How do you resize a Lua 5.1 userdata object in C during runtime?
I would like to change the size of the NumArray structure described in Roberto Ierusalimschy's book Programming in Lua, 2nd edition, pp. 260, from the Lua 5.1 console.
My NumArray userdata can either store unsigned chars or lua_Numbers. I played around with calling the unmodified luaM_realloc_ function defined in lmem.c, but finally any call to C's realloc
(via l_alloc) function just returns NULL, so I get a not enough memory error message.
Would somebody please help me ?
--- lapi.c -------------------------------------------------------------------------------------------------
LUA_API void *agn_resizeud (lua_State *L, void *block, size_t osize, size_t nsize) {
Udata *u = (Udata *)luaM_realloc_(L, block, osize + sizeof(Udata), nsize + sizeof(Udata));
u->uv.len = nsize;
if (u == NULL)
luaL_error(L, "Error in " LUA_QS ": failed to allocate memory.", "API/agn_resizeud");
return u;
}
--- numarray.c ---------------------------------------------------------------------------------------------
#define checkarray(L, n) (NumArray *)luaL_checkudata(L, n, "numarray")
[...]
/* Auxiliary C function to finally call luaM_realloc_ via the Lua C API */
void *reallocud (lua_State *L, void *block, size_t o, size_t n, size_t sizeofelem, size_t sizeofnumarray) {
if (n + 1 > MAX_SIZET/sizeofelem)
luaL_error(L, "Error in " LUA_QS ": memory allocation error: block too big.", "(reallocud)", n);
else
return agn_resizeud(L, block, sizeofnumarray + o*sizeofelem, sizeofnumarray + n*sizeofelem);
}
/* function to resize a NumArray */
static int numarray_resize (lua_State *L) {
size_t i;
global_State *g = G(L);
NumArray *a = checkarray(L, 1);
int n = luaL_checkinteger(L, 2);
if (n < 1)
luaL_error(L, "Error in " LUA_QS ": new size %d is non-positive.", "numarray.resize", n);
if (n == a->size) { /* do nothing and do not complain */
lua_pushinteger(L, a->size);
return 1;
}
if (n > a->size) { /* extend */
a = reallocud(L, a, a->size, n, a->isnumber ? sizeof(lua_Number) : sizeof(char), sizeof(NumArray));
if (a->isnumber) {
for (i=a->size; i < n; i++) a->data.n[i] = 0;
} else {
for (i=a->size; i < n; i++) a->data.c[i] = 0;
}
} else { /* reduce */
if (a->isnumber) {
for (i=a->size - 1; i > n - 1; i--) a->data.n[i] = 0;
} else {
for (i=a->size - 1; i > n - 1; i--) a->data.c[i] = 0;
}
a = reallocud(L, a, a->size, n, a->isnumber ? sizeof(lua_Number) : sizeof(char), sizeof(NumArray));
}
a->size = n;
lua_pushnumber(L, a->size);
return 1;
}
[...]
typedef struct NumArray {
size_t size; /* number of slots */
char isnumber;
union data {
lua_Number n[1]; /* pointer to the various lua_Number values */
unsigned char c[1];
} data;
} NumArray;
There is no need to mess with the internals of Lua.
But you need to change how the userdata is created: instead of allocating the struct and the data in the same memory block, allocate them separately, using lua_newuserdata to allocate the struct and malloc to allocate the data.
Then, to resize the data part, just update the size in the struct and realloc the data part.
you have:
static int numarray_resize (lua_State *L) {
[...]
if (n > a->size) { /* extend */
a = realloc(L, a, a->size, n, a->isnumber ? sizeof(lua_Number) : sizeof(char), sizeof(NumArray));
if (a->isnumber) {
[...]
But realloc takes 2 arguments, you want reallocud.
reallocing L with size a is likely to be a giant number and L may not have ever been allocated by the alloc family so there a couple reasons to get NULL unexpectedly.
Related
The following code is trying to count word frequency in a document, by using hashset and vector.
#include <assert.h>
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
/** ==================================== VECTOR ======================================= */
typedef enum {
true, false
} bool;
typedef int (*VectorCompareFunction)(const void *elemAddr1, const void *elemAddr2);
typedef void (*VectorFreeFunction)(void *elemAddr);
typedef struct {
int elemSize; //how many byte for each element
int elemNum; //number of current element in vector
int capacity; //maximum number of element vector can hold
void *elems; //pointer to data memory
VectorFreeFunction freefn; //pointer to the function used to free each element
} vector;
/**
* Reallocate a new memory of twice of original size
* return 1 if reallocation success, otherwise return -1.
*/
static void DoubleMemory(vector *v) {
void *tmp = realloc(v->elems, v->capacity * v->elemSize * 2);
assert(tmp != NULL);
v->elems = tmp;
v->capacity *= 2;
}
/**
* Constructor
*/
void VectorNew(vector *v, int elemSize, VectorFreeFunction freefn, int initialAllocation) {
v->elems = malloc(initialAllocation * elemSize);
assert(v->elems != NULL);
v->elemSize = elemSize;
v->elemNum = 0;
v->capacity = initialAllocation;
v->freefn = freefn;
}
/**
* Appends a new element to the end of the specified vector.
*/
void VectorAppend(vector *v, const void *elemAddr) {
/* double size if neccessary */
if (v->elemNum == v->capacity) DoubleMemory(v);
memcpy((char *)v->elems + v->elemNum * v->elemSize, elemAddr, v->elemSize);
v->elemNum++;
}
/**
* Search the specified vector for an element whose contents match the element passed as the key.
*/
int VectorSearch(const vector *v, const void *key, VectorCompareFunction searchfn, int startIndex, bool isSorted) {
assert(key && searchfn);
if (v->elemNum == 0) return -1;
assert(startIndex >= 0 && startIndex < v->elemNum);
if (isSorted == true) {
/* binary search */
void *startAddr = (char *)v->elems + startIndex * v->elemSize;
int size = v->elemNum - startIndex;
void *resAddr = bsearch(key, startAddr, size, v->elemSize, searchfn);
return (resAddr != NULL)? ((char *)resAddr - (char *)v->elems) / v->elemSize : -1;
} else {
/* linear search */
for (int i = 0; i < v->elemNum; i++) {
if (searchfn((char *)v->elems + i * v->elemSize, key) == 0) {
return i;
}
}
return -1;
}
}
/**
* Overwrites the element at the specified position.
