import matplotlib
matplotlib.use('Agg')
import matplotlib.pyplot as plt
import numpy as np

plt.rcParams['font.family'] = 'sans-serif'
for f in ['WenQuanYi Micro Hei', 'Noto Sans CJK SC', 'SimHei', 'DejaVu Sans']:
    try:
        plt.rcParams['font.sans-serif'] = [f]
        break
    except:
        continue
plt.rcParams['axes.unicode_minus'] = False

ranges = ['0分钟\n(无时长数据)', '5-10', '10-15', '15-20', '20-30', '30-45', '45-60', '60-90', '90-120', '120+']
l1 = [0, 0, 7, 195, 1036, 749, 225, 118, 37, 28]
l2 = [1403, 4, 14, 45, 172, 582, 238, 115, 31, 22]

fig, axes = plt.subplots(1, 2, figsize=(18, 7))

x = np.arange(len(ranges))
w = 0.35

# === Left: Side-by-side bar ===
b1 = axes[0].bar(x - w/2, l1, w, color='#4472C4', label='L1 (2,395组合)', edgecolor='white', linewidth=0.5)
b2 = axes[0].bar(x + w/2, l2, w, color='#ED7D31', label='L2 (2,626组合)', edgecolor='white', linewidth=0.5)

# Labels for bars > 0
for bar, v, c in zip(b1, l1, ['#4472C4']*len(l1)):
    if v > 20:
        axes[0].text(bar.get_x()+bar.get_width()/2, v+max(max(l1),max(l2))*0.02,
                    str(v), ha='center', fontsize=8, color=c)
for bar, v, c in zip(b2, l2, ['#ED7D31']*len(l2)):
    if v > 20:
        axes[0].text(bar.get_x()+bar.get_width()/2, v+max(max(l1),max(l2))*0.02,
                    str(v), ha='center', fontsize=8, color=c)

axes[0].set_xticks(x)
axes[0].set_xticklabels(ranges, fontsize=10, rotation=0)
axes[0].set_ylabel('重复学习组合数', fontsize=12)
axes[0].set_title('L1 vs L2 总学习时长分布对比', fontsize=14, fontweight='bold')
axes[0].legend(fontsize=11)
axes[0].grid(axis='y', alpha=0.2, linestyle='--')
axes[0].set_ylim(0, max(max(l1), max(l2))*1.18)

# === Right: Percentage within each level ===
l1_total = sum(l1)  # 2395
l2_total = sum(l2)  # 2626
l1_pct = [v/l1_total*100 if l1_total > 0 else 0 for v in l1]
l2_pct = [v/l2_total*100 if l2_total > 0 else 0 for v in l2]

bp1 = axes[1].bar(x - w/2, l1_pct, w, color='#4472C4', label=f'L1 ({l1_total}组合)', edgecolor='white', linewidth=0.5)
bp2 = axes[1].bar(x + w/2, l2_pct, w, color='#ED7D31', label=f'L2 ({l2_total}组合)', edgecolor='white', linewidth=0.5)

for bar, v in zip(bp1, l1_pct):
    if v > 2:
        axes[1].text(bar.get_x()+bar.get_width()/2, v+max(max(l1_pct),max(l2_pct))*0.02,
                    f'{v:.1f}%', ha='center', fontsize=8, color='#4472C4')
for bar, v in zip(bp2, l2_pct):
    if v > 2:
        axes[1].text(bar.get_x()+bar.get_width()/2, v+max(max(l1_pct),max(l2_pct))*0.02,
                    f'{v:.1f}%', ha='center', fontsize=8, color='#ED7D31')

axes[1].set_xticks(x)
axes[1].set_xticklabels(ranges, fontsize=10, rotation=0)
axes[1].set_ylabel('占比 (%)', fontsize=12)
axes[1].set_title('L1 vs L2 总学习时长占比对比', fontsize=14, fontweight='bold')
axes[1].legend(fontsize=11)
axes[1].grid(axis='y', alpha=0.2, linestyle='--')
axes[1].set_ylim(0, max(max(l1_pct), max(l2_pct))*1.18)

fig.suptitle('最近3个月（2026.03-05）重复学习时长分布 L1 vs L2', 
             fontsize=16, fontweight='bold', y=1.02)

fig.text(0.5, 0.01,
    '⚠️ L2有1,403个组合(53.4%)无组件级时长数据(0分钟) | L1数据完整 | '
    'L1峰值20-30分钟(43%), L2峰值30-45分钟(去除零值后占48%)',
    ha='center', fontsize=10, color='#C0392B')

plt.tight_layout(rect=[0, 0.06, 1, 0.95])
plt.savefig('/root/.openclaw/workspace/output/repeat_duration_l1l2_3m.png', dpi=150, bbox_inches='tight')
print('Saved.')