From 5ab0bc15bc4dc1a241408bf42cb9289d1c5a71a2 Mon Sep 17 00:00:00 2001
From: Zengtudor <zengtudor@outlook.com>
Date: Sat, 21 Sep 2024 22:32:34 +0800
Subject: [PATCH] add pybind surport

---
 dna.pyi       |  1 +
 py_example.py |  5 ++++
 src/main.cpp  | 79 ++++++++++++++++++++++++++++++++++++++-------------
 xmake.lua     | 19 ++++++++++++-
 4 files changed, 83 insertions(+), 21 deletions(-)
 create mode 100644 dna.pyi
 create mode 100644 py_example.py

diff --git a/dna.pyi b/dna.pyi
new file mode 100644
index 0000000..dd5a57d
--- /dev/null
+++ b/dna.pyi
@@ -0,0 +1 @@
+def dna_reverse(input_file_path: str, output_file_path: str):...
\ No newline at end of file
diff --git a/py_example.py b/py_example.py
new file mode 100644
index 0000000..2d82acc
--- /dev/null
+++ b/py_example.py
@@ -0,0 +1,5 @@
+import dna
+
+help(dna)
+
+dna.dna_reverse("filteredReads.txt","reversedSequence.txt")
\ No newline at end of file
diff --git a/src/main.cpp b/src/main.cpp
index 2fadc05..5094479 100644
--- a/src/main.cpp
+++ b/src/main.cpp
@@ -1,5 +1,12 @@
 #include "dna.hpp"
 
+#ifdef DNA_IS_PYBIND
+#include <pybind11/pybind11.h>
+// #include <pybind11/stl_bind.h>
+namespace py = pybind11;
+#endif
+
+
 //注意，这个函数会被并行执行，请只访问begin<=i<end之间的内容，以免出现数据竞争导致程序异常
 void reverseComplement(char *begin, char *end) 
 {
@@ -14,7 +21,7 @@ void reverseComplement(char *begin, char *end)
     std::reverse(begin, end); //翻转DNA序列
 
     for (ptrdiff_t i = 0; i < (end - begin); ++i) {
-        static int _ = (zt::print(NAME_VALUE(omp_get_num_threads()),"\n"),0); // 打印线程数量
+        // static int _ = (zt::print(NAME_VALUE(omp_get_num_threads()),"\n"),0); // 打印线程数量
         auto it = complement.find(begin[i]);
         if (it != complement.end()) {
             begin[i] = it->second;
@@ -22,27 +29,59 @@ void reverseComplement(char *begin, char *end)
     }
 }
 
+#ifndef DNA_IS_PYBIND
 int main()
 {
 	try{
-        //原理是这里定义处理函数，将函数传入dna::open_file_and_calculate，在open_file_and_calculate中会调用传入的函数
-        //参数列表 <文件分块内存大小，单个DNA序列最长大小>("输入文件名","输出文件名",序列处理函数);
-        //这个函数在src/tools/dna里面
-        dna::open_file_and_calculate<(size_t)4*1024 * 1024 *1024 , (size_t)5e4+5>("filteredReads.txt", "reversedSequence.txt",reverseComplement);
-        // dna::open_file_and_calculate<(size_t)30, (size_t)5e4 + 5>("test.txt", "test.out.txt", reverseComplement);
+       //原理是这里定义处理函数，将函数传入dna::open_file_and_calculate，在open_file_and_calculate中会调用传入的函数
+       //参数列表 <文件分块内存大小，单个DNA序列最长大小>("输入文件名","输出文件名",序列处理函数);
+       //这个函数在src/tools/dna里面
+       dna::open_file_and_calculate<(size_t)4*1024 * 1024 *1024 , (size_t)5e4+5>("filteredReads.txt", "reversedSequence.txt",reverseComplement);
+       // dna::open_file_and_calculate<(size_t)30, (size_t)5e4 + 5>("test.txt", "test.out.txt", reverseComplement);
 
-    }catch(const std::exception &e){
-        zt::eprint(
-            "Caught an error because:\n",
-            "\t",NAME_VALUE(e.what()),"\n"
-            "Closing\n"
-        );
-        throw  e;
-    }catch(...){
-        zt::eprint(
-            "Caught an unknown error :\n",
-            "Closing\n"
-        );
-        throw;
-    }
+   }catch(const std::exception &e){
+       zt::eprint(
+           "Caught an error because:\n",
+           "\t",NAME_VALUE(e.what()),"\n"
+           "Closing\n"
+       );
+       throw  e;
+   }catch(...){
+       zt::eprint(
+           "Caught an unknown error :\n",
+           "Closing\n"
+       );
+       throw;
+   }
 }
+
+#else
+
+void bind_func(std::string input_file_name,std::string out_put_file_name){
+    try{
+       dna::open_file_and_calculate<(size_t)4*1024 * 1024 *1024 , (size_t)5e4+5>(input_file_name, out_put_file_name,reverseComplement);
+    // dna::open_file_and_calculate<(size_t)1 * 1024 *1024 , (size_t)5e4+5>(input_file_name, out_put_file_name,reverseComplement);
+   }catch(const std::exception &e){
+       zt::eprint(
+           "Caught an error because:\n",
+           "\t",NAME_VALUE(e.what()),"\n"
+           "Closing\n"
+       );
+       throw  e;
+   }catch(...){
+       zt::eprint(
+           "Caught an unknown error :\n",
+           "Closing\n"
+       );
+       throw;
+   }
+}
+
+PYBIND11_MODULE(dna, m){
+    m.doc() = "DNASequence processing functions"; // optional module docstring
+
+    m.def("dna_reverse", &bind_func, "DNA is double-stranded and complementary to each other, and when sequencing a DNA sample you can't be sure which strand is being measured, so the complementary strands of all the DNA fragments are counted and assembled together with the original file."
+        ,py::arg("input_file_path"),py::arg("output_file_path"));
+}
+
+#endif
\ No newline at end of file
diff --git a/xmake.lua b/xmake.lua
index 4bc8bf7..30aa8b0 100644
--- a/xmake.lua
+++ b/xmake.lua
@@ -1,6 +1,8 @@
 add_rules("mode.debug","mode.release")
 set_languages("c++23")
 
+add_requires("pybind11")
+
 if is_mode("release")then
     set_optimize("aggressive") --这里使用了激进的优化，可能会导致浮点数计算不准确，考虑到本项目没有浮点计算，酌情考虑开启
     -- set_optimize("fastest") --上面和下面的二选一，--表示注释
@@ -24,5 +26,20 @@ end
 add_includedirs("src/tools")
 set_rundir("./")
 
-target("dna")
+target("dna_pybind")
+    add_defines("DNA_IS_PYBIND")
+    add_packages("pybind11")
+    set_kind("shared")
+    set_extension(".pyd")
+    add_files("src/main.cpp")
+    set_basename("dna")
+    after_build(
+        function (target) 
+            print(target:targetdir())
+            os.cp("*.py*",target:targetdir())
+        end
+    )
+
+target("dna")
+    set_kind("binary")
     add_files("src/main.cpp")