Mercurial > hg > sv-dependency-builds

annotate src/zlib-1.2.8/examples/zlib_how.html @ 43:5ea0608b923f

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Current zlib source
author Chris Cannam
date Tue, 18 Oct 2016 14:33:52 +0100
parents
children
rev   line source
Chris@43 1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"
Chris@43 2 "http://www.w3.org/TR/REC-html40/loose.dtd">
Chris@43 3 <html>
Chris@43 4 <head>
Chris@43 5 <meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1">
Chris@43 6 <title>zlib Usage Example</title>
Chris@43 7 <!-- Copyright (c) 2004, 2005 Mark Adler. -->
Chris@43 8 </head>
Chris@43 9 <body bgcolor="#FFFFFF" text="#000000" link="#0000FF" vlink="#00A000">
Chris@43 10 <h2 align="center"> zlib Usage Example </h2>
Chris@43 11 We often get questions about how the <tt>deflate()</tt> and <tt>inflate()</tt> functions should be used.
Chris@43 12 Users wonder when they should provide more input, when they should use more output,
Chris@43 13 what to do with a <tt>Z_BUF_ERROR</tt>, how to make sure the process terminates properly, and
Chris@43 14 so on. So for those who have read <tt>zlib.h</tt> (a few times), and
Chris@43 15 would like further edification, below is an annotated example in C of simple routines to compress and decompress
Chris@43 16 from an input file to an output file using <tt>deflate()</tt> and <tt>inflate()</tt> respectively. The
Chris@43 17 annotations are interspersed between lines of the code. So please read between the lines.
Chris@43 18 We hope this helps explain some of the intricacies of <em>zlib</em>.
Chris@43 19 <p>
Chris@43 20 Without further adieu, here is the program <a href="zpipe.c"><tt>zpipe.c</tt></a>:
Chris@43 21 <pre><b>
Chris@43 22 /* zpipe.c: example of proper use of zlib's inflate() and deflate()
Chris@43 23 Not copyrighted -- provided to the public domain
Chris@43 24 Version 1.4 11 December 2005 Mark Adler */
Chris@43 25
Chris@43 26 /* Version history:
Chris@43 27 1.0 30 Oct 2004 First version
Chris@43 28 1.1 8 Nov 2004 Add void casting for unused return values
Chris@43 29 Use switch statement for inflate() return values
Chris@43 30 1.2 9 Nov 2004 Add assertions to document zlib guarantees
Chris@43 31 1.3 6 Apr 2005 Remove incorrect assertion in inf()
Chris@43 32 1.4 11 Dec 2005 Add hack to avoid MSDOS end-of-line conversions
Chris@43 33 Avoid some compiler warnings for input and output buffers
Chris@43 34 */
Chris@43 35 </b></pre><!-- -->
Chris@43 36 We now include the header files for the required definitions. From
Chris@43 37 <tt>stdio.h</tt> we use <tt>fopen()</tt>, <tt>fread()</tt>, <tt>fwrite()</tt>,
Chris@43 38 <tt>feof()</tt>, <tt>ferror()</tt>, and <tt>fclose()</tt> for file i/o, and
Chris@43 39 <tt>fputs()</tt> for error messages. From <tt>string.h</tt> we use
Chris@43 40 <tt>strcmp()</tt> for command line argument processing.
Chris@43 41 From <tt>assert.h</tt> we use the <tt>assert()</tt> macro.
Chris@43 42 From <tt>zlib.h</tt>
Chris@43 43 we use the basic compression functions <tt>deflateInit()</tt>,
Chris@43 44 <tt>deflate()</tt>, and <tt>deflateEnd()</tt>, and the basic decompression
Chris@43 45 functions <tt>inflateInit()</tt>, <tt>inflate()</tt>, and
Chris@43 46 <tt>inflateEnd()</tt>.
