1 // Protocol Buffers - Google's data interchange format
2 // Copyright 2008 Google Inc.  All rights reserved.
3 // https://developers.google.com/protocol-buffers/
4 //
5 // Redistribution and use in source and binary forms, with or without
6 // modification, are permitted provided that the following conditions are
7 // met:
8 //
9 //     * Redistributions of source code must retain the above copyright
10 // notice, this list of conditions and the following disclaimer.
11 //     * Redistributions in binary form must reproduce the above
12 // copyright notice, this list of conditions and the following disclaimer
13 // in the documentation and/or other materials provided with the
14 // distribution.
15 //     * Neither the name of Google Inc. nor the names of its
16 // contributors may be used to endorse or promote products derived from
17 // this software without specific prior written permission.
18 //
19 // THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
20 // "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
21 // LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
22 // A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
23 // OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
24 // SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
25 // LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
26 // DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
27 // THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
28 // (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
29 // OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
30 
31 // Author: brianolson@google.com (Brian Olson)
32 //
33 // This file contains the definition for classes GzipInputStream and
34 // GzipOutputStream.
35 //
36 // GzipInputStream decompresses data from an underlying
37 // ZeroCopyInputStream and provides the decompressed data as a
38 // ZeroCopyInputStream.
39 //
40 // GzipOutputStream is an ZeroCopyOutputStream that compresses data to
41 // an underlying ZeroCopyOutputStream.
42 
43 #ifndef GOOGLE_PROTOBUF_IO_GZIP_STREAM_H__
44 #define GOOGLE_PROTOBUF_IO_GZIP_STREAM_H__
45 
46 #include <zlib.h>
47 
48 #include <google/protobuf/stubs/common.h>
49 #include <google/protobuf/io/zero_copy_stream.h>
50 
51 namespace google {
52 namespace protobuf {
53 namespace io {
54 
55 // A ZeroCopyInputStream that reads compressed data through zlib
56 class LIBPROTOBUF_EXPORT GzipInputStream : public ZeroCopyInputStream {
57  public:
58   // Format key for constructor
59   enum Format {
60     // zlib will autodetect gzip header or deflate stream
61     AUTO = 0,
62 
63     // GZIP streams have some extra header data for file attributes.
64     GZIP = 1,
65 
66     // Simpler zlib stream format.
67     ZLIB = 2,
68   };
69 
70   // buffer_size and format may be -1 for default of 64kB and GZIP format
71   explicit GzipInputStream(
72       ZeroCopyInputStream* sub_stream,
73       Format format = AUTO,
74       int buffer_size = -1);
75   virtual ~GzipInputStream();
76 
77   // Return last error message or NULL if no error.
ZlibErrorMessage()78   inline const char* ZlibErrorMessage() const {
79     return zcontext_.msg;
80   }
ZlibErrorCode()81   inline int ZlibErrorCode() const {
82     return zerror_;
83   }
84 
85   // implements ZeroCopyInputStream ----------------------------------
86   bool Next(const void** data, int* size);
87   void BackUp(int count);
88   bool Skip(int count);
89   int64 ByteCount() const;
90 
91  private:
92   Format format_;
93 
94   ZeroCopyInputStream* sub_stream_;
95 
96   z_stream zcontext_;
97   int zerror_;
98 
99   void* output_buffer_;
100   void* output_position_;
101   size_t output_buffer_length_;
102 
103   int Inflate(int flush);
104   void DoNextOutput(const void** data, int* size);
105 
106   GOOGLE_DISALLOW_EVIL_CONSTRUCTORS(GzipInputStream);
107 };
108 
109 
110 class LIBPROTOBUF_EXPORT GzipOutputStream : public ZeroCopyOutputStream {
111  public:
112   // Format key for constructor
113   enum Format {
114     // GZIP streams have some extra header data for file attributes.
115     GZIP = 1,
116 
117     // Simpler zlib stream format.
118     ZLIB = 2,
119   };
120 
121   struct Options {
122     // Defaults to GZIP.
123     Format format;
124 
125     // What size buffer to use internally.  Defaults to 64kB.
126     int buffer_size;
127 
128     // A number between 0 and 9, where 0 is no compression and 9 is best
129     // compression.  Defaults to Z_DEFAULT_COMPRESSION (see zlib.h).
130     int compression_level;
131 
132     // Defaults to Z_DEFAULT_STRATEGY.  Can also be set to Z_FILTERED,
133     // Z_HUFFMAN_ONLY, or Z_RLE.  See the documentation for deflateInit2 in
134     // zlib.h for definitions of these constants.
135     int compression_strategy;
136 
137     Options();  // Initializes with default values.
138   };
139 
140   // Create a GzipOutputStream with default options.
141   explicit GzipOutputStream(ZeroCopyOutputStream* sub_stream);
142 
143   // Create a GzipOutputStream with the given options.
144   GzipOutputStream(
145       ZeroCopyOutputStream* sub_stream,
146       const Options& options);
147 
148   virtual ~GzipOutputStream();
149 
150   // Return last error message or NULL if no error.
ZlibErrorMessage()151   inline const char* ZlibErrorMessage() const {
152     return zcontext_.msg;
153   }
ZlibErrorCode()154   inline int ZlibErrorCode() const {
155     return zerror_;
156   }
157 
158   // Flushes data written so far to zipped data in the underlying stream.
159   // It is the caller's responsibility to flush the underlying stream if
160   // necessary.
161   // Compression may be less efficient stopping and starting around flushes.
162   // Returns true if no error.
163   //
164   // Please ensure that block size is > 6. Here is an excerpt from the zlib
165   // doc that explains why:
166   //
167   // In the case of a Z_FULL_FLUSH or Z_SYNC_FLUSH, make sure that avail_out
168   // is greater than six to avoid repeated flush markers due to
169   // avail_out == 0 on return.
170   bool Flush();
171 
172   // Writes out all data and closes the gzip stream.
173   // It is the caller's responsibility to close the underlying stream if
174   // necessary.
175   // Returns true if no error.
176   bool Close();
177 
178   // implements ZeroCopyOutputStream ---------------------------------
179   bool Next(void** data, int* size);
180   void BackUp(int count);
181   int64 ByteCount() const;
182 
183  private:
184   ZeroCopyOutputStream* sub_stream_;
185   // Result from calling Next() on sub_stream_
186   void* sub_data_;
187   int sub_data_size_;
188 
189   z_stream zcontext_;
190   int zerror_;
191   void* input_buffer_;
192   size_t input_buffer_length_;
193 
194   // Shared constructor code.
195   void Init(ZeroCopyOutputStream* sub_stream, const Options& options);
196 
197   // Do some compression.
198   // Takes zlib flush mode.
199   // Returns zlib error code.
200   int Deflate(int flush);
201 
202   GOOGLE_DISALLOW_EVIL_CONSTRUCTORS(GzipOutputStream);
203 };
204 
205 }  // namespace io
206 }  // namespace protobuf
207 
208 }  // namespace google
209 #endif  // GOOGLE_PROTOBUF_IO_GZIP_STREAM_H__
210