cgi.c

boa：Linux系统中的轻量级Web服务

     * into words, according to the following rules:

       search-string = search-word *( "+" search-word )
       search-word   = 1*schar
       schar         = xunreserved | escaped | xreserved
       xunreserved   = alpha | digit | xsafe | extra
       xsafe         = "$" | "-" | "_" | "."
       xreserved     = ";" | "/" | "?" | ":" | "@" | "&"

       After parsing, each word is URL-decoded, optionally encoded in a system
       defined manner, and then the argument list
       is set to the list of words.


      Thus, schar is alpha|digit|"$"|"-"|"_"|"."|";"|"/"|"?"|":"|"@"|"&"

      As of this writing, escape.pl escapes the following chars:

       "-", "_", ".", "!", "~", "*", "'", "(", ")",
       "0".."9", "A".."Z", "a".."z",
       ";", "/", "?", ":", "@", "&", "=", "+", "\$", ","

      Which therefore means
       "=", "+", "~", "!", "*", "'", "(", ")", ","
       are *not* escaped and should be?
      Wait, we don't do any escaping, and nor should we.
      According to the RFC draft, we unescape and then re-escape
      in a "system defined manner" (here: none).

      The CGI/1.1 draft (03, latest is 1999???) is very unclear here.

      I am using the latest published RFC, 2396, for what does and does
      not need escaping.

      Since boa builds the argument list and does not call /bin/sh,
      (boa uses execve for CGI)
     */

    if (q && !strchr(q, '=')) {
        /* we have an 'index' style */
        q = strdup(q);
        if (!q) {
            WARN("unable to strdup 'q' in create_argv!");
        }
        for (aargc = 1; q && (aargc < CGI_ARGC_MAX);) {
            r = q;
            /* for an index-style CGI, + is used to seperate arguments
             * an escaped '+' is of no concern to us
             */
            if ((p = strchr(q, '+'))) {
                *p = '\0';
                q = p + 1;
            } else {
                q = NULL;
            }
            if (unescape_uri(r, NULL)) {
                /* printf("parameter %d: %s\n",aargc,r); */
                aargv[aargc++] = r;
            }
        }
        aargv[aargc] = NULL;
    } else {
        aargv[1] = NULL;
    }
}

/*
 * Name: init_cgi
 *
 * Description: Called for GET/POST requests that refer to ScriptAlias
 * directories or application/x-httpd-cgi files.  Ties stdout to socket,
 * stdin to data if POST, and execs CGI.
 * stderr remains tied to our log file; is this good?
 *
 * Returns:
 * 0 - error or NPH, either way the socket is closed
 * 1 - success
 */

int init_cgi(request * req)
{
    int child_pid;
    int pipes[2];
    int use_pipes = 0;

    SQUASH_KA(req);

    if (req->is_cgi) {
        if (complete_env(req) == 0) {
            return 0;
        }
    }
#ifdef FASCIST_LOGGING
    {
        int i;
        for (i = 0; i < req->cgi_env_index; ++i)
            fprintf(stderr, "%s - environment variable for cgi: \"%s\"\n",
                    __FILE__, req->cgi_env[i]);
    }
#endif

    if (req->is_cgi == CGI || 1) {
        use_pipes = 1;
        if (pipe(pipes) == -1) {
            log_error_time();
            perror("pipe");
            return 0;
        }

        /* set the read end of the socket to non-blocking */
        if (set_nonblock_fd(pipes[0]) == -1) {
            log_error_time();
            perror("cgi-fcntl");
            close(pipes[0]);
            close(pipes[1]);
            return 0;
        }
    }

    child_pid = fork();
    switch(child_pid) {
    case -1:
        /* fork unsuccessful */
        log_error_time();
        perror("fork");

        if (use_pipes) {
            close(pipes[0]);
            close(pipes[1]);
        }
        send_r_error(req);
        /* FIXME: There is aproblem here. send_r_error would work
           for NPH and CGI, but not for GUNZIP.  Fix that. */
        /* i'd like to send_r_error, but.... */
        return 0;
        break;
    case 0:
        /* child */
        if (req->is_cgi == CGI || req->is_cgi == NPH) {
            char *foo = strdup(req->pathname);
            char *c;