Chris@43 47 <pre><b>
Chris@43 48 #include &lt;stdio.h&gt;
Chris@43 49 #include &lt;string.h&gt;
Chris@43 50 #include &lt;assert.h&gt;
Chris@43 51 #include "zlib.h"
Chris@43 52 </b></pre><!-- -->
Chris@43 53 This is an ugly hack required to avoid corruption of the input and output data on
Chris@43 54 Windows/MS-DOS systems. Without this, those systems would assume that the input and output
Chris@43 55 files are text, and try to convert the end-of-line characters from one standard to
Chris@43 56 another. That would corrupt binary data, and in particular would render the compressed data unusable.
Chris@43 57 This sets the input and output to binary which suppresses the end-of-line conversions.
Chris@43 58 <tt>SET_BINARY_MODE()</tt> will be used later on <tt>stdin</tt> and <tt>stdout</tt>, at the beginning of <tt>main()</tt>.
Chris@43 59 <pre><b>
Chris@43 60 #if defined(MSDOS) || defined(OS2) || defined(WIN32) || defined(__CYGWIN__)
Chris@43 61 # include &lt;fcntl.h&gt;
Chris@43 62 # include &lt;io.h&gt;
Chris@43 63 # define SET_BINARY_MODE(file) setmode(fileno(file), O_BINARY)
Chris@43 64 #else
Chris@43 65 # define SET_BINARY_MODE(file)
Chris@43 66 #endif
Chris@43 67 </b></pre><!-- -->
Chris@43 68 <tt>CHUNK</tt> is simply the buffer size for feeding data to and pulling data
Chris@43 69 from the <em>zlib</em> routines. Larger buffer sizes would be more efficient,
Chris@43 70 especially for <tt>inflate()</tt>. If the memory is available, buffers sizes
Chris@43 71 on the order of 128K or 256K bytes should be used.
Chris@43 72 <pre><b>
Chris@43 73 #define CHUNK 16384
Chris@43 74 </b></pre><!-- -->
Chris@43 75 The <tt>def()</tt> routine compresses data from an input file to an output file. The output data
Chris@43 76 will be in the <em>zlib</em> format, which is different from the <em>gzip</em> or <em>zip</em>
Chris@43 77 formats. The <em>zlib</em> format has a very small header of only two bytes to identify it as
Chris@43 78 a <em>zlib</em> stream and to provide decoding information, and a four-byte trailer with a fast
Chris@43 79 check value to verify the integrity of the uncompressed data after decoding.
Chris@43 80 <pre><b>
Chris@43 81 /* Compress from file source to file dest until EOF on source.
Chris@43 82 def() returns Z_OK on success, Z_MEM_ERROR if memory could not be
Chris@43 83 allocated for processing, Z_STREAM_ERROR if an invalid compression
Chris@43 84 level is supplied, Z_VERSION_ERROR if the version of zlib.h and the
Chris@43 85 version of the library linked do not match, or Z_ERRNO if there is
Chris@43 86 an error reading or writing the files. */
Chris@43 87 int def(FILE *source, FILE *dest, int level)
Chris@43 88 {
Chris@43 89 </b></pre>
Chris@43 90 Here are the local variables for <tt>def()</tt>. <tt>ret</tt> will be used for <em>zlib</em>
Chris@43 91 return codes. <tt>flush</tt> will keep track of the current flushing state for <tt>deflate()</tt>,
Chris@43 92 which is either no flushing, or flush to completion after the end of the input file is reached.
Chris@43 93 <tt>have</tt> is the amount of data returned from <tt>deflate()</tt>. The <tt>strm</tt> structure
Chris@43 94 is used to pass information to and from the <em>zlib</em> routines, and to maintain the
Chris@43 95 <tt>deflate()</tt> state. <tt>in</tt> and <tt>out</tt> are the input and output buffers for
Chris@43 96 <tt>deflate()</tt>.
Chris@43 97 <pre><b>
Chris@43 98 int ret, flush;
Chris@43 99 unsigned have;
Chris@43 100 z_stream strm;
Chris@43 101 unsigned char in[CHUNK];
Chris@43 102 unsigned char out[CHUNK];
Chris@43 103 </b></pre><!-- -->
Chris@43 104 The first thing we do is to initialize the <em>zlib</em> state for compression using
Chris@43 105 <tt>deflateInit()</tt>. This must be done before the first use of <tt>deflate()</tt>.