*/
void VectorReplace(vector *v, const void *elemAddr, int position) {
assert(position >= 0 && position < v->elemNum);
void *posAddr = (char *)v->elems + position * v->elemSize;
/* free the memory of old element first */
if (v->freefn != NULL) v->freefn(posAddr);
memcpy(posAddr, elemAddr, v->elemSize);
}
/** ==================================== HASHSET ======================================= */
typedef int (*HashSetHashFunction)(const void *elemAddr, int numBuckets);
typedef int (*HashSetCompareFunction)(const void *elemAddr1, const void *elemAddr2);
typedef void (*HashSetFreeFunction)(void *elemAddr);
typedef struct {
int elemNum; //current element number
int bucketNum; //number of hash bucket
int elemSize; //how many byte each element has
vector *buckets; //array of vector
HashSetHashFunction hashfn;
HashSetCompareFunction compfn;
HashSetFreeFunction freefn;
} hashset;
void HashSetNew(hashset *h, int elemSize, int numBuckets,
HashSetHashFunction hashfn, HashSetCompareFunction comparefn, HashSetFreeFunction freefn) {
assert(elemSize > 0 && numBuckets > 0 && hashfn != NULL && comparefn != NULL);
h->buckets = (vector *)malloc(numBuckets * sizeof(vector));
assert(h->buckets != NULL);
for (int i = 0; i < numBuckets; i++) {
vector *bucket = (vector *)((char *)h->buckets + i * sizeof(vector));
VectorNew(bucket, elemSize, freefn, 4);
}
h->bucketNum = numBuckets;
h->elemSize = elemSize;
h->elemNum = 0;
h->hashfn = hashfn;
h->compfn = comparefn;
h->freefn = freefn;
}
void HashSetEnter(hashset *h, const void *elemAddr) {
int hash = h->hashfn(elemAddr, h->bucketNum);
vector *bucket = (vector *)((char *)h->buckets + hash * sizeof(vector));
// search in the hash set first
int pos = VectorSearch(bucket, elemAddr, h->compfn, 0, false);
if (pos != -1) {
// replace the old one if find a match
VectorReplace(bucket, elemAddr, pos);
} else {
// otherwise insert the new one
VectorAppend(bucket, elemAddr);
h->elemNum++;
}
}
/** ==================================== DOC_FREQ & WORD_INDEX ======================================= */
/****************************************************************
*
* doc_freq is a key-value pair of [documentid, frequency]
* It's not supposed to be exposed to user or search engine.
* -----------------------------------------------------------
* It looks like:
* [1611742826915764000] [4 ]
* |-------------------| |-------|
* docid freq
***************************************************************/
typedef struct {
long docid;
int freq;
} doc_freq;
static void new_docfreq(doc_freq *df, long docid, int freq) {
df->docid = docid;
df->freq = freq;
}
/**
* HashSetHashFunction<doc_freq>
*/
static int hash_docfreq(const void *elemAddr, int numBuckets) {
doc_freq *df = (doc_freq *)elemAddr;
return (int)(df->docid % numBuckets);
}
/**
* HashSetCompareFunction<doc_freq>
*/
static int comp_docfreq(const void *elemAddr1, const void *elemAddr2) {
long id1 = ((doc_freq *)elemAddr1)->docid;
long id2 = ((doc_freq *)elemAddr2)->docid;
if (id1 < id2) {
return -1;
} else if (id1 > id2) {
return 1;
} else { // id1 == id2
return 0;
}
}
/**
* word_index is a index of a single word.
* ---------------------------------------
* A typical word_index looks like:
* [apple]: [doc1, 5], [doc3, 10], [doc5, 7]
* |-----| |------------------------------|
* word freqs
*/
typedef struct {
char *word;
hashset *freqs; // hashset<doc_freq>
} word_index;
static const size_t kWordIndexHashSetBuckets = 64;
static void new_wordindex(word_index *wi, const char *word) {
hashset h;
HashSetNew(&h, sizeof(doc_freq), kWordIndexHashSetBuckets, hash_docfreq, comp_docfreq, NULL);
wi->freqs = &h;
size_t wordlen = strlen(word);
wi->word = (char *)malloc(wordlen + 1); // +1 for null-termination
strcpy(wi->word, word);
(wi->word)[wordlen] = '\0';
}
/**
* Mainly used to build a word_index.
*/
void add_docfreq(word_index *wi, const long docid, const int frequency) {
doc_freq df;
new_docfreq(&df, docid, frequency);
HashSetEnter(wi->freqs, &df);
}
/** ==================================== UNIT-TEST ======================================= */
int main(void) {
/* apple: [1611742826915764000, 5][1611742826915538000, 10] */
word_index *apple = (word_index *)malloc(sizeof(word_index));
new_wordindex(apple, "apple");
add_docfreq(apple, 1611742826915764000L, 5);
add_docfreq(apple, 1611742826915538000L, 10);
}
It gave me a segmentation fault:
[1] 84309 segmentation fault testindexer
lldb find the problem occured when hashset try to callback the given pointer of function hashfn. I don't quite understand what is EXC_BAD_ACCESS (code=EXC_I386_GPFLT) here. I have done several unit test on hashset before, the HashSetEnter() function worked well with hashfn. Another unit test was conducted on hash_docfreq() function, it can also calculate correctly the hash number. I'm a little bit confused. Anyone can help? Thanks!
Process 89962 stopped
* thread #1, queue = 'com.apple.main-thread', stop reason = EXC_BAD_ACCESS (code=EXC_I386_GPFLT)
frame #0: 0x0000000100003b83 testnothing`HashSetEnter(h=0x00007ffeefbff620, elemAddr=0x00007ffeefbff638) at test_nothing.c:130:13
127 }
128
129 void HashSetEnter(hashset *h, const void *elemAddr) {
-> 130 int hash = h->hashfn(elemAddr, h->bucketNum);
131 vector *bucket = (vector *)((char *)h->buckets + hash * sizeof(vector));
132 // search in the hash set first
133 int pos = VectorSearch(bucket, elemAddr, h->compfn, 0, false);
Target 0: (testnothing) stopped.
(lldb) bt
* thread #1, queue = 'com.apple.main-thread', stop reason = EXC_BAD_ACCESS (code=EXC_I386_GPFLT)
* frame #0: 0x0000000100003b83 testnothing`HashSetEnter(h=0x00007ffeefbff620, elemAddr=0x00007ffeefbff638) at test_nothing.c:130:13
frame #1: 0x0000000100003c37 testnothing`add_docfreq(wi=0x0000000100306060, docid=1611742826915764000, frequency=5) at test_nothing.c:222:2
frame #2: 0x0000000100003cae testnothing`main at test_nothing.c:235:2
frame #3: 0x00007fff70df0cc9 libdyld.dylib`start + 1
(lldb)
Running under gdb, after the fault, doing a tb command to get a stack traceback, we see:
#0 0x00000005004016e6 in ?? ()
#1 0x000000000040163a in HashSetEnter (h=0x7fffffffdc10,
elemAddr=0x7fffffffdc40) at orig.c:150
#2 0x0000000000401834 in add_docfreq (wi=0x405260, docid=1611742826915764000,
frequency=5) at orig.c:266
#3 0x0000000000401879 in main () at orig.c:278
(gdb) frame 1
#1 0x000000000040163a in HashSetEnter (h=0x7fffffffdc10,
elemAddr=0x7fffffffdc40) at orig.c:150
150 int hash = h->hashfn(elemAddr, h->bucketNum);
You are segfaulting in HashSetEnter, at the line:
int hash = h->hashfn(elemAddr, h->bucketNum);
This is because h is not valid at this point.
