int search_negated_elements (ft_doc_index *doc_idx, List *negated_elements)
{
ListElement *cursor;
void *element;
ft_word_info word_info;
ASSERT (negated_elements);
for (element = List_getFirst (negated_elements, &cursor);
element;
element = List_getNext (&cursor))
{
if (*(int *)element == FT_WORD_TYPE)
{
if (search_word (doc_idx, element, &word_info) == TMAN_FOUND)
{
return TMAN_FOUND;
}
}
else if (*(int *)element == FT_PHRASE_TYPE)
{
if (search_phrase (doc_idx, element) == TMAN_FOUND)
{
return TMAN_FOUND;
}
}
}
return TMAN_NOT_FOUND;
}
void search_word(int start, int sub_index)
{
int i;
//printf("[%s] start %d sub_index %d\n", __FUNCTION__, start, sub_index);
if(start > input_size)
{
//printf("over input_size\n");
return;
}
if(sub_index > sub_size)
{
//printf("over sub_size\n");
return;
}
for(i = start; i < input_size; i++)
{
if(input_str[i] == sub_str[sub_index])
{
if(sub_index == sub_size -1)
{
// printf("match! start %d sub_index %d i %d\n", start, sub_index, i);
count++;
}
else
{
search_word(i+1, sub_index + 1);
}
}
}
}
int asm_cmd_pop(FILE* strin, var* vars, double* codes, int* cur_code)
{
if (strin == NULL) return ASM_STRIN_BAD;
if (codes == NULL) return ASM_BAD_CODES_POINTER;
if (cur_code == NULL) return ASM_BAD_CUR_CODE_POINTER;
char pop_word[CMD_MAXLEN] = {};
int cond = fscanf_s(strin, "%s", pop_word, CMD_MAXLEN);
if (cond <= 0) return ASM_SCAN_FAILED;
double reg = 0;
int ret = asm_str_convert(_strlwr(pop_word), ®);
if (ret != ASM_OK) return ASM_BAD_POP_STR_TOKEN;
if (reg == STR_var)
{
char pop_word_var[CMD_MAXLEN] = {};
int cond = fscanf_s(strin, "%s", pop_word_var, CMD_MAXLEN);
if (cond <= 0) return ASM_SCAN_FAILED;
ret = search_word(vars, pop_word_var);
if (ret < 0) return ASM_NO_SUCH_VAR;
SEC_RUN_(asm_push_code(codes, cur_code, reg));
SEC_RUN_(asm_push_code(codes, cur_code, ret));
return ASM_OK;
}
SEC_RUN_(asm_push_code(codes, cur_code, reg));
SEC_RUN_(asm_push_code(codes, cur_code, 0));
return ASM_OK;
}
int main(int argc, char** argv) {
init_dict_search();
char * targets[] = {
"yuppify", "bird-watcher", "arthritis", "ali, muhammad", "-monger",
"yukon", "women's studies", "not exits", "go", "zoo"
};
for(int i = 0; i < 10; ++i) {
char* result = search_word(targets[i]);
printf("%s, %p\n",targets[i], result);
}
}
static NJ_INT16 njd_search_word(NJ_CLASS *iwnn, NJ_CURSOR *cursor, NJ_UINT8 comp_flg,
NJ_UINT8 *exit_flag) {
NJ_INT16 ret;
ret = check_search_cursor(iwnn, cursor);
if (ret != 0) {
return ret;
}
return search_word(iwnn, cursor, comp_flg, exit_flag);
}
static double match_score_or_basic_pred (ft_doc_index *doc_idx, ft_basic_pred *ft_pred)
{
int num_non_negated_words, i;
double result;
ft_word_info *non_negated_words_info;
ft_word_info non_neg_words_info_static[10] = {'\0'};
word_syntree *word;
ft_phrase_info *phrase_info = NULL;
ListElement *cursor;
ASSERT (ft_pred);
// There should be no negated word in this basic predicate.
