Class: BloomFilter

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

Constant Summary collapse

VERSION =
rb_str_new2(RUBY_BLOOM_FILTER_VERSION)

Class Method Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(*args) ⇒ Object



41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
# File 'ext/bloom_filter.c', line 41

VALUE bloom_initialize(int argc, VALUE *argv, VALUE self) {
    double error;
    size_t nbits, nhash, nmax;

    VALUE max_size, error_rate, bitmap_size, hash_count, options;
    BloomFilter *filter = 0;

    rb_scan_args(argc, argv, "01", &options);
    if (!NIL_P(options) && TYPE(options) != T_HASH)
        rb_raise(rb_eArgError, "invalid options, expect hash");

    if (NIL_P(options)) {
        nbits = 1000000;
        nhash = 4;
    }
    else {
        max_size    = rb_hash_aref(options, ID2SYM(rb_intern("size")));
        error_rate  = rb_hash_aref(options, ID2SYM(rb_intern("error_rate")));
        bitmap_size = rb_hash_aref(options, ID2SYM(rb_intern("bits")));
        hash_count  = rb_hash_aref(options, ID2SYM(rb_intern("hashes")));

        nhash = NIL_P(hash_count) ? 4 : NUM2ULONG(hash_count);

        if (!NIL_P(bitmap_size))
            nbits = NUM2ULONG(bitmap_size);
        else if (!NIL_P(max_size)) {
            nmax  = NUM2ULONG(max_size);
            error = NIL_P(error_rate) ? 0.01 : NUM2DBL(error_rate);
            nbits = ceil(fabs(log(error) * (double)nmax / pow(log(2), 2)));
            nhash = ceil(0.7 * (double)nbits / (double)nmax);
        }
        else
            rb_raise(rb_eArgError, "requires either size & error_rate or bits & hashes");
    }

    filter = bloom_filter_new(nbits, string_nocase_hash, nhash);

    if (!filter)
        rb_raise(rb_eNoMemError, "unable to allocate memory for BloomFilter");

    DATA_PTR(self) = filter;
    return self;
}

Class Method Details

.load(file) ⇒ Object



194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
# File 'ext/bloom_filter.c', line 194

VALUE bloom_load(VALUE klass, VALUE file) {
    int fd;
    void *buffer;
    size_t nbits, bytes;
    FileHeader header;
    BloomFilter *filter;
    VALUE instance;

    fd = open(CSTRING(file), O_RDONLY);
    if (fd == -1)
        rb_raise(rb_eIOError, "unable to open file: %s", strerror(errno));

    if (read(fd, &header, sizeof(header)) != sizeof(header)) {
        close(fd);
        rb_raise(rb_eIOError, "unable to read file, header corrupted\n");
    }

    nbits  = (header.table_size + 7) / 8;
    buffer = malloc(nbits);
    if (!buffer) {
        close(fd);
        rb_raise(rb_eNoMemError, "out of memory dumping BloomFilter");
    }

    bytes = read(fd, buffer, nbits);
    if (bytes != nbits) {
        free(buffer);
        close(fd);
        rb_raise(rb_eStandardError, "unable to load BloomFilter, expected %ld but got %ld bytes", nbits, bytes);
    }

    filter = bloom_filter_new(header.table_size, string_nocase_hash, header.num_functions);
    bloom_filter_load(filter, buffer);
    instance = Data_Wrap_Struct(klass, 0, bloom_free, filter);

    free(buffer);
    close(fd);
    return instance;
}

Instance Method Details

#binaryObject



168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
# File 'ext/bloom_filter.c', line 168

VALUE bloom_binary(VALUE klass) {
    BloomFilter *filter = bloom_handle(klass);

    VALUE bitmap;
    unsigned char *buffer;

    int nbytes = (filter->table_size + 7) / 8;
    buffer = (unsigned char *)malloc(nbytes);

    if (!buffer)
        rb_raise(rb_eNoMemError, "out of memory dumping BloomFilter");

    bloom_filter_read(filter, buffer);

    bitmap = rb_str_new((const char *)buffer, nbytes);
    free(buffer);
    return bitmap;
}

#binary=(buffer) ⇒ Object



187
188
189
190
191
192
# File 'ext/bloom_filter.c', line 187

VALUE bloom_binary_set(VALUE klass, VALUE buffer) {
    BloomFilter *filter = bloom_handle(klass);
    unsigned char* ptr = (unsigned char *) RSTRING_PTR(buffer);
    bloom_filter_load(filter, ptr);
    return Qtrue;
}

#bitsObject

not reachable



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
# File 'ext/bloom_filter.c', line 141

VALUE bloom_bits(VALUE klass) {
    BloomFilter *filter = bloom_handle(klass);

    VALUE bitmap;
    char *buffer;
    unsigned char b;
    int i = 0, bit, nbits = filter->table_size;

    buffer = (char *)malloc(nbits);
    if (!buffer)
        rb_raise(rb_eNoMemError, "out of memory dumping BloomFilter");

    for (i = 0; i < nbits; i++) {
        b = filter->table[i / 8];
        bit = 1 << (i % 8);

        if ((b & bit) == 0)
            buffer[i] = '0';
	    else
	        buffer[i] = '1';
    }

    bitmap = rb_str_new(buffer, nbits);
    free(buffer);
    return bitmap;
}

#dump(file) ⇒ Object



97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
# File 'ext/bloom_filter.c', line 97

VALUE bloom_dump(VALUE klass, VALUE file) {
    int fd;
    void *buffer;
    uint64_t nbits;
    FileHeader header;
    BloomFilter *filter = bloom_handle(klass);

    nbits  = (filter->table_size + 7) / 8;
    buffer = malloc(nbits);

    if (!buffer)
        rb_raise(rb_eNoMemError, "out of memory dumping BloomFilter");

    bloom_filter_read(filter, buffer);

    fd = open(CSTRING(file), O_WRONLY|O_CREAT|O_TRUNC, S_IRUSR|S_IWUSR);
    if (fd == -1) {
        free(buffer);
        rb_raise(rb_eIOError, "unable to open file. %s", strerror(errno));
    }

    header.table_size    = filter->table_size;
    header.num_functions = filter->num_functions;

    if (write(fd, &header, sizeof(header)) == -1) {
        free(buffer);
        close(fd);
        rb_raise(rb_eIOError, "error dumping BloomFilter: %s\n", strerror(errno));
    }

    if (write(fd, buffer, nbits) != -1) {
        free(buffer);
        close(fd);
        return Qtrue;
    }
    else {
        free(buffer);
        close(fd);
        rb_raise(rb_eIOError, "error dumping BloomFilter: %s\n", strerror(errno));
    }

    return Qfalse; // not reachable
}

#include?(string) ⇒ Boolean

Returns:

  • (Boolean)


92
93
94
95
# File 'ext/bloom_filter.c', line 92

VALUE bloom_include(VALUE klass, VALUE string) {
    BloomFilter *filter = bloom_handle(klass);
    return bloom_filter_query(filter, (BloomFilterValue)CSTRING(string)) ? Qtrue : Qfalse;
}

#insert(string) ⇒ Object



86
87
88
89
90
# File 'ext/bloom_filter.c', line 86

VALUE bloom_insert(VALUE klass, VALUE string) {
    BloomFilter *filter = bloom_handle(klass);
    bloom_filter_insert(filter, (BloomFilterValue)CSTRING(string));
    return Qtrue;
}