Unix pipe - reading data from stdin in the child descriptor

问题

I'm trying to implement unix piping in c (i.e. execute ls | wc). I have found a related solution to my problem (C Unix Pipes Example) however, I am not sure why a specific portion of the solved code snippet works.

Here's the code:

/* Run WC. */
int filedes[2];
pipe(filedes);

/* Run LS. */
pid_t pid = fork();
if (pid == 0) {
    /* Set stdout to the input side of the pipe, and run 'ls'. */
    dup2(filedes[1], 1);
    char *argv[] = {"ls", NULL};
    execv("/bin/ls", argv);
 } else {
    /* Close the input side of the pipe, to prevent it staying open. */
     close(filedes[1]);
 }

 /* Run WC. */
 pid = fork();
 if (pid == 0) {
      dup2(filedes[0], 0);
      char *argv[] = {"wc", NULL};
      execv("/usr/bin/wc", argv);
  }

In the child process that executes the wc command, though it attaches stndin to a file descriptor, it seems that we are not explicitly reading the output produced by ls in the first child process. Thus, to me it seems that ls is run independently and wc is running independently as we not explicitly using the output of ls when executing wc. How then does this code work (i.e. it executes ls | wc)?

回答1:

The code shown just about works (it cuts a number of corners, but it works) because the forked children ensure that the the file descriptor that the executed process will write to (in the case of ls) and read from (in the case of wc) is the appropriate end of the pipe. You don't have to do any more; standard input is file descriptor 0, so wc with no (filename) arguments reads from standard input. ls always writes to standard output, file descriptor 1, unless it is writing an error message.

There are three processes in the code snippet; the parent process and two children, one from each fork(). The parent process should be closing both its ends of the pipe too; it only closes one.

In general, after you do a dup() or dup2() call on a pipe file descriptor, you should close both ends of the pipe. You get away with it here because ls generates data and terminates; you wouldn't in all circumstances.

The comment:

/* Set stdout to the input side of the pipe, and run 'ls'. */

is inaccurate; you're setting stdout to the output side of the pipe, not the input side.

You should have an error exit after the execv() calls; if they fail, they return, and the process can wreak havoc (for example, if the ls fails, you end up with two copies of wc running.

An SSCCE

Note the careful closing of both ends of the pipe in each of the processes. The parent process has no use for the pipe once it has launched both children. I left the code which closes filedes[1] early in place (but removed it from an explicit else block since the following code was also only executed if the else was executed). I might well have kept pairs of closes() in each of the three code paths where files need to be closed.

#include <stdio.h>
#include <stdlib.h>
#include <sys/wait.h>
#include <unistd.h>

int main(void)
{
    int filedes[2];
    int corpse;
    int status;
    pipe(filedes);

    /* Run LS. */
    pid_t pid = fork();
    if (pid == 0)
    {
        /* Set stdout to the output side of the pipe, and run 'ls'. */
        dup2(filedes[1], 1);
        close(filedes[1]);
        close(filedes[0]);
        char *argv[] = {"ls", NULL};
        execv("/bin/ls", argv);
        fprintf(stderr, "Failed to execute /bin/ls\n");
        exit(1);
    }
    /* Close the input side of the pipe, to prevent it staying open. */
    close(filedes[1]);

    /* Run WC. */
    pid = fork();
    if (pid == 0)
    {
        /* Set stdin to the input side of the pipe, and run 'wc'. */
        dup2(filedes[0], 0);
        close(filedes[0]);
        char *argv[] = {"wc", NULL};
        execv("/usr/bin/wc", argv);
        fprintf(stderr, "Failed to execute /usr/bin/wc\n");
        exit(1);
    }

    close(filedes[0]);

    while ((corpse = waitpid(-1, &status, 0)) > 0)
        printf("PID %d died 0x%.4X\n", corpse, status);
    return(0);

}

Example output:

$ ./pipes-14312939
      32      32     389
PID 75954 died 0x0000
PID 75955 died 0x0000
$

来源：https://stackoverflow.com/questions/14312939/unix-pipe-reading-data-from-stdin-in-the-child-descriptor

标签

unix

piping