char *test_findHashElementByKey() {
struct Symbol *symbol = calloc(1, sizeof(struct Symbol));
//simple hash function
struct hash *symbolTable = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
createHashElement(symbolTable, "blue", symbol);
mu_assert("Could not find element that should be table.",
findHashElementByKey(symbolTable, "blue") != NULL);
mu_assert("Found element that should not be table.",
findHashElementByKey(symbolTable, "green") == NULL);
destroyHash(&symbolTable);
mu_assert("Call to destroyHash does not seg fault.", 1);
//normal hash function
symbolTable = createHash(&getHashedKeyNormal);
mu_assert("Call to createHash does not seg fault.", 1);
createHashElement(symbolTable, "teekkekkke", symbol);
mu_assert("Could not find element that should be table.",
findHashElementByKey(symbolTable, "teekkekkke") != NULL);
mu_assert("Found element that should not be table.",
findHashElementByKey(symbolTable, "green") == NULL);
destroyHash(&symbolTable);
mu_assert("Call to destroyHash does not seg fault.", 1);
return NULL;
}
char *test_createHashElement() {
struct Symbol *symbol = calloc(1, sizeof(struct Symbol));
symbol->lvl = 14;
//simple hash function
struct hash *symbolTable = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
mu_assert("Could not created element in empty table.",
createHashElement(symbolTable, "blue", symbol) == 0);
mu_assert("Could not create element where bucket collison happened.",
createHashElement(symbolTable, "boo", symbol) == 0);
mu_assert("Level of symbol not same as current lex level.",
createHashElement(symbolTable, "blue", symbol) == 1);
setLexLevel(symbolTable, 14);
mu_assert("Head of symbol linked list hash same lex level as appending symbol.",
createHashElement(symbolTable, "blue", symbol) == 3);
struct Symbol *symbol2 = calloc(1, sizeof(struct Symbol));
symbol2->lvl = 1;
setLexLevel(symbolTable, 1);
mu_assert("Appending symbol is at a lower lex level then list head.",
createHashElement(symbolTable, "blue", symbol2) == 4);
symbol2->lvl = 15;
setLexLevel(symbolTable, 15);
mu_assert("Could not append symbol where expected.",
createHashElement(symbolTable, "blue", symbol2) == 0);
// createHashElement(symbolTable, "green", symbol) ;
// createHashElement(symbolTable, "red", symbol) ;
// dumpHash(symbolTable);
destroyHash(&symbolTable);
mu_assert("Call to destroyHash does not seg fault.", 1);
//normal hash function
symbolTable = createHash(&getHashedKeyNormal);
mu_assert("Call to createHash does not seg fault.", 1);
mu_assert("Could not created element in empty table.",
createHashElement(symbolTable, "blue", symbol) == 0);
mu_assert("Could not create element where bucket collison happened.",
createHashElement(symbolTable, "boo", symbol) == 0);
destroyHash(&symbolTable);
mu_assert("Call to destroyHash does not seg fault.", 1);
return NULL;
}
char *test_deleteHashElement_single() {
// simple hash function
struct Symbol *symbol = calloc(1, sizeof(struct Symbol));
struct hash *hash = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
createHashElement(hash, "GREEN", symbol);
createHashElement(hash, "bb", symbol);
createHashElement(hash, "bbb", symbol);
createHashElement(hash, "bbbb", symbol);
int index = getHashIndex(hash, "G");
struct hashElement *element = findHashElementByKey(hash, "GREEN");
mu_assert("unexpected element at in single bucket list.",
hash->elements[index] == element);
mu_assert("Delete function did not remove single bucket element.",
deleteHashElement(hash, "GREEN") == 0);
mu_assert("Single element not deleted.",
hash->elements[index] == NULL);
destroyHash(&hash);
mu_assert("Call to destroyHash does not seg fault.", 1);
//normal hash function
hash = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
createHashElement(hash, "GREEN", symbol);
createHashElement(hash, "bb", symbol);