            if (!foo) {
                WARN("unable to strdup pathname for req->pathname");
                _exit(1);
            }
            c = strrchr(foo, '/');
            if (c) {
                ++c;
                *c = '\0';
            } else {
                /* we have a serious problem */
                log_error_time();
                perror("chdir");
                if (use_pipes)
                    close(pipes[1]);
                _exit(1);
            }
            if (chdir(foo) != 0) {
                log_error_time();
                perror("chdir");
                if (use_pipes)
                    close(pipes[1]);
                _exit(1);
            }
        }
        if (use_pipes) {
            close(pipes[0]);
            /* tie cgi's STDOUT to it's write end of pipe */
            if (dup2(pipes[1], STDOUT_FILENO) == -1) {
                log_error_time();
                perror("dup2 - pipes");
                close(pipes[1]);
                _exit(1);
            }
            close(pipes[1]);
            if (set_block_fd(STDOUT_FILENO) == -1) {
                log_error_time();
                perror("cgi-fcntl");
                _exit(1);
            }
        } else {
            /* tie stdout to socket */
            if (dup2(req->fd, STDOUT_FILENO) == -1) {
                log_error_time();
                perror("dup2 - fd");
                _exit(1);
            }
            /* Switch socket flags back to blocking */
            if (set_block_fd(req->fd) == -1) {
                log_error_time();
                perror("cgi-fcntl");
                _exit(1);
            }
        }
        /* tie post_data_fd to POST stdin */
        if (req->method == M_POST) { /* tie stdin to file */
            lseek(req->post_data_fd, SEEK_SET, 0);
            dup2(req->post_data_fd, STDIN_FILENO);
            close(req->post_data_fd);
        }
        /* Close access log, so CGI program can't scribble
         * where it shouldn't
         */
        close_access_log();

        /*
         * tie STDERR to cgi_log_fd
         * cgi_log_fd will automatically close, close-on-exec rocks!
         * if we don't tied STDERR (current log_error) to cgi_log_fd,
         *  then we ought to close it.
         */
        if (!cgi_log_fd)
            dup2(devnullfd, STDERR_FILENO);
        else
            dup2(cgi_log_fd, STDERR_FILENO);

        if (req->is_cgi) {
            char *aargv[CGI_ARGC_MAX + 1];
            create_argv(req, aargv);
            execve(req->pathname, aargv, req->cgi_env);
        } else {
            if (req->pathname[strlen(req->pathname) - 1] == '/')
                execl(dirmaker, dirmaker, req->pathname, req->request_uri,
                      NULL);
#ifdef GUNZIP
            else
                execl(GUNZIP, GUNZIP, "--stdout", "--decompress",
                      req->pathname, NULL);
#endif
        }
        /* execve failed */
        WARN(req->pathname);
        _exit(1);
        break;

    default:
        /* parent */
        /* if here, fork was successful */
        if (verbose_cgi_logs) {
            log_error_time();
            fprintf(stderr, "Forked child \"%s\" pid %d\n",
                    req->pathname, child_pid);
        }

        if (req->method == M_POST) {
            close(req->post_data_fd); /* child closed it too */
            req->post_data_fd = 0;
        }

        /* NPH, GUNZIP, etc... all go straight to the fd */
        if (!use_pipes)
            return 0;

        close(pipes[1]);
        req->data_fd = pipes[0];

        req->status = PIPE_READ;
        if (req->is_cgi == CGI) {
            req->cgi_status = CGI_PARSE; /* got to parse cgi header */
            /* for cgi_header... I get half the buffer! */
            req->header_line = req->header_end =
                (req->buffer + BUFFER_SIZE / 2);
        } else {
            req->cgi_status = CGI_BUFFER;
            /* I get all the buffer! */
            req->header_line = req->header_end = req->buffer;
        }

        /* reset req->filepos for logging (it's used in pipe.c) */
        /* still don't know why req->filesize might be reset though */
        req->filepos = 0;
        break;
    }

    return 1;
}