#!/bin/bash
# 合并并行预测的结果
# 用法: bash merge_results.sh

set -e

TEMP_DIR="Data/fragment/Frags-Enamine-18M_temp"
OUTPUT_FILE="Data/fragment/Frags-Enamine-18M_predicted.csv"

echo "============================================================"
echo "📦 合并预测结果"
echo "============================================================"

# 检查临时目录是否存在
if [ ! -d "$TEMP_DIR" ]; then
    echo "❌ 临时目录不存在: $TEMP_DIR"
    exit 1
fi

# 检查所有部分文件是否存在
MISSING=0
for i in 0 1 2 3; do
    if [ ! -f "${TEMP_DIR}/part_${i}.csv" ]; then
        echo "❌ 缺少文件: part_${i}.csv"
        MISSING=1
    else
        LINES=$(wc -l < "${TEMP_DIR}/part_${i}.csv")
        echo "✓ part_${i}.csv: $LINES 行"
    fi
done

if [ $MISSING -eq 1 ]; then
    echo "❌ 有文件缺失，请等待所有进程完成"
    exit 1
fi

echo ""
echo "合并文件..."

# 合并 CSV 文件（保留第一个文件的表头，跳过其他文件的表头）
cat "${TEMP_DIR}/part_0.csv" > "$OUTPUT_FILE"
for i in 1 2 3; do
    tail -n +2 "${TEMP_DIR}/part_${i}.csv" >> "$OUTPUT_FILE"
done

# 统计结果
TOTAL_LINES=$(wc -l < "$OUTPUT_FILE")
TOTAL_MOLECULES=$((TOTAL_LINES - 1))

echo ""
echo "============================================================"
echo "✓ 合并完成"
echo "============================================================"
echo "输出文件: $OUTPUT_FILE"
echo "总分子数: $TOTAL_MOLECULES"
echo ""

# 统计广谱抗菌分子
if command -v python3 &> /dev/null; then
    python3 << EOF
import pandas as pd
df = pd.read_csv("$OUTPUT_FILE")
n_broad = df['broad_spectrum'].sum()
print(f"广谱抗菌: {n_broad:,} 个 ({n_broad/len(df)*100:.2f}%)")
print(f"非广谱: {len(df)-n_broad:,} 个 ({(len(df)-n_broad)/len(df)*100:.2f}%)")
print("")
print("抑制菌株数分布:")
for threshold in [0, 5, 10, 15, 20, 30]:
    n = (df['ginhib_total'] >= threshold).sum()
    print(f"  ≥{threshold:2d} 个菌株: {n:,} ({n/len(df)*100:.2f}%)")
EOF
fi

echo "============================================================"