from serenitas.utils.exchange import ExchangeMessage
import pandas as pd
from serenitas.utils.db2 import dbconn
from psycopg.errors import UniqueViolation
import datetime
import logging
import argparse
import re
from serenitas.ops.trade_dataclasses import CreditSwaptionDeal, desc_str
from serenitas.ops.dataclass_mapping import Fund

parser = argparse.ArgumentParser()
parser.add_argument(
    "workdate",
    nargs="?",
    default=str(datetime.date.today()),
    type=datetime.date.fromisoformat,
)
args = parser.parse_args()

logger = logging.getLogger(__name__)

columns = [
    "option_recap",
    "index_buyer",
    "index_seller",
    "trade_date",
    "effective_date",
    "execution_time",
    "notional",
    "price",
    "mid_price",
    "strike",
    "expiry_date",
    "option_type",
    "exercise_clearing_house",
    "premium_fee",
    "premium_fee_pay_date",
    "trade_id",
    "venue_mic",
]


# Counterparty Ticket to Database
_baml_funds: dict[str, Fund] = {
    "LMIL-Boston Patriot Bowdoin St LLC": "BOWDST",
    "Iso Selene Inc.": "ISOSEL",
    "Serenitas Credit Gamma Master Fund LP": "SERCGMAST",
    "Lmcg Investments, Llc": "SERCGMAST",
    "Reep-Ofc One Bowdoin Square Ma Llc Et Al (Cob)": "BOWDST",
}


def regex_baml_numbers(text: str):
    return float(re.search(r"[\d,]+", text)[0].replace(",", ""))


def from_baml_email(d: dict, conn):
    buysell = "Bank Of America" in d["Index Seller"]
    index, series = re.search(r"(HY|IG)\.([\d]{2})", d["OPTION RECAP"]).groups()
    security_desc = desc_str(index, series, "5")
    trade_date = datetime.datetime.strptime(d["Trade Date"], "%d-%b-%Y")
    price = float(d["Price"]) / 100
    sql_str = (
        "SELECT * FROM index_desc WHERE INDEX=%s AND series=%s "
        "AND lastdate >= %s AND tenor='5yr'"
    )
    with conn.cursor() as c:
        c.execute(sql_str, (index, series, trade_date))
        row = c.fetchone()

    return CreditSwaptionDeal(
        buysell=buysell,
        fund=_baml_funds[d[f"Index {'Buyer' if buysell else 'Seller'}"]],
        cp_code="BAMSNY",
        portfolio="OPTIONS",
        folder=index + ("PAYER" if d["Option Type"] == "Payer" else "REC"),
        trade_date=trade_date,
        settle_date=datetime.datetime.strptime(d["Premium Fee Pay Date"], "%d-%b-%Y"),
        expiration_date=datetime.datetime.strptime(d["Expiry Date"], "%d-%b-%Y"),
        notional=regex_baml_numbers(d["Notional"]),
        option_type=d["Option Type"].upper(),
        strike=d["Strike"],
        swap_type="CD_INDEX_OPTION",
        fixed_rate=row.coupon / 100,
        maturity=row.maturity,
        security_desc=security_desc,
        security_id=row.redindexcode,
        currency="USD",
        price=price,
        initial_margin_percentage=(
            regex_baml_numbers(d["Collateral"])
            / regex_baml_numbers(d["Premium Fee"])
            * price
        )
        if d.get("Collateral")
        else None,
    )


if __name__ == "__main__":
    dawndb = dbconn("dawndb")
    em = ExchangeMessage()
    start = datetime.datetime.combine(args.workdate, datetime.time.min).replace(
        tzinfo=em._account.default_timezone
    )
    for msg in em.get_msgs(
        path=["AutoBook", "BAML Swaption"], datetime_received__gte=start
    ):
        dfs = pd.read_html(msg.body)
        tickets = []
        for ticket in dfs:
            if len(ticket[0]) > 5:
                ticket = ticket.set_index(0).to_dict()[1]
                if "Block" in ticket or "Unwind Price" in ticket:
                    continue
                tickets.append(ticket)
                trade = from_baml_email(ticket, dawndb)
                trade.stage()

        df = pd.DataFrame.from_dict(tickets)
        df.columns = df.columns.str.lower().str.replace(" ", "_")
        if "collateral" in df.columns:
            additional_columns = ["collateral"]
        else:
            additional_columns = []
        df = df[columns + additional_columns]
        place_holders = ",".join(["%s"] * len(df.columns))
        try:
            with dawndb.cursor() as c:
                c.executemany(
                    f"INSERT INTO baml_swaption_ticket({','.join(df.columns)})"
                    f"VALUES ({place_holders})",
                    df.itertuples(index=False),
                )
        except UniqueViolation as e:
            logger.warning(str(e))
            CreditSwaptionDeal._insert_queue.clear()
        else:
            CreditSwaptionDeal.commit()