将利用30行X算法求解数独的python程序转成DuckDB自定义函数并比较性能

将利用DeepSeek辅助改造30行DLX求解数独python程序和调用python函数的不同方法效率对比测试的代码整合在一起

只要把代码粘贴到一个文件，比如duckudf.py，然后增加如下语句

python 复制代码

duckdb.create_function("solve_single_sudoku", solve_single_sudoku)  #, [VARCHAR, [[VARCHAR]], BIGINT)

s="""
select solve_single_sudoku('000080304250100090000003000040020030006090800700310009000000080602000053004002900');
"""
t=time.time();print(duckdb.sql(s));print( time.time()-t)

并修改以下函数的定义，添加参数和返回值类型

python 复制代码

def solve_single_sudoku(sudoku_str: str) -> str:

并删除原有的自定义函数及全局的代码块

python 复制代码

if __name__ == "__main__":
    # 运行文档测试
    import doctest
    doctest.testmod()
    
    # 运行主函数
    main()

然后用如下命令行测试

复制代码

C:\d>timer64 python duckudf.py
1
┌──────────────────────────────────────────────────────────────────────────────────────────────────────────┐
│ solve_single_sudoku('000080304250100090000003000040020030006090800700310009000000080602000053004002900') │
│                                                 varchar                                                  │
├──────────────────────────────────────────────────────────────────────────────────────────────────────────┤
│ 167589324253174698498263175941827536326495817785316249579631482612948753834752961                        │
└──────────────────────────────────────────────────────────────────────────────────────────────────────────┘

0.25064992904663086


Kernel  Time =     0.156 =   35%
User    Time =     0.234 =   52%
Process Time =     0.390 =   87%    Virtual  Memory =    541 MB
Global  Time =     0.445 =  100%    Physical Memory =     90 MB

C:\d>timer64 python pydlx4path.txt
[[0, 0, 0, 0, 8, 0, 3, 0, 4], [2, 5, 0, 1, 0, 0, 0, 9, 0], [0, 0, 0, 0, 0, 3, 0, 0, 0], [0, 4, 0, 0, 2, 0, 0, 3, 0], [0, 0, 6, 0, 9, 0, 8, 0, 0], [7, 0, 0, 3, 1, 0, 0, 0, 9], [0, 0, 0, 0, 0, 0, 0, 8, 0], [6, 0, 2, 0, 0, 0, 0, 5, 3], [0, 0, 4, 0, 0, 2, 9, 0, 0]]
[[1, 6, 7, 5, 8, 9, 3, 2, 4], [2, 5, 3, 1, 7, 4, 6, 9, 8], [4, 9, 8, 2, 6, 3, 1, 7, 5], [9, 4, 1, 8, 2, 7, 5, 3, 6], [3, 2, 6, 4, 9, 5, 8, 1, 7], [7, 8, 5, 3, 1, 6, 2, 4, 9], [5, 7, 9, 6, 3, 1, 4, 8, 2], [6, 1, 2, 9, 4, 8, 7, 5, 3], [8, 3, 4, 7, 5, 2, 9, 6, 1]]


Kernel  Time =     0.000 =    0%
User    Time =     0.031 =   87%
Process Time =     0.031 =   87%    Virtual  Memory =      7 MB
Global  Time =     0.035 =  100%    Physical Memory =     12 MB

C:\d>

第一个命令行是调用duckdb执行自定义函数，第二个是直接执行python程序。可见调用duckdb执行自定义函数的用时是后者15倍，内存占用也放大了很多倍。