ASSERT (ft_pred->negated_query_elements == NULL);
num_non_negated_words = va_get_size (ft_pred->non_negated_words);
if (num_non_negated_words <= 10)
{
non_negated_words_info = non_neg_words_info_static;
}
else
{
non_negated_words_info = (ft_word_info *) tman_malloc (num_non_negated_words * sizeof(ft_word_info));
memset (non_negated_words_info, '\0', num_non_negated_words * sizeof(ft_word_info));
}
for (i = 0; i < num_non_negated_words; i++)
{
word = (word_syntree *)va_get_element (ft_pred->non_negated_words, i);
search_word (doc_idx, word, &(non_negated_words_info[i]));
}
// Do phrase matching.
if (ft_pred->phrases_words_info)
{
for (phrase_info = List_getFirst (ft_pred->phrases_words_info, &cursor); phrase_info;
phrase_info = List_getNext (&cursor))
{
phrase_pos_match (doc_idx, phrase_info, non_negated_words_info, num_non_negated_words);
}
}
// Score the basic predicate.
result = calculate_score (non_negated_words_info, num_non_negated_words);
if (num_non_negated_words > 10)
{
tman_free (non_negated_words_info);
}
return result;
}
/*
* 缓存KEYS在idx中的偏移信息，以便加快search_word的搜索速度
*/
void cache_idx(long *idx_cache)
{
int i;
long *p=idx_cache;
unsigned int unused;
rewind(pidx);
for(i=0; i<MAX_KEYS; i++){
if( search_word(KEY[i], &unused, &unused)){
p[i]=ftell(pidx);
}else p[i]=0;
}
}
int main()
{
int m,n,w,a,b,i,c,d,r,count;
r = 0;
//freopen("in6","r",stdin);
while (scanf("%d%d%d",&m,&n,&w) != -1)
{
if( r++ != 0 )
printf("\n");
for( i = 0; i < m; i++ )
scanf("%s",puzzle[i]);
memset(info,0,sizeof(info));
for( i = 0; i < w; i++ )
{
scanf("%s",words[i]);
info[i].size = strlen(words[i]);
}
count = w;
for( a = 0; a < m; a++ )
for( b = 0; b < n; b++ )
{
if( count == 0 )
goto escape;
for( c = 0; c < w; c++ )
{
if( info[c].find == true )
continue;
if( (d = search_word(m,n,a,b,c)) >= 0 )
{
count -= 1;
info[c].find = true;
info[c].dir = direct[d];
info[c].x = a;
info[c].y = b;
}
}
}
escape:
for( i = 0; i < w; i++ )
printf("%d %d %c\n",info[i].x,info[i].y,info[i].dir);
}
return 0;
}
//look for word in the given tree and saves result string to result
void search_word(TREE_NODE *root, W_TOKEN* word, wchar_t *result){
if(!root){//word not found
wcscpy(result, L"Consulta: ");
wcscat(result, ( (W_TOKEN *)word )->word);
wcscat(result, L" Palavra não encontrada\n");
return;
}
else if(lex_order(( (W_TOKEN *)( root->data ) ), word) == 0){//found word
wcscpy(result, L"Consulta: ");
wcscat(result, ( (W_TOKEN *)word )->word);
wcscat(result, L" Palavra encontrada nas linhas ");
wchar_t line[10000];
line[0] = '\0';
int first_number = 1;//used to remove comma from last number
LIST_NODE *node = ( (W_TOKEN *)( root->data ) )->list;
while(node){//print all lines on which word was found
if(!first_number)
wcscat(result, L", ");
swprintf(line, 1000, L"%d", *(int *)node->data);
wcscat(result, line);
node = node->next;
first_number = 0;
}
wcscat(result, L"\n");
return;
}
else if(lex_order(( (W_TOKEN *)( root->data ) ), word) < 0){//word bigger than root
search_word(root->right, word, result);
}
else if(lex_order(( (W_TOKEN *)( root->data ) ), word) > 0){//word smaller than root
search_word(root->left, word, result);
}
return;
}
void search(gdsl_bstree_t dict){
char *buffer=NULL;
size_t len;
gdsl_element_t result;
getchar();
printf("Input a word to search: ");
getline(&buffer,&len,stdin);
result = search_word(dict,buffer);
if(result==NULL){
printf("Could not find word: %s",buffer);
}else{
printf("That word is in the dictionary.\n");
