Segmentation fault in parsing into linked list - c

Question

I have this program that parses and RSS feed into a linked list.
#include"util.h"
#include<stdio.h>
#include<stdlib.h>
#include<sys/stat.h>
#include<string.h>
void parse_tc(){
struct node *head = NULL;
char *bytes = 0;
struct stat st;
stat("techcrunch.txt", &st);
int size = st.st_size;
FILE *f = fopen("techcrunch.txt", "rb");
bytes = (char*)malloc(size + 1);
size_t nread = fread(bytes,1,size,f);
bytes[nread] = 0;
fclose(f);
struct node *temp = (struct node*)malloc(sizeof(struct node));
temp->position = 1;
printf("%d. ", temp->position);
char *a = title_parser_tc(bytes,temp);
head = temp;
for(int i = 2; i<21; i++){
temp = (struct node*)malloc(sizeof(struct node));
temp->position = i;
printf("%d. ", temp->position);
a = title_parser_tc(a, temp);
struct node* temp1 = head;
while(temp1->link != NULL)
{
temp1 = temp1->link;
}
temp1->link = temp;
}
free(bytes);
int holder = 0;
int check = 0;
do {
printf("Enter a number: ");
scanf("%d", &holder);
if(holder<1 || holder > 20){
puts("Invalid input");
check = 1;
}
else{
check = 0;
}
} while(check);
get_feed(holder, head);
}
char* title_parser_tc(char *bytes, struct node *temp){
char *ptr = strstr(bytes, "<title>");
if (ptr) {
ptr += 7;
char *ptr2 = strstr(ptr, "</title>");
if (ptr2) {
char* output = malloc(ptr2 - ptr + 1);
memcpy(output, ptr, ptr2 - ptr);
output[ptr2 - ptr] = 0;
if(strcmp(output,"TechCrunch")!=0){
temp->title = output;
puts(temp->title);
temp->link = NULL;free(output);
char *load = pubdate_parser_tc(ptr2, temp);
return load;
}
else{
char *load = title_parser_tc(ptr2, temp);
free(output);
return load;
}
}
}
return NULL;
}
char* pubdate_parser_tc(char *bytes, struct node *temp){
char *ptr = strstr(bytes, "<pubDate>");
if (ptr) {
ptr += 9;
char *ptr2 = strstr(ptr, "</pubDate>");
if (ptr2) {
char* output = malloc(ptr2 - ptr + 1);
memcpy(output, ptr, ptr2 - ptr);
output[ptr2 - ptr] = 0;
temp->pubdate = output;
free(output);
char *load = description_parser_tc(ptr2, temp);
return load;
}
}
return NULL;
}
char* description_parser_tc(char *bytes, struct node *temp){
char *ptr = strstr(bytes, "<description>");
if (ptr) {
ptr += 13;
char *ptr2 = strstr(ptr, "</description>");
if (ptr2){
char* output = malloc(ptr2 - ptr + 1);
memcpy(output, ptr, ptr2 - ptr);
output[ptr2 - ptr] = 0;
description_cleaner_tc(output, temp);
free(output);
char *load = url_parser_tc(ptr2, temp);
return load;
}
}
return NULL;
}
void description_cleaner_tc(char *bytes, struct node *temp){
char *ptr = strstr(bytes, "&nbsp;");
if (ptr) {
ptr += 10;
char *ptr2 = strstr(ptr, "<a ");
if (ptr2) {
char* output = malloc(ptr2 - ptr + 1);
memcpy(output, ptr, ptr2 - ptr);
output[ptr2 - ptr] = 0;
temp->description = output;
puts(temp->description);
free(output);
}
}
}
char* url_parser_tc(char *bytes, struct node *temp){
char *ptr = strstr(bytes, "href");
if (ptr) {
ptr += 6;
char *ptr2 = strstr(ptr, ">");
if (ptr2) {
char* output = (char*)malloc(ptr2 - ptr);
memcpy(output, ptr, ptr2 - ptr - 1);
output[ptr2 - ptr - 1] = 0;
temp->url = output;
puts(temp->pubdate);
puts(temp->url);
puts("");
free(output);
return ptr2;
}
}
return NULL;
}
My problem is that for this file textcrunch.txt, my program occurs a segmentation fault at around the 10th loop in parse_tc(). The program works for another file but this file gives me an error. Any solution?
The code is basically the same function repeated for different strings to be parsed.

Answer 1

Segmentation fault usually means dereferencing a null-pointer (or a pointer to uninitialized memory). If you are using GCC or Clang, you can recompile with the -g flag and run the resulting program through gdb:
gdb --args ....
r
bt
The first line starts gdb, supply your command as you would usually run it. r starts the run and bt gives a backtrace from the point your program stops with the segmentation fault. At least this gives you the location in the code the problem occurs. You could start futher debugging with print statements or add some defensive coding there.