Module: Snappy

Defined in:
lib/snappy.rb,
ext/snappy/snappy.c

Defined Under Namespace

Classes: Error

Constant Summary collapse

VERSION = 
"1.0.0"

Class Method Summary collapse

Class Method Details

.compress(*args) ⇒ String

Compress a string using Snappy compression

Parameters:

  • data (String) —

    the data to compress

  • options (Hash) —

    options hash

Returns:

  • (String) —

    compressed data



22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
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
 
# File 'ext/snappy/snappy.c', line 22

static VALUE rb_snappy_compress(int argc, VALUE *argv, VALUE self) {
    VALUE data, options;
    int level = DEFAULT_COMPRESSION_LEVEL;

    rb_scan_args(argc, argv, "11", &data, &options);

    // Validate input
    Check_Type(data, T_STRING);

    // Parse options
    if (!NIL_P(options)) {
        Check_Type(options, T_HASH);
        VALUE level_val = rb_hash_aref(options, ID2SYM(rb_intern("level")));
        if (!NIL_P(level_val)) {
            level = NUM2INT(level_val);
            if (level < MIN_COMPRESSION_LEVEL || level > MAX_COMPRESSION_LEVEL) {
                rb_raise(rb_eArgError, "compression level must be between %d and %d",
                        MIN_COMPRESSION_LEVEL, MAX_COMPRESSION_LEVEL);
            }
        }
    }

    const char *input = RSTRING_PTR(data);
    size_t input_length = RSTRING_LEN(data);

    // Get maximum compressed length
    size_t max_compressed_length = snappy_max_compressed_length(input_length);

    // Allocate buffer for compressed data
    char *compressed = ALLOC_N(char, max_compressed_length);
    size_t compressed_length = max_compressed_length;

    // Perform compression
    snappy_status status = snappy_compress(input, input_length, compressed, &compressed_length);

    if (status != SNAPPY_OK) {
        xfree(compressed);
        rb_raise(rb_eSnappyError, "compression failed with status: %d", status);
    }

    // Create Ruby string with compressed data
    VALUE result = rb_str_new(compressed, compressed_length);
    xfree(compressed);

    return result;
}

.compress_file(*args) ⇒ Object

Compress a file using Snappy compression

Parameters:

  • input_path (String) —

    path to input file

  • output_path (String) —

    path to output file

  • options (Hash) —

    options hash



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
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
 
# File 'ext/snappy/snappy.c', line 126

static VALUE rb_snappy_compress_file(int argc, VALUE *argv, VALUE self) {
    VALUE input_path, output_path, options;
    int level = DEFAULT_COMPRESSION_LEVEL;

    rb_scan_args(argc, argv, "21", &input_path, &output_path, &options);

    Check_Type(input_path, T_STRING);
    Check_Type(output_path, T_STRING);

    // Parse options
    if (!NIL_P(options)) {
        Check_Type(options, T_HASH);
        VALUE level_val = rb_hash_aref(options, ID2SYM(rb_intern("level")));
        if (!NIL_P(level_val)) {
            level = NUM2INT(level_val);
            if (level < MIN_COMPRESSION_LEVEL || level > MAX_COMPRESSION_LEVEL) {
                rb_raise(rb_eArgError, "compression level must be between %d and %d",
                        MIN_COMPRESSION_LEVEL, MAX_COMPRESSION_LEVEL);
            }
        }
    }

    const char *input_file = StringValueCStr(input_path);
    const char *output_file = StringValueCStr(output_path);

    // Open input file
    FILE *in = fopen(input_file, "rb");
    if (!in) {
        rb_sys_fail(input_file);
    }

    // Get file size
    fseek(in, 0, SEEK_END);
    long file_size = ftell(in);
    fseek(in, 0, SEEK_SET);

    // Read file contents
    char *input_data = ALLOC_N(char, file_size);
    size_t bytes_read = fread(input_data, 1, file_size, in);
    fclose(in);

    if (bytes_read != file_size) {
        xfree(input_data);
        rb_raise(rb_eIOError, "failed to read complete file");
    }

    // Compress data
    size_t max_compressed_length = snappy_max_compressed_length(file_size);
    char *compressed = ALLOC_N(char, max_compressed_length);
    size_t compressed_length = max_compressed_length;

    snappy_status status = snappy_compress(input_data, file_size, compressed, &compressed_length);
    xfree(input_data);

    if (status != SNAPPY_OK) {
        xfree(compressed);
        rb_raise(rb_eSnappyError, "compression failed with status: %d", status);
    }

