We Keep Coding

Why 0xFEFF appear in recv data

i try to create a client written in c++ and a server using python flask. The task is simple, client connect and get data from server, then display it on console. I have two piece of code:
Server:
from flask import Flask
app = Flask(__name__)
#app.route('/hello')
def hello():
return "hello".encode("utf-16")
if __name__ == "__main__":
app.run(debug=True,host="127.0.0.1")
Client:
BOOL bResults = FALSE;
bool ret = false;
DWORD dwDownloaded = 0;
WCHAR wDataRecv[1024] = {0};
HINTERNET hConnect = NULL;
HINTERNET hRequest = NULL;
DWORD dwSize = 0;
HINTERNET hSession = WinHttpOpen(L"WinHTTP Example/1.0",
WINHTTP_ACCESS_TYPE_DEFAULT_PROXY,
WINHTTP_NO_PROXY_NAME,
WINHTTP_NO_PROXY_BYPASS, 0); # open session to connect
hConnect = WinHttpConnect(hSession, L"localhost", (port == 0) ? INTERNET_DEFAULT_HTTP_PORT : port, 0); # connect to localhost with port 80 or custom port (this time i use port 5000)
if (!hConnect)
goto Free_And_Exit;
hRequest = WinHttpOpenRequest(hConnect, L"GET", L"/hello", NULL, WINHTTP_NO_REFERER, WINHTTP_DEFAULT_ACCEPT_TYPES, 0);
if (hRequest) {
bResults = WinHttpSendRequest(hRequest, NULL, 0,WINHTTP_NO_REQUEST_DATA, 0,0, 0);# send GET request
}
if (bResults)
bResults = WinHttpReceiveResponse(hRequest, NULL);
if (bResults)
{
dwSize = 0;
if (!WinHttpQueryDataAvailable(hRequest, &dwSize)) {
std::cout << "WinHttpQueryDataAvailable failed with code: " << GetLastError();
}
if (!WinHttpReadData(hRequest, wDataRecv, dwSize, &dwDownloaded)) {
std::cout << "WinHttpReadData failed with code: " << GetLastError();
}
std::wcout << wDataRecv; # wcout wont print anything to console
}
Free_And_Exit: #clean up
if (hRequest) WinHttpCloseHandle(hRequest);
if (hConnect) WinHttpCloseHandle(hConnect);
return ret;
I noticed that the data return from server like:
b'\xfe\xff\hello'
Why 0xFEFF is there ?

Its is BOM (byte order mark). I just need to decode from the client or use:
return "hello".encode('UTF-16LE') # not UTF-16