}
//getchar();
}
//search for all words given in file_name and save results to output_file_name
void search_words(TREE_NODE *root, char * file_name, char * output_file_name){
FILE *file, *output_file;
file = fopen(file_name,"r");
output_file = fopen(output_file_name,"w");
if(!file){
printf("arquivo de entrada invalido\n");
return;
}
if(!output_file){
printf("arquivo de saida invalido\n");
return;
}
wchar_t line [ 10000 ];
wchar_t *result = malloc(sizeof(wchar_t) * 10000);
//start benchmark
double start_time, end_time, time_elapsed;
start_time = (double)clock();
while ( fgetws ( line, sizeof line, file ) != NULL ){//read line by line
if(line[wcslen(line) -1] == '\n') line[wcslen(line) -1] = '\0';//remove new line char
if(line[wcslen(line) -1] == ' ') line[wcslen(line) -1] = '\0';
W_TOKEN *token = malloc(sizeof(W_TOKEN));
str_to_lower(line);
token->word = line;
search_word(root, token, result);
fwprintf(output_file, L"%ls",result);
fputws ( result, stdout );
}
end_time = (double)clock();
time_elapsed = ( end_time - start_time )/CLOCKS_PER_SEC;
//end benchmark
fwprintf(output_file, L"%s","\n");
fwprintf(output_file, L"O tempo gasto na busca foi de %fms.\n", time_elapsed*1000);
printf("\n");
printf("O tempo gasto na busca foi de %fms.\n", time_elapsed*1000);
fclose(file);
fclose(output_file);
}
int main(){
struct dictionary *d = NULL, *tmp;
char *text = "Ciao a tutti ciao ciao a tutti tutti ciao ciao ciao tutti ciao a a";
char *word;
do{
word = next_word(text);
if(d==NULL)
d=new_dictionary_element(word, 1);
if( (tmp = search_word(d, word)) != NULL){
tmp->occurences++;
} else {
insert_word(d, word, 1);
}
if(text=strchr(text, ' '))
text++;
}while( text != NULL );
print_dictionary(d);
}
void search_word(gnode_t *pnode,int i,char tmp[])
{
i++;
while(pnode != NULL)
{
if(word_count >= win_x - 6 || word_count >= MAX_WORD)
return;
tmp[i - 1] = pnode->ch;
if(pnode->pchinese != NULL)
{
memcpy(word[word_count],tmp,i);
word_count++;
}
search_word(pnode->psub,i,tmp);
pnode = pnode->pnext;
}
}
void show_word(dict_t *pdict,char *english)
{
fill(6,2,win_x - 6,win_y / 3 - 2,' ');
fill(6,win_y / 3 + 1,win_x - 6,win_y - win_y / 3 - 1,' ');
gnode_t *pnode = search_node(pdict->phead,english);
char tmp[MAX_CH];
memset(word,0,MAX_WORD * MAX_CH);
if(pnode != NULL)
{
word_count = 1;
search_word(pnode->psub,0,tmp);
show_english(english);
show_chinese(pnode->pchinese,english);
}
}
/*
* 主要查询函数
*/
void consult()
{
char word[MAX_WORD+1];
char *data=NULL;
long idx[MAX_KEYS];
unsigned int offset, length;
cache_idx(idx);
while(1){
printf("INPUT A WORD OR PHRASE: ");
if(get_input(word, MAX_WORD)){
fseek(pidx,locate_idx(word,idx),SEEK_SET);
if(search_word(word,&offset,&length))
{
data=get_data(offset,length);
display_data(data,length);
free(data);
data=NULL;
}else fprintf(stderr,"SORRY, '%s' CANNOT BE FOUND!\n", word);
printf("\n----------------------------------------\n\n");
}else break;
}
}
int main(void)
{
int num_case;
scanf("%d", &num_case);
while(num_case)
{
scanf("%s", &input_str);
//printf("%s\n", input_str);
scanf("%s", &sub_str);
// printf("%s %d\n", sub_str, strlen(sub_str));
input_size = strlen(input_str);
sub_size = strlen(sub_str);
//algorithm
count = 0;
search_word(0, 0);
printf("%d\n",count);
num_case--;
}
return 0;
}
static int to_something(buffer *b, int (to_first)(int), int (to_rest)(int)) {
assert_buffer(b);
/* If we are after the end of the line, just return ERROR. */