createHashElement(hash, "bbb", symbol);
createHashElement(hash, "bbbb", symbol);
index = getHashIndex(hash, "G");
element = findHashElementByKey(hash, "GREEN");
mu_assert("unexpected element at in single bucket list.",
hash->elements[index] == element);
mu_assert("Delete function did not remove single bucket element.",
deleteHashElement(hash, "GREEN") == 0);
mu_assert("Single element not deleted.",
hash->elements[index] == NULL);
destroyHash(&hash);
mu_assert("Call to destroyHash does not seg fault.", 1);
return NULL;
}
char *test_getSizeOfBucket() {
//simple hash
struct Symbol *symbol = calloc(1, sizeof(struct Symbol));
struct hash *hash = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
mu_assert("Expected empty bucket.",
getSizeOfBucket(hash, "blue") == 0);
createHashElement(hash, "blue", symbol);
mu_assert("Expected bucket size of 1.",
getSizeOfBucket(hash, "blue") == 1);
createHashElement(hash, "red", symbol);
createHashElement(hash, "ruby", symbol);
createHashElement(hash, "rouse", symbol);
createHashElement(hash, "rose", symbol);
mu_assert("Expected bucket size of 4.",
getSizeOfBucket(hash, "r") == 4);
destroyHash(&hash);
mu_assert("Call to destroyHash does not seg fault.", 1);
//normal hash
hash = createHash(&getHashedKeyNormal);
mu_assert("Call to createHash does not seg fault.", 1);
mu_assert("Expected empty bucket.",
getSizeOfBucket(hash, "blue") == 0);
createHashElement(hash, "blue", symbol);
mu_assert("Expected bucket size of 1.",
getSizeOfBucket(hash, "blue") == 1);
createHashElement(hash, "red", symbol);
createHashElement(hash, "rose", symbol);
mu_assert("Expected bucket size of 0.",
getSizeOfBucket(hash, "r") == 0);
mu_assert("Expected bucket size of 4.",
getSizeOfBucket(hash, "red") == 1);
destroyHash(&hash);
mu_assert("Call to destroyHash does not seg fault.", 1);
return NULL;
}
char *test_createHash() {
//simple hash function
struct hash *symbolTable = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
int index = (*(symbolTable->hashFunction))("b");
mu_assert("Hash function does not caluculate expected value for key 'b'.",
index == 98);
index = (*(symbolTable->hashFunction))("a");
mu_assert("Hash function does not caluculate expected value for key 'a'.",
index == 97);
for (int i = 0; i < TABLE_SIZE; ++i) {
mu_assert("Empty table bucket does not equal NULL.",
symbolTable->elements[i] == NULL);
}
mu_assert("Expected initial lex level of 0",
symbolTable->lexLevel == 0);
destroyHash(&symbolTable);
mu_assert("Call to destroyHash does not seg fault.", 1);
//normal hash function
symbolTable = createHash(&getHashedKeyNormal);
mu_assert("Call to createHash does not seg fault.", 1);
index = (*(symbolTable->hashFunction))("A");
mu_assert("Hash function does not caluculate expected value for key 'A'.",
index == 638);
index = (*(symbolTable->hashFunction))("aaa");
mu_assert("Hash function does not caluculate expected value for key 'a'.",
index == 928);
for (int i = 0; i < TABLE_SIZE; ++i) {
mu_assert("Empty table bucket does not equal NULL.",
symbolTable->elements[i] == NULL);
}
mu_assert("Expected initial lex level of 0",
symbolTable->lexLevel == 0);
destroyHash(&symbolTable);
mu_assert("Call to destroyHash does not seg fault.", 1);
return NULL;
}
//can't test on normal hash function. can't find collisons
char *test_deleteHashElement_end() {
struct Symbol *symbol = calloc(1, sizeof(struct Symbol));
struct hash *hash = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
createHashElement(hash, "GREEN", symbol);
createHashElement(hash, "bb", symbol);
createHashElement(hash, "bbb", symbol);
createHashElement(hash, "bbbb", symbol);
struct hashElement *element = findHashElementByKey(hash, "bbbb");
struct hashElement *newTail = element->prev;
mu_assert("unexpected element at end of list.",