Examinining the source, the place that sets the value that is ultimately invalid, it is set in new_wordindex.
In new_wordindex, you are saving [and returning] the address of h.
h is a function scoped variable here, so it is no longer valid after the function returns.
You have to use malloc for this. And, later, you need to be able to free this pointer during cleanup.
Here's the refactored code for the incorrect function.
Note that to show old/original code vs. new/corrected code, I'm using preprocessor conditionals:
#if 0
// old/original code
// NOTE: this is _not_ compiled in
#else
// new/corrected code
// NOTE: this _is_ compiled in
#endif
The code under #if 0 can be elided/removed, leaving just the #else code.
static void
new_wordindex(word_index * wi, const char *word)
{
// NOTE/BUG: h is function scoped -- this can _not_ be saved and returned
// because it ceases to be valid when we return
#if 0
hashset h;
HashSetNew(&h, sizeof(doc_freq), kWordIndexHashSetBuckets, hash_docfreq, comp_docfreq, NULL);
wi->freqs = &h;
#else
hashset *h = malloc(sizeof(*h));
HashSetNew(h, sizeof(doc_freq), kWordIndexHashSetBuckets, hash_docfreq, comp_docfreq, NULL);
wi->freqs = h;
#endif
size_t wordlen = strlen(word);
wi->word = (char *) malloc(wordlen + 1); // +1 for null-termination
strcpy(wi->word, word);
(wi->word)[wordlen] = '\0';
}
I've just encountered a very strange bug. I was doing unit-test for a simple function as below.
UPDATE: Thanks #Bodo, here's the minimal working example. You can simply compile and run tokenizer.c.
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <ctype.h>
#include <assert.h>
/* ============================= BOOL =============================== */
#ifndef _BOOL_
#define _BOOL_
typedef enum {
true, false
} bool;
#endif // _BOOL_
/* ============================= STACK =============================== */
#ifndef _STACK_
#define _STACK_
typedef void (*stack_freefn)(void *elemAddr);
typedef struct {
size_t size; // number of element allowed
int ite; // point to the current last element
size_t elemSize; // size of each element (how many bytes)
void *elems; // stockage of elements
stack_freefn freefn; // free memory allocated for each element if necessary
} stack;
/* constructor */
void new_stack(stack *s, const size_t size, const size_t elemSize, stack_freefn freefn) {
s->size = size;
s->ite = 0;
s->elemSize = elemSize;
s->elems = malloc(size * elemSize);
s->freefn = freefn;
}
/* free memory */
void dispose_stack(stack *s) {
if (s->freefn != NULL) {
while (s->ite > 0) {
void *elemAddr = (char *)s->elems + --s->ite * s->elemSize;
s->freefn(elemAddr);
}
}
free(s->elems);
s->elems = NULL;
}
/* push one new element on the top */
void push_stack(stack *s, const void *value, const size_t elemSize) {
if (s->ite == s->size) {
s->size *= 2;
s->elems = realloc(s->elems, s->size * s->elemSize);
}
void *elemAddr = (char *)s->elems + s->elemSize * s->ite++;
memcpy(elemAddr, value, s->elemSize);
}
/* pop our the element on the top */
void pop_stack(stack *s, void *res) {
if (s->ite > 0) {
void *elemAddr = (char *)s->elems + ((s->ite - 1) * s->elemSize);
memcpy(res, elemAddr, s->elemSize);
s->ite--;
}
}
void clear_stack(stack *s) {
if (s->freefn != NULL) {
while (s->ite > 0) {
void *elemAddr = (char *)s->elems + --s->ite * s->elemSize;
s->freefn(elemAddr);
}
} else {
s->ite = 0;
}
}
size_t stack_size(stack *s) {
return s->ite;
}
#endif // _STACK_
/* ============================= VECTOR =============================== */
#ifndef _VECTOR_
#define _VECTOR_
typedef int (*VectorCompareFunction)(const void *elemAddr1, const void *elemAddr2);
typedef void (*VectorFreeFunction)(void *elemAddr);
typedef struct {
int elemSize; //how many byte for each element
int elemNum; //number of current element in vector
int capacity; //maximum number of element vector can hold
void *elems; //pointer to data memory
VectorFreeFunction freefn; //pointer to the function used to free each element
} vector;
/**
* Reallocate a new memory of twice of original size
* return 1 if reallocation success, otherwise return -1.
*/
static void DoubleMemory(vector *v) {
void *tmp = realloc(v->elems, v->capacity * v->elemSize * 2);
assert(tmp != NULL);
v->elems = tmp;
v->capacity *= 2;
}
/**
* Constructor
*/
void VectorNew(vector *v, int elemSize, VectorFreeFunction freefn, int initialAllocation) {
v->elems = malloc(initialAllocation * elemSize);
assert(v->elems != NULL);
v->elemSize = elemSize;
v->elemNum = 0;
v->capacity = initialAllocation;
v->freefn = freefn;
}
/**
* Frees up all the memory of the specified vector.
*/
void VectorDispose(vector *v) {
if (v->freefn != NULL) {
for (; v->elemNum > 0; v->elemNum--) {
void *elemAddr = (char *)v->elems + (v->elemNum - 1) * v->elemSize;
v->freefn(elemAddr);
}
}
free(v->elems);
v->elems = NULL;
}
/**
* Returns the logical length of the vector.
*/
int VectorLength(const vector *v) {
return v->elemNum;
}
/**
* Appends a new element to the end of the specified vector.
*/
void VectorAppend(vector *v, const void *elemAddr) {
/* double size if neccessary */
if (v->elemNum == v->capacity) DoubleMemory(v);
memcpy((char *)v->elems + v->elemNum * v->elemSize, elemAddr, v->elemSize);
v->elemNum++;
}
/**
* Search the specified vector for an element whose contents match the element passed as the key.