C_Python not releasing buffer memory

I'm writing C code for python (Python C API), and I noticed that python is not releasing the memory of the file, I'm wondering if the issue is in my code.
I want to simplify as much as passable, but I hope that no details will be missing.
The file is a binary file with buffers, first 4 bytes is the buffer size, then the buffer.
The binary file (big_file.comp):
du ~/Desktop/TEST_FILES/big_file.comp
4175416 ~/Desktop/TEST_FILES/big_file.comp
The python code (test.py):
#!/usr/bin/env python3
from struct import unpack_from
from psutil import Process
from os import getpid
import decomplib
def file_handler(file_name):
with open(file_name, 'rb') as reader:
while True:
next_4_bytes = reader.read(4)
if next_4_bytes == b'':
break
next_size, *_ = unpack_from("I", next_4_bytes)
buffer = reader.read(next_size)
yield buffer, next_size
def main():
args = _parse_args()
decompress = decomplib.Decompress()
for buf, buf_size in file_handler(args.file):
for msg in decompress.decompress_buffer(buf, buf_size):
print(msg)
if __name__ == "__main__":
pid = getpid()
ps = Process(pid)
main()
print(ps.memory_info())
Some of the C code simplified:
#include <Python.h>
#include "structmember.h"
typedef struct {
PyObject_HEAD
uint32_t arr_size;
} DecompressObject;
static int Decompress_init(DecompressObject *self, PyObject *args, PyObject *kwds){
return 0;
}
static PyObject* Decompress_handle_buffer(DecompressObject* self, PyObject* args){
uint32_t buf_size = 0;
uint8_t *buf = NULL;
// get buffer and buffer length from python function
if(!PyArg_ParseTuple(args, "y*i", &buf, &buf_size)){
PyErr_SetString(PyExc_Exception, "Failed to parse function arguments");
return NULL;
}
self->arr_size = 10;
Py_XINCREF(self);
return (PyObject *) self;
}
static PyObject* Decompress_next(DecompressObject *self, PyObject *Py_UNUSED(ignored)){
static uint32_t seq_index = 0;
if (seq_index < self->arr_size) {
seq_index++;
Py_RETURN_NONE;
}
seq_index = 0;
return NULL;
}
static void Decompress_dealloc(DecompressObject *self){
Py_TYPE(self)->tp_free((PyObject *) self);
}
static PyMethodDef Decompress_methods[] = {
{"decompress_buffer", (PyCFunction) Decompress_handle_buffer, METH_VARARGS, "Decompress a buffer to asc data."},
{NULL} /* Sentinel */
};
static PyTypeObject DecompressType = {
PyVarObject_HEAD_INIT(NULL, 0)
.tp_name = "decomplib.Decompress",
.tp_doc = "Decompress object",
.tp_basicsize = sizeof(DecompressObject),
.tp_itemsize = 0,
.tp_flags = Py_TPFLAGS_DEFAULT | Py_TPFLAGS_BASETYPE,
.tp_alloc = PyType_GenericAlloc,
.tp_new = PyType_GenericNew,
.tp_iter = PyObject_SelfIter,
.tp_init = (initproc) Decompress_init,
.tp_dealloc = (destructor) Decompress_dealloc,
.tp_iternext = (iternextfunc) Decompress_next,
.tp_methods = Decompress_methods,
};
static PyModuleDef Decompressmodule = {
PyModuleDef_HEAD_INIT,
.m_name = "decomplib",
.m_doc = "Decompress an compressed file.",
.m_size = -1,
};
PyMODINIT_FUNC PyInit_decomplib(void){
PyObject *d;
if (PyType_Ready(&DecompressType) < 0)
return NULL;
d = PyModule_Create(&Decompressmodule);
if (d == NULL)
return NULL;
Py_INCREF(&DecompressType);
if (PyModule_AddObject(d, "Decompress", (PyObject *) &DecompressType) < 0) {
Py_DECREF(&DecompressType);
Py_DECREF(d);
return NULL;
}
return d;
}
As a result, I got the following output:
./test.py -f ~/Desktop/TEST_CAN_OPT/big_fie.comp
None
None
None
...
None
None
None
pmem(rss=4349915136, vms=4412583936, shared=6270976, text=2867200, lib=0, data=4344135680, dirty=0)
While playing around I noticed that if I change in the C function Decompress_handle_buffer the call to the function PyArg_ParseTuple the second argument from "y*i" to "Si", Python do cleanup the memory...
./test.py -f ~/Desktop/TEST_CAN_OPT/big_fie.comp
None
None
None
...
None
None
None
pmem(rss=22577152, vms=84869120, shared=6361088, text=2867200, lib=0, data=16420864, dirty=0)
However, The buffer is NOT correctly read.
Any ideas?!
Extra Info:
I'm using a virtual machine (VMware Workstation 15)
OS Ubuntu 18.4
Python 3.6.9

y* does not correspond to uint8_t like you're using it. As stated in the documentation, it fills a Py_buffer struct that you're supposed to provide.
You need to actually provide a Py_buffer, and when you're done with it, you need to release the buffer with PyBuffer_Release.

How to get a string from C++ to python when using ctypes and wchar_t?

I can:
Get an integer from C++ and use it in python
Send a python string (as a wchar_t) to C++ and do some logic with it
I cannot
Step 2 in opposite direction.
Here is my C++ code (compiled with clion and cygwin as a shared library using C++14).
#include <iostream>
wchar_t aa[2];
extern "C" {
int DoA()
{
return 10;
}
int DoB(wchar_t * in)
{
if (in[1] == 'a')
{
return 25;
}
return 30;
}
wchar_t * DoC()
{
aa[0] = 'a';
aa[1] = 'b';
return aa;
}
}
Here is my python 3.6.1 code that shows what I can and what I cannot do. So how should I get my string and do things with it in python? I expect to use the address with wstring_at to get the value, but it is not working.
from ctypes import *
import os.path
print('Hello')
itExist = os.path.exists('C:/Users/Daan/CLionProjects/stringproblem/cmake-build-release/cygstringproblem.dll')
print(itExist)
lib = cdll.LoadLibrary('C:/Users/Daan/CLionProjects/stringproblem/cmake-build-release/cygstringproblem.dll')
print('dll loaded')
A = lib.DoA()
print(A)
Bx = lib.DoB(c_wchar_p('aaa'))
print(Bx)
By = lib.DoB(c_wchar_p('bbb'))
print(By)
Ca = lib.DoC()
print(Ca)
print('Issue is coming')
Cb = wstring_at(Ca,2)
print(Cb)
Here is the output with error.
Hello
True
dll loaded
10
25
30
-1659080704
Issue is coming
Traceback (most recent call last):
File "ShowProblem.py", line 19, in <module>
Cb = wstring_at(Ca,2)
File "C:\Users\Daan\AppData\Local\Programs\Python\Python36\lib\ctypes\__init__.py", line 504, in wstring_at
return _wstring_at(ptr, size)
OSError: exception: access violation reading 0xFFFFFFFF9D1C7000