element->next == NULL);
mu_assert("Delete function did not remove end element.",
deleteHashElement(hash, "bbbb") == 0);
mu_assert("End element not deleted.",
newTail->next == NULL);
mu_assert("Did not delete end of list propertly.",
newTail->prev != NULL);
destroyHash(&hash);
mu_assert("Call to destroyHash does not seg fault.", 1);
return NULL;
}
//can't test on normal hash function. can't find collisons
char *test_deleteHashElement_begining() {
struct Symbol *symbol = calloc(1, sizeof(struct Symbol));
struct hash *hash = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
createHashElement(hash, "GREEN", symbol);
createHashElement(hash, "bb", symbol);
createHashElement(hash, "bbb", symbol);
createHashElement(hash, "bbbb", symbol);
struct hashElement *element = findHashElementByKey(hash, "bb");
struct hashElement *newHead = element->next;
int index = getHashIndex(hash, "b");
mu_assert("unexpected element at head of bucket list.",
hash->elements[index] == element);
mu_assert("Delete function did not remove begining element.",
deleteHashElement(hash, "bb") == 0);
mu_assert("Begining element not deleted.",
hash->elements[index] == newHead);
mu_assert("Did not delete front of list propertly.",
newHead->prev == NULL);
mu_assert("Resulting bucket should have more than one element.",
newHead->next != NULL);
destroyHash(&hash);
mu_assert("Call to destroyHash does not seg fault.", 1);
return NULL;
}
//can't test on normal hash function. can't find collisons
char *test_isKeyInBucket() {
struct Symbol *symbol = calloc(1, sizeof(struct Symbol));
struct hash *symbolTable = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
createHashElement(symbolTable, "blue", symbol);
createHashElement(symbolTable, "boo", symbol);
createHashElement(symbolTable, "bobby", symbol);
mu_assert("Could not find expected key 'blue' in bucket.",
isKeyInBucket(symbolTable, "blue") == 1 );
mu_assert("Could not find expected key 'blue' in bucket.",
isKeyInBucket(symbolTable, "bobby") == 1 );
mu_assert("Could not find expected key 'blue' in bucket.",
isKeyInBucket(symbolTable, "gree") == 0 );
mu_assert("Could not find expected key 'blue' in bucket.",
isKeyInBucket(symbolTable, "boo") == 1 );
destroyHash(&symbolTable);
mu_assert("Call to destroyHash does not seg fault.", 1);
return NULL;
}
//can't test on normal hash function. can't find collisons
char *test_deleteHashElement_middle() {
struct Symbol *symbol = calloc(1, sizeof(struct Symbol));
struct hash *hash = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
createHashElement(hash, "GREEN", symbol);
createHashElement(hash, "bb", symbol);
createHashElement(hash, "bbb", symbol);
createHashElement(hash, "bbbb", symbol);
struct hashElement *element = findHashElementByKey(hash, "bbb");
struct hashElement *head = findHashElementByKey(hash, "bb");
struct hashElement *tail = findHashElementByKey(hash, "bbbb");
mu_assert("unexpected element after head.",
head->next == element);
mu_assert("unexpected element before tail.",
tail->prev == element);
mu_assert("Delete function did not remove middle element.",
deleteHashElement(hash, "bbb") == 0);
mu_assert("Did not reset next in list propertly.",
head->next == tail);
mu_assert("Did not reset prev in list propertly",
tail->prev == head);
destroyHash(&hash);
mu_assert("Call to destroyHash does not seg fault.", 1);
return NULL;
}
char *test_dumpHash() {
struct hash *hash = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
dumpHash(hash);
mu_assert("Hash dump seg faults.", 1);
/*test 2: Copy stdout to to buffer and compare out of empty symbol table.*/
char buffer[100] = {0};
int out_pipe[2];
int saved_stdout = dup(STDOUT_FILENO); /* save stdout for display later */
pipe(out_pipe); /* make a pipe */
dup2(out_pipe[1], STDOUT_FILENO); /* redirect stdout to the pipe */