*/
int VectorSearch(const vector *v, const void *key, VectorCompareFunction searchfn, int startIndex, bool isSorted) {
assert(key && searchfn);
if (v->elemNum == 0) return -1;
assert(startIndex >= 0 && startIndex < v->elemNum);
if (isSorted == true) {
/* binary search */
void *startAddr = (char *)v->elems + startIndex * v->elemSize;
int size = v->elemNum - startIndex;
void *resAddr = bsearch(key, startAddr, size, v->elemSize, searchfn);
return (resAddr != NULL)? ((char *)resAddr - (char *)v->elems) / v->elemSize : -1;
} else {
/* linear search */
for (int i = 0; i < v->elemNum; i++) {
if (searchfn((char *)v->elems + i * v->elemSize, key) == 0) {
return i;
}
}
return -1;
}
}
#endif // _VECTOR_
/* ============================= TOKENIZER =============================== */
/**
* Dump current string into vector as a new word.
* Strings are null-terminated.
*/
static void dumpstack(stack *s, vector *v) {
size_t len = stack_size(s);
char *word = (char *)malloc((len + 1) * sizeof(char)); // +1 for null-terminator
for (int i = len - 1; i >= 0; i--) {
pop_stack(s, word + i * sizeof(char));
}
word[len] = '\0';
VectorAppend(v, &word);
clear_stack(s);
}
static const size_t kTokenStackDefaultSize = 64;
static void tokenize(vector *words, char *stream) {
stack s;
new_stack(&s, kTokenStackDefaultSize, sizeof(char), NULL);
size_t len = strlen(stream);
bool begin = false;
char c;
for (int i = 0; i < len; i++) {
c = stream[i];
/* =============================== My printf() is here ============================== */
// printf("char c = [%c]\n", c);
/* =============================== My printf() is here ============================== */
if (isalpha(c) || isdigit(c)) {
if (begin == false) begin = true;
char lower = tolower(c);
push_stack(&s, &lower, sizeof(char));
} else if (c == '-') {
if (begin == true) { // case: covid-19
push_stack(&s, &c, sizeof(char));
} else {
if (i < len - 1 && isdigit(stream[i + 1])) { // case: -9
begin = true;
push_stack(&s, &c, sizeof(char));
} else {
if (begin == true) {
dumpstack(&s, words);
begin = false;
}
}
}
} else if (c == '.' && begin == true) { // case: 3.14
if (isdigit(stream[i - 1])) {
push_stack(&s, &c, sizeof(char));
} else {
if (begin == true) {
dumpstack(&s, words);
begin = false;
}
}
} else {
if (begin == true) {
dumpstack(&s, words);
begin = false;
}
}
}
if (begin == true) {
dumpstack(&s, words);
begin = false;
}
dispose_stack(&s);
}
/* ============================= UNIT-TEST =============================== */
/**
* HashSetFreeFunction<char *>
*/
static void freestr(void *elemAddr) {
char *str = *(char **)elemAddr;
free(str);
}
/**
* HashSetCompareFunction<char *>
*/
static int compstr(const void *elemAddr1, const void *elemAddr2) {
char *str1 = *(char **)elemAddr1;
char *str2 = *(char **)elemAddr2;
return strcmp(str1, str2);
}
static void test_tokenize(void) {
printf("Testing Tokenizer.c::tokenize() ...\n");
char *sentence = "Covid-19: Top adviser warns France at 'emergency' virus moment - BBC News\nPi = 3.14\n-1 is negative.";
vector words;
VectorNew(&words, sizeof(char *), freestr, 256);
tokenize(&words, sentence);
char *musthave[] = {"covid-19", "top", "3.14", "-1"};
char *musthavenot[] = {"-", "1"};
assert(VectorLength(&words) == 16);
for (int i = 0; i < sizeof(musthave)/sizeof(char *); i++) {
assert(VectorSearch(&words, &musthave[i], compstr, 0, false) != -1);
}
for (int i = 0; i < sizeof(musthavenot)/sizeof(char *); i++) {
assert(VectorSearch(&words, &musthavenot[i], compstr, 0, false) == -1);
}
VectorDispose(&words);
printf("[ALL PASS]\n");
}
int main(void) {
test_tokenize();
}
I've got segmentation fault at first.
[1] 4685 segmentation fault testtokenizer
But when I add a printf() to debug, the segmentation fault was gone and the test passed. After comment out the printf, the function still works. I was so confused.
Just recall that before this test, I tested some memory dispose function, and perhaps had left some unfreed blocks in memory. Will that be the reason for fleeting segmentation fault? Thx bros.
UPDATE:
Now I can't even reproduce this bug myself. tokenizer.c above can pass the unit-test. I thought it might caused by makefile prerequisite rules. gcc didn't re-compile some object files when source code is changed.
Thanks #Steve Summit, you make it clear that unfreed memory will not cause segmentation fault.
Thanks #schwern for code review, it's really helpful.
But when I add a printf() to debug, the segmentation fault was gone and the test passed. After comment out the printf, the function still works. I was so confused.
They call it undefined behavior, because its behavior is undefined. Seemingly unrelated operations might nudge things just a bit to make the code "work" but they're only tangentially related to the problem.
I tested some memory dispose function, and perhaps had left some unfreed blocks in memory. Will that be the reason for fleeting segmentation fault?
No. It does mean the memory is unreferencable and "leaked". The memory will be freed to the operating system when the process exits.
The problem must lie elsewhere. Without seeing your whole program we can't say for sure, but two fishy things stand out.
You're defining a fixed sized stack, but you're pushing onto it an indeterminate number of times. Unless push_stack has protection against this, you will walk off your allocated memory.
You're storing references to variables on the stack. lower, c
char lower = tolower(c);
push_stack(&s, &lower, sizeof(char));
Once lower goes out of scope it will automatically be freed and the memory reused. &lower is invalid once tokenize returns. This seems to be fine if your stack only lasts the length of the function, but it's worth noting.
And it's possible new_stack, push_stack, or dumpstack are doing something incorrect.
I have written the code for the circular buffer in C and it works well until some extent. I took the size of the buffer being equal to 10. When I fill the buffer till element 6 - it works fine. But at the moment when I fill the 7-th element - I get the result "The size of the buffer is equal to 767". For the element 8 - it does not work. I use "head" to write and "tail" to extract values. Could you please help me with this?