I reproduced your problem on Linux and corrected it by defining the return type from your DoC function:
from ctypes import *
print('Hello')
lib = cdll.LoadLibrary(PATH_TO_TOUR_LIB)
print('dll loaded')
# this line solved the issue for me
lib.DoC.restype = c_wchar_p
A = lib.DoA()
print(A)
Bx = lib.DoB(c_wchar_p('aaa'))
print(Bx)
By = lib.DoB(c_wchar_p('bbb'))
print(By)
Ca = lib.DoC()
print(Ca)
print('Issue is coming')
Cb = wstring_at(Ca,2)
print(Cb)
I also allocated the memory dynamically (some Python expert might comment on this, I guess that this causes a memory leak):
extern "C" {
int DoA()
{
return 10;
}
int DoB(wchar_t * in)
{
if (in[1] == 'a')
{
return 25;
}
return 30;
}
wchar_t * DoC()
{
wchar_t* aa = new wchar_t[2];
aa[0] = 'a';
aa[1] = 'b';
return aa;
}
}
Let me know if it works on Windows.

If you set the .argtypes and .restype of your wrapped functions, you can call them more naturally. To handle an output string, it will be thread safe if you allocate the buffer in Python instead of using a global variable, or just return a wide string constant. Here's an example coded for the Microsoft compiler:
test.c
#include <wchar.h>
#include <string.h>
__declspec(dllexport) int DoA(void) {
return 10;
}
__declspec(dllexport) int DoB(const wchar_t* in) {
if(wcslen(in) > 1 && in[1] == 'a') // Make sure not indexing past the end.
return 25;
return 30;
}
// This version good if variable data is returned.
// Need to pass a buffer of sufficient length.
__declspec(dllexport) int DoC(wchar_t* aa, size_t length) {
if(length < 3)
return 0;
aa[0] = 'a';
aa[1] = 'b';
aa[2] = '\0';
return 1;
}
// Safe to return a constant. No memory leak.
__declspec(dllexport) wchar_t* DoD(void) {
return L"abcdefg";
}
test.py
from ctypes import *
# Set up the arguments and return type
lib = CDLL('test')
lib.DoA.argtypes = None
lib.DoA.restype = c_int # default, but just to be thorough.
lib.DoB.argtypes = [c_wchar_p]
lib.DoB.restype = c_int
lib.DoC.argtypes = [c_wchar_p,c_size_t]
lib.DoC.restype = c_int
lib.DoD.argtypes = None
lib.DoD.restype = c_wchar_p
# Map to local namespace functions
DoA = lib.DoA
DoB = lib.DoB
DoD = lib.DoD
# Do some pre- and post-processing to hide the memory details.
def DoC():
tmp = create_unicode_buffer(3) # Writable array of wchar_t.
lib.DoC(tmp,sizeof(tmp))
return tmp.value # return a Python string instead of the ctypes array.
print(DoA())
print(DoB('aaa'))
print(DoB('bbb'))
print(DoC())
print(DoD())
Output:
10
25
30
ab
abcdefg

