exit(1)

# === PROCESS ===
df = df.sort_values("datetime")
df["cum_profit"] = df["profit"].cumsum()

# === SAVE ===
df.to_csv(OUTPUT, index=False)
print(f"✅ Saved {len(df)} rows → {OUTPUT}")
print(f"⚠️ Skipped lines: {errors}")

print("Games:", len(df))
print("Total profit:", df["profit"].sum())
print("ROI:", df["profit"].sum() / df["buy_in"].sum())
print("Win rate:", (df["position"] == "1").mean())
print("ITM:", (df["prize"] > 0).mean())

# === GRAPH ===
plt.figure()
plt.plot(df["datetime"], df["cum_profit"])
plt.xlabel("Time")
plt.ylabel("€")
plt.title("Cumulative Profit")
plt.xticks(rotation=45)
plt.tight_layout()
plt.show()