Chris@43 106 The <tt>zalloc</tt>, <tt>zfree</tt>, and <tt>opaque</tt> fields in the <tt>strm</tt>
Chris@43 107 structure must be initialized before calling <tt>deflateInit()</tt>. Here they are
Chris@43 108 set to the <em>zlib</em> constant <tt>Z_NULL</tt> to request that <em>zlib</em> use
Chris@43 109 the default memory allocation routines. An application may also choose to provide
Chris@43 110 custom memory allocation routines here. <tt>deflateInit()</tt> will allocate on the
Chris@43 111 order of 256K bytes for the internal state.
Chris@43 112 (See <a href="zlib_tech.html"><em>zlib Technical Details</em></a>.)
Chris@43 113 <p>
Chris@43 114 <tt>deflateInit()</tt> is called with a pointer to the structure to be initialized and
Chris@43 115 the compression level, which is an integer in the range of -1 to 9. Lower compression
Chris@43 116 levels result in faster execution, but less compression. Higher levels result in
Chris@43 117 greater compression, but slower execution. The <em>zlib</em> constant Z_DEFAULT_COMPRESSION,
Chris@43 118 equal to -1,
Chris@43 119 provides a good compromise between compression and speed and is equivalent to level 6.
Chris@43 120 Level 0 actually does no compression at all, and in fact expands the data slightly to produce
Chris@43 121 the <em>zlib</em> format (it is not a byte-for-byte copy of the input).
Chris@43 122 More advanced applications of <em>zlib</em>
Chris@43 123 may use <tt>deflateInit2()</tt> here instead. Such an application may want to reduce how
Chris@43 124 much memory will be used, at some price in compression. Or it may need to request a
Chris@43 125 <em>gzip</em> header and trailer instead of a <em>zlib</em> header and trailer, or raw
Chris@43 126 encoding with no header or trailer at all.
Chris@43 127 <p>
Chris@43 128 We must check the return value of <tt>deflateInit()</tt> against the <em>zlib</em> constant
Chris@43 129 <tt>Z_OK</tt> to make sure that it was able to
Chris@43 130 allocate memory for the internal state, and that the provided arguments were valid.
Chris@43 131 <tt>deflateInit()</tt> will also check that the version of <em>zlib</em> that the <tt>zlib.h</tt>
Chris@43 132 file came from matches the version of <em>zlib</em> actually linked with the program. This
Chris@43 133 is especially important for environments in which <em>zlib</em> is a shared library.
Chris@43 134 <p>
Chris@43 135 Note that an application can initialize multiple, independent <em>zlib</em> streams, which can
Chris@43 136 operate in parallel. The state information maintained in the structure allows the <em>zlib</em>
Chris@43 137 routines to be reentrant.
Chris@43 138 <pre><b>
Chris@43 139 /* allocate deflate state */
Chris@43 140 strm.zalloc = Z_NULL;
Chris@43 141 strm.zfree = Z_NULL;
Chris@43 142 strm.opaque = Z_NULL;
Chris@43 143 ret = deflateInit(&amp;strm, level);
Chris@43 144 if (ret != Z_OK)
Chris@43 145 return ret;
Chris@43 146 </b></pre><!-- -->
Chris@43 147 With the pleasantries out of the way, now we can get down to business. The outer <tt>do</tt>-loop
Chris@43 148 reads all of the input file and exits at the bottom of the loop once end-of-file is reached.
Chris@43 149 This loop contains the only call of <tt>deflate()</tt>. So we must make sure that all of the
Chris@43 150 input data has been processed and that all of the output data has been generated and consumed
Chris@43 151 before we fall out of the loop at the bottom.