Problems communicating between C and Python programs

I am trying to implement a UDP communication protocol between a C program and a python program. The C program has a structure that it sends through the UDP port (tx_port) as binary data. This program also listens on another port (rx_port) for any received data, and then prints the received binary output to the screen.
The python program listens on tx_port and unpacks the received data and prints it to the screen. Then it repacks the data and sends it back through UDP port (rx_port).
Here are the C and Python programs that I used.
C program
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include <arpa/inet.h>
#include <sys/socket.h>
#include <unistd.h>
#include <pthread.h>
#define BUFLEN 4096
#define RX_PORT 8888
#define TX_PORT 8889
// Structure data
struct data {
long frame_number;
double time;
} tx_data, rx_data;
int dlen = sizeof(tx_data);
struct sockaddr_in si_me, si_other;
int tx_soc;
int slen = sizeof(si_other);
int recv_len;
char* buf;
pthread_t rx_thread;
void* receiver_thread(void *arg)
{
int i =0;
while (1) {
recv_len = recvfrom(tx_soc, buf, sizeof(rx_data), 0, (struct sockaddr *) &si_other, &slen);
printf("\nReceived data : %d\n", recv_len);
for (i = 0; i < recv_len; i++) {
printf("%x ", buf[i]);
}
printf("\n");
fflush(stdout);
};
}
void data_init(void) {
tx_data.frame_number = 0;
tx_data.time = 0;
};
int main(void)
{
// Initialize data
data_init();
//create a UDP socket
if ((tx_soc=socket(AF_INET, SOCK_DGRAM, IPPROTO_UDP)) == -1)
{
printf("Socket error!");
exit(0);
}
// zero out the structure
memset((char *) &si_me, 0, sizeof(si_other));
memset((char *) &si_other, 0, sizeof(si_other));
// Host socket address
si_me.sin_family = AF_INET;
si_me.sin_port = htons(RX_PORT);
si_me.sin_addr.s_addr = htonl(INADDR_ANY);
// Remote socket address
si_other.sin_family = AF_INET;
si_other.sin_port = htons(TX_PORT);
si_other.sin_addr.s_addr = htonl(INADDR_ANY);
//bind sockets to the ports
if( bind(tx_soc, (struct sockaddr*)&si_me, sizeof(si_me) ) == -1)
{
printf("Binding error!");
}
// Start reader thread.
if (pthread_create(&rx_thread, NULL, &receiver_thread, NULL) != 0) {
printf("\ncan't create thread");
}
//keep listening for data
while(1)
{
// Allocate memory for receive buffer.
buf = (char*) malloc(sizeof(rx_data));
// Update data value.
tx_data.frame_number++;
printf("\nFrame numner: %ld", tx_data.frame_number);
fflush(stdout);
// Send data.
if (sendto(tx_soc, (char*)&tx_data, dlen, 0, (struct sockaddr*) &si_other, slen) == -1)
{
printf("Sending error!");
}
sleep(1);
}
close(tx_soc);
return 0;
}
Python program
from twisted.internet.protocol import DatagramProtocol
from twisted.internet import reactor
import struct
# Packet format string
packet_fmt = ''.join(['i', # Frame number
'd', # Frame time stamp
])
s = struct.Struct(packet_fmt)
class Echo(DatagramProtocol):
def datagramReceived(self, data, (host, port)):
new_data = s.unpack(data)
print new_data
echo_data = s.pack(*new_data)
self.transport.write(echo_data, (host, port))
reactor.listenUDP(8889, Echo())
reactor.run()
When I execute the two programs, I am able to receive data on both sides. I am able to unpack data in python, print it, repack and send it.
But on the C side, the received data does not match the sent data. I have checked on the python side to make sure the repacked data matches the original data.
Here is a sample output from the C and Python programs. I started the python programs first, and then the C program.
What is the mistake I might be making?

How to execute Python script from CreateProcess in C on Windows?