close(out_pipe[1]);
dumpHash(hash); /* anything sent to printf should now go down the pipe */
fflush(stdout);
read(out_pipe[0], buffer, 100); //read from pipe into buffer
dup2(saved_stdout, STDOUT_FILENO); /* reconnect stdout for testing */
char expectedOutput[33] = "\n\nDUMPING HASH:\nDUMP COMPLETE.\n\n";
mu_assert("Hash dumper gives unexpected output on empty hash.",
strcmp(expectedOutput, buffer) == 0);
destroyHash(&hash);
mu_assert("Call to destroyHash does not seg fault.", 1);
hash = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
// createHashElement(hash, "blue", 123);
// createHashElement(hash, "boo", 23456);
// createHashElement(hash, "bobby", 123);
// dumpHash(hash);
destroyHash(&hash);
mu_assert("Call to destroyHash does not seg fault.", 1);
return NULL;
}
char *test_destroyHash() {
struct hash *symbolTable = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
for (int i = 0; i < TABLE_SIZE; ++i) {
mu_assert("destroyed bucket does not equal NULL.",
symbolTable->elements[i] == NULL);
}
destroyHash(&symbolTable);
mu_assert("Call to destroyHash does not seg fault.", 1);
return NULL;
}
char *test_getHashIndex() {
//simple hash function
struct hash *symbolTable = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
int index = getHashIndex(symbolTable, "b");
mu_assert("getHashIndex does not return expected value on key 'b'",
index == 98);
index = getHashIndex(symbolTable, "W");
mu_assert("getHashIndex does not return expected value on key 'b'",
index == 87);
index = getHashIndex(symbolTable, "=");
mu_assert("getHashIndex does not return expected value on key 'b'",
index == 61);
destroyHash(&symbolTable);
mu_assert("Call to destroyHash does not seg fault.", 1);
//normal hash function
symbolTable = createHash(&getHashedKeyNormal);
mu_assert("Call to createHash does not seg fault.", 1);
index = getHashIndex(symbolTable, "b");
mu_assert("getHashIndex does not return expected value on key 'b'",
index == 671);
index = getHashIndex(symbolTable, "W");
mu_assert("getHashIndex does not return expected value on key 'b'",
index == 660);
destroyHash(&symbolTable);
mu_assert("Call to destroyHash does not seg fault.", 1);
return NULL;
}
//not testing on normal hash to hard to find collisons
char *test_isKeyCollison() {
struct Symbol *symbol = calloc(1, sizeof(struct Symbol));
struct hash *symbolTable = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
mu_assert("Collision on empty symbol table.",
isKeyCollison(symbolTable, "blue") == 0);
createHashElement(symbolTable, "blue", symbol);
mu_assert("Could not find key collision where one should exist.",
isKeyCollison(symbolTable, "blue") == 1);
mu_assert("Found key collision where one should not exist.",
isKeyCollison(symbolTable, "green") == 0);
destroyHash(&symbolTable);
mu_assert("Call to destroyHash does not seg fault.", 1);
return NULL;
}
//can't test on normal hash function. can't find collisons
char *test_getHashBucketHead() {
struct Symbol *symbol = calloc(1, sizeof(struct Symbol));
struct hash *symbolTable = createHash(&getHashedKeySimple);
mu_assert("Call to createHash does not seg fault.", 1);
createHashElement(symbolTable, "blue", symbol);
createHashElement(symbolTable, "boo", symbol);
createHashElement(symbolTable, "bobby", symbol);
mu_assert("Returned pointer when NULL was expected in getHashBucketHead.",
getHashBucketHead(symbolTable, "gree") == NULL );
mu_assert("Returned bucket head is incorrect.",
getHashBucketHead(symbolTable, "b") == findHashElementByKey(symbolTable, "blue") );
mu_assert("Returned bucket head is incorrect.",
getHashBucketHead(symbolTable, "bobby") != findHashElementByKey(symbolTable, "bobby") );
destroyHash(&symbolTable);
mu_assert("Call to destroyHash does not seg fault.", 1);
return NULL;
}
//This helper function resizes the data structure Hash.
