"""Inspect a chain in the pinned free sample. No dependencies beyond Python 3.10+."""
import argparse
from collections import Counter
import csv
from decimal import Decimal
import hashlib
import io
import json
from pathlib import Path
import platform
import zipfile
from download_options_sample import SHA256, digest

HEADER = 'contract,underlying,expiration,type,strike,style,quote_date,bid,bid_size,ask,ask_size,quote_time,volume,open_interest,open,high,low,close,trade_vwap,transactions,multileg_volume,active_minutes,last_trade_date,underlying_close,settlement_time,iv_bid,iv_ask,iv,iv_flag,delta,gamma,theta,vega,rho'.split(',')

def number(cell):
    # Blank is unavailable. A recorded zero remains zero.
    if cell == '':
        return None
    value = Decimal(cell)
    if not value.is_finite():
        raise ValueError('Unexpected non-finite numeric value')
    return value

def analyze(archive, symbol, day, expiration, low, high, option_type):
    if digest(archive) != SHA256:
        raise ValueError('Sample SHA-256 does not match tutorial revision 1')
    name = day + '_options.csv'
    with zipfile.ZipFile(archive) as z:
        manifest_bytes = z.read('day_by_date/manifest.json')
        manifest = json.loads(manifest_bytes)
        entries = [e for e in manifest['files'] if e['name'] == name]
        if len(entries) != 1:
            raise ValueError('Select exactly one published sample date')
        entry = entries[0]
        raw = z.read('day_by_date/' + name)
        if hashlib.sha256(raw).hexdigest() != entry['sha256'] or len(raw) != entry['bytes']:
            raise ValueError('Selected CSV differs from manifest')
    reader = csv.DictReader(io.StringIO(raw.decode('utf-8'), newline=''))
    if reader.fieldnames != HEADER:
        raise ValueError('Unexpected column order')
    rows = list(reader)
    if len(rows) != entry['rows'] or any(None in r or None in r.values() for r in rows):
        raise ValueError('Row count or column structure differs from manifest')
    # Exact symbol matching, without merging contract roots or renaming securities.
    selected = [r for r in rows if r['underlying'] == symbol and r['quote_date'] == day
                and r['expiration'] == expiration and r['type'] == option_type
                and low <= number(r['strike']) <= high]
    counts = Counter()
    flags = Counter()
    mids = []
    ivs = []
    seen = set()
    for r in selected:
        if r['contract'] in seen:
            raise ValueError('Duplicate selected contract')
        seen.add(r['contract'])
        bid, ask = number(r['bid']), number(r['ask'])
        volume, iv = number(r['volume']), number(r['iv'])
        flags[r['iv_flag']] += 1
        counts['missing_bid'] += bid is None
        counts['missing_ask'] += ask is None
        counts['missing_volume'] += volume is None
        counts['zero_volume'] += volume == 0
        counts['missing_iv'] += iv is None
        # Do not invent zero quotes or turn crossed/one-sided quotes into a midpoint.
        usable = bid is not None and ask is not None and 0 < bid <= ask
        counts['positive_noncrossed_quotes'] += usable
        if usable:
            mids.append((bid + ask) / 2)
        # Flag 0 uses parity forward; flag 1 is the documented fallback forward.
        if r['iv_flag'] in ('0', '1') and iv is not None:
            ivs.append(iv)
        counts['zero_iv_bid'] += number(r['iv_bid']) == 0
    def bounds(values):
        return [str(min(values)), str(max(values))] if values else None
    return {'tutorial_revision': 1, 'python': platform.python_version(), 'sample_url': 'https://historicaldata.net/file/options_sample_2022H2.zip',
            'sample_sha256': SHA256, 'manifest_sha256': hashlib.sha256(manifest_bytes).hexdigest(),
            'csv': name, 'csv_sha256': entry['sha256'], 'csv_rows': len(rows), 'columns': len(HEADER),
            'selection': {'underlying': symbol, 'quote_date': day, 'expiration': expiration, 'type': option_type,
                          'strike_min_inclusive': str(low), 'strike_max_inclusive': str(high)},
            'selected_contracts': len(selected), 'counts': dict(counts), 'iv_flags': dict(sorted(flags.items())),
            'quote_unit': 'index_points' if symbol in ('SPX', 'SPXW') else 'USD_per_share',
            'midpoint_range': bounds(mids), 'solved_iv_decimal_range': bounds(ivs)}

if __name__ == '__main__':
    parser = argparse.ArgumentParser(description=__doc__)
    parser.add_argument('--sample', type=Path, default=Path('options_sample_2022H2.zip'))
    parser.add_argument('--symbol', choices=['TSLA', 'KO', 'SPY', 'SPX', 'SPXW'], default='SPY')
    parser.add_argument('--date', default='2022-08-25')
    parser.add_argument('--expiration', default='2022-09-16')
    parser.add_argument('--strike-min', type=Decimal, default=Decimal('400'))
    parser.add_argument('--strike-max', type=Decimal, default=Decimal('440'))
    parser.add_argument('--type', choices=['call', 'put'], default='call')
    a = parser.parse_args()
    if a.strike_min > a.strike_max:
        parser.error('strike-min must not exceed strike-max')
    print(json.dumps(analyze(a.sample, a.symbol, a.date, a.expiration, a.strike_min, a.strike_max, a.type), indent=2))