Chris@43 152 <pre><b>
Chris@43 153 /* compress until end of file */
Chris@43 154 do {
Chris@43 155 </b></pre>
Chris@43 156 We start off by reading data from the input file. The number of bytes read is put directly
Chris@43 157 into <tt>avail_in</tt>, and a pointer to those bytes is put into <tt>next_in</tt>. We also
Chris@43 158 check to see if end-of-file on the input has been reached. If we are at the end of file, then <tt>flush</tt> is set to the
Chris@43 159 <em>zlib</em> constant <tt>Z_FINISH</tt>, which is later passed to <tt>deflate()</tt> to
Chris@43 160 indicate that this is the last chunk of input data to compress. We need to use <tt>feof()</tt>
Chris@43 161 to check for end-of-file as opposed to seeing if fewer than <tt>CHUNK</tt> bytes have been read. The
Chris@43 162 reason is that if the input file length is an exact multiple of <tt>CHUNK</tt>, we will miss
Chris@43 163 the fact that we got to the end-of-file, and not know to tell <tt>deflate()</tt> to finish
Chris@43 164 up the compressed stream. If we are not yet at the end of the input, then the <em>zlib</em>
Chris@43 165 constant <tt>Z_NO_FLUSH</tt> will be passed to <tt>deflate</tt> to indicate that we are still
Chris@43 166 in the middle of the uncompressed data.
Chris@43 167 <p>
Chris@43 168 If there is an error in reading from the input file, the process is aborted with
Chris@43 169 <tt>deflateEnd()</tt> being called to free the allocated <em>zlib</em> state before returning
Chris@43 170 the error. We wouldn't want a memory leak, now would we? <tt>deflateEnd()</tt> can be called
Chris@43 171 at any time after the state has been initialized. Once that's done, <tt>deflateInit()</tt> (or
Chris@43 172 <tt>deflateInit2()</tt>) would have to be called to start a new compression process. There is
Chris@43 173 no point here in checking the <tt>deflateEnd()</tt> return code. The deallocation can't fail.
Chris@43 174 <pre><b>
Chris@43 175 strm.avail_in = fread(in, 1, CHUNK, source);
Chris@43 176 if (ferror(source)) {
Chris@43 177 (void)deflateEnd(&amp;strm);
Chris@43 178 return Z_ERRNO;
Chris@43 179 }
Chris@43 180 flush = feof(source) ? Z_FINISH : Z_NO_FLUSH;
Chris@43 181 strm.next_in = in;
Chris@43 182 </b></pre><!-- -->
Chris@43 183 The inner <tt>do</tt>-loop passes our chunk of input data to <tt>deflate()</tt>, and then
Chris@43 184 keeps calling <tt>deflate()</tt> until it is done producing output. Once there is no more
Chris@43 185 new output, <tt>deflate()</tt> is guaranteed to have consumed all of the input, i.e.,
Chris@43 186 <tt>avail_in</tt> will be zero.
Chris@43 187 <pre><b>
Chris@43 188 /* run deflate() on input until output buffer not full, finish
Chris@43 189 compression if all of source has been read in */
Chris@43 190 do {
Chris@43 191 </b></pre>
Chris@43 192 Output space is provided to <tt>deflate()</tt> by setting <tt>avail_out</tt> to the number
Chris@43 193 of available output bytes and <tt>next_out</tt> to a pointer to that space.
Chris@43 194 <pre><b>
Chris@43 195 strm.avail_out = CHUNK;
Chris@43 196 strm.next_out = out;
Chris@43 197 </b></pre>
Chris@43 198 Now we call the compression engine itself, <tt>deflate()</tt>. It takes as many of the
Chris@43 199 <tt>avail_in</tt> bytes at <tt>next_in</tt> as it can process, and writes as many as
Chris@43 200 <tt>avail_out</tt> bytes to <tt>next_out</tt>. Those counters and pointers are then
Chris@43 201 updated past the input data consumed and the output data written. It is the amount of
Chris@43 202 output space available that may limit how much input is consumed.
Chris@43 203 Hence the inner loop to make sure that
Chris@43 204 all of the input is consumed by providing more output space each time. Since <tt>avail_in</tt>
Chris@43 205 and <tt>next_in</tt> are updated by <tt>deflate()</tt>, we don't have to mess with those
Chris@43 206 between <tt>deflate()</tt> calls until it's all used up.