if (b->cur_line == b->num_lines -1 && b->cur_pos >= b->cur_line_desc->line_len) return ERROR;
int64_t pos = b->cur_pos;
int c;
/* First of all, we search for the word start, if we're not over it. */
if (pos >= b->cur_line_desc->line_len || !ne_isword(c = get_char(&b->cur_line_desc->line[pos], b->encoding), b->encoding))
if (search_word(b, 1) != OK)
return ERROR;
bool changed = false;
int64_t new_len = 0;
pos = b->cur_pos;
const int64_t cur_char = b->cur_char;
const int cur_x = b->cur_x;
/* Then, we compute the word position extremes, length of the result (which
may change because of casing). */
while (pos < b->cur_line_desc->line_len && ne_isword(c = get_char(&b->cur_line_desc->line[pos], b->encoding), b->encoding)) {
const int new_c = new_len ? to_rest(c) : to_first(c);
changed |= (c != new_c);
if (b->encoding == ENC_UTF8) new_len += utf8seqlen(new_c);
else new_len++;
pos = next_pos(b->cur_line_desc->line, pos, b->encoding);
}
const int64_t len = pos - b->cur_pos;
if (!len) {
char_right(b);
return OK;
}
if (changed) {
/* We actually perform changes only if some character was case folded. */
char * word = malloc(new_len * sizeof *word);
if (!word) return OUT_OF_MEMORY;
pos = b->cur_pos;
new_len = 0;
/* Second pass: we actually build the transformed word. */
while (pos < b->cur_line_desc->line_len && ne_isword(c = get_char(&b->cur_line_desc->line[pos], b->encoding), b->encoding)) {
if (b->encoding == ENC_UTF8) new_len += utf8str(new_len ? to_rest(c) : to_first(c), word + new_len);
else {
word[new_len] = new_len ? to_rest(c) : to_first(c);
new_len++;
}
pos = next_pos(b->cur_line_desc->line, pos, b->encoding);
}
start_undo_chain(b);
delete_stream(b, b->cur_line_desc, b->cur_line, b->cur_pos, len);
if (new_len) insert_stream(b, b->cur_line_desc, b->cur_line, b->cur_pos, word, new_len);
free(word);
end_undo_chain(b);
if (cur_char < b->attr_len) b->attr_len = cur_char;
update_line(b, b->cur_line_desc, b->cur_y, cur_x, false);
need_attr_update = true;
}
return search_word(b, 1);
}
static double match_score_accrue_basic_pred (ft_doc_index *doc_idx, ft_basic_pred *ft_pred)
{
int num_non_negated_words = 0, status, i;
double result;
ft_word_info *non_negated_words_info = NULL;
ft_word_info non_neg_words_info_static[10] = {'\0'};
word_syntree *word;
ft_phrase_info *phrase_info = NULL;
ListElement *cursor;
ASSERT (ft_pred);
// First check for negated words, if any.
if (ft_pred->negated_query_elements != NULL)
{
status = search_negated_elements (doc_idx, ft_pred->negated_query_elements);
// If even one negated element was found, bail out.
if (status == TMAN_FOUND)
{
result = 0.0;
goto cleanup;
}
}
num_non_negated_words = va_get_size (ft_pred->non_negated_words);
if (num_non_negated_words <= 10)
{
non_negated_words_info = non_neg_words_info_static;
}
else
{
non_negated_words_info = (ft_word_info *) tman_malloc (num_non_negated_words * sizeof(ft_word_info));
memset (non_negated_words_info, '\0', num_non_negated_words * sizeof(ft_word_info));
}
for (i = 0; i < num_non_negated_words; i++)
{
word = (word_syntree *)va_get_element (ft_pred->non_negated_words, i);
search_word (doc_idx, word, &(non_negated_words_info[i]));
}
// Do phrase matching.
if (ft_pred->phrases_words_info)
{
for (phrase_info = List_getFirst (ft_pred->phrases_words_info, &cursor); phrase_info;
phrase_info = List_getNext (&cursor))
{
phrase_pos_match (doc_idx, phrase_info, non_negated_words_info, num_non_negated_words);
}
}
// Score the basic predicate.
