I am writing a UNIX paste clone. However I keep getting "encountered a breakpoint" messages, but VS won't tell me on what line it happened.
#include <stdio.h>
#include <stdlib.h>
#define INITALLOC 16
#define STEP 8
int main(int argc, char *argv[])
{
if (horzmerge(argc - 1, argv + 1) == 0) {
perror("horzmerge");
return EXIT_FAILURE;
}
getchar();
return EXIT_SUCCESS;
}
int horzmerge(int nfiles, const char **filenames)
{
FILE **files;
char *line;
int i;
if ((files = malloc(nfiles * sizeof (FILE *))) == NULL)
return 0;
for (i = 0; i < nfiles; ++i)
if ((files[i] = fopen(filenames[i], "r")) == NULL)
return 0;
do {
for (i = 0; i < nfiles; ++i) {
if (getline(files[i], &line) == 0)
return 0;
fprintf(stdout, "%s", line);
free(line);
}
putchar('\n');
} while (!feof(files[0])); /* we can still get another line */
for (i = 0; i < nfiles; ++i)
fclose(files[i]);
free(files);
return 1;
}
int getline(FILE *fp, char **dynline)
{
size_t nalloced = INITALLOC;
int c, i;
if ((*dynline = calloc(INITALLOC, sizeof(char))) == NULL)
return 0;
for (i = 0; (c = getc(fp)) != EOF && c != '\n'; ++i) {
if (i == nalloced)
if ((*dynline = realloc(*dynline, nalloced += STEP)) == NULL)
return 0;
(*dynline)[i] = c;
}
(*dynline)[i] = '\0';
if (c == EOF)
return EOF;
return i;
}
I placed breakpoints, and saw that it was the free(line) statement in horzmerge. But sometimes the program runs fine. Sometimes it doesn't. Sometimes I get a "Heap corrupted" in getline. I've been working on this code for a week, still can't find the bug(s).
It looks to me like the line where you null-terminate the input string is capable of overrunning the buffer you calloced or realloced. That has the potential of corrupting your heap when you free that buffer.
Dont't forget to leave room for the null character at the end of the string when you allocate memory.
Null-terminated strings are like disco. They still suck forty years later.
Related
I'm trying to read the following file line by line into an array of strings where each line is an element of the array:
AATGC
ATGCC
GCCGT
CGTAC
GTACG
TACGT
ACGTA
CGTAC
GTACG
TACGA
ACGAA
My code is as follows:
void **get_genome(char *filename) {
FILE *file = fopen(filename, "r");
int c;
int line_count = 0;
int line_length = 0;
for (c = getc(file); c != EOF; c = getc(file)) {
if (c == '\n') line_count++;
else line_length++;
}
line_length /= line_count;
rewind(file);
char **genome = calloc(line_length * line_count, sizeof(char));
for (int i = 0; i < line_count; i++) {
genome[i] = calloc(line_length, sizeof(char));
fscanf(file, "%s\n", genome[i]);
}
printf("%d lines of %d length\n", line_count, line_length);
for (int i = 0; i < line_count; i++)
printf("%s\n", genome[i]);
}
However, for some reason I get garbage output for the first 2 elements of the array. The following is my output:
`NP��
�NP��
GCCGT
CGTAC
GTACG
TACGT
ACGTA
CGTAC
GTACG
TACGA
ACGAA
You seem to assume that all lines have the same line length. If such is the case, you still have some problems:
the memory for the row pointers is allocated incorrectly, it should be
char **genome = calloc(line_count, sizeof(char *));
or better and less error prone:
char **genome = calloc(line_count, sizeof(*genome));
the memory for each row should be one byte longer the the null terminator.
\n is the fscanf() format string matches any sequence of whitespace characters. It is redundant as %s skips those anyway.
it is safer to count items separated by white space to avoid miscounting the items if the file contains any blank characters.
you do not close file.
you do not return the genome at the end of the function
you do not check for errors.