#include<stdio.h>
#include<stdint.h>
#include <stdbool.h>
typedef struct RingBuffer {
uint16_t* buffer;
size_t head;
size_t tail;
size_t max;
bool full;
}*cbuf_handle_t;
cbuf_handle_t init_RingBuffer (uint8_t* buffer, size_t size){
cbuf_handle_t cbuf = malloc (sizeof(cbuf_handle_t));
cbuf->buffer = buffer;
cbuf->max = size;
return cbuf;
}
void RingBuffer_free(cbuf_handle_t cbuf){
free(cbuf);
}
void RingBuffer_reset(cbuf_handle_t cbuf){
cbuf->head = 0;
cbuf->tail = 0;
cbuf->full = false;
}
bool RingBuffer_full (cbuf_handle_t cbuf){
return cbuf->full;
}
bool RingBuffer_empty(cbuf_handle_t cbuf){
return (!cbuf->full && (cbuf->tail == cbuf->head));
}
size_t RingBuffer_Capacity(cbuf_handle_t cbuf){
return cbuf->max;
}
size_t RingBuffer_size(cbuf_handle_t cbuf){
size_t size = cbuf->max;
if (!cbuf->full){
if (cbuf->head >= cbuf->tail)
{
size = (cbuf->head - cbuf->tail);}
else
{
size = (cbuf->head - cbuf->tail + cbuf->max);
}
}
return size;
}
void RingBuffer_AdvancePointer(cbuf_handle_t cbuf){
if (cbuf->full){
cbuf->tail = (cbuf->tail+1)%cbuf->max;
}
cbuf->head = (cbuf->head + 1)%cbuf->max;
cbuf->full = (cbuf->head == cbuf->tail);
}
void RingBuffer_retreatPointer (cbuf_handle_t cbuf){
cbuf->full = false;
cbuf->tail = (cbuf->tail + 1)%cbuf->max;
}
void RingBuffer_addValue (cbuf_handle_t cbuf, uint8_t data){
cbuf->buffer[cbuf->head] = data;
RingBuffer_AdvancePointer(cbuf);
}
int RingBuffer_Remove (cbuf_handle_t cbuf, uint8_t *data){
int r = -1;
if (!RingBuffer_empty(cbuf)){
*data = cbuf->buffer[cbuf->tail];
RingBuffer_retreatPointer(cbuf);
r = 0;
}
return r;
}
int main (){
uint8_t arr[10];
cbuf_handle_t cpt = init_RingBuffer(arr, 10);
//initialzie the buffer, tail and head and max
int i = 0;
RingBuffer_reset(cpt);
for ( i = 0 ; i< 6; i++){
RingBuffer_addValue(cpt, i);
}
size_t size = RingBuffer_size(cpt);
printf("The size of the buffer %d", size);
}
Thank you in advance!
Regards
Rostyslav
As said in comments, the declaration of the structure as a pointer is generally not recommended. However you can fix that bug by changing the way you allocate it using malloc :
cbuf_handle_t cbuf = malloc (sizeof(*cbuf));
This is because, cbuf being a pointer to the structure, if you dereference it you get the structure and thus its real size when you pass it to sizeof.
This question is unlikely to help any future visitors; it is only relevant to a small geographic area, a specific moment in time, or an extraordinarily narrow situation that is not generally applicable to the worldwide audience of the internet. For help making this question more broadly applicable, visit the help center.
Closed 10 years ago.
I seem to be going in circles with this project! I get so many errors of 'dereferencing pointer to incomplete type', and have quite a few others. it seems like when I fix one another one will pop up to take its place!
It's my first time using hash tables, and I admit that I am rather lost but I think I made a very good start at least. Any input as to how to solve my 'dereferencing pointer to incomplete type' problems would be amazing!
htable.c
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include "htable.h"
struct htablerec {
int size;
int num_entries;
hashing_t method;
char **keys;
int *freqs;
int *stats;
};
void *emalloc(size_t s) {
void *result = malloc(s);
if (NULL == result) {
fprintf(stderr, "Memory allocation failed!\n");
exit(EXIT_FAILURE);
}
return result;
}
/*moves pointer to point to something appropriate*/
htable htable_new(int capacity) {
int i;
htable ht = emalloc(sizeof *ht);
ht->size = size;
ht->method = method;
ht->num_keys = 0;
ht->keys = emalloc(size * sizeof ht->keys[0]);
ht->freqs = emalloc(size * sizeof ht->freqs[0]);
ht->stats = emalloc(size * sizeof ht->stats[0]);
for(i = 0; i<size; i++){
ht->keys[i] = NULL;
ht->freqs[i] = 0;
ht->stats[i] = 0;
}
return ht;
}
static unsigned int htable_step(htable h, unsigned int i_key){
return 1 + (i_key % (h->size - 1));
}
static unsigned int htable_wtoi(char *word){
unsigned int result = 0;
while(*word != '\0') result = (*word++ +31 * result);
return result;
}
static unsigned int htable_hash(htable h, unsigned int i_key){
return i_key % h->size;
}
void htable_free(htable h) {
int i;
for(i = 0; i<h->size; i++){
if(h->keys[i] != NULL){
free(h->keys[i]);
}
}
if(h->keys != NULL){
free(h->keys);
}
if(h->freqs != NULL){
free(h->freqs);
}
free(h);
}
static unsigned int htable_word_to_int(char *word) {
unsigned int result = 0;
while (*word != '\0') {
result = (*word++ + 31 * result);
}
return result;
}
int htable_insert(htable h, char *str) {
int num_collisions = 0;
int i_key = htable_wtoi(key);
int pos = htable_hash(h, i_key);
int step = 1;
if(h->method == DOUBLE)
step = htable_step(h, i_key);
while(h->keys[pos]!=NULL &&
strcmp(h->keys[pos],key)!=0 &&
num_collisions < h->size ){
pos = htable_hash(h, pos + step);
num_collisions++;
}
if(h->keys[pos] == NULL){
h->keys[pos] = emalloc((strlen(key)+1) * sizeof h->keys[0][0]);
strcpy(h->keys[pos],key);
h->stats[h->num_keys] = num_collisions;
h->num_keys++;
}
if(num_collisions >= h->size) /* We must be full, so return zero.*/
return 0;
return ++(h->freqs[pos]);
}
static int htable_search(htable h, char *key){
int num_collisions = 0;
int i_key = htable_wtoi(key);
int pos = htable_hash(h, i_key);
int step = 1;
if(h->method == DOUBLE)
step = htable_step(h, i_key);
while(h->keys[pos]!=NULL &&
strcmp(h->keys[pos],key)!=0 &&
num_collisions < h->size ){
pos = htable_hash(h, pos + step);
num_keys++;
}
if(num_keys >= h->size)
return 0;
else
return h->freqs[pos];
}
void htable_print(htable h, FILE *stream){
int i;
for(i = 0; i<h->size; i++){
if(h->keys[i] != NULL)
fprintf(stream, "%d\t%s\n",i, h->freqs[i], h->keys[i]);
}
}
void htable_print_entire_table(htable h, FILE *stream) {
int i;
for (i=0; loop < h->capacity; i++) {
if (h->key[i] != NULL) {
fprintf("%d\t%s\n", h->freqs[i], h->key[i]);
}
}
}
/**
* Prints a line of data indicating the state of the hash table when
* it is a given percentage full.
