extbrotli 0.0.1.PROTOTYPE

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (68) hide show
  1. checksums.yaml +7 -0
  2. data/LICENSE +28 -0
  3. data/README.md +67 -0
  4. data/Rakefile +158 -0
  5. data/contrib/brotli/LICENSE +202 -0
  6. data/contrib/brotli/README.md +18 -0
  7. data/contrib/brotli/dec/bit_reader.c +55 -0
  8. data/contrib/brotli/dec/bit_reader.h +256 -0
  9. data/contrib/brotli/dec/context.h +260 -0
  10. data/contrib/brotli/dec/decode.c +1573 -0
  11. data/contrib/brotli/dec/decode.h +160 -0
  12. data/contrib/brotli/dec/dictionary.h +9494 -0
  13. data/contrib/brotli/dec/huffman.c +325 -0
  14. data/contrib/brotli/dec/huffman.h +77 -0
  15. data/contrib/brotli/dec/port.h +148 -0
  16. data/contrib/brotli/dec/prefix.h +756 -0
  17. data/contrib/brotli/dec/state.c +149 -0
  18. data/contrib/brotli/dec/state.h +185 -0
  19. data/contrib/brotli/dec/streams.c +99 -0
  20. data/contrib/brotli/dec/streams.h +100 -0
  21. data/contrib/brotli/dec/transform.h +315 -0
  22. data/contrib/brotli/dec/types.h +36 -0
  23. data/contrib/brotli/enc/backward_references.cc +769 -0
  24. data/contrib/brotli/enc/backward_references.h +50 -0
  25. data/contrib/brotli/enc/bit_cost.h +147 -0
  26. data/contrib/brotli/enc/block_splitter.cc +418 -0
  27. data/contrib/brotli/enc/block_splitter.h +78 -0
  28. data/contrib/brotli/enc/brotli_bit_stream.cc +884 -0
  29. data/contrib/brotli/enc/brotli_bit_stream.h +149 -0
  30. data/contrib/brotli/enc/cluster.h +290 -0
  31. data/contrib/brotli/enc/command.h +140 -0
  32. data/contrib/brotli/enc/context.h +185 -0
  33. data/contrib/brotli/enc/dictionary.h +9485 -0
  34. data/contrib/brotli/enc/dictionary_hash.h +4125 -0
  35. data/contrib/brotli/enc/encode.cc +715 -0
  36. data/contrib/brotli/enc/encode.h +196 -0
  37. data/contrib/brotli/enc/encode_parallel.cc +354 -0
  38. data/contrib/brotli/enc/encode_parallel.h +37 -0
  39. data/contrib/brotli/enc/entropy_encode.cc +492 -0
  40. data/contrib/brotli/enc/entropy_encode.h +88 -0
  41. data/contrib/brotli/enc/fast_log.h +179 -0
  42. data/contrib/brotli/enc/find_match_length.h +87 -0
  43. data/contrib/brotli/enc/hash.h +686 -0
  44. data/contrib/brotli/enc/histogram.cc +76 -0
  45. data/contrib/brotli/enc/histogram.h +100 -0
  46. data/contrib/brotli/enc/literal_cost.cc +172 -0
  47. data/contrib/brotli/enc/literal_cost.h +38 -0
  48. data/contrib/brotli/enc/metablock.cc +544 -0
  49. data/contrib/brotli/enc/metablock.h +88 -0
  50. data/contrib/brotli/enc/port.h +151 -0
  51. data/contrib/brotli/enc/prefix.h +85 -0
  52. data/contrib/brotli/enc/ringbuffer.h +108 -0
  53. data/contrib/brotli/enc/static_dict.cc +441 -0
  54. data/contrib/brotli/enc/static_dict.h +40 -0
  55. data/contrib/brotli/enc/static_dict_lut.h +12063 -0
  56. data/contrib/brotli/enc/streams.cc +127 -0
  57. data/contrib/brotli/enc/streams.h +129 -0
  58. data/contrib/brotli/enc/transform.h +250 -0
  59. data/contrib/brotli/enc/write_bits.h +91 -0
  60. data/ext/extbrotli.cc +24 -0
  61. data/ext/extbrotli.h +73 -0
  62. data/ext/extconf.rb +35 -0
  63. data/ext/lldecoder.c +220 -0
  64. data/ext/llencoder.cc +433 -0
  65. data/gemstub.rb +21 -0
  66. data/lib/extbrotli.rb +243 -0
  67. data/lib/extbrotli/version.rb +3 -0
  68. metadata +140 -0
@@ -0,0 +1,91 @@
1
+ // Copyright 2010 Google Inc. All Rights Reserved.