I have managed to get C code calling Python scripts happily on Unix using PIPES within the C code. I now need to do the same on Windows.
Essentially I would like to write scripts in different scripting languages like Python / Lua etc on Windows and be able to execute them using STDIN / STDOUT etc.
I have been looking at the "CreateProcess" call at:
http://msdn.microsoft.com/en-us/library/ms682425(VS.85).aspx
and although I can get it to work with a "child written in C", I cannot get it to call a Python script.
Below is the "parent / sender code" on my windows box:
#include<windows.h>
#include <stdio.h>
#include <stdlib.h>
#pragma comment(lib, "User32.lib")
void DisplayError(char *pszAPI);
void readFromPipe(HANDLE hPipeRead);
void createChildProcess(char *commandLine,
HANDLE hChildStdOut,
HANDLE hChildStdIn,
HANDLE hChildStdErr);
DWORD WINAPI writeToPipe(LPVOID lpvThreadParam);
HANDLE hChildProcess = NULL;
HANDLE hStdIn = NULL;
BOOL bRunThread = TRUE;
char *inputStream;
int main(int argc, char *argv[]){
HANDLE hOutputReadTmp,hOutputRead,hOutputWrite;
HANDLE hInputWriteTmp,hInputRead,hInputWrite;
HANDLE hErrorWrite;
HANDLE hThread;
DWORD ThreadId;
SECURITY_ATTRIBUTES sa;
int streamLen;
sa.nLength= sizeof(SECURITY_ATTRIBUTES);
sa.lpSecurityDescriptor = NULL;
sa.bInheritHandle = TRUE;
if (!CreatePipe(&hOutputReadTmp,&hOutputWrite,&sa,0))
return 1;
if (!DuplicateHandle(GetCurrentProcess(),hOutputWrite,
GetCurrentProcess(),&hErrorWrite,0,
TRUE,DUPLICATE_SAME_ACCESS))
return 1;
if (!CreatePipe(&hInputRead,&hInputWriteTmp,&sa,0))
return 1;
if (!DuplicateHandle(GetCurrentProcess(),hOutputReadTmp,
GetCurrentProcess(),
&hOutputRead,
0,FALSE,
DUPLICATE_SAME_ACCESS))
return 1;
if (!DuplicateHandle(GetCurrentProcess(),hInputWriteTmp,
GetCurrentProcess(),
&hInputWrite,
0,FALSE,
DUPLICATE_SAME_ACCESS))
return 1;
if (!CloseHandle(hOutputReadTmp)) return 1;;
if (!CloseHandle(hInputWriteTmp)) return 1;;
if ( (hStdIn = GetStdHandle(STD_INPUT_HANDLE)) == INVALID_HANDLE_VALUE )
return 1;
if (argc == 2){
createChildProcess(argv[1], hOutputWrite,hInputRead,hErrorWrite);
}else{
puts("No process name / input stream specified\n");
return 1;
}
if (!CloseHandle(hOutputWrite)) return 1;;
if (!CloseHandle(hInputRead )) return 1;;
if (!CloseHandle(hErrorWrite)) return 1;;
hThread = CreateThread(NULL,0,writeToPipe,
(LPVOID)hInputWrite,0,&ThreadId);
if (hThread == NULL)
return 1;;
readFromPipe(hOutputRead);
if (!CloseHandle(hStdIn))
return 1;
bRunThread = FALSE;
if (WaitForSingleObject(hThread,INFINITE) == WAIT_FAILED)
return 1;;
if (!CloseHandle(hOutputRead)) return 1;;
if (!CloseHandle(hInputWrite)) return 1;;
}
void createChildProcess(char *commandLine,
HANDLE hChildStdOut,
HANDLE hChildStdIn,
HANDLE hChildStdErr){
PROCESS_INFORMATION pi;
STARTUPINFO si;
ZeroMemory(&si,sizeof(STARTUPINFO));
si.cb = sizeof(STARTUPINFO);
si.dwFlags = STARTF_USESTDHANDLES;
si.hStdOutput = hChildStdOut;
si.hStdInput = hChildStdIn;
si.hStdError = hChildStdErr;
if (!CreateProcess(NULL,commandLine,NULL,NULL,TRUE,
NULL,NULL,NULL,&si,&pi))
hChildProcess = pi.hProcess;
if (!CloseHandle(pi.hThread)) return 1;;
}
void readFromPipe(HANDLE hPipeRead)
{
CHAR lpBuffer[256];
DWORD nBytesRead;
DWORD nCharsWritten;
while(TRUE)
{
if (!ReadFile(hPipeRead,lpBuffer,sizeof(lpBuffer),
&nBytesRead,NULL) || !nBytesRead)
{
if (GetLastError() == ERROR_BROKEN_PIPE)
break; // pipe done - normal exit path.
else
return 1; // Something bad happened.
}
if (!WriteConsole(GetStdHandle(STD_OUTPUT_HANDLE),lpBuffer,
nBytesRead,&nCharsWritten,NULL))
return 1;;
}
}
DWORD WINAPI writeToPipe(LPVOID lpvThreadParam)
{
CHAR read_buff[256];
DWORD nBytesRead,nBytesWrote;
HANDLE hPipeWrite = (HANDLE)lpvThreadParam;
while (bRunThread){
nBytesRead = 21;
strncpy(read_buff, "hello from the paren\n",21);
read_buff[nBytesRead] = '\0';
if (!WriteFile(hPipeWrite,read_buff,nBytesRead,&nBytesWrote,NULL)){
if (GetLastError() == ERROR_NO_DATA)
break; //Pipe was closed (normal exit path).
else
return 1;;
}
}
return 1;
}
Quite a bit of the above code is "hardcoded" just for testing purposes...essentially I passing some text like "hello from the paren" to be sent to a "child.exe"....
Here is the code for the child.c...a simple ECHO of what is sent to it
#include<windows.h>
#include<stdio.h>
#include<string.h>
void main (){
CHAR szInput[1024];
ZeroMemory(szInput,1024);
gets(szInput);
puts(szInput);
fflush(NULL);
}
To run the app I send "CallSubProcess.exe Child.exe" and it works 100%
Next I want to change "child.c" to be a PYTHON SCRIPT...
import sys
if __name__ == "__main__":
inStream = sys.stdin.read()
outStream = inStream
sys.stdout.write(outStream)
sys.stdout.flush()
So how can I change the CreateProcess call to execute this script?
if (!CreateProcess("C:\\Python26\\python.exe", "echo.py",NULL, NULL,FALSE, 0,NULL,NULL,&si,&pi)){
But it never works.
Any ideas how I can get this to work? Any help will be greatly appreciated.