Here is a modified version:
void **get_genome(const char *filename) {
FILE *file = fopen(filename, "r");
if (file == NULL)
return NULL;
int line_count = 1;
int item_count = 0;
int item_length = -1;
int length = 0;
int c;
while ((c = getc(file)) != EOF) {
if (isspace(c)) {
if (length == 0)
continue; // ignore subsequent whitespace
item_count++;
if (item_length < 0) {
item_length = length;
} else
if (item_length != length) {
printf("inconsistent item length on line %d\", line_count);
fclose(file);
return NULL;
}
length = 0;
} else {
length++;
}
}
if (length) {
printf("line %d truncated\n", line_count);
fclose(file);
return NULL;
}
rewind(file);
char **genome = calloc(item_count, sizeof(*genome));
if (genome == NULL) {
printf("out of memory\n");
fclose(file);
return NULL;
}
for (int i = 0; i < item_count; i++) {
genome[i] = calloc(item_length + 1, sizeof(*genome[i]));
if (genome[i] == NULL) {
while (i > 0) {
free(genome[i]);
}
free(genome);
printf("out of memory\n");
fclose(file);
return NULL;
}
fscanf(file, "%s", genome[i]);
}
fclose(file);
printf("%d items of %d length on %d lines\n",
item_count, item_length, line_count);
for (int i = 0; i < item_count; i++)
printf("%s\n", genome[i]);
return genome;
}
char **genome = calloc(line_length * line_count, sizeof(char));
must be
char **genome = calloc(line_count, sizeof(char*));
or more 'secure'
char **genome = calloc(line_count, sizeof(*genome));
in case you change the type of genome
else the allocated block if not enough long if you are in 64b because line_count is 5 rather than 8, so you write out of it with an undefined behavior
You also need to return genome at the end of the function
It was also possible to not count the number of lines and to use realloc to increment your array when reading the file
As I see the lines have the same length. Your function should inform the caller how many lines have been read. There is no need of reading the file twice. There is no need of calloc (which is more expensive function). Always check the result of the memory allocation functions.
Here is a bit different version of the function:
char **get_genome(char *filename, size_t *line_count) {
FILE *file = fopen(filename, "r");
int c;
size_t line_length = 0;
char **genome = NULL, **tmp;
*line_count = 0;
if(file)
{
while(1)
{
c = getc(file);
if( c == EOF || c == '\n') break;
line_length++;
}
rewind(file);
while(1)
{
char *line = malloc(line_length + 1);
if(line)
{
if(!fgets(line, line_length + 1, file))
{
free(line);
break;
}
line[line_length] = 0;
tmp = realloc(genome, (*line_count + 1) * sizeof(*genome));
if(tmp)
{
genome = tmp;
genome[*line_count] = line;
*line_count += 1;
}
else
{
// do some memory free magic
}
}
}
fclose(file);
}
return genome;
}
So I'm trying to create a function that takes in a text file, which contains a bunch of words separated by the newline character, and reads the text file into a char** array.
When I run this code in netbeans on windows, it works fine but if I run it in Linux, I get a segmentation fault error.
// globals
FILE *words_file;
char **dic;
int num_words = 0;
void read_to_array() {
words_file = fopen("words.txt", "r");
char *line = NULL;
int i = 0;
size_t len = 0;
dic = (char **)malloc(99999 * sizeof(char *));
// read dic to array
while (getline(&line, &len, words_file) != -1) {
dic[i] = (char*)malloc(len);
strcpy(dic[i], line);
// get rid of \n after word
if (dic[i][strlen(dic[i]) - 1] == '\n') {
dic[i][strlen(dic[i]) - 1] = '\0';
}
++i;
num_words++;
}
//printf("%s", dic[i][strlen(dic[i]) - 1]); //testing
fclose(words_file);
dic[i] = NULL;
}
What am I missing here?
There are some problems in your program that may cause the undefined behavior that you observe:
You do not test if the file was open successfully, causing undefined behavior if the file is not where you expect it or has a different name.
You do not limit the number of lines read into the array, causing undefined behavior if the file contains more than 99998 lines, which may be be the case in linux as /usr/share/dict/words has 139716 lines on my system, for example.
Your memory allocation scheme is suboptimal but correct: you should compute the length of the word and strip the newline before allocating the copy. As coded, you allocate too much memory. Yet you should free line before returning from read_to_array and you should avoid using global variables.