//Handles collisions by using chaining as a means of rehashing.
//Returns void.
static void grow(Hash h){
Hash h2;
struct hash_table swap; //Temporary structure for the swaps
struct element *elt;
int i;
h2 = createHashIntern((h -> size) * GROWTH_FACTOR);
for (i = 0; i < h -> size; i++){
for (elt = h -> table[i]; elt != 0; elt = elt -> next){
insertElement(h2, elt -> key, elt -> value, elt -> length); //Recopies everything
}
}
//Swap the contents of the hash table h and h2.
swap = *h;
*h = *h2;
*h2 = swap;
destroyHash(h2);
}
int main(int argc, char* argv[])
{
//local variables
HashTable *inverted_index, *test_index;
char *target_dir, *results_file1, *results_file2, *rewritten_file;
target_dir = results_file1 = results_file2 = rewritten_file=NULL;
// 1. Program parameter processing
//./indexer [TARGET_DIRECTORY] [RESULTS FILENAME1] [RESULTS FILENAME2] [REWRITEN FILENAME]
//or
// ./indexer [TARGET_DIRECTORY] [RESULTS FILENAME]
if(argc == 3 || argc == 5) {
target_dir = calloc(strlen(argv[1])+1, sizeof(char));
if(target_dir == NULL) {
LOG(stdout,"Error: failed to calloc the variable target_dir\n");
return 1;
}
if( strcpy(target_dir, argv[1]) == NULL) {
LOG(stdout,"Error: failed to strcpy to the variable target_dir\n");
return 1;
}
//check if the directory is valid
if(IsDir(target_dir)) {
printf("%s is a directory\n", target_dir);
}
else {
printf("Error: %s is NOT a directory\n", target_dir);
free(target_dir);
return 1;
}
results_file1 = calloc(strlen(argv[2])+1, sizeof(char));
if(results_file1 == NULL) {
LOG(stdout,"Error: failed to calloc the variable results_file1\n");
return 1;
}
if( strcpy(results_file1, argv[2]) == NULL) {
LOG(stdout,"Error: failed to strcpy to the variable results_file1\n");
return 1;
}
if(argc == 5) {
results_file2 = calloc(strlen(argv[3])+1, sizeof(char));
if(results_file2 == NULL) {
LOG(stdout,"Error: failed to calloc the varaible results_file2\n");
free(target_dir);
free(results_file1);
return 1;
}
if( strcpy(results_file2, argv[3]) == NULL) {
LOG(stdout,"Error: failed to strcpy to variable results_file2\n");
free(target_dir);
free(results_file1);
return 1;
}
rewritten_file = calloc(strlen(argv[4])+1, sizeof(char));
if(rewritten_file == NULL) {
LOG(stdout,"Error: failed to calloc rewritten_file\n");
free(target_dir);
free(results_file1);
free(results_file2);
return 1;
}
if( strcpy(rewritten_file, argv[4]) == NULL) {
LOG(stdout,"Error: failed to strcpy to variable rewritten_file\n");
free(target_dir);
free(results_file1);
free(results_file2);
return 1;
}
}
}
else {
//wrong input parameters, inform the user about it.
