File size: 5,726 Bytes
21f3d42
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
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
68
69
70
71
72
73
74
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
117
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
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
// Copyright 2002 Google LLC
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
//
//     http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
//
// Author: Neal Cardwell
//

#include "sparse_matmul/zlib_wrapper/gzipheader.h"

#include <algorithm>

#include "absl/base/macros.h"
#include "glog/logging.h"
#include "zlib.h"  // for Z_DEFAULT_COMPRESSION

namespace csrblocksparse {

const uint8_t GZipHeader::magic[] = {0x1f, 0x8b};

// ----------------------------------------------------------------------
// GZipHeader::ReadMore()
//    Attempt to parse the beginning of the given buffer as a gzip
//    header. If these bytes do not constitute a complete gzip header,
//    return INCOMPLETE_HEADER. If these bytes do not constitute a
//    *valid* gzip header, return INVALID_HEADER. If we find a
//    complete header, return COMPLETE_HEADER and set the pointer
//    pointed to by header_end to the first byte beyond the gzip header.
// ----------------------------------------------------------------------

GZipHeader::Status GZipHeader::ReadMore(const char* inbuf, int inbuf_len,
                                        const char** header_end) {
  CHECK_GE(inbuf_len, 0);
  const uint8_t* pos = reinterpret_cast<const uint8_t*>(inbuf);
  const uint8_t* const end = pos + inbuf_len;

  while (pos < end) {
    switch (state_) {
      case IN_HEADER_ID1:
        if (*pos != magic[0]) return INVALID_HEADER;
        pos++;
        state_++;
        break;
      case IN_HEADER_ID2:
        if (*pos != magic[1]) return INVALID_HEADER;
        pos++;
        state_++;
        break;
      case IN_HEADER_CM:
        if (*pos != Z_DEFLATED) return INVALID_HEADER;
        pos++;
        state_++;
        break;
      case IN_HEADER_FLG:
        flags_ =
            (*pos) & (FLAG_FHCRC | FLAG_FEXTRA | FLAG_FNAME | FLAG_FCOMMENT);
        pos++;
        state_++;
        break;

      case IN_HEADER_MTIME_BYTE_0:
        pos++;
        state_++;
        break;
      case IN_HEADER_MTIME_BYTE_1:
        pos++;
        state_++;
        break;
      case IN_HEADER_MTIME_BYTE_2:
        pos++;
        state_++;
        break;
      case IN_HEADER_MTIME_BYTE_3:
        pos++;
        state_++;
        break;

      case IN_HEADER_XFL:
        pos++;
        state_++;
        break;

      case IN_HEADER_OS:
        pos++;
        state_++;
        break;

      case IN_XLEN_BYTE_0:
        if (!(flags_ & FLAG_FEXTRA)) {
          state_ = IN_FNAME;
          break;
        }
        // We have a two-byte little-endian length, followed by a
        // field of that length.
        extra_length_ = *pos;
        pos++;
        state_++;
        break;
      case IN_XLEN_BYTE_1:
        extra_length_ += *pos << 8;
        pos++;
        state_++;
        // If we have a zero-length FEXTRA, we want to check to notice that
        // we're done reading the FEXTRA before we exit this loop...
        ABSL_FALLTHROUGH_INTENDED;

      case IN_FEXTRA: {
        // Grab the rest of the bytes in the extra field, or as many
        // of them as are actually present so far.
        const int num_extra_bytes = std::min<int>(extra_length_, (end - pos));
        pos += num_extra_bytes;
        extra_length_ -= num_extra_bytes;
        if (extra_length_ == 0) {
          state_ = IN_FNAME;  // advance when we've seen extra_length_ bytes
          flags_ &= ~FLAG_FEXTRA;  // we're done with the FEXTRA stuff
        }
        break;
      }

      case IN_FNAME:
        if (!(flags_ & FLAG_FNAME)) {
          state_ = IN_FCOMMENT;
          break;
        }
        // See if we can find the end of the \0-terminated FNAME field.
        pos = reinterpret_cast<const uint8_t*>(memchr(pos, '\0', (end - pos)));
        if (pos != nullptr) {
          pos++;                  // advance past the '\0'
          flags_ &= ~FLAG_FNAME;  // we're done with the FNAME stuff
          state_ = IN_FCOMMENT;
        } else {
          pos = end;  // everything we have so far is part of the FNAME
        }
        break;

      case IN_FCOMMENT:
        if (!(flags_ & FLAG_FCOMMENT)) {
          state_ = IN_FHCRC_BYTE_0;
          break;
        }
        // See if we can find the end of the \0-terminated FCOMMENT field.
        pos = reinterpret_cast<const uint8_t*>(memchr(pos, '\0', (end - pos)));
        if (pos != nullptr) {
          pos++;                     // advance past the '\0'
          flags_ &= ~FLAG_FCOMMENT;  // we're done with the FCOMMENT stuff
          state_ = IN_FHCRC_BYTE_0;
        } else {
          pos = end;  // everything we have so far is part of the FNAME
        }
        break;

      case IN_FHCRC_BYTE_0:
        if (!(flags_ & FLAG_FHCRC)) {
          state_ = IN_DONE;
          break;
        }
        pos++;
        state_++;
        break;

      case IN_FHCRC_BYTE_1:
        pos++;
        flags_ &= ~FLAG_FHCRC;  // we're done with the FHCRC stuff
        state_++;
        break;

      case IN_DONE:
        *header_end = reinterpret_cast<const char*>(pos);
        return COMPLETE_HEADER;
    }
  }

  if ((state_ > IN_HEADER_OS) && (flags_ == 0)) {
    *header_end = reinterpret_cast<const char*>(pos);
    return COMPLETE_HEADER;
  } else {
    return INCOMPLETE_HEADER;
  }
}

}  // namespace csrblocksparse