Here is a modified version:
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
char **read_to_array(const char *filename, int *countp) {
FILE *words_file;
char *line = NULL;
size_t line_size = 0;
char **dic = NULL;
int dic_size = 0;
int i = 0;
words_file = fopen(filename, "r");
if (words_file == NULL) {
fprintf(stderr, "cannot open dictionary file %s\n", filename);
return NULL;
}
dic_size = 99999;
dic = malloc(dic_size * sizeof(char *));
if (dic == NULL) {
fprintf(stderr, "cannot allocate dictionary array\n");
fclose(words_file);
return NULL;
}
// read dic to array
while (getline(&line, &line_size, words_file) != -1) {
size_t len = strlen(line);
/* strip the newline if any */
if (len > 0 && line[len - 1] == '\n') {
line[--len] = '\0';
}
if (i >= dic_size - 1) {
/* too many lines: should reallocate the dictionary */
fprintf(stderr, "too many lines\n");
break;
}
dic[i] = malloc(len + 1);
if (dic[i] == NULL) {
/* out of memory: report the error */
fprintf(stderr, "cannot allocate memory for line %d\n", i);
break;
}
strcpy(dic[i], line);
i++;
}
dic[i] = NULL;
*countp = i;
fclose(words_file);
free(line);
return dic;
}
int main(int argc, char **argv) {
const char *filename = (argc > 1) ? argv[1] : "words.txt";
int num_words;
char **dic = read_to_array(filename, &num_words);
if (dic != NULL) {
printf("dictionary loaded: %d lines\n", num_words);
while (num_words > 0)
free(dic[--num_words]);
free(dic);
}
return 0;
}
Output:
chqrlie> readdic /usr/share/dict/words
too many lines
dictionary loaded: 99998 lines
I've been trying to run this code on VS2017. The code is compiling and running, but not in the way I want it too. So, I try to use the debugger and it says:
Debug Assertion Failed!
Program:
File: minkernel\crts\ucrt\src\appcrt\stdio\fgets.cpp
Line:33
Expression: stream.valid()
From past questions I understood that it may happen because of mishandling the opening of files, but I think that my code does take care of it.
Any help would be greatly appreciated!
(my relevant code):
int main(int argc, char *argv[]) {
int i, count_commands, PC_A, lastLine;
int *PC = &PC_A;
FILE *memin;
FILE *memout;
FILE *regout;
FILE *trace;
FILE *count;
assert(argc == 6);
*PC = 0;
count_commands = 0;
//allocationg memory for registers content
char **regs = (char **)(malloc(sizeof(char *) * 16));
for (i = 0; i < 16; i++) {
regs[i] = (char *)(malloc(sizeof(char) * 9));
for (int j = 0; j < 8; j++) {
regs[i][j] = '0';
}
regs[i][8] = '\0';
}
//allocationg memory for the memory image we have
char **memory = (char **)(malloc(sizeof(char *) * 4096));
for (i = 0; i < 4096; i++) {
memory[i] = (char *)(malloc(sizeof(char) * 9));
memory[i][0] = '\0';
}
//load memin image into memory
char *line = (char *)malloc(sizeof(char) * 8);
memin = fopen(argv[1], "r");
if (memin != NULL) {
perror(strerror(errno));
}
int j = 0;
while ((line = fgets(line, 10, (FILE *)memin)) != NULL) {
strcpy(memory[j], line);
memory[j][8] = '\0';
j++;
}
After opening the file, in OP's code there is this check:
if (memin != NULL) {
perror(strerror(errno));
}
So, if the opening succeeded an error string is printed. In my implementation, it reports:
Success: Success
No action is taken if it fails to open the file.
When it comes to the actual reading of all the lines in the file, there are some other issues. A buffer (char array) named line of size 8 is dinamically allocated and passed to fgets:
while ((line = fgets(line, 10, (FILE *)memin)) != NULL) {
// ^^
Note that 10 is also passed, as size of the buffer, which is wrong, because it allows fgets to write out of the bounds of the allocated array.
Also, given OP's compiler is MSVC 2017, I assume this code is running on Windows, so chances are that in the file, the lines are terminated by a "\r\n" sequence, rather then a single '\n'. Even if OP is confident that each line is a 8 char string, fgets needs a buffer of at least size 8 + 3 (8 + '\r' + '\n' + '\0') to read them safely.
Consider how those suggestions are implemented in this snippet:
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <assert.h>
#define MEM_SIZE 1024u
#define LINE_SIZE 128u
#define STR_SIZE 8u
int main(int argc, char *argv[])
{
// Try to open the input file
if (argc < 2) {
fprintf(stderr, "Missing file name in command line.\n");
return EXIT_FAILURE;
}
FILE *memin = fopen(argv[1], "r");
if (memin == NULL) {
fprintf(stderr, "Unable to open file [%s].\n", argv[1]);
return EXIT_FAILURE;
}
// I'd use plain arrays to store the lines
char memory[MEM_SIZE][STR_SIZE + 1] = {{'\0'}};
char line[LINE_SIZE] = {'\0'};
size_t count = 0;
while ( count < MEM_SIZE && fgets(line, LINE_SIZE, memin) ) {
size_t length = strcspn(line, "\r\n");
if (length > STR_SIZE) {
fprintf(stdout, "Warning, line too long: %zu.\n", count);
length = STR_SIZE;
}
memcpy(memory[count], line, length);
memory[count][STR_SIZE] = '\0';
++count;
}
for ( size_t i = 0; i < count; ++i ) {
printf("[%s]\n", memory[i]);
}
}
I want to read the txt line by line and store it to the array;
I am like success to store line by line in array;
I use printf("%s", loadtext[i]) with i= 0,1,2,3-20 respectively to check that things in stored in array;
but I realize that the for loop had done 1912 times when I
type printf("%d", i); right behind the for loops,
Suppose my txt is store like this:
I am a jerk
I am a noob
I am an idiot
I am done
I have another program to add new lines to the text file while this program is running.
