Class: MultiCompress::Inflater

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

Instance Method Summary collapse

Constructor Details

#initialize(*args) ⇒ Object 



1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
 
# File 'ext/multi_compress/multi_compress.c', line 1993

static VALUE inflater_initialize(int argc, VALUE *argv, VALUE self) {
    VALUE opts;
    rb_scan_args(argc, argv, "0:", &opts);

    inflater_t *inf;
    TypedData_Get_Struct(self, inflater_t, &inflater_type, inf);

    VALUE algo_sym = Qnil, dict_val = Qnil;
    if (!NIL_P(opts)) {
        algo_sym = rb_hash_aref(opts, ID2SYM(rb_intern("algo")));
        dict_val = rb_hash_aref(opts, ID2SYM(rb_intern("dictionary")));
    }

    inf->algo = NIL_P(algo_sym) ? ALGO_ZSTD : sym_to_algo(algo_sym);
    inf->closed = 0;
    inf->finished = 0;

    dictionary_t *dict = NULL;
    if (!NIL_P(dict_val)) {
        if (inf->algo == ALGO_LZ4) {
            rb_raise(eUnsupportedError, "LZ4 does not support dictionaries");
        }
        TypedData_Get_Struct(dict_val, dictionary_t, &dictionary_type, dict);
        rb_ivar_set(self, rb_intern("@dictionary"), dict_val);
    }

    switch (inf->algo) {
    case ALGO_ZSTD:
        inf->ctx.zstd = ZSTD_createDStream();
        if (!inf->ctx.zstd)
            rb_raise(eMemError, "zstd: failed to create dstream");
        if (dict) {
            ZSTD_DCtx_reset(inf->ctx.zstd, ZSTD_reset_session_only);
            size_t r = ZSTD_DCtx_loadDictionary(inf->ctx.zstd, dict->data, dict->size);
            if (ZSTD_isError(r))
                rb_raise(eError, "zstd dict load: %s", ZSTD_getErrorName(r));
        } else {
            ZSTD_initDStream(inf->ctx.zstd);
        }
        break;
    case ALGO_BROTLI:
        inf->ctx.brotli = BrotliDecoderCreateInstance(NULL, NULL, NULL);
        if (!inf->ctx.brotli)
            rb_raise(eMemError, "brotli: failed to create decoder");
        if (dict) {
            BrotliDecoderAttachDictionary(inf->ctx.brotli, BROTLI_SHARED_DICTIONARY_RAW, dict->size,
                                          dict->data);
        }
        break;
    case ALGO_LZ4:
        inf->lz4_buf.cap = 16 * 1024;
        inf->lz4_buf.buf = ALLOC_N(char, inf->lz4_buf.cap);
        inf->lz4_buf.len = 0;
        inf->lz4_buf.offset = 0;
        break;
    }

    return self;
}

Instance Method Details

#close ⇒ Object 



2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
 
# File 'ext/multi_compress/multi_compress.c', line 2292

static VALUE inflater_close(VALUE self) {
    inflater_t *inf;
    TypedData_Get_Struct(self, inflater_t, &inflater_type, inf);
    if (inf->closed)
        return Qnil;

    switch (inf->algo) {
    case ALGO_ZSTD:
        if (inf->ctx.zstd) {
            ZSTD_freeDStream(inf->ctx.zstd);
            inf->ctx.zstd = NULL;
        }
        break;
    case ALGO_BROTLI:
        if (inf->ctx.brotli) {
            BrotliDecoderDestroyInstance(inf->ctx.brotli);
            inf->ctx.brotli = NULL;
        }
        break;
    case ALGO_LZ4:
        break;
    }
    inf->closed = 1;
    return Qnil;
}

#closed? ⇒ Boolean

Returns:

  • (Boolean) 


2318
2319
2320
2321
2322
 
# File 'ext/multi_compress/multi_compress.c', line 2318

static VALUE inflater_closed_p(VALUE self) {
    inflater_t *inf;
    TypedData_Get_Struct(self, inflater_t, &inflater_type, inf);
    return inf->closed ? Qtrue : Qfalse;
}

#finish ⇒ Object 



2240
2241
2242
2243
2244
2245
2246
2247
 
# File 'ext/multi_compress/multi_compress.c', line 2240

static VALUE inflater_finish(VALUE self) {
    inflater_t *inf;
    TypedData_Get_Struct(self, inflater_t, &inflater_type, inf);
    if (inf->closed)
        rb_raise(eStreamError, "stream is closed");
    inf->finished = 1;
    return rb_binary_str_new("", 0);
}

#reset ⇒ Object 



2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
 
# File 'ext/multi_compress/multi_compress.c', line 2249

static VALUE inflater_reset(VALUE self) {
    inflater_t *inf;
    TypedData_Get_Struct(self, inflater_t, &inflater_type, inf);