*
* The data is printed out right justified (with the given field widths,
* and decimal places) in this order:
*
* - How full the hash-table is as a percentage (4)
* - How many keys are in the hash-table at that point (11)
* - What percentage of those keys were placed 'at home' (12, 1 dp)
* - The average number of collisions per key placed (12, 2 dp)
* - The maximum number of collisions while placing a key (12)
*
* #param h the hash-table to get data from.
* #param stream the place to send output to.
* #param percent_full the point at which to print the statistics.
* If the hashtable is less full than that, then
* nothing will be printed.
*/
static void print_stats_line(htable h, FILE *stream, int percent_full) {
int current_entries = h->capacity * percent_full / 100;
double average_collisions = 0.0;
int at_home = 0;
int max_collisions = 0;
int i = 0;
if (current_entries > 0 && current_entries <= h->num_keys) {
for (i = 0; i < current_entries; i++) {
if (h->stats[i] == 0) {
at_home++;
}
if (h->stats[i] > max_collisions) {
max_collisions = h->stats[i];
}
average_collisions += h->stats[i];
}
fprintf(stream, "%4d%11d%12.1f%12.2f%12d\n", percent_full,
current_entries, at_home * 100.0 / current_entries,
average_collisions / current_entries, max_collisions);
}
}
void htable_print_stats(htable ht, FILE *stream, int num_stats) {
int i;
fprintf(stream, "Percent Current Percent Average Maximum\n");
fprintf(stream, " Full Entries At Home Collisions Collisions\n");
fprintf(stream, "-----------------------------------------------------\n");
for (i = 1; i <= num_stats; i++) {
print_stats_line(ht, stream, 100 * i / num_stats);
}
fprintf(stream, "-----------------------------------------------------\n\n");
}
htable.h
#ifndef HTABLE_H
#define HTABLE_H
#include <stdio.h>
typedef struct hashtable *htable;
typedef enum hashing_e { LINEAR, DOUBLE } hashing_t;
extern htable htable_new(int size);
extern void htable_destroy(htable ht);
extern int htable_insert(htable h, char *key);
extern int htable_search(htable h, char *key);
extern void htable_print(htable h, FILE *stream);
extern void htable_print_stats(htable ht, FILE *stream, int num_stats);
#endif
main.c (given by tutor for project to fit)
/**
* #file main.c
* #author Iain Hewson
* #date August 2012
*
* This program is written to test the hash table ADT specified in
* Cosc242 assignment two. It creates a hash table which can use
* linear-probing or double-hashing as a collision resolution
* strategy. Various options are provided which make it possible to
* examine a hash table as well as see how it performs while being
* filled.
*/
#include <stdio.h>
#include <stdlib.h>
#include <getopt.h>
#include "mylib.h"
#include "htable.h"
/* A boolean type which can be TRUE or FALSE */
typedef enum bool_e {FALSE, TRUE} bool_t;
/* function declarations */
static void usage(char *progname);
static void setup(int argc, char **argv, bool_t *double_hashing,
bool_t *entire_table, bool_t *print_stats,
int *snapshots, int *tablesize);
/**
*
* Creates a hash-table and inserts words into it read from stdin.
* Arguments on the command line alter the behaviour of the program
* as follows:
* - -d Use double hashing (linear probing is the default)
* - -e Display entire contents of hash-table on stderr
* - -n NUM Show NUM statistics snapshots (if -p is used)
* - -p Print stats info instead of frequencies & words
* - -s SIZE Use the first prime >= SIZE as htable size
* - -h Display this message
*
* By default each word and it's frequency are printed to stdout.
*
* #param argc the number of command-line arguments.
* #param argv an array of strings containing the command-line arguments.
*
* #return EXIT_SUCCESS if the program is successful.
*/
int main(int argc,char **argv) {
bool_t entire_table = FALSE, double_hashing = FALSE, print_stats = FALSE;
int tablesize = 0, snapshots = 0;
char word[256];
htable ht;
setup(argc, argv, &double_hashing, &entire_table, &print_stats,
&snapshots, &tablesize);
ht = htable_new(tablesize, (double_hashing) ? DOUBLE_H : LINEAR_P);
while (getword(word, sizeof word, stdin) != EOF) {
htable_insert(ht, word);
}
if (entire_table) {
htable_print_entire_table(ht, stderr);
}
if (print_stats) {
htable_print_stats(ht, stdout, snapshots);
} else {
htable_print(ht, stdout); /* print words and frequencies */
}
htable_free(ht);
return EXIT_SUCCESS;
}
/**
* Prints out a usage message to stderr outlining all of the options.
* #param prog_name the name of the program to include in usage message.
*/
static void usage(char *prog_name) {
fprintf(stderr, "Usage: %s [OPTION]... <STDIN>\n\n%s%s", prog_name,
"Perform various operations using a hash-table. By default read\n"
"words from stdin and print them with their frequencies to stdout.\n\n"
" -d Use double hashing (linear probing is the default)\n"
" -e Display entire contents of hash-table on stderr\n",
" -n NUM Show NUM stats snapshots (if -p is used)\n"
" -p Print stats info instead of frequencies & words\n"
" -s SIZE Use the first prime >= SIZE as htable size\n\n"
" -h Display this message\n\n");
}
/**
* Handle options given on the command-line by setting a number of
* variables appropriately. May call usage() if incorrect arguments
* or -h given.
*
* #param argc the number of command-line arguments.
* #param argv an array of strings contain the command-line arguments.
* #param double_hashing set to TRUE if -d given
* #param entire_table set to TRUE if -e given
* #param snapshots set to NUM if -n NUM given and NUM > 0 else set to 10
* #param print_stats set to TRUE if -p given
* #param tablesize set to SIZE if -t SIZE given and SIZE > 0 else set to 113
*/
static void setup(int argc, char **argv, bool_t *double_hashing,
bool_t *entire_table, bool_t *print_stats,
int *snapshots, int *tablesize) {
const char *optstring = "dehpn:s:";
char option;
while ((option = getopt(argc, argv, optstring)) != EOF) {
switch (option) {
case 'd':
*double_hashing = TRUE;
break;
case 'e':
*entire_table = TRUE;
break;
case 'p':
*print_stats = TRUE;
break;
case 'n':
*snapshots = atoi(optarg);
break;
case 's':
*tablesize = atoi(optarg);
break;
case 'h':
default:
usage(argv[0]);
exit(EXIT_SUCCESS);
}
}
/* set default values if nothing sensible entered */
if (*tablesize < 1) *tablesize = 113;
if (*snapshots < 1) *snapshots = 10;
}
mylib.c
#include <stdlib.h>
#include "mylib.h"
/**************************
* *
* emalloc *
* *
**************************
Used to handle new memory allocation to a pointer and handle exceptions
which may arrise if memory allocation fails, which happens all the time
when you try and enter negative values.
