问题
The only part I am confused on thus far is how to set up execv with the first parameter as the current working directory. I've tried both "." and "~", neither are executing anything to the screen; same for "/." and "/~". I'm confused on how to have execv run something like this:
$ ./prog ls -t -al
And have it execute the commands after the program execution (which are stored into argv) in the current directory, or the same directory as the file is in (which will vary based on who is using it.)
My code:
#include <stdlib.h>
#include <stdio.h>
#include <unistd.h>
#include <sys/types.h>
void main(int argc, char *argv[])
{
int pid;
int count = 0;
char *argv2[argc+1];
for(count = 0; count < argc-1; count++){
argv2[count] = argv[count+1];
printf("Argv2: %s\n", argv2[count]); //just double checking
argv2[argc-1] = NULL;
}
pid = fork();
if(pid == 0){
printf("Child's PID is %d. Parent's PID is %d\n", (int)getpid, (int)getppid());
execv(".", argv2); //<---- confused here
}
else{
wait(pid);
exit(0);
}
}
Some sample output:
$ ./prog ls -t -al
Argv2: ls
Argv2: -t
Argv2: -al
Child's PID is 19194. Parent's PID is 19193
回答1:
I guess execv is what is required to be used. execvp is a lot nicer since it will look for commands in your PATH setting.
execv(".", argv2); //<---- confused here
...
#include <errno.h>
#include <string.h>
if ( execv(argv2[0],argv2) )
{
printf("execv failed with error %d %s\n",errno,strerror(errno));
return 254;
}
wait(pid);
...
pid_t wait_status = wait(&pid);
回答2:
The first argument, by convention, should point to the filename associated with the file being executed.
You want to exec ls
, so the first argument should be /bin/ls
,that means the code is
execv("/bin/ls", argv2);
You could try it
回答3:
/* main() returns int */
int main(int argc, char *argv[])
{
int pid;
pid = fork();
if(pid == 0){
printf("Child's PID is %d. Parent's PID is %d\n"
, (int)getpid, (int)getppid());
execv(argv[1], argv+1);
}
else{
wait(NULL);
exit(0);
}
return 0;
}
UPDATE: execv() needs the absolute path of the executable; for files in the current directory you would have to construct that path (eg via pwd()). If you want the executable to seached via the $PATH environment variable, you could use execvp(), which does all the seaching for you.
回答4:
The first arg to execv() is not the directory in which to run but rather the path to the file you want to exec. By default, your exec-ed program will run in the context of the caller's current directory.
回答5:
The first argument is the absolute path of the program you want to run; in other words, the first part of the command you type in a terminal. You can find where the program using the command whereis
. Also, execv
function can look up the path for you, which makes your code more portable.
回答6:
Code sample for your reference:
#include <stdio.h>
int main() {
int ret = fork();
if(ret == 0)
{
char *params[4] = {"/bin/ls", "-l",0}; //cmd params filled
int res = execv( "/bin/ls" , params); //parameters for cmd
//int res = execv( "/bin/ls" , NULL); //this is fail case (when child-exit with -1 status can be seen)
printf("\n child exiting (%d) .. \n", res); //on successful execution of cmd, this exit never appears
}
else
{
waitpid(ret,1,0);
printf("parent exiting\n");
}
return 1;
}
来源:https://stackoverflow.com/questions/12575826/using-execv-c-language-to-run-commands-from-a-linux-command-prompt