Chris@43 207 <p>
Chris@43 208 The parameters to <tt>deflate()</tt> are a pointer to the <tt>strm</tt> structure containing
Chris@43 209 the input and output information and the internal compression engine state, and a parameter
Chris@43 210 indicating whether and how to flush data to the output. Normally <tt>deflate</tt> will consume
Chris@43 211 several K bytes of input data before producing any output (except for the header), in order
Chris@43 212 to accumulate statistics on the data for optimum compression. It will then put out a burst of
Chris@43 213 compressed data, and proceed to consume more input before the next burst. Eventually,
Chris@43 214 <tt>deflate()</tt>
Chris@43 215 must be told to terminate the stream, complete the compression with provided input data, and
Chris@43 216 write out the trailer check value. <tt>deflate()</tt> will continue to compress normally as long
Chris@43 217 as the flush parameter is <tt>Z_NO_FLUSH</tt>. Once the <tt>Z_FINISH</tt> parameter is provided,
Chris@43 218 <tt>deflate()</tt> will begin to complete the compressed output stream. However depending on how
Chris@43 219 much output space is provided, <tt>deflate()</tt> may have to be called several times until it
Chris@43 220 has provided the complete compressed stream, even after it has consumed all of the input. The flush
Chris@43 221 parameter must continue to be <tt>Z_FINISH</tt> for those subsequent calls.
Chris@43 222 <p>
Chris@43 223 There are other values of the flush parameter that are used in more advanced applications. You can
Chris@43 224 force <tt>deflate()</tt> to produce a burst of output that encodes all of the input data provided
Chris@43 225 so far, even if it wouldn't have otherwise, for example to control data latency on a link with
Chris@43 226 compressed data. You can also ask that <tt>deflate()</tt> do that as well as erase any history up to
Chris@43 227 that point so that what follows can be decompressed independently, for example for random access
Chris@43 228 applications. Both requests will degrade compression by an amount depending on how often such
Chris@43 229 requests are made.
Chris@43 230 <p>
Chris@43 231 <tt>deflate()</tt> has a return value that can indicate errors, yet we do not check it here. Why
Chris@43 232 not? Well, it turns out that <tt>deflate()</tt> can do no wrong here. Let's go through
Chris@43 233 <tt>deflate()</tt>'s return values and dispense with them one by one. The possible values are
Chris@43 234 <tt>Z_OK</tt>, <tt>Z_STREAM_END</tt>, <tt>Z_STREAM_ERROR</tt>, or <tt>Z_BUF_ERROR</tt>. <tt>Z_OK</tt>
Chris@43 235 is, well, ok. <tt>Z_STREAM_END</tt> is also ok and will be returned for the last call of
Chris@43 236 <tt>deflate()</tt>. This is already guaranteed by calling <tt>deflate()</tt> with <tt>Z_FINISH</tt>
Chris@43 237 until it has no more output. <tt>Z_STREAM_ERROR</tt> is only possible if the stream is not
Chris@43 238 initialized properly, but we did initialize it properly. There is no harm in checking for
Chris@43 239 <tt>Z_STREAM_ERROR</tt> here, for example to check for the possibility that some
Chris@43 240 other part of the application inadvertently clobbered the memory containing the <em>zlib</em> state.
Chris@43 241 <tt>Z_BUF_ERROR</tt> will be explained further below, but
Chris@43 242 suffice it to say that this is simply an indication that <tt>deflate()</tt> could not consume
Chris@43 243 more input or produce more output. <tt>deflate()</tt> can be called again with more output space
Chris@43 244 or more available input, which it will be in this code.
Chris@43 245 <pre><b>
Chris@43 246 ret = deflate(&amp;strm, flush); /* no bad return value */
Chris@43 247 assert(ret != Z_STREAM_ERROR); /* state not clobbered */
Chris@43 248 </b></pre>
Chris@43 249 Now we compute how much output <tt>deflate()</tt> provided on the last call, which is the
Chris@43 250 difference between how much space was provided before the call, and how much output space
Chris@43 251 is still available after the call. Then that data, if any, is written to the output file.