2
+ //
3
+ // Licensed under the Apache License, Version 2.0 (the "License");
4
+ // you may not use this file except in compliance with the License.
5
+ // You may obtain a copy of the License at
6
+ //
7
+ // http://www.apache.org/licenses/LICENSE-2.0
8
+ //
9
+ // Unless required by applicable law or agreed to in writing, software
10
+ // distributed under the License is distributed on an "AS IS" BASIS,
11
+ // WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
12
+ // See the License for the specific language governing permissions and
13
+ // limitations under the License.
14
+ //
15
+ // Write bits into a byte array.
16
+
17
+ #ifndef BROTLI_ENC_WRITE_BITS_H_
18
+ #define BROTLI_ENC_WRITE_BITS_H_
19
+
20
+ #include <assert.h>
21
+ #include <stdint.h>
22
+ #include <stdio.h>
23
+
24
+ #include "./port.h"
25
+
26
+ namespace brotli {
27
+
28
+ //#define BIT_WRITER_DEBUG
29
+
30
+ // This function writes bits into bytes in increasing addresses, and within
31
+ // a byte least-significant-bit first.
32
+ //
33
+ // The function can write up to 56 bits in one go with WriteBits
34
+ // Example: let's assume that 3 bits (Rs below) have been written already:
35
+ //
36
+ // BYTE-0 BYTE+1 BYTE+2
37
+ //
38
+ // 0000 0RRR 0000 0000 0000 0000
39
+ //
40
+ // Now, we could write 5 or less bits in MSB by just sifting by 3
41
+ // and OR'ing to BYTE-0.
42
+ //
43
+ // For n bits, we take the last 5 bits, OR that with high bits in BYTE-0,
44
+ // and locate the rest in BYTE+1, BYTE+2, etc.
45
+ inline void WriteBits(int n_bits,
46
+ uint64_t bits,
47
+ int * __restrict pos,
48
+ uint8_t * __restrict array) {
49
+ #ifdef BIT_WRITER_DEBUG
50
+ printf("WriteBits %2d 0x%016llx %10d\n", n_bits, bits, *pos);
51
+ #endif
52
+ assert(bits < 1ULL << n_bits);
53
+ #ifdef IS_LITTLE_ENDIAN
54
+ // This branch of the code can write up to 56 bits at a time,
55
+ // 7 bits are lost by being perhaps already in *p and at least
56
+ // 1 bit is needed to initialize the bit-stream ahead (i.e. if 7
57
+ // bits are in *p and we write 57 bits, then the next write will
58
+ // access a byte that was never initialized).
59
+ uint8_t *p = &array[*pos >> 3];
60
+ uint64_t v = *p;
61
+ v |= bits << (*pos & 7);
62
+ BROTLI_UNALIGNED_STORE64(p, v); // Set some bits.
