Class: CBloomFilter

Inherits:
Object
  • Object
show all
Defined in:
ext/cbloomfilter/cbloomfilter.c

Instance Method Summary collapse

Constructor Details

#initialize(*args) ⇒ Object



118
119
120
121
122
123
124
125
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
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
# File 'ext/cbloomfilter/cbloomfilter.c', line 118

static VALUE bf_initialize(int argc, VALUE *argv, VALUE self) {
    struct BloomFilter *bf;
    VALUE arg1, arg2;
    long m_value, k_value;
    int m, k;

    bf = bf_ptr(self);

    if (argc > 2) {
        rb_error_arity(argc, 0, 2);
    }

    /* defaults */
    arg1 = INT2FIX(1000);
    arg2 = INT2FIX(4);

    switch (argc) {
        case 2:
      arg2 = argv[1];
        case 1:
      arg1 = argv[0];
      break;
    }

    m_value = NUM2LONG(arg1);
    k_value = NUM2LONG(arg2);

    if (m_value > INT_MAX - 15)
        rb_raise(rb_eRangeError, "bit length is too large");
    if (k_value > INT_MAX)
        rb_raise(rb_eRangeError, "hash length is too large");

    m = (int) m_value;
    k = (int) k_value;

    if (m < 1)
        rb_raise(rb_eArgError, "bit length must be >= 1");
    if (k < 1)
        rb_raise(rb_eArgError, "hash length must be >= 1");
    if (k > salts_length)
        rb_raise(rb_eArgError, "hash length must be <= %d", salts_length);

    bf->m = m;
    bf->k = k;

    ruby_xfree(bf->ptr);
    bf->ptr = NULL;
    bf->bytes = 0;
    /* Preserve the existing serialized bitmap length, including one padding byte. */
    bf->bytes = (m + 15) / 8;
    bf->ptr = ALLOC_N(unsigned char, bf->bytes);

    /* initialize the bits with zeros */
    memset(bf->ptr, 0, bf->bytes);

    return self;
}

Instance Method Details

#&(other) ⇒ Object



248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
# File 'ext/cbloomfilter/cbloomfilter.c', line 248

static VALUE bf_and(VALUE self, VALUE other) {
    struct BloomFilter *bf = bf_ptr(self);
    struct BloomFilter *bf_other = bf_ptr(other);
    struct BloomFilter *target;
    VALUE klass, obj, args[5];
    int i;

    bf_ensure_compatible(bf, bf_other);

    args[0] = INT2FIX(bf->m);
    args[1] = INT2FIX(bf->k);
    klass = rb_funcall(self,rb_intern("class"),0);
    obj = rb_class_new_instance(2, args, klass);
    target = bf_ptr(obj);
    for (i = 0; i < bf->bytes; i++){
        target->ptr[i] = bf->ptr[i] & bf_other->ptr[i];
    }

    return obj;
}

#add(key) ⇒ Object



209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
# File 'ext/cbloomfilter/cbloomfilter.c', line 209

static VALUE bf_add(VALUE self, VALUE key) {
    VALUE skey;
    unsigned long hash;
    int index;
    int i, len, m, k;
    char *ckey;
    struct BloomFilter *bf = bf_ptr(self);

    skey = rb_obj_as_string(key);
    ckey = StringValuePtr(skey);
    len = (int) (RSTRING_LEN(skey)); /* length of the string in bytes */

    m = bf->m;
    k = bf->k;

    hash = (unsigned long) djb2(ckey, len);
    for (i = 0; i <= k - 1; i++) {
        index = (int) ((hash ^ salts[i]) % (unsigned int) (m));

        /*  set a bit at the index */
        bucket_set(bf, index);
    }

    return Qnil;
}

#bitmapObject



318
319
320
321
322
323
324
325
326
327
# File 'ext/cbloomfilter/cbloomfilter.c', line 318

static VALUE bf_bitmap(VALUE self) {
    struct BloomFilter *bf = bf_ptr(self);

    VALUE str = rb_str_new(0, bf->bytes);
    unsigned char* ptr = (unsigned char *) RSTRING_PTR(str);

    memcpy(ptr, bf->ptr, bf->bytes);

    return str;
}

#clearObject



176
177
178
179
180
# File 'ext/cbloomfilter/cbloomfilter.c', line 176

static VALUE bf_clear(VALUE self) {
    struct BloomFilter *bf = bf_ptr(self);
    memset(bf->ptr, 0, bf->bytes);
    return Qtrue;
}

