import csv
from datetime import datetime, timedelta
from collections import defaultdict
import statistics

# Load CGM data
cgm = []
with open('/home/ben/downloads/Blood Glucose-2026-02-19-2026-03-17.csv') as f:
    reader = csv.DictReader(f)
    for row in reader:
        dt = datetime.strptime(row['Date/Time'], '%Y-%m-%d %H:%M:%S')
        bg = int(row['Blood Glucose (mg/dL)'])
        cgm.append((dt, bg))

cgm.sort(key=lambda x: x[0])
daily = defaultdict(list)
for dt, bg in cgm:
    daily[dt.strftime('%Y-%m-%d')].append((dt, bg))

# Pre-Lent (Feb 19-23) vs Early Lent (Feb 24 - Mar 2) vs Mid Lent (Mar 3-9) vs Late Lent (Mar 10-17)
periods = {
    'Pre-Lent (Feb 19-23)': ('2026-02-19', '2026-02-23'),
    'Early Lent OMAD (Feb 24 - Mar 2)': ('2026-02-24', '2026-03-02'),
    'Mid Lent (Mar 3-9)': ('2026-03-03', '2026-03-09'),
    'Late Lent (Mar 10-17)': ('2026-03-10', '2026-03-17'),
}

print("="*80)
print("PERIOD COMPARISON")
print("="*80)

for name, (start, end) in periods.items():
    readings = []
    for date_str in sorted(daily.keys()):
        if start <= date_str <= end:
            readings.extend([bg for dt, bg in daily[date_str]])
    if not readings:
        continue
    mean = statistics.mean(readings)
    tir = sum(1 for bg in readings if 70 <= bg <= 140) / len(readings) * 100
    above_140 = sum(1 for bg in readings if bg > 140) / len(readings) * 100
    above_160 = sum(1 for bg in readings if bg > 160) / len(readings) * 100
    cv = (statistics.stdev(readings) / mean) * 100
    
    # Overnight only (4-6am)
    overnight = []
    for date_str in sorted(daily.keys()):
        if start <= date_str <= end:
            overnight.extend([bg for dt, bg in daily[date_str] if 4 <= dt.hour < 6])
    on_mean = statistics.mean(overnight) if overnight else 0
    
    print(f"\n{name}:")
    print(f"  Mean: {mean:.0f}, CV: {cv:.1f}%, TIR: {tir:.0f}%")
    print(f"  >140: {above_140:.1f}%, >160: {above_160:.1f}%")
    print(f"  Min: {min(readings)}, Max: {max(readings)}")
    print(f"  Overnight (4-6am): {on_mean:.0f}")
    print(f"  Readings: {len(readings)}")

# Reactive hypoglycemia frequency
print("\n\n" + "="*80)
print("REACTIVE HYPOGLYCEMIA ANALYSIS")
print("="*80)

# Find all drops below 90 that follow a meal spike
reactive_events = []
for date_str in sorted(daily.keys()):
    day_data = sorted(daily[date_str], key=lambda x: x[0])
    for i, (dt, bg) in enumerate(day_data):
        if bg < 90:
            # Check if there was a reading >130 in the prior 4 hours
            prior_high = max(
                (bg2 for dt2, bg2 in day_data 
                 if timedelta(0) < (dt - dt2) < timedelta(hours=4)),
                default=0
            )
            if prior_high > 125:
                reactive_events.append((dt, bg, prior_high))

print(f"\nTotal reactive drops (<90 after >125 spike): {len(reactive_events)}")
for dt, low, high in reactive_events:
    print(f"  {dt.strftime('%b %d %H:%M')}: dropped to {low} (prior high: {high})")

# Look at fasting duration vs pre-meal glucose correlation
print("\n\n" + "="*80)
print("FASTING DURATION vs PRE-MEAL GLUCOSE")
print("="*80)

fasting_data = [
    (21.9, 103), (19.2, 115), (17.0, 112), (14.6, 126), (18.6, 115),
    (19.4, 103), (14.3, 120), (10.8, 115), (11.8, 115), (18.7, 108),
    (13.6, 110), (14.3, 101), (18.8, 98),
]

# Simple correlation
n = len(fasting_data)
sum_x = sum(h for h, g in fasting_data)
sum_y = sum(g for h, g in fasting_data)
sum_xy = sum(h * g for h, g in fasting_data)
sum_x2 = sum(h ** 2 for h, g in fasting_data)
sum_y2 = sum(g ** 2 for h, g in fasting_data)

r = (n * sum_xy - sum_x * sum_y) / ((n * sum_x2 - sum_x**2) * (n * sum_y2 - sum_y**2)) ** 0.5
print(f"\nCorrelation (fast hours vs pre-meal BG): r = {r:.3f}")
print("(negative = longer fasts → lower pre-meal glucose)")

for h, g in sorted(fasting_data, key=lambda x: x[0]):
    print(f"  {h:.1f}h fast → {g} mg/dL pre-meal")

# Best and worst days analysis
print("\n\n" + "="*80)
print("WHAT MAKES A GOOD vs BAD DAY")
print("="*80)

good_days = ['2026-02-28', '2026-03-05', '2026-03-06', '2026-03-14', '2026-03-15', '2026-03-16']
bad_days = ['2026-02-19', '2026-03-07', '2026-02-23', '2026-02-24']

print("\nGood days (mean <115):")
for d in good_days:
    readings = [bg for dt, bg in daily.get(d, [])]
    if readings:
        print(f"  {d}: mean {statistics.mean(readings):.0f}, max {max(readings)}")

print("\nBad days (mean >125 or max >180):")
for d in bad_days:
    readings = [bg for dt, bg in daily.get(d, [])]
    if readings:
        print(f"  {d}: mean {statistics.mean(readings):.0f}, max {max(readings)}")