Chris@43 252 We can then reuse the output buffer for the next call of <tt>deflate()</tt>. Again if there
Chris@43 253 is a file i/o error, we call <tt>deflateEnd()</tt> before returning to avoid a memory leak.
Chris@43 254 <pre><b>
Chris@43 255 have = CHUNK - strm.avail_out;
Chris@43 256 if (fwrite(out, 1, have, dest) != have || ferror(dest)) {
Chris@43 257 (void)deflateEnd(&amp;strm);
Chris@43 258 return Z_ERRNO;
Chris@43 259 }
Chris@43 260 </b></pre>
Chris@43 261 The inner <tt>do</tt>-loop is repeated until the last <tt>deflate()</tt> call fails to fill the
Chris@43 262 provided output buffer. Then we know that <tt>deflate()</tt> has done as much as it can with
Chris@43 263 the provided input, and that all of that input has been consumed. We can then fall out of this
Chris@43 264 loop and reuse the input buffer.
Chris@43 265 <p>
Chris@43 266 The way we tell that <tt>deflate()</tt> has no more output is by seeing that it did not fill
Chris@43 267 the output buffer, leaving <tt>avail_out</tt> greater than zero. However suppose that
Chris@43 268 <tt>deflate()</tt> has no more output, but just so happened to exactly fill the output buffer!
Chris@43 269 <tt>avail_out</tt> is zero, and we can't tell that <tt>deflate()</tt> has done all it can.
Chris@43 270 As far as we know, <tt>deflate()</tt>
Chris@43 271 has more output for us. So we call it again. But now <tt>deflate()</tt> produces no output
Chris@43 272 at all, and <tt>avail_out</tt> remains unchanged as <tt>CHUNK</tt>. That <tt>deflate()</tt> call
Chris@43 273 wasn't able to do anything, either consume input or produce output, and so it returns
Chris@43 274 <tt>Z_BUF_ERROR</tt>. (See, I told you I'd cover this later.) However this is not a problem at
Chris@43 275 all. Now we finally have the desired indication that <tt>deflate()</tt> is really done,
Chris@43 276 and so we drop out of the inner loop to provide more input to <tt>deflate()</tt>.
Chris@43 277 <p>
Chris@43 278 With <tt>flush</tt> set to <tt>Z_FINISH</tt>, this final set of <tt>deflate()</tt> calls will
Chris@43 279 complete the output stream. Once that is done, subsequent calls of <tt>deflate()</tt> would return
Chris@43 280 <tt>Z_STREAM_ERROR</tt> if the flush parameter is not <tt>Z_FINISH</tt>, and do no more processing
Chris@43 281 until the state is reinitialized.
Chris@43 282 <p>
Chris@43 283 Some applications of <em>zlib</em> have two loops that call <tt>deflate()</tt>
Chris@43 284 instead of the single inner loop we have here. The first loop would call
Chris@43 285 without flushing and feed all of the data to <tt>deflate()</tt>. The second loop would call
Chris@43 286 <tt>deflate()</tt> with no more
Chris@43 287 data and the <tt>Z_FINISH</tt> parameter to complete the process. As you can see from this
Chris@43 288 example, that can be avoided by simply keeping track of the current flush state.
Chris@43 289 <pre><b>
Chris@43 290 } while (strm.avail_out == 0);
Chris@43 291 assert(strm.avail_in == 0); /* all input will be used */
Chris@43 292 </b></pre><!-- -->
Chris@43 293 Now we check to see if we have already processed all of the input file. That information was
Chris@43 294 saved in the <tt>flush</tt> variable, so we see if that was set to <tt>Z_FINISH</tt>. If so,
Chris@43 295 then we're done and we fall out of the outer loop. We're guaranteed to get <tt>Z_STREAM_END</tt>
Chris@43 296 from the last <tt>deflate()</tt> call, since we ran it until the last chunk of input was
Chris@43 297 consumed and all of the output was generated.