63
+ *pos += n_bits;
64
+ #else
65
+ // implicit & 0xff is assumed for uint8_t arithmetics
66
+ uint8_t *array_pos = &array[*pos >> 3];
67
+ const int bits_reserved_in_first_byte = (*pos & 7);
68
+ bits <<= bits_reserved_in_first_byte;
69
+ *array_pos++ |= bits;
70
+ for (int bits_left_to_write = n_bits - 8 + bits_reserved_in_first_byte;
71
+ bits_left_to_write >= 1;
72
+ bits_left_to_write -= 8) {
73
+ bits >>= 8;
74
+ *array_pos++ = bits;
75
+ }
76
+ *array_pos = 0;
77
+ *pos += n_bits;
78
+ #endif
79
+ }
80
+
81
+ inline void WriteBitsPrepareStorage(int pos, uint8_t *array) {
82
+ #ifdef BIT_WRITER_DEBUG
83
+ printf("WriteBitsPrepareStorage %10d\n", pos);
84
+ #endif
85
+ assert((pos & 7) == 0);
86
+ array[pos >> 3] = 0;
87
+ }
88
+
89
+ } // namespace brotli
90
+
91
+ #endif // BROTLI_ENC_WRITE_BITS_H_
data/ext/extbrotli.cc ADDED
@@ -0,0 +1,24 @@
1
+ #include "extbrotli.h"
2
+
3
+ VALUE mBrotli;
4
+ VALUE mConst;
5
+ VALUE eError;
6
+ VALUE eNeedsMoreInput;
7
+ VALUE eNeedsMoreOutput;
8
+
9
+ EXTBROTLI_CEXTERN
10
+ void
11
+ Init_extbrotli(void)
12
+ {
13
+ mBrotli = rb_define_module("Brotli");
14
+
15
+ mConst = rb_define_module_under(mBrotli, "Constants");
16
+ rb_include_module(mBrotli, mConst);
17
+
18
+ eError = rb_define_class_under(mBrotli, "Error", rb_eRuntimeError);
19
+ eNeedsMoreInput = rb_define_class_under(mBrotli, "NeedsMoreInput", eError);
20
+ eNeedsMoreOutput = rb_define_class_under(mBrotli, "NeedsMoreOutput", eError);
21
+
22
+ extbrotli_init_lowlevelencoder();
23
+ extbrotli_init_lowleveldecoder();
24
+ }
data/ext/extbrotli.h ADDED
@@ -0,0 +1,73 @@
1
+ #ifndef EXTBROTLI_H
2
+ #define EXTBROTLI_H 1
3
+
4
+ #include <ruby.h>
5
+
6
+ #ifdef __cplusplus
7
+ # define EXTBROTLI_CEXTERN extern "C"
8
+ # define EXTBROTLI_BEGIN_CEXTERN EXTBROTLI_CEXTERN {
9
+ # define EXTBROTLI_END_CEXTERN }
10
+ #else
11
+ # define EXTBROTLI_CEXTERN
12
+ # define EXTBROTLI_BEGIN_CEXTERN
13
+ # define EXTBROTLI_END_CEXTERN
14
+ #endif
15
+
16
+ #define RDOCFAKE(...)
17
+
18
+ EXTBROTLI_BEGIN_CEXTERN
19
+
20
+ extern VALUE mBrotli;
21
+ RDOCFAKE(mBrotli = rb_define_module("Brotli"));
22
+
23
+ extern VALUE mConst;
24
+ RDOCFAKE(mConst = rb_define_module_under(mBrotli, "Constants"));
25
+
26
+ extern VALUE eError;
27
+ RDOCFAKE(eError = rb_define_class_under(mBrotli, "Error", rb_eRuntimeError));
28
+
29
+ extern VALUE eNeedsMoreInput;
30
+ RDOCFAKE(eNeedsMoreInput = rb_define_class_under(mBrotli, "NeedsMoreInput", eError));
31
+
32
+ extern VALUE eNeedsMoreOutput;
33
+ RDOCFAKE(eNeedsMoreOutput = rb_define_class_under(mBrotli, "NeedsMoreOutput", eError));
34
+
35
+ void extbrotli_init_lowlevelencoder(void);
36
+ void extbrotli_init_lowleveldecoder(void);
37
+
38
+ static inline void
39
+ referr(VALUE obj)
40
+ {
41
+ rb_raise(rb_eTypeError,
42
+ "invalid reference - #<%s:%p>",
43
+ rb_obj_classname(obj), (void *)obj);
44
+ }
45
+
46
+ static inline void
47
+ reiniterror(VALUE obj)
48
+ {
49