#include?(key) ⇒ Boolean

Returns:

  • (Boolean)


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
# File 'ext/cbloomfilter/cbloomfilter.c', line 290

static VALUE bf_include(VALUE self, VALUE key) {
    VALUE skey;
    unsigned long hash;
    int index;
    int i, len, m, k;
    char *ckey;
    struct BloomFilter *bf = bf_ptr(self);

    skey = rb_obj_as_string(key);
    ckey = StringValuePtr(skey);
    len = (int) (RSTRING_LEN(skey)); /* length of the string in bytes */

    m = bf->m;
    k = bf->k;

    hash = (unsigned long) djb2(ckey, len);
    for (i = 0; i <= k - 1; i++) {
        index = (int) ((hash ^ salts[i]) % (unsigned int) (m));

        /* check the bit at the index */
        if (!bucket_check(bf, index)) {
            return Qfalse; /* i.e., it is a new entry ; escape the loop */
        }
    }

    return Qtrue;
}

#kObject



187
188
189
190
# File 'ext/cbloomfilter/cbloomfilter.c', line 187

static VALUE bf_k(VALUE self) {
    struct BloomFilter *bf = bf_ptr(self);
    return INT2FIX(bf->k);
}

#load(bitmap) ⇒ Object



329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
# File 'ext/cbloomfilter/cbloomfilter.c', line 329

static VALUE bf_load(VALUE self, VALUE bitmap) {
    struct BloomFilter *bf = bf_ptr(self);
    VALUE bitmap_string = StringValue(bitmap);
    unsigned char* ptr;

    if (RSTRING_LEN(bitmap_string) != bf->bytes) {
        rb_raise(rb_eArgError, "bitmap length must be %d bytes", bf->bytes);
    }

    ptr = (unsigned char *) RSTRING_PTR(bitmap_string);

    memcpy(bf->ptr, ptr, bf->bytes);
    bf_clear_padding_bits(bf);

    return Qnil;
}

#mObject



182
183
184
185
# File 'ext/cbloomfilter/cbloomfilter.c', line 182

static VALUE bf_m(VALUE self) {
    struct BloomFilter *bf = bf_ptr(self);
    return INT2FIX(bf->m);
}

#merge(other) ⇒ Object



235
236
237
238
239
240
241
242
243
244
245
246
# File 'ext/cbloomfilter/cbloomfilter.c', line 235

static VALUE bf_merge(VALUE self, VALUE other) {
    struct BloomFilter *bf = bf_ptr(self);
    struct BloomFilter *target = bf_ptr(other);
    int i;

    bf_ensure_compatible(bf, target);

    for (i = 0; i < bf->bytes; i++) {
        bf->ptr[i] |= target->ptr[i];
    }
    return Qnil;
}

#set_bitsObject



192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
# File 'ext/cbloomfilter/cbloomfilter.c', line 192

static VALUE bf_set_bits(VALUE self){
    struct BloomFilter *bf = bf_ptr(self);
    int i, count = 0;

    for (i = 0; i < bf->bytes; i++) {
        unsigned char byte = bf->ptr[i];

        /* Brian Kernighan’s bit-count loop a*/
        while (byte != 0) {
            byte &= (unsigned char) (byte - 1);
            count++;
        }
    }

    return INT2FIX(count);
}

#|(other) ⇒ Object



269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
# File 'ext/cbloomfilter/cbloomfilter.c', line 269

static VALUE bf_or(VALUE self, VALUE other) {
    struct BloomFilter *bf = bf_ptr(self);
    struct BloomFilter *bf_other = bf_ptr(other);
    struct BloomFilter *target;
    VALUE klass, obj, args[5];
    int i;

    bf_ensure_compatible(bf, bf_other);

    args[0] = INT2FIX(bf->m);
    args[1] = INT2FIX(bf->k);
    klass = rb_funcall(self,rb_intern("class"),0);
    obj = rb_class_new_instance(2, args, klass);
    target = bf_ptr(obj);
    for (i = 0; i < bf->bytes; i++){
        target->ptr[i] = bf->ptr[i] | bf_other->ptr[i];
    }

    return obj;
}