    VALUE dict_val = rb_attr_get(self, rb_intern("@dictionary"));
    dictionary_t *dict = NULL;
    if (!NIL_P(dict_val)) {
        TypedData_Get_Struct(dict_val, dictionary_t, &dictionary_type, dict);
    }

    switch (inf->algo) {
    case ALGO_ZSTD:
        if (inf->ctx.zstd) {
            ZSTD_DCtx_reset(inf->ctx.zstd, ZSTD_reset_session_only);
            if (dict) {
                size_t r = ZSTD_DCtx_loadDictionary(inf->ctx.zstd, dict->data, dict->size);
                if (ZSTD_isError(r))
                    rb_raise(eError, "zstd dict reload on reset: %s", ZSTD_getErrorName(r));
            }
        }
        break;
    case ALGO_BROTLI:
        if (inf->ctx.brotli) {
            BrotliDecoderDestroyInstance(inf->ctx.brotli);
            inf->ctx.brotli = BrotliDecoderCreateInstance(NULL, NULL, NULL);
            if (!inf->ctx.brotli)
                rb_raise(eMemError, "brotli: failed to recreate decoder");
            if (dict) {
                BrotliDecoderAttachDictionary(inf->ctx.brotli, BROTLI_SHARED_DICTIONARY_RAW,
                                              dict->size, dict->data);
            }
        }
        break;
    case ALGO_LZ4:
        inf->lz4_buf.len = 0;
        inf->lz4_buf.offset = 0;
        break;
    }
    inf->closed = 0;
    inf->finished = 0;
    return self;
}

#write(chunk) ⇒ Object 



2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
 
# File 'ext/multi_compress/multi_compress.c', line 2053

static VALUE inflater_write(VALUE self, VALUE chunk) {
    inflater_t *inf;
    TypedData_Get_Struct(self, inflater_t, &inflater_type, inf);
    if (inf->closed)
        rb_raise(eStreamError, "stream is closed");
    StringValue(chunk);

    const char *src = RSTRING_PTR(chunk);
    size_t slen = RSTRING_LEN(chunk);
    if (slen == 0)
        return rb_binary_str_new("", 0);

    switch (inf->algo) {
    case ALGO_ZSTD: {
        ZSTD_inBuffer input = {src, slen, 0};
        size_t out_cap = ZSTD_DStreamOutSize();
        size_t result_cap = out_cap > slen * 2 ? out_cap : slen * 2;
        VALUE result = rb_binary_str_buf_reserve(result_cap);
        size_t result_len = 0;
        VALUE scheduler = current_fiber_scheduler();

        while (input.pos < input.size) {
            if (result_len + out_cap > result_cap) {
                result_cap = result_cap * 2;
                rb_str_resize(result, result_cap);
            }

            ZSTD_outBuffer output = {RSTRING_PTR(result) + result_len, out_cap, 0};
            size_t ret;

            if (scheduler != Qnil && (input.size - input.pos) >= FIBER_STREAM_THRESHOLD) {
                zstd_decompress_stream_chunk_args_t args = {
                    .dstream = inf->ctx.zstd,
                    .output  = &output,
                    .input   = &input,
                    .result  = 0,
                };
                run_via_fiber_worker(scheduler, zstd_decompress_stream_chunk_nogvl, &args);
                ret = args.result;
            } else {
                ret = ZSTD_decompressStream(inf->ctx.zstd, &output, &input);
            }

            if (ZSTD_isError(ret))
                rb_raise(eDataError, "zstd decompress stream: %s", ZSTD_getErrorName(ret));
            result_len += output.pos;
            if (ret == 0)
                break;
        }
        rb_str_set_len(result, result_len);
        RB_GC_GUARD(chunk);
        return result;
    }
    case ALGO_BROTLI: {
        size_t available_in = slen;
        const uint8_t *next_in = (const uint8_t *)src;
        size_t result_cap = slen * 2;
        if (result_cap < 1024)
            result_cap = 1024;
        VALUE result = rb_binary_str_buf_reserve(result_cap);
        size_t result_len = 0;
        VALUE scheduler = current_fiber_scheduler();

        while (available_in > 0 || BrotliDecoderHasMoreOutput(inf->ctx.brotli)) {
            size_t available_out = 0;
            uint8_t *next_out = NULL;
            BrotliDecoderResult res;