+ rb_raise(rb_eTypeError,
50
+ "initialized object already - #<%s:%p>",
51
+ rb_obj_classname(obj), (void *)obj);
52
+ }
53
+
54
+ static inline void *
55
+ getrefp(VALUE obj, const rb_data_type_t *type)
56
+ {
57
+ void *p;
58
+ TypedData_Get_Struct(obj, void, type, p);
59
+ return p;
60
+ }
61
+
62
+ static inline void *
63
+ getref(VALUE obj, const rb_data_type_t *type)
64
+ {
65
+ void *p = getrefp(obj, type);
66
+ if (!p) { referr(obj); }
67
+ return p;
68
+ }
69
+
70
+
71
+ EXTBROTLI_END_CEXTERN
72
+
73
+ #endif /* EXTBROTLI_H */
data/ext/extconf.rb ADDED
@@ -0,0 +1,35 @@
1
+ #!ruby
2
+ #vim: set fileencoding:utf-8
3
+
4
+ require "mkmf"
5
+
6
+ dir = File.dirname(__FILE__).gsub(/[\[\{\?\*]/, "[\\0]")
7
+
8
+ encfiles = Dir.glob(File.join(dir, "../contrib/brotli/enc/*.cc")).sort
9
+ encsrc = "brotli_encoder.cc"
10
+ File.write encsrc, (encfiles.map {|n| "#include \"#{n}\"" }.join("\n") << "\n")
11
+ decfiles = Dir.glob(File.join(dir, "../contrib/brotli/dec/*.c")).sort
12
+ decsrc = "brotli_decoder.c"
13
+ File.write decsrc, (decfiles.map {|n| "#include \"#{n}\"" }.join("\n") << "\n")
14
+
15
+ filepattern = "*.c{,c,pp}"
16
+ target = File.join(dir, filepattern)
17
+ files = Dir.glob(target).map { |n| File.basename n }
18
+ rejects = (RbConfig::CONFIG["arch"] =~ /mswin|mingw/) ? /_pthreads_/ : /_win32_/
19
+ files.reject! { |n| n =~ rejects }
20
+ $srcs = files + [encsrc, decsrc]
21
+
22
+ #$VPATH.push "$(srcdir)/../contrib"
23
+
24
+ #find_header "brotli/enc/encode.h", "$(srcdir)/../contrib" or abort
25
+ find_header "brotli/dec/decode.h", "$(srcdir)/../contrib" or abort
26
+
27
+ if RbConfig::CONFIG["arch"] =~ /mingw/
28
+ $CPPFLAGS << " -D__forceinline=__attribute__\\(\\(always_inline\\)\\)"
29
+ $CXXFLAGS << " -std=gnu++11"
30
+ $LDFLAGS << " -static-libgcc -static-libstdc++"
31
+ end
32
+
33
+ try_link "void main(void){}", " -Wl,-Bsymbolic " and $LDFLAGS << " -Wl,-Bsymbolic "
34
+
35
+ create_makefile("extbrotli")
data/ext/lldecoder.c ADDED
@@ -0,0 +1,220 @@
1
+ #include "extbrotli.h"
2
+ #include <brotli/dec/decode.h>
3
+
4
+ static VALUE cLLDecoder; /* class Brotli::LowLevelDecoder */
5
+
6
+ static void
7
+ lldec_free(void *pp)
8
+ {
9
+ if (pp) {
10
+ BrotliState *p = (BrotliState *)pp;
11
+ BrotliStateCleanup(p);
12
+ }
13
+ }
14
+
15
+ static const rb_data_type_t lldecoder_type = {
16
+ .wrap_struct_name = "extbrotli.lowleveldecoder",
17
+ .function.dmark = NULL,
18
+ .function.dfree = lldec_free,
19
+ .function.dsize = NULL,
20
+ };
21
+
22
+ static inline BrotliState *
23
+ getlldecoderp(VALUE obj)
24
+ {
25
+ return (BrotliState *)getrefp(obj, &lldecoder_type);
26
+ }
27
+
28
+ static inline BrotliState *
29
+ getlldecoder(VALUE obj)
30
+ {
31
+ return (BrotliState *)getref(obj, &lldecoder_type);
32
+ }
33
+
34
+ static VALUE
35
+ lldec_alloc(VALUE mod)
36
+ {
37
+ return TypedData_Wrap_Struct(mod, &lldecoder_type, NULL);
38
+ }
39
+
40
+ static VALUE
41
+ lldec_init(VALUE dec)
42
+ {