How can I detect I am done or the new line added later is the last line to not allow the for loop do so many times?
Here is my code
#include <stdio.h>
#include <stdlib.h>
#include <math.h>
#include <ctype.h>
#include <string.h>
int main(){
FILE *fp = fopen("abc.txt","r");
int i = 0,j=0,k=0;
char ch[10000];
char loadtext[100][100];
for (i=0; ch[i] != EOF; i++){
ch[i] = fgetc(fp);
if (ch[i] != '\n'){
loadtext[j][k] = ch[i];
k++;
}
if(ch[i] == '\n'){
loadtext[j][k] = ch[i];
k=0;
j++;
}
}
printf("%s", loadtext[0]);
printf("%s", loadtext[1]);
printf("%s", loadtext[2]);
fclose(fp);
return 0;
}
To read a whole file into an "array" of pointers to char which represent the lines, you can do:
#include <stddef.h> // size_t
#include <stdlib.h> // EXIT_SUCCESS, EXIT_FAILURE
#include <stdio.h> // malloc(), realloc(), free(), fgets()
#include <string.h> // strlen(), strcpy()
enum { BUFFER_SIZE = 30 }; // whatever line length you suspect the input file to be + 1
int main(void)
{
char const *filename = "test.txt";
FILE *is = fopen(filename, "r");
if (!is) {
fprintf(stderr, "Couldn't open \"%s\" for reading :(\n\n", filename);
return EXIT_FAILURE;
}
int result = EXIT_SUCCESS; // assume a happy ending
char buffer[BUFFER_SIZE];
size_t num_lines = 0;
char **lines = NULL;
while (fgets(buffer, sizeof(buffer), is)) {
++num_lines;
char **temp = realloc(lines, num_lines * sizeof(*lines));
if (!temp) {
fputs("Not enough memory :(\n\n", stderr);
fclose(is);
result = EXIT_FAILURE;
goto cleanup;
}
lines = temp;
size_t length = strlen(buffer);
length = strlen(buffer);
// remove a trailing newline if any:
if (length && buffer[length - 1] == '\n')
buffer[--length] = '\0';
size_t line_length = length;
lines[num_lines - 1] = malloc((length + 1) * sizeof(*lines));
if (!lines[num_lines - 1]) {
fputs("Not enough memory :(\n\n", stderr);
fclose(is);
result = EXIT_FAILURE;
goto cleanup;
}
strcpy(lines[num_lines - 1], buffer);
// as long as the buffer has been filled completely by the previous
// call to fgets() and a next call to fgets() also succeeds:
while (length == BUFFER_SIZE - 1 && fgets(buffer, sizeof(buffer), is)) {
length = strlen(buffer);
// remove a trailing newline if any:
if (length && buffer[length - 1] == '\n')
buffer[--length] = '\0';
char *temp = realloc(lines[num_lines - 1], line_length + length + 1);
if (!temp) {
fputs("Not enough memory :(\n\n", stderr);
fclose(is);
result = EXIT_FAILURE;
goto cleanup;
}
lines[num_lines - 1] = temp;
strcpy(lines[num_lines - 1] + line_length, buffer);
line_length += length;
}
}
fclose(is);
// use lines:
for (size_t i = 0; i < num_lines; ++i)
puts(lines[i]);
cleanup:
for (size_t i = 0; i < num_lines; ++i)
free(lines[i]);
free(lines);
return result;
}
Using only a fixed-size two-dimensional array and fgetc()*):
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
enum { MAX_LINES = 100, MAX_LINE_LENGTH = 100 };
int main(void)
{
char const *filename = "test.txt";
FILE *is = fopen(filename, "r");
if (!is) {
fprintf(stderr, "Couldn't open \"%s\" for reading :(\n\n", filename);
return EXIT_FAILURE;
}
char text[MAX_LINES][MAX_LINE_LENGTH + 1] = { 0 }; // zero-initialize the array
// so we don't have to care
size_t num_lines = 0; // about zero terminating
size_t current_column = 0; // every line
int ch;
// as long as we are still inside the bounds of the fixed size array
// and fgetc() doesn't return EOF
while (num_lines < MAX_LINES && current_column < MAX_LINE_LENGTH &&
(ch = fgetc(is)) != EOF)
{
if (ch == '\n') { // "move" num_lines and current_column to the next
++num_lines; // line.