PARAMETERS: s = calculated size of required memory.
RETURN VALUE: a pointer of any type.
*/
void *emalloc(size_t s){
void *result = malloc(s);
if(NULL == result){
fprintf(stderr, "Memory allocation failed!\n");
exit(EXIT_FAILURE);
}
return result;
}
/**************************
* *
* erealloc *
* *
**************************
Handles the reallocation of an updated amount of memory to an existing
with existing data attached.
PARAMETERS: p = the existing pointer we would like additional memory
allocated to.
s = calculated size of required memory.
RETURN VALUE: a pointer of any type.
*/
void *erealloc(void *p, size_t s){
void *result = realloc(p,s);
if(NULL == result){
fprintf(stderr, "Memory allocation failed!\n");
exit(EXIT_FAILURE);
}
return result;
}
/**************************
* *
* getword *
* *
**************************
Is used to read input from the designated file stream (standard in for the
assignment). Getword removes white space with the first while loop. And only
returns one word. Maintaining continous input is therefore the responsibility
of the calling function.
PARAMETERS: s = the pointer to the character array.
limit = the maximum size a word can be.
stream = where to read from.
RETURN VALUE: the integer value of the character at s[0]. The string s does
not need to be returned, since it is an array and is passed as
a memory address. If no chars were read into the string s, then
s[0] would have the '\0' [NULL] value which equates to 0 if used
in a boolean equation.
*/
int getword(char *s, int limit, FILE *stream){
int c;
while(!isalnum( c = getc(stream)) && c != EOF);
if(c == EOF)
return EOF;
else
*s++ = tolower(c);
while(--limit > 0){
if(isalnum(c = getc(stream)))
*s++ = tolower(c);
else if('\'' == c) continue;
else break;
}
*s = '\0';
return s[0];
}
/**************************
* *
* stoi *
* *
**************************
Not using this function now.
PARAMETERS: s = string representation of a number.
RETURN VALUE: the integer value.
*/
int stoi(char *s){
int i,j,r;
int sign = 1;
i=1;
r=0;
for(j=my_strlen(s)-1; j>=0; j--){
if(j == 0 && s[j] == '-')
sign = -1;
else {
if(s[j]>='0' && s[j]<='9'){
r+=((s[j]-'0')*i);
i*=10;
} else {
fprintf(stderr, "Invalid input for String to int\n");
exit(EXIT_FAILURE);
}
}
}
return r * sign;
}
/**************************
* *
* my_strlen *
* *
**************************
I am using my own string length function, but I wrote it with the stoi
function, so am using it here. Perhaps not as safe as the library
functions?
PARAMETERS: s = a string delimited by the NULL character.
RETURN VALUE: the number of characters in the string.
*/
int my_strlen(char *s){
int i=0;
while(s[i]!='\0')
i++;
return i;
}
/**************************
* *
* factors *
* *
**************************
Another unrequired function used to calculate the possiblity of factorisation.
PARAMETERS: x = An integer to be factored towards.
RETURN VALUE: 0 if x has factors, 1 if x is a prime number.
*/
static int factors(int x){
int f = 2;
while(f*f < x){
if(x % f == 0){
return 0;
} else {
f++;
}
}
return 1;
}
/**************************
* *
* prime_gt *
* *
**************************
Used in conjunction with factors to find factorless integers. We increment
bound until it is truely prime.
Bound - We start with bound, sending it to the factors function. If it is
a prime number, then stop searching. Otherwise loop until we find
an prime integer larger than the input integer.
PARAMETERS: s = the input integer.
RETURN VALUE: Bound, for it is now a prime number.
*/
static int prime_gt(int s){
int bound = s;
while(bound > 0){
if(factors(bound))
break;
else
bound++;
}
return bound;
}
/**************************
* *
* relative_prime *
* *
**************************
Decides on a prime number to use to set the table size to.
PARAMETERS: s = the required size of the table.
RETURN VALUE: the newer beter prime number size for the table.
*/
unsigned int relative_prime(int s){
return prime_gt(s);
}
Sorry for it being so big, it's ok if it's just a complete unfixable jumble.
You don't seem to have defined struct hashtable anywhere. You need to say somewhere what fields that struct should actually contain, at the moment there is only a forward declaration in htable.h.
Such a forward declaration just says that the type exists, but not how it exactly looks like. Therefore it is considered an incomplete type, until the compiler sees a full definition of it.
Cause the definition of struct hashtable does not exist, the following line creates an alias of an incomplete pointer type (htable).
typedef struct hashtable *htable;
It is impossible to deference such pointer (the compiler doesn't know the different fields of your structure).
In your file htable.c, maybe you meant :
/* instead of struct htablered */
struct htable {
int size;
int num_entries;
hashing_t method;
char **keys;
int *freqs;
int *stats;
};
I am trying to write a Huffman encoding program to compress a text file. Upon completetion, the program will terminate at the return statement, or when I attempt to close a file I was reading from. I assume I have memory leaks, but I cannot find them. If you can spot them, let me know (and a method for fixing them would be appreciated!).
