i changed chunk.c file which is coded with high-level i/o function to low-level system calls.
this is chunk.c coded with high-level i/o function
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#define DEFAULT_PREFIX "x"
#define DEFAULT_CHUNK_SIZE 1000
#define ALPHABET_SIZE 26
#define MAX_DIGITS 2
void print_usage() {
printf("Usage: chunk [-l line_count | -w word_count | -c character_count] [-p prefix] [-s suffix] [-f filename.txt | < filename.txt]\n");
}
int main(int argc, char *argv[]) {
char *prefix = DEFAULT_PREFIX;
int chunk_size = DEFAULT_CHUNK_SIZE;
int suffix_start = 0;
char *filename = NULL;
// Parse command line arguments
int opt;
while ((opt = getopt(argc, argv, "l:w:c:p:s:f:")) != -1) {
switch (opt) {
case 'l':
chunk_size = atoi(optarg);
break;
case 'w':
// TODO: implement word count option
printf("Error: -w option not implemented\n");
print_usage();
return 1;
case 'c':
chunk_size = atoi(optarg);
break;
case 'p':
prefix = optarg;
break;
case 's':
suffix_start = atoi(optarg);
break;
case 'f':
filename = optarg;
break;
default:
print_usage();
return 1;
}
}
// Open input file
FILE *input_file = NULL;
input_file = fopen(filename, "r");
if (input_file == NULL) {
printf("Error: could not open file '%s'\n", filename);
return -1;
}
// Read input file and write output files
int line_count = 0;
int chunk_count = 0;
char suffix[MAX_DIGITS + 1];
suffix[MAX_DIGITS] = '\0';
FILE *output_file = NULL;
while (!feof(input_file)) {
if (line_count == 0) {
// Close previous output file
if (output_file != NULL) {
fclose(output_file);
output_file = NULL;
}
// Open new output file
snprintf(suffix, MAX_DIGITS + 1, "%0*d", MAX_DIGITS, suffix_start + chunk_count);
char *filename_buffer = (char *) malloc(strlen(prefix) + MAX_DIGITS + 1);
snprintf(filename_buffer, strlen(prefix) + MAX_DIGITS + 1, "%s%s", prefix, suffix);
output_file = fopen(filename_buffer, "w");
free(filename_buffer);
if (output_file == NULL) {
printf("Error: could not create output file '%s'\n", filename_buffer);
return -1;
}
chunk_count++;
}
// Read input line
char line_buffer[1024];
fgets(line_buffer, 1024, input_file);
if (feof(input_file)) {
break;
}
// Write output line
fputs(line_buffer, output_file);
line_count++;
if (line_count == chunk_size) {
line_count = 0;
}
}
// Close last output file
if (output_file != NULL) {
fclose(output_file);
}
}
and this is the changed one which coded with low-level system calls.
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#include <fcntl.h>
#include <errno.h>
#define DEFAULT_PREFIX "x"
#define DEFAULT_CHUNK_SIZE 1000
#define ALPHABET_SIZE 26
#define MAX_DIGITS 2
void print_usage() {
printf("Usage: chunk [-l line_count | -w word_count | -c character_count] [-p prefix] [-s suffix] [-f filename.txt | < filename.txt]\n");
}
int main(int argc, char *argv[]) {
char *prefix = DEFAULT_PREFIX;
int chunk_size = DEFAULT_CHUNK_SIZE;
int suffix_start = 0;
char *filename = NULL;
// Parse command line arguments
int opt;
while ((opt = getopt(argc, argv, "l:w:c:p:s:f:")) != -1) {
switch (opt) {
case 'l':
chunk_size = atoi(optarg);
break;
case 'w':
// TODO: implement word count option
printf("Error: -w option not implemented\n");
print_usage();
return 1;
case 'c':
chunk_size = atoi(optarg);
break;
case 'p':
prefix = optarg;
break;
case 's':
suffix_start = atoi(optarg);
break;
case 'f':
filename = optarg;
break;
default:
print_usage();
return 1;
}
}
// Open input file
int input_fd = STDIN_FILENO;
if (filename != NULL) {
input_fd = open(filename, O_RDONLY);
if (input_fd == -1) {
printf("Error: could not open file '%s': %s\n", filename, strerror(errno));
return -1;
}
}
// Read input file and write output files
int line_count = 0;
int chunk_count = 0;
char suffix[MAX_DIGITS + 1];
suffix[MAX_DIGITS] = '\0';
int output_fd = -1;
while (1) {
if (line_count == 0) {
// Close previous output file
if (output_fd != -1) {
close(output_fd);
output_fd = -1;
}
// Open new output file
snprintf(suffix, MAX_DIGITS + 1, "%0*d", MAX_DIGITS, suffix_start + chunk_count);
char *filename_buffer = (char *) malloc(strlen(prefix) + MAX_DIGITS + 1);
snprintf(filename_buffer, strlen(prefix) + MAX_DIGITS + 1, "%s%s", prefix, suffix);
output_fd = open(filename_buffer, O_CREAT | O_WRONLY | O_TRUNC, S_IRUSR | S_IWUSR);
free(filename_buffer);
if (output_fd == -1) {
printf("Error: could not create output file '%s': %s\n", filename_buffer, strerror(errno));
return -1;
}
chunk_count++;
}
// Read input line
char line_buffer[1024];
ssize_t bytes_read = read(input_fd, line_buffer, 1024);
if (bytes_read == -1) {
printf("Error: could not read input file: %s\n", strerror(errno));
return -1;
}
if (bytes_read == 0) {
break;
}
// Write output line
ssize_t bytes_written = write(output_fd, line_buffer, bytes_read);
if (bytes_written == -1) {
printf("Error: could not write output file: %s\n", strerror(errno));
return -1;
}
line_count++;
if (line_count == chunk_size) {
line_count = 0;
}
}
// Close last output file
if (output_fd != -1) {
close(output_fd);
}
}
both codes are not complete yet but anyway, the expected result ls
$ chunk -l 100 -f z_answer.jok.txt -p part- -s 00
$ echo $? # check exit status
0
$ wc *part* z_answer.jok.txt
100 669 4052 part-00
100 725 4221 part-01
100 551 3373 part-02
100 640 3763 part-03
100 588 3685 part-04
100 544 3468 part-05
90 473 3017 part-06
690 4190 25579 z_answer.jok.txt
1380 8380 51158 total
when i run the chunk.c which is coded with high-level i/o function, i can get a correct results like the above.
but when i run the chunk.c which is coded with low-level system call, then the result is like
$ chunk -l 100 -f z_answer.jok.txt -p part- -s 00
$ echo $? # check exit status
0
$ wc *part* z_answer.jok.txt
100 669 4052 part-00
means chunk.c file does not work properly. I thought the problem is in the while loop(or the condition), but still cannot find out the problem. what should i change to make it run properly?