Chris@43 298 <pre><b>
Chris@43 299 /* done when last data in file processed */
Chris@43 300 } while (flush != Z_FINISH);
Chris@43 301 assert(ret == Z_STREAM_END); /* stream will be complete */
Chris@43 302 </b></pre><!-- -->
Chris@43 303 The process is complete, but we still need to deallocate the state to avoid a memory leak
Chris@43 304 (or rather more like a memory hemorrhage if you didn't do this). Then
Chris@43 305 finally we can return with a happy return value.
Chris@43 306 <pre><b>
Chris@43 307 /* clean up and return */
Chris@43 308 (void)deflateEnd(&amp;strm);
Chris@43 309 return Z_OK;
Chris@43 310 }
Chris@43 311 </b></pre><!-- -->
Chris@43 312 Now we do the same thing for decompression in the <tt>inf()</tt> routine. <tt>inf()</tt>
Chris@43 313 decompresses what is hopefully a valid <em>zlib</em> stream from the input file and writes the
Chris@43 314 uncompressed data to the output file. Much of the discussion above for <tt>def()</tt>
Chris@43 315 applies to <tt>inf()</tt> as well, so the discussion here will focus on the differences between
Chris@43 316 the two.
Chris@43 317 <pre><b>
Chris@43 318 /* Decompress from file source to file dest until stream ends or EOF.
Chris@43 319 inf() returns Z_OK on success, Z_MEM_ERROR if memory could not be
Chris@43 320 allocated for processing, Z_DATA_ERROR if the deflate data is
Chris@43 321 invalid or incomplete, Z_VERSION_ERROR if the version of zlib.h and
Chris@43 322 the version of the library linked do not match, or Z_ERRNO if there
Chris@43 323 is an error reading or writing the files. */
Chris@43 324 int inf(FILE *source, FILE *dest)
Chris@43 325 {
Chris@43 326 </b></pre>
Chris@43 327 The local variables have the same functionality as they do for <tt>def()</tt>. The
Chris@43 328 only difference is that there is no <tt>flush</tt> variable, since <tt>inflate()</tt>
Chris@43 329 can tell from the <em>zlib</em> stream itself when the stream is complete.
Chris@43 330 <pre><b>
Chris@43 331 int ret;
Chris@43 332 unsigned have;
Chris@43 333 z_stream strm;
Chris@43 334 unsigned char in[CHUNK];
Chris@43 335 unsigned char out[CHUNK];
Chris@43 336 </b></pre><!-- -->
Chris@43 337 The initialization of the state is the same, except that there is no compression level,
Chris@43 338 of course, and two more elements of the structure are initialized. <tt>avail_in</tt>
Chris@43 339 and <tt>next_in</tt> must be initialized before calling <tt>inflateInit()</tt>. This
Chris@43 340 is because the application has the option to provide the start of the zlib stream in
Chris@43 341 order for <tt>inflateInit()</tt> to have access to information about the compression
Chris@43 342 method to aid in memory allocation. In the current implementation of <em>zlib</em>
Chris@43 343 (up through versions 1.2.x), the method-dependent memory allocations are deferred to the first call of
Chris@43 344 <tt>inflate()</tt> anyway. However those fields must be initialized since later versions
Chris@43 345 of <em>zlib</em> that provide more compression methods may take advantage of this interface.
Chris@43 346 In any case, no decompression is performed by <tt>inflateInit()</tt>, so the
Chris@43 347 <tt>avail_out</tt> and <tt>next_out</tt> fields do not need to be initialized before calling.
Chris@43 348 <p>
Chris@43 349 Here <tt>avail_in</tt> is set to zero and <tt>next_in</tt> is set to <tt>Z_NULL</tt> to
Chris@43 350 indicate that no input data is being provided.