43
+ BrotliState *p = getlldecoderp(dec);
44
+ if (p) { reiniterror(dec); }
45
+ DATA_PTR(dec) = p = ALLOC(BrotliState);
46
+ BrotliStateInit(p);
47
+ return dec;
48
+ }
49
+
50
+ /*
51
+ * call-seq:
52
+ * decode(src, dest, maxdest, finish) -> [result_code, src, dest]
53
+ */
54
+ static VALUE
55
+ lldec_decode(VALUE dec, VALUE src, VALUE dest, VALUE maxdest, VALUE finish)
56
+ {
57
+ BrotliState *p = getlldecoder(dec);
58
+
59
+ char *srcp, *destp;
60
+ size_t srcsize, destsize;
61
+ if (NIL_P(src)) {
62
+ srcp = NULL;
63
+ srcsize = 0;
64
+ } else {
65
+ rb_obj_infect(dec, src);
66
+ rb_check_type(src, RUBY_T_STRING);
67
+ RSTRING_GETMEM(src, srcp, srcsize);
68
+ }
69
+ if (NIL_P(dest)) {
70
+ destp = NULL;
71
+ destsize = 0;
72
+ } else {
73
+ rb_check_type(dest, RUBY_T_STRING);
74
+ destsize = NUM2SIZET(maxdest);
75
+ rb_str_modify(dest);
76
+ rb_str_set_len(dest, 0);
77
+ rb_str_modify_expand(dest, destsize);
78
+ destp = RSTRING_PTR(dest);
79
+ }
80
+ char *srcpp = srcp, *destpp = destp;
81
+ size_t total;
82
+ BrotliResult s = BrotliDecompressBufferStreaming(
83
+ &srcsize, (const uint8_t **)&srcp, NUM2INT(finish),
84
+ &destsize, (uint8_t **)&destp, &total, p);
85
+ switch (s) {
86
+ case BROTLI_RESULT_ERROR:
87
+ rb_raise(eError, "failed BrotliDecompressBufferStreaming() - #<%s:%p>", rb_obj_classname(dec), (void *)dec);
88
+ case BROTLI_RESULT_SUCCESS:
89
+ case BROTLI_RESULT_NEEDS_MORE_INPUT:
90
+ case BROTLI_RESULT_NEEDS_MORE_OUTPUT:
91
+ if (destp > destpp) {
92
+ rb_obj_infect(dest, dec);
93
+ rb_str_set_len(dest, destp - destpp);
94
+ }
95
+ if (srcsize > 0) {
96
+ src = rb_str_substr(src, srcp - srcpp, srcsize);
97
+ } else {
98
+ src = Qnil;
99
+ }
100
+ break;
101
+ default:
102
+ rb_raise(eError,
103
+ "failed BrotliDecompressBufferStreaming() - unknown result code (%d) - #<%s:%p>",
104
+ s, rb_obj_classname(dec), (void *)dec);
105
+ }
106
+
107
+ const VALUE vec[3] = { INT2NUM(s), src, dest };
108
+ return rb_ary_new4(3, vec);
109
+ }
110
+
111
+ static void
112
+ lldec_s_decode_args(int argc, VALUE argv[], VALUE *src, VALUE *dest, size_t *srcsize, size_t *destsize, int *partial)
113
+ {
114
+ if (argc > 0) {
115
+ *src = argv[0];
116
+ rb_check_type(*src, RUBY_T_STRING);
117
+ *srcsize = RSTRING_LEN(*src);
118
+
119
+ int s;
120
+ switch (argc) {
121
+ case 1:
122
+ s = BrotliDecompressedSize(*srcsize, (const uint8_t *)RSTRING_PTR(*src), destsize);
123
+ if (s == 0) { rb_raise(eError, "failed BrotliDecompressedSize()"); }
124
+ *dest = rb_str_buf_new(*destsize);
125
+ *partial = 0;
126
+ return;
127
+ case 2:
128
+ *dest = argv[1];
129
+ if (rb_type_p(*dest, RUBY_T_STRING)) {
130
+ s = BrotliDecompressedSize(*srcsize, (const uint8_t *)RSTRING_PTR(*src), destsize);
131
+ if (s == 0) { rb_raise(eError, "failed BrotliDecompressedSize()"); }
132
+ rb_check_type(*dest, RUBY_T_STRING);
133
+ rb_str_modify(*dest);
134
+ rb_str_set_len(*dest, 0);
135
+ rb_str_modify_expand(*dest, *destsize);