(note: small1.txt is any standard text file)
Here is the main program
#include<stdio.h>
#include<string.h>
#include<stdlib.h>
#define ASCII 255
struct link {
int freq;
char ch[ASCII];
struct link* right;
struct link* left;
};
typedef struct link node;
typedef char * string;
FILE * ofp;
FILE * ifp;
int writebit(unsigned char);
void sort(node *[], int);
node* create(char[], int);
void sright(node *[], int);
void Assign_Code(node*, int[], int, string *);
void Delete_Tree(node *);
int main(int argc, char *argv[]) {
//Hard-coded variables
//Counters
int a, b, c = 0;
//Arrays
char *key = (char*) malloc(ASCII * sizeof(char*));
int *value = (int*) malloc(ASCII * sizeof(int*));
//File pointers
FILE *fp = fopen(argv[1], "r");
if (fp == NULL) {
fprintf(stderr, "can't open %s\n", argv[1]);
return 0;
}
//Nodes
node* ptr;//, *head;
node* array[ASCII];
//
int u, carray[ASCII];
char str[ASCII];
//Variables
char car = 0;
int inList = 0;
int placeinList = -1;
int numofKeys;
if (argc < 2) {
printf("Usage: huff <.txt file> \n");
return 0;
}
for (a = 0; a < ASCII; a++) {
key[a] = -1;
value[a] = 0;
}
car = fgetc(fp);
while (!feof(fp)) {
for (a = 0; a < ASCII; a++) {
if (key[a] == car) {
inList = 1;
placeinList = a;
}
}
if (inList) {
//increment value array
value[placeinList]++;
inList = 0;
} else {
for (b = 0; b < ASCII; b++) {
if (key[b] == -1) {
key[b] = car;
break;
}
}
}
car = fgetc(fp);
}
fclose(fp);
c = 0;
for (a = 0; a < ASCII; a++) {
if (key[a] != -1) {
array[c] = create(&key[a], value[a]);
numofKeys = c;
c++;
}
}
string code_string[numofKeys];
while (numofKeys > 1) {
sort(array, numofKeys);
u = array[0]->freq + array[1]->freq;
strcpy(str, array[0]->ch);
strcat(str, array[1]->ch);
ptr = create(str, u);
ptr->right = array[1];
ptr->left = array[0];
array[0] = ptr;
sright(array, numofKeys);
numofKeys--;
}
Assign_Code(array[0], carray, 0, code_string);
ofp = fopen("small1.txt.huff", "w");
ifp = fopen("small1.txt", "r");
car = fgetc(ifp);
while (!feof(ifp)) {
for (a = 0; a < ASCII; a++) {
if (key[a] == car) {
for (b = 0; b < strlen(code_string[a]); b++) {
if (code_string[a][b] == 48) {
writebit(0);
} else if (code_string[a][b] == 49) {
writebit(1);
}
}
}
}
car = fgetc(ifp);
}
writebit(255);
fclose(ofp);
ifp = fopen("small1.txt", "r");
fclose(ifp);
free(key);
//free(value);
//free(code_string);
printf("here1\n");
return 0;
}
int writebit(unsigned char bitval) {
static unsigned char bitstogo = 8;
static unsigned char x = 0;
if ((bitval == 0) || (bitval == 1)) {
if (bitstogo == 0) {
fputc(x, ofp);
x = 0;
bitstogo = 8;
}
x = (x << 1) | bitval;
bitstogo--;
} else {
x = (x << bitstogo);
fputc(x, ofp);
}
return 0;
}
void Assign_Code(node* tree, int c[], int n, string * s) {
int i;
static int cnt = 0;
string buf = malloc(ASCII);
if ((tree->left == NULL) && (tree->right == NULL)) {
for (i = 0; i < n; i++) {
sprintf(buf, "%s%d", buf, c[i]);
}
s[cnt] = buf;
cnt++;
} else {
c[n] = 1;
n++;
Assign_Code(tree->left, c, n, s);
c[n - 1] = 0;
Assign_Code(tree->right, c, n, s);
}
}
node* create(char a[], int x) {
node* ptr;
ptr = (node *) malloc(sizeof(node));
ptr->freq = x;
strcpy(ptr->ch, a);
ptr->right = ptr->left = NULL;
return (ptr);
}
void sort(node* a[], int n) {
int i, j;
node* temp;
for (i = 0; i < n - 1; i++)
for (j = i; j < n; j++)
if (a[i]->freq > a[j]->freq) {
temp = a[i];
a[i] = a[j];
a[j] = temp;
}
}
void sright(node* a[], int n) {
int i;
for (i = 1; i < n - 1; i++)
a[i] = a[i + 1];
}
If your program is crashing on what is otherwise a valid operation (like returning from a function or closing a file), I'll near-guarantee it's a buffer overflow problem rather than a memory leak.
Memory leaks just generally mean your mallocs will eventually fail, they do not mean that other operations will be affected. A buffer overflow of an item on the stack (for example) will most likely corrupt other items on the stack near it (such as a file handle variable or the return address from main).
Probably your best bet initially is to set up a conditional breakpoint on writes to the file handles. This should happen in the calls to fopen and nowhere else. If you detect a write after the fopen calls are finished, that will be where your problem occurred, so just examine the stack and the executing line to find out why.
Your first problem (this is not necessarily the only one) lies here:
c = 0;
for (a = 0; a < ASCII; a++) {
if (key[a] != -1) {
array[c] = create(&key[a], value[a]);
numofKeys = c; // DANGER,
c++; // WILL ROBINSON !!
}
}
string code_string[numofKeys];
You can see that you set the number of keys before you increment c. That means the number of keys is one less than you actually need so that, when you access the last element of code_string, you're actually accessing something else (which is unlikely to be a valid pointer).
Swap the numofKeys = c; and c++; around. When I do that, I at least get to the bit printing here1 and exit without a core dump. I can't vouch for the correctness of the rest of your code but this solves the segmentation violation so anything else should probably go in your next question (if need be).
I can see one problem:
strcpy(str, array[0]->ch);
strcat(str, array[1]->ch);
the ch field of struct link is a char array of size 255. It is not NUL terminated. So you cannot copy it using strcpy.
Also you have:
ofp = fopen("small1.txt.huff", "w");
ifp = fopen("small1.txt", "r");
If small1.txt.huff does not exist, it will be created. But if small1.txt it will not be created and fopen will return NULL, you must check the return value of fopen before you go and read from the file.
Just from counting, you have 4 separate malloc calls, but only one free call.
I would also be wary of your sprintf call, and how you are actually mallocing.
You do an sprintf(buf, "%s%d", buf, c[i]) but that can potentially be a buffer overflow if your final string is longer than ASCII bytes.
I advise you to step through with a debugger to see where it's throwing a segmentation fault, and then debug from there.
i compiled the program and ran it with it's source as that small1.txt file and got "can't open (null)" if the file doesn't exist or the file exist and you give it on the command like ./huf small1.txt the program crashes with:
Program terminated with signal 11, Segmentation fault.
#0 0x08048e47 in sort (a=0xbfd79688, n=68) at huf.c:195
195 if (a[i]->freq > a[j]->freq) {
(gdb) backtrace
#0 0x08048e47 in sort (a=0xbfd79688, n=68) at huf.c:195
#1 0x080489ba in main (argc=2, argv=0xbfd79b64) at huf.c:99
to get this from gdb you run
ulimit -c 100000000
./huf
gdb --core=./core ./huf
and type backtrace
You have various problems in your Code:
1.- mallocs (must be):
//Arrays
char *key = (char*) malloc(ASCII * sizeof(char));
int *value = (int*) malloc(ASCII * sizeof(int));
sizeof(char) == 1, sizeof(char *) == 4 or 8 (if 64 bits compiler is used).
2.- Buffer sizes 255 (ASCII) is too short to receive the contents of array[0]->ch + array[1]->ch + '\0'.
3.- Use strncpy instead of strcpy and strncat instead of strcat.
4.- key is an array of individuals chars or is a null terminated string ?, because you are using this variable in both ways in your code. In the characters counting loop you are using this variables as array of individuals chars, but in the creation of nodes you are passing the pointer of the array and copying as null terminated array.
5.- Finally always check your parameters before used it, you are checking if argc < 2 after trying to open argv[1].