current_column = 0;
continue;
}
text[num_lines][current_column++] = ch;
}
if (ch != EOF) {
fputs("The file is too big :(\n\n", stderr);
return EXIT_FAILURE;
}
for (size_t i = 0; i <= num_lines; ++i)
puts(text[i]);
}
*) Could be done with fgets() too.
to read the txt line by line and store it to the array
Code has various problems:
ch[i] != EOF accesses ch[i] before it is assigned! Undefined Behavior (UB).
char can be signed or unsigned. EOF is some negative. When char is unsigned the below is an infinite loop. When char is signed, ch[i] != EOF could exit the loop early.
printf("%s", loadtext[0]); attempts to print loadtext[0] as if it was a string. But lacking a certain null character, and thus not a string, this leading to more UB.
Lack of buffer index checks: loadtext[j][k] = ch[i]; k++; may increment k to beyond 100. Similar weak code with i.
Code may fail to open the file and no check provided.
Instead save the result from fgetc() as an int, test for for EOF, test for '\n' and append a null character to form strings.
Some ideas for alternate code:
#define LINES_N 100
#define LINE_SIZE 100
int main(void) {
FILE *fp = fopen("abc.txt", "r");
if (fp) {
// char ch[10000];
char loadtext[LINES_N][LINE_SIZE];
int ch_index = 0;
int line_count = 0;
int character;
int previous_character = '\n';
while ((character = fgetc(fp)) != EOF) {
// Starting a new line?
if (previous_character == '\n') {
if (++line_count > LINES_N) {
printf("Too many lines\n");
return EXIT_FAILURE;
}
}
loadtext[line_count - 1][ch_index++] = (char) character;
loadtext[line_count - 1][ch_index] = '\0';
if (ch_index + 1 >= LINE_SIZE) {
printf("Too long a line\n");
return EXIT_FAILURE;
}
previous_character = character;
}
if (line_count >= 1) printf("%s", loadtext[0]);
if (line_count >= 2) printf("%s", loadtext[1]);
if (line_count >= 3) printf("%s", loadtext[2]);
fclose(fp);
}
return 0;
}
I try to read in a file multiple times instead of just once.
While trying that I got alot of segementation faults. The part of the program with the while loop looks like this:
#include <stdio.h>
#include <unistd.h>
#include <stdlib.h>
#include <string.h>
/* General use buffer */
#define STRLEN 8196
char string[STRLEN];
int lines = 1024;
char **line;
int linemax;
int longest=0;
int main(){
int len,i;
int zwei = 1;
FILE * fp;
char *s;
int debug = 0;
line=(char **)malloc(sizeof(char *) * 1024);
do{
if ( (fp = fopen("rhel7_160731_0606.nmon", "r")) == NULL) {
perror("failed to open file");
perror("fopen");
exit(75);
}
printf("where is the problem1,3\n");
for (i = 0; fgets(string, STRLEN, fp) != NULL; i++) {
if (i >= lines) {
lines += 1024;
line = (char **)realloc((void *)line, sizeof(char *) * lines);
}
if (string[strlen(string)-1] == '\n')
string[strlen(string)-1] = 0;
if (string[strlen(string)-1] == '\r')
string[strlen(string)-1] = 0;
if (string[strlen(string)-1] == ' ')
string[strlen(string)-1] = 0;
if (string[strlen(string)-1] == ',')
string[strlen(string)-1] = 0;
len = strlen(string) + 1;
if (len > longest)
longest = len;
s = malloc(len);
strcpy(s, string);
line[i] = (char *)s;
}
linemax = i;
lines = i;
if (debug)
for (i = 0; i < linemax; i++)
printf("line %d lastline %s\n", i, line[i-1]);
zwei++;
}while(zwei<4);
return 0;
}
It hangs nothing or ends with a segmentation fault.
You seem to have forgotten to allocate memory for line. It fails here: line[i] = (char *)s. I think you need to set lines to zero, since you reallocate line only if your iterator i gets grower than lines.
Also, fix this: while(zwei > 4) to while(zwei < 4). And, you need to free the memory you allocate - because you store all the pointers in line, it is not going to be complicated - just one loop.