Class: WordScoop
- Inherits:
-
Object
- Object
- WordScoop
- Defined in:
- lib/word_scoop.rb,
lib/word_scoop/version.rb,
ext/word_scoop/word_scoop.c
Constant Summary collapse
- VERSION =
'2.1.2'
Instance Attribute Summary collapse
-
#link_url ⇒ Object
Returns the value of attribute link_url.
Class Method Summary collapse
-
.new(*args) ⇒ Object
new.
Instance Method Summary collapse
-
#add(str) ⇒ Object
(also: #<<)
add.
-
#filter_html(str) ⇒ Object
filter_html.
-
#search(str) ⇒ Object
search.
Instance Attribute Details
#link_url ⇒ Object
Returns the value of attribute link_url.
7 8 9 |
# File 'lib/word_scoop.rb', line 7 def link_url @link_url end |
Class Method Details
.new(*args) ⇒ Object
new
104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 |
# File 'ext/word_scoop/word_scoop.c', line 104
static VALUE t_new(int argc, VALUE *argv, VALUE klass)
{
node root;
VALUE obj, array, string;
root = initialize_node(NULL_CHAR);
obj = Data_Make_Struct(klass, struct _node, NULL, destroy_node, root);
if (argc == 1) {
array = argv[0];
while((string = rb_ary_shift(argv[0])) != Qnil) {
t_add(obj, string);
}
}
return obj;
}
|
Instance Method Details
#add(str) ⇒ Object Also known as: <<
add
126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 |
# File 'ext/word_scoop/word_scoop.c', line 126
static VALUE t_add(VALUE self, VALUE str)
{
node root, now;
char *keyword;
int i, len;
keyword = StringValuePtr(str);
len = strlen(keyword);
while(keyword[len - 1] == CR || keyword[len - 1] == LF ||
keyword[len - 1] == TAB || keyword[len - 1] == SPACE) {
len--;
}
if (len < 1) {
return Qfalse;
}
Data_Get_Struct(self, struct _node, root);
now = root;
for(i = 0; i < len; i++) {
now = search_child_or_create(now, keyword[i]);
}
now->end_flag = true;
return str;
}
|
#filter_html(str) ⇒ Object
filter_html
216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 |
# File 'ext/word_scoop/word_scoop.c', line 216
static VALUE t_filter_html(VALUE self, VALUE str)
{
node root, now, ret;
bool in_tag;
char *text, *inner_tag;
int i, head_i, tail_i, copy_head_i, total_len;
VALUE change_str, url_base, word;
rb_encoding *enc;
change_str = rb_str_new2(EMPTY_STRING);
enc = rb_enc_get(str);
text = StringValuePtr(str);
Data_Get_Struct(self, struct _node, root);
url_base = rb_iv_get(self, LINK_URL_VARIABLE);
if (url_base == Qnil) {
url_base = rb_str_new2(DEAULT_LINK_URL);
}
now = root;
total_len = strlen(text);
head_i = -1;
tail_i = -1;
copy_head_i = 0;
in_tag = false;
inner_tag = NULL;
for(i = 0; i <= total_len; i++) {
if (!in_tag && text[i] == BEGIN_TAG) {
in_tag = true;
if (strncasecmp(&text[i + 1], A_TAG, strlen(A_TAG)) == 0) {
inner_tag = A_TAG;
} else if (strncasecmp(&text[i + 1], SCRIPT_TAG, strlen(SCRIPT_TAG)) == 0) {
inner_tag = SCRIPT_TAG;
} else if (strncasecmp(&text[i + 1], PRE_TAG, strlen(PRE_TAG)) == 0) {
inner_tag = PRE_TAG;
} else if (strncasecmp(&text[i + 1], IFRAME_TAG, strlen(IFRAME_TAG)) == 0) {
inner_tag = IFRAME_TAG;
} else if (strncasecmp(&text[i + 1], OBJECT_TAG, strlen(OBJECT_TAG)) == 0) {
inner_tag = OBJECT_TAG;
}
continue;
}
if (in_tag && !inner_tag && text[i] == END_TAG) {
in_tag = false;
continue;
}
if (inner_tag && text[i] == BEGIN_TAG) {
if (strncasecmp(&text[i + 2], inner_tag, strlen(inner_tag)) == 0) {
inner_tag = NULL;
continue;
}
}
if (in_tag) {
continue;
}
ret = search_child(now, text[i]);
if (ret && i != total_len) {
if (head_i == -1) {
head_i = i;
}
if (ret->end_flag) {
tail_i = i;
}
now = ret;
} else {
if (head_i != -1) {
if (tail_i != -1) {
if (copy_head_i < head_i) {
rb_funcall(
change_str,
rb_intern("concat"),
1,
add_encode(rb_str_new(&text[copy_head_i], (head_i - copy_head_i)), enc)
);
}
word = rb_str_new(&text[head_i], (tail_i - head_i + 1));
rb_funcall(
change_str,
rb_intern("concat"),
1,
add_encode(rb_funcall(url_base, rb_intern("%"), 1, rb_assoc_new(word, word)), enc)
);
i = tail_i;
copy_head_i = tail_i + 1;
tail_i = -1;
} else {
i = head_i;
}
head_i = -1;
}
now = root;
}
}
if (copy_head_i == 0) {
return str;
} else {
rb_funcall(
change_str,
rb_intern("concat"),
1,
add_encode(rb_str_new(&text[copy_head_i], (total_len - copy_head_i)), enc)
);
return change_str;
}
}
|
#search(str) ⇒ Object
search
159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 |
# File 'ext/word_scoop/word_scoop.c', line 159
static VALUE t_search(VALUE self, VALUE str)
{
node root, now, ret;
char *text;
int i, head_i, tail_i, total_len;
VALUE array;
rb_encoding *enc;
array = rb_ary_new();
enc = rb_enc_get(str);
text = StringValuePtr(str);
Data_Get_Struct(self, struct _node, root);
now = root;
total_len = strlen(text);
head_i = -1;
tail_i = -1;
for(i = 0; i <= total_len; i++) {
ret = search_child(now, text[i]);
if (ret && i != total_len) {
if (head_i == -1) {
head_i = i;
}
if (ret->end_flag) {
tail_i = i;
}
now = ret;
} else {
if (head_i != -1) {
if (tail_i != -1) {
rb_funcall(
array,
rb_intern("push"),
1,
add_encode(rb_str_new(&text[head_i], (tail_i - head_i + 1)), enc)
);
i = tail_i;
tail_i = -1;
} else {
i = head_i;
}
head_i = -1;
}
now = root;
}
}
return array;
}
|