            if (scheduler != Qnil && available_in >= FIBER_STREAM_THRESHOLD) {
                brotli_decompress_stream_args_t sargs = {
                    .dec           = inf->ctx.brotli,
                    .available_in  = &available_in,
                    .next_in       = &next_in,
                    .available_out = &available_out,
                    .next_out      = &next_out,
                    .result        = BROTLI_DECODER_RESULT_ERROR,
                };
                run_via_fiber_worker(scheduler, brotli_decompress_stream_nogvl, &sargs);
                res = sargs.result;
            } else {
                res = BrotliDecoderDecompressStream(
                    inf->ctx.brotli, &available_in, &next_in, &available_out, &next_out, NULL);
            }
            if (res == BROTLI_DECODER_RESULT_ERROR)
                rb_raise(eDataError, "brotli decompress stream: %s",
                         BrotliDecoderErrorString(BrotliDecoderGetErrorCode(inf->ctx.brotli)));
            const uint8_t *out_data;
            size_t out_size = 0;
            out_data = BrotliDecoderTakeOutput(inf->ctx.brotli, &out_size);
            if (out_size > 0) {
                if (result_len + out_size > result_cap) {
                    result_cap = (result_len + out_size) * 2;
                    rb_str_resize(result, result_cap);
                }

                memcpy(RSTRING_PTR(result) + result_len, out_data, out_size);
                result_len += out_size;
            }
            if (res == BROTLI_DECODER_RESULT_SUCCESS)
                break;
            if (res == BROTLI_DECODER_RESULT_NEEDS_MORE_INPUT && available_in == 0)
                break;
        }
        rb_str_set_len(result, result_len);
        RB_GC_GUARD(chunk);
        return result;
    }
    case ALGO_LZ4: {
        size_t data_len = inf->lz4_buf.len - inf->lz4_buf.offset;
        size_t needed = data_len + slen;

        if (inf->lz4_buf.offset > 0 && needed > inf->lz4_buf.cap) {
            if (data_len > 0)
                memmove(inf->lz4_buf.buf, inf->lz4_buf.buf + inf->lz4_buf.offset, data_len);
            inf->lz4_buf.offset = 0;
            inf->lz4_buf.len = data_len;
        } else if (inf->lz4_buf.offset > inf->lz4_buf.cap / 2) {
            if (data_len > 0)
                memmove(inf->lz4_buf.buf, inf->lz4_buf.buf + inf->lz4_buf.offset, data_len);
            inf->lz4_buf.offset = 0;
            inf->lz4_buf.len = data_len;
        }

        needed = inf->lz4_buf.len + slen;
        if (needed > inf->lz4_buf.cap) {
            inf->lz4_buf.cap = needed * 2;
            REALLOC_N(inf->lz4_buf.buf, char, inf->lz4_buf.cap);
        }
        memcpy(inf->lz4_buf.buf + inf->lz4_buf.len, src, slen);
        inf->lz4_buf.len += slen;

        size_t result_cap = slen * 4;
        if (result_cap < 256)
            result_cap = 256;
        VALUE result = rb_binary_str_buf_new(result_cap);
        size_t result_len = 0;
        int use_fiber = has_fiber_scheduler();
        size_t fiber_counter = 0;

        size_t pos = inf->lz4_buf.offset;
        while (pos + 4 <= inf->lz4_buf.len) {
            const uint8_t *p = (const uint8_t *)(inf->lz4_buf.buf + pos);
            uint32_t orig_size = (uint32_t)p[0] | ((uint32_t)p[1] << 8) | ((uint32_t)p[2] << 16) |
                                 ((uint32_t)p[3] << 24);
            if (orig_size == 0) {
                inf->finished = 1;
                pos += 4;
                break;
            }
            if (pos + 8 > inf->lz4_buf.len)
                break;
            uint32_t comp_size = (uint32_t)p[4] | ((uint32_t)p[5] << 8) | ((uint32_t)p[6] << 16) |
                                 ((uint32_t)p[7] << 24);
            if (pos + 8 + comp_size > inf->lz4_buf.len)
                break;
            if (orig_size > 64 * 1024 * 1024)
                rb_raise(eDataError, "lz4 stream: block too large (%u)", orig_size);

            if (result_len + orig_size > result_cap) {
                result_cap = (result_len + orig_size) * 2;
                rb_str_resize(result, result_cap);
            }

            int dsize = LZ4_decompress_safe(inf->lz4_buf.buf + pos + 8,
                                            RSTRING_PTR(result) + result_len, (int)comp_size,
                                            (int)orig_size);
            if (dsize < 0)
                rb_raise(eDataError, "lz4 stream decompress block failed");

            result_len += dsize;
            pos += 8 + comp_size;
            if (use_fiber) {
                int did_yield = 0;
                fiber_counter = fiber_maybe_yield(fiber_counter, (size_t)dsize, &did_yield);
                (void)did_yield;
            }
        }

        inf->lz4_buf.offset = pos;
        rb_str_set_len(result, result_len);
        RB_GC_GUARD(chunk);
        return result;
    }
    }
    return rb_binary_str_new("", 0);
}