printf("Usage: ./indexer [TARGET_DIRECTORY] [RESULTS FILENAME]\n");
printf(" or ./indexer [TARGET_DIRECTORY] [RESULTS FILENAME1] [RESULTS FILENAME2] [REWRITEN FILENAME]\n");
printf("You have entered %d parameters\n", (argc - 1));
return 1;
}
// 2. Initialize data structures
// allocate Inverted_index, zero it, and set links to NULL.
inverted_index = calloc(1, sizeof(HashTable));
if(inverted_index == NULL) {
LOG(stdout,"Error: failed to calloc the variable inverted_index\n");
return 1;
}
// 3.print to stdout that the building of index is starting
//printf("\nStarting to build the index\n");
LOG(stdout,"Starting to build the index\n");
// 4. Build the inverted index
if(buildIndexFromDirectory (inverted_index, target_dir)) {
LOG(stdout, "Error: buildIndexFromDirectory() failed\n");
return 1;
}
LOG(stdout,"Building the inverted index\n");
// 5. Save the inverted index
if(!saveFile(inverted_index, results_file1)) {
LOG(stdout,"Successfully saved the inverted index\n");
}
else {
LOG(stdout, "Error: saveFile() failed\n");
}
// 6. Free appropriate memory
destroyHash(inverted_index);
free(target_dir);
free(results_file1);
// 7. If 4 input parameters are detected
if(5 == argc) {
test_index = calloc(1, sizeof(HashTable));
if(test_index == NULL) {
LOG(stdout, "Error: failed to calloc the variable test_index\n");
free(results_file2);
free(rewritten_file);
destroyHash(test_index);
return 1;
}
if(!IsFile(results_file2)) {
printf("Error: %s is not a file\n", results_file2);
free(results_file2);
free(rewritten_file);
destroyHash(test_index);
return 1;
}
// 8. Inform user that testing has started
LOG(stdout, "Starting the testing process\n");
// 9. Reload the index from the file and rewrite it to a new file
if(readFile(test_index, results_file2)) {
LOG(stdout,"Error: readFile failed\n");
free(results_file2);
free(rewritten_file);
destroyHash(test_index);
return 1;
}
// 10. save the reconstructed index
if(saveFile(test_index, rewritten_file)) {
LOG(stdout,"Error: saveFile(test_index, rewritten_file) failed\n");
free(results_file2);
free(rewritten_file);
destroyHash(test_index);
return 1;
}
else {
LOG(stdout,"Successfully saved the inverted index\n");
}
// 11. free appropriate memory
free(results_file2);
free(rewritten_file);
destroyHash(test_index);
LOG(stdout,"Test is complete\n");
}
// 12. Finished
LOG(stdout,"indexer.c has finished\n");
return 0;
}
int main (int argc, char *argv[]) {
if (argc != 4) return 3;
int M = strtol(argv[1], NULL, 10); //get numerical value from string
//open input file
FILE *in = fopen(argv[2], "r");
if (!in) return 2;
//open output file
FILE *out = fopen(argv[3], "w");
if (!out) {
fclose(in);
return 2;
}
//buffer input which will be used to parse commands
//aux is used to check return value of function fgets
//aux tells us when to stop reading
char *aux = NULL, *buffer = NULL;
buffer = malloc(BSIZE * sizeof(char));
if (!buffer) {
fclose(in);
fclose(out);
return 2;
}
THash *H = initHash(M);
if (!H) {
fclose(in);
fclose(out);
free(buffer);
return 2;
}
aux = fgets(buffer, BSIZE, in);
while (aux) {
if (buffer[0] == '\n') break; //limit case when end of file reached (almost) and last line is '\n'
buffer[strlen(buffer) - 1] = '\0';
//try to parse commands
//if command is put and memory cannot be allocated then exit program (clean)
if (parse(H, buffer, out) == -1) {
fclose(in);
fclose(out);
free(buffer);
destroyHash(&H, freeInf);
return 2;
}
aux = fgets(buffer, BSIZE, in);
}
//free everything
destroyHash(&H, freeInf);
fclose(in);
fclose(out);
free(buffer);
return 0;
}