result = calculate_score (non_negated_words_info, num_non_negated_words);
cleanup:
if (num_non_negated_words > 10)
{
tman_free (non_negated_words_info);
}
return result;
}
int main(void)
{
FILE *fp = fopen("dict.txt", "r"); //以读的方式打开文件“dict.tx t”
if(fp == NULL)
{
perror("open file dict.txt\n");
exit(0);
}
while(1)
{
char content[1000];
char Ess[1000];
char *token;
link p = make_node();
if(!fgets(content, 1000, fp)) //从文件读一行（单词），如果到末尾则跳出循环
break;
if(!fgets(Ess, 1000, fp)) //从文件读一行（内容），如果到末尾了跳出寻环
break;
if(head == NULL) //初始化链表
{
head = p;
tail = head;
}
else
{
tail -> next = p;
tail = tail -> next;
}
// fgets(content,100,fp);
if(content[strlen(content) - 1] == '\n') //如果数组到最后一个元素时候将‘\n’变为'\0'
content[strlen(content) - 1] = '\0';
strcpy(p -> word, content + 1); //由于第一个为#所以把第二个以后的元素复制给p -> word
// printf("%s\n",p->word);
// fgets(Ess,100,fp);
token = strtok(Ess, ":"); //截取：以前的内容
// printf("%s\n",token);
p->count = 0; //初始化
while((token = strtok(NULL, "@"))!=NULL)
{
p -> Explain[p -> count] = (char *)malloc(strlen(token) + 1);
strcpy(p -> Explain[p -> count], token);
p->count++;
}
print_word(p);
}
fclose(fp);
int i = 0;
while(1)
{
printf("take a choice:\n0 printf_word\n1 add a new word\n2 delete a word\n3 search a word\n4 write a 2file\n5 read_2file\n6 exit\n");
scanf("%d", &i);
switch(i)
{
case 0 :
printf_word(head);
break;
case 1:
add_word(head);
break;
case 2:
head = delete_word(head);
break;
case 3:
search_word(head);
break;
case 4:
write_2file(head);
break;
case 5:
read_2file();
break;
case 6:
printf("exit...\n");
save_word(head);
return 0;
break;
default:
continue;
break;
}
}
return 0;
}
static int search_phrase (ft_doc_index *doc_idx, phrase_syntree *phrase)
{
ft_word_info *words_info;
ft_word_info words_info_static[10] = {'\0'};
int words_pos_static[10] = {0};
ListElement *cursor;
word_syntree *word;
int i;
ft_phrase_info phrase_info;
int result = TMAN_FOUND;
ASSERT (phrase);
phrase_info.num_words = phrase->num_words;
phrase_info.type = FT_PHRASE_INFO_TYPE;
if (phrase->num_words > 10)
{
words_info = (ft_word_info *)tman_malloc (phrase->num_words * sizeof(ft_word_info));
memset (words_info, '\0', phrase->num_words * sizeof(ft_word_info));
phrase_info.word_positions = (int *)tman_malloc (phrase->num_words * sizeof (int));
memset (phrase_info.word_positions, 0, phrase->num_words * sizeof (int));
}
else
{
words_info = words_info_static;
phrase_info.word_positions = words_pos_static;
}
// First search all the words in the phrase. If there is even one word that is not present in the document,
// the search should fail.
for (word = List_getFirst (phrase->word_list, &cursor), i = 0; word; word = List_getNext (&cursor), i++)
{
// Word positions are the same as they appear in the word list in phrase node.
phrase_info.word_positions[i] = i;
if (search_word (doc_idx, word, &(words_info[i])) == TMAN_NOT_FOUND)
{
result = TMAN_NOT_FOUND;
break;
}
}
if (result == TMAN_NOT_FOUND)
{
goto cleanup;
}
else
{
// If all the words are present in the document, search for their sequencing.
if (phrase_pos_match (doc_idx, &phrase_info, words_info, phrase->num_words) == TMAN_NOMATCH)
{
result = TMAN_NOT_FOUND;
}
}
cleanup:
if (phrase->num_words > 10)
{
tman_free (words_info);
tman_free (phrase_info.word_positions);
}
return (result);
}