    // Write compressed data to output file
    FILE *out = fopen(output_file, "wb");
    if (!out) {
        xfree(compressed);
        rb_sys_fail(output_file);
    }

    size_t bytes_written = fwrite(compressed, 1, compressed_length, out);
    fclose(out);
    xfree(compressed);

    if (bytes_written != compressed_length) {
        rb_raise(rb_eIOError, "failed to write complete compressed data");
    }

    return Qnil;
}

.decompress(data) ⇒ String

Decompress Snappy compressed data

Parameters:

  • data (String) —

    the compressed data

Returns:

  • (String) —

    decompressed data



75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
 
# File 'ext/snappy/snappy.c', line 75

static VALUE rb_snappy_decompress(VALUE self, VALUE data) {
    Check_Type(data, T_STRING);

    const char *compressed = RSTRING_PTR(data);
    size_t compressed_length = RSTRING_LEN(data);

    // Get uncompressed length
    size_t uncompressed_length;
    snappy_status status = snappy_uncompressed_length(compressed, compressed_length, &uncompressed_length);

    if (status != SNAPPY_OK) {
        rb_raise(rb_eSnappyError, "failed to get uncompressed length, status: %d", status);
    }

    // Allocate buffer for decompressed data
    char *uncompressed = ALLOC_N(char, uncompressed_length);

    // Perform decompression
    status = snappy_uncompress(compressed, compressed_length, uncompressed, &uncompressed_length);

    if (status != SNAPPY_OK) {
        xfree(uncompressed);
        rb_raise(rb_eSnappyError, "decompression failed with status: %d", status);
    }

    // Create Ruby string with decompressed data
    VALUE result = rb_str_new(uncompressed, uncompressed_length);
    xfree(uncompressed);

    // Try to set UTF-8 encoding if the data is valid UTF-8
    // Otherwise, keep it as binary (ASCII-8BIT)
    int utf8_index = rb_utf8_encindex();
    rb_enc_associate_index(result, utf8_index);

    // Check if the string is valid in the UTF-8 encoding
    // If not, fall back to binary encoding
    if (rb_enc_str_coderange(result) == ENC_CODERANGE_BROKEN) {
        rb_enc_associate_index(result, rb_ascii8bit_encindex());
    }

    return result;
}

.decompress_file(input_path, output_path) ⇒ Object

Decompress a Snappy compressed file

Parameters:

  • input_path (String) —

    path to compressed input file

  • output_path (String) —

    path to decompressed output file



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
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
 
# File 'ext/snappy/snappy.c', line 209

static VALUE rb_snappy_decompress_file(VALUE self, VALUE input_path, VALUE output_path) {
    Check_Type(input_path, T_STRING);
    Check_Type(output_path, T_STRING);

    const char *input_file = StringValueCStr(input_path);
    const char *output_file = StringValueCStr(output_path);

    // Open input file
    FILE *in = fopen(input_file, "rb");
    if (!in) {
        rb_sys_fail(input_file);
    }

    // Get file size
    fseek(in, 0, SEEK_END);
    long file_size = ftell(in);
    fseek(in, 0, SEEK_SET);

    // Read compressed file contents
    char *compressed_data = ALLOC_N(char, file_size);
    size_t bytes_read = fread(compressed_data, 1, file_size, in);
    fclose(in);

    if (bytes_read != file_size) {
        xfree(compressed_data);
        rb_raise(rb_eIOError, "failed to read complete file");
    }

    // Get uncompressed length
    size_t uncompressed_length;
    snappy_status status = snappy_uncompressed_length(compressed_data, file_size, &uncompressed_length);

    if (status != SNAPPY_OK) {
        xfree(compressed_data);
        rb_raise(rb_eSnappyError, "failed to get uncompressed length, status: %d", status);
    }

    // Decompress data
    char *uncompressed = ALLOC_N(char, uncompressed_length);
    status = snappy_uncompress(compressed_data, file_size, uncompressed, &uncompressed_length);
    xfree(compressed_data);

    if (status != SNAPPY_OK) {
        xfree(uncompressed);
        rb_raise(rb_eSnappyError, "decompression failed with status: %d", status);
    }

    // Write decompressed data to output file
    FILE *out = fopen(output_file, "wb");
    if (!out) {
        xfree(uncompressed);
        rb_sys_fail(output_file);
    }

    size_t bytes_written = fwrite(uncompressed, 1, uncompressed_length, out);
    fclose(out);
    xfree(uncompressed);

    if (bytes_written != uncompressed_length) {
        rb_raise(rb_eIOError, "failed to write complete decompressed data");
    }

    return Qnil;
}