Using a consortium to send and receive floating-point numbers via serial

Here are two methods, one is to use pointers, one is to use the Union, time-valuable spectator can jump directly to the second methodThis is just a discussion of the processing of data before and after receiving, not discussing the specific sending process method One: pointer

UART Serial port has a disadvantage, that is, send and receive is a byte of a byte of the receive, if the number of floating points sent how to do ah?

Someone would say, that's a byte to send a byte. Well, I'll first define a double data32, a double type of 8 bytes, which means that data needs to split 8 blocks to send through the UART how? When you learn C, you must have read this sentence: The pointer is the essence of C language then use the pointer! How to use it? I don't know, there's only one variable, so just declare a pointer to the variable.

Double *= &data;

P will take 8 bytes as a whole so that p+1 cannot point to the next byte address so if I turn P strong into unsigned char *? Or we declare a unsigned char * pointing to this data

unsigned  char  *= (unsigned  char *) p;

Then the serial port is sent with a for loop is enough

 for (i=0;i<8; i++) {    Send (* (q+i));}

The receiver first uses an array to start receiving data from the low and then forcibly points to the array header using a double pointer. The following code is the process of simulating the data splitting combination before and after the serial port is sent VS2010
Method Two: ConsortiumMethod one is too cumbersome, is there a simple way? How simple is the use of a consortium (community)? Directly on the code

#include"stdafx.h"#defineMax_lenth 8Union u1{CharS[max_lenth]; DoubleD;}; Union u2{CharS[max_lenth]; DoubleD;}; int_tmain (intARGC, _tchar*argv[])    {U1 u1;    U2 U2; intI=0; U1.D=2.111; U2.D=3.00; printf ("u1.d =%lf\n", U1.D); printf ("u2.d =%lf\n", U2.D); printf ("Send Data ...");  for(i=0; i<max_lenth;i++) {U2.s[i]=U1.s[i]; } printf ("u2.d =%lf\n", U2.D);  while(GetChar () = ='Q'); return 0;}

Method One is simply wasting brain power!

VS2010 Results

Why would a consortium be so magical?because all its members have an offset of 0 relative to the base addressThis means that the double-precision floating-point address &u1.d is the same as the first address of the array u1.s "Hindsight": in fact, method one method and the idea of the consortium is the same as the use of the array to save the received data (the essence of the data is the memory of the binary), a pointer to a double point to the first address of the array, then we have the value of the symbol * will be the data intact read out Ah! Only the address of double in the Union is the first address of the array, do not need to be as strong as method one!

Using a consortium to send and receive floating-point numbers via serial