My application posts a string to a python script, and the python script posts the string back to the c
application. It works well.
//c code
#pragma comment(lib, "json_vc71_libmtd.lib")
#include <windows.h>
#include <iostream>
#include <io.h>
#include "./json/json.h"
using namespace std;
DWORD WINAPI threadproc(PVOID pParam);
HANDLE hRead, hWrite, hRead1, hWrite1;
int main()
{
SECURITY_ATTRIBUTES sa;
sa.nLength = sizeof(SECURITY_ATTRIBUTES);
sa.lpSecurityDescriptor = NULL;
sa.bInheritHandle = TRUE;
if (!CreatePipe(&hRead, &hWrite, &sa, 0)){
::MessageBox(NULL, L"can't create pipe", L"error", MB_OK);
return -1;
}
if (!CreatePipe(&hRead1, &hWrite1, &sa, 0)){
::MessageBox(NULL, L"can't create pipe1", L"error", MB_OK);
return -1;
}
STARTUPINFO si;
PROCESS_INFORMATION pi;
GetStartupInfo(&si);
si.cb = sizeof(STARTUPINFO);
si.hStdError = hWrite;
si.hStdOutput = hWrite;
si.hStdInput = hRead1;
si.wShowWindow = SW_SHOW;
si.dwFlags = STARTF_USESHOWWINDOW | STARTF_USESTDHANDLES;
WCHAR szCmdLine[] = L"\"D:\\tools\\python\\python.exe\" D:\\code\\test\\pipeCallCore\\pipeCallCore\\json_wraper.py";
if (!CreateProcess(NULL, szCmdLine, NULL, NULL, TRUE, CREATE_NEW_CONSOLE, NULL, NULL, &si, &pi)){
::MessageBox(NULL, L"can't create process", L"error", MB_OK);
return -1;
}
CloseHandle(hWrite);
CloseHandle(hRead1);
const int cBufferSize = 4096;
char buffer[cBufferSize] = {0};
DWORD bytes;
int i = 0;
while (true){
cout << "come !" << endl;
ZeroMemory(buffer, sizeof(buffer));
sprintf(buffer, "{\"write\":%d}\n", i ++);
if (NULL == WriteFile(hWrite1, buffer, strlen(buffer), &bytes, NULL)){
::MessageBox(NULL, L"write file failed!", L"error", MB_OK);
break;
}
ZeroMemory(buffer, sizeof(buffer));
if (NULL == ReadFile(hRead, buffer, cBufferSize - 1, &bytes, NULL)){
::MessageBox(NULL, L"readfile failed", L"error", MB_OK);
return -1;
}
cout <<"yes " << buffer << endl;
Sleep(2000);
}
return 0;
}
//python code
#!/usr/bin/env python
# -*- coding: utf-8 -*-
import sys
while True:
try:
s = sys.stdin.readline()
sys.stdout.write(s)
sys.stdout.flush()
except EOFError, KeyboardInterrupt:
break

Maybe
if (!CreateProcess("C:\\Python26\\python.exe",
"echo.py 'hello from parent'",
NULL, NULL, FALSE, 0, NULL, NULL, &si, &pi)) {

CreateProcess is kind of tricky to use.
From the MSDN documentation:
If both lpApplicationName and lpCommandLine are non-NULL, ... lpApplicationName specifies the module to execute, and ... lpCommandLine specifies the command line.... Console processes written in C can use the argc and argv arguments to parse the command line. Because argv[0] is the module name, C programmers generally repeat the module name as the first token in the command line.
To avoid the weirdness, I recommend always passing NULL for the first argument and to pass the full command-line as the second:
CreateProcess(NULL, "\"C:\\Python26\\python.exe\" echo.py", ...);