136
+ } else {
137
+ *destsize = NUM2SIZET(*dest);
138
+ *dest = rb_str_buf_new(*destsize);
139
+ }
140
+ *partial = 0;
141
+ return;
142
+ case 3:
143
+ *destsize = NUM2SIZET(argv[1]);
144
+ *dest = argv[2];
145
+ if (*dest == Qtrue || *dest == Qfalse) {
146
+ *partial = RTEST(*dest) ? 1 : 0;
147
+ *dest = rb_str_buf_new(*destsize);
148
+ } else {
149
+ rb_check_type(*dest, RUBY_T_STRING);
150
+ rb_str_modify(*dest);
151
+ rb_str_set_len(*dest, 0);
152
+ rb_str_modify_expand(*dest, *destsize);
153
+ *partial = 0;
154
+ }
155
+ return;
156
+ case 4:
157
+ *destsize = NUM2SIZET(argv[1]);
158
+ *dest = argv[2];
159
+ rb_check_type(*dest, RUBY_T_STRING);
160
+ rb_str_modify(*dest);
161
+ rb_str_set_len(*dest, 0);
162
+ rb_str_modify_expand(*dest, *destsize);
163
+ *partial = RTEST(argv[3]) ? 1 : 0;
164
+ return;
165
+ }
166
+ }
167
+
168
+ rb_error_arity(argc, 1, 4);
169
+ }
170
+
171
+ /*
172
+ * call-seq:
173
+ * decode(src, dest = "") -> decoded binary string
174
+ * decode(src, destsize, partial = false) -> decoded binary string
175
+ * decode(src, destsize, dest = "", partial = false) -> decoded binary string
176
+ */
177
+ static VALUE
178
+ lldec_s_decode(int argc, VALUE argv[], VALUE mod)
179
+ {
180
+ VALUE src, dest;
181
+ size_t srcsize, destsize;
182
+ int partial;
183
+ lldec_s_decode_args(argc, argv, &src, &dest, &srcsize, &destsize, &partial);
184
+
185
+ int s = BrotliDecompressBuffer(srcsize, (const uint8_t *)RSTRING_PTR(src),
186
+ &destsize, (uint8_t *)RSTRING_PTR(dest));
187
+
188
+ if (s == BROTLI_RESULT_NEEDS_MORE_INPUT) {
189
+ if (!partial) {
190
+ rb_raise(eNeedsMoreInput, "failed BrotliDecompressBuffer()");
191
+ }
192
+ } else if (s == BROTLI_RESULT_NEEDS_MORE_OUTPUT) {
193
+ if (!partial) {
194
+ rb_raise(eNeedsMoreOutput, "failed BrotliDecompressBuffer()");
195
+ }
196
+ } else if (s != BROTLI_RESULT_SUCCESS) {
197
+ rb_raise(eError, "failed BrotliDecompressBuffer() - status=%d", s);
198
+ }
199
+
200
+ rb_obj_infect(dest, src);
201
+ rb_str_set_len(dest, destsize);
202
+ return dest;
203
+ }
204
+
205
+ void
206
+ extbrotli_init_lowleveldecoder(void)
207
+ {
208
+ cLLDecoder = rb_define_class_under(mBrotli, "LowLevelDecoder", rb_cObject);
209
+ rb_define_singleton_method(cLLDecoder, "decode", RUBY_METHOD_FUNC(lldec_s_decode), -1);
210
+
211
+ rb_define_alloc_func(cLLDecoder, lldec_alloc);
212
+ rb_define_method(cLLDecoder, "initialize", RUBY_METHOD_FUNC(lldec_init), 0);
213
+ rb_define_method(cLLDecoder, "decode", RUBY_METHOD_FUNC(lldec_decode), 4);
214
+
215
+ rb_include_module(cLLDecoder, mConst);
216
+ rb_define_const(mConst, "RESULT_ERROR", INT2FIX(BROTLI_RESULT_ERROR));
217
+ rb_define_const(mConst, "RESULT_SUCCESS", INT2FIX(BROTLI_RESULT_SUCCESS));
218
+ rb_define_const(mConst, "RESULT_NEEDS_MORE_INPUT", INT2FIX(BROTLI_RESULT_NEEDS_MORE_INPUT));
219
+ rb_define_const(mConst, "RESULT_NEEDS_MORE_OUTPUT", INT2FIX(BROTLI_RESULT_NEEDS_MORE_OUTPUT));
220
+ }