Chris@43 351 <pre><b>
Chris@43 352 /* allocate inflate state */
Chris@43 353 strm.zalloc = Z_NULL;
Chris@43 354 strm.zfree = Z_NULL;
Chris@43 355 strm.opaque = Z_NULL;
Chris@43 356 strm.avail_in = 0;
Chris@43 357 strm.next_in = Z_NULL;
Chris@43 358 ret = inflateInit(&amp;strm);
Chris@43 359 if (ret != Z_OK)
Chris@43 360 return ret;
Chris@43 361 </b></pre><!-- -->
Chris@43 362 The outer <tt>do</tt>-loop decompresses input until <tt>inflate()</tt> indicates
Chris@43 363 that it has reached the end of the compressed data and has produced all of the uncompressed
Chris@43 364 output. This is in contrast to <tt>def()</tt> which processes all of the input file.
Chris@43 365 If end-of-file is reached before the compressed data self-terminates, then the compressed
Chris@43 366 data is incomplete and an error is returned.
Chris@43 367 <pre><b>
Chris@43 368 /* decompress until deflate stream ends or end of file */
Chris@43 369 do {
Chris@43 370 </b></pre>
Chris@43 371 We read input data and set the <tt>strm</tt> structure accordingly. If we've reached the
Chris@43 372 end of the input file, then we leave the outer loop and report an error, since the
Chris@43 373 compressed data is incomplete. Note that we may read more data than is eventually consumed
Chris@43 374 by <tt>inflate()</tt>, if the input file continues past the <em>zlib</em> stream.
Chris@43 375 For applications where <em>zlib</em> streams are embedded in other data, this routine would
Chris@43 376 need to be modified to return the unused data, or at least indicate how much of the input
Chris@43 377 data was not used, so the application would know where to pick up after the <em>zlib</em> stream.
Chris@43 378 <pre><b>
Chris@43 379 strm.avail_in = fread(in, 1, CHUNK, source);
Chris@43 380 if (ferror(source)) {
Chris@43 381 (void)inflateEnd(&amp;strm);
Chris@43 382 return Z_ERRNO;
Chris@43 383 }
Chris@43 384 if (strm.avail_in == 0)
Chris@43 385 break;
Chris@43 386 strm.next_in = in;
Chris@43 387 </b></pre><!-- -->
Chris@43 388 The inner <tt>do</tt>-loop has the same function it did in <tt>def()</tt>, which is to
Chris@43 389 keep calling <tt>inflate()</tt> until has generated all of the output it can with the
Chris@43 390 provided input.
Chris@43 391 <pre><b>
Chris@43 392 /* run inflate() on input until output buffer not full */
Chris@43 393 do {
Chris@43 394 </b></pre>
Chris@43 395 Just like in <tt>def()</tt>, the same output space is provided for each call of <tt>inflate()</tt>.
Chris@43 396 <pre><b>
Chris@43 397 strm.avail_out = CHUNK;
Chris@43 398 strm.next_out = out;
Chris@43 399 </b></pre>
Chris@43 400 Now we run the decompression engine itself. There is no need to adjust the flush parameter, since
Chris@43 401 the <em>zlib</em> format is self-terminating. The main difference here is that there are
Chris@43 402 return values that we need to pay attention to. <tt>Z_DATA_ERROR</tt>
Chris@43 403 indicates that <tt>inflate()</tt> detected an error in the <em>zlib</em> compressed data format,
Chris@43 404 which means that either the data is not a <em>zlib</em> stream to begin with, or that the data was
Chris@43 405 corrupted somewhere along the way since it was compressed. The other error to be processed is
Chris@43 406 <tt>Z_MEM_ERROR</tt>, which can occur since memory allocation is deferred until <tt>inflate()</tt>
Chris@43 407 needs it, unlike <tt>deflate()</tt>, whose memory is allocated at the start by <tt>deflateInit()</tt>.
Chris@43 408 <p>
Chris@43 409 Advanced applications may use
Chris@43 410 <tt>deflateSetDictionary()</tt> to prime <tt>deflate()</tt> with a set of likely data to improve the
Chris@43 411 first 32K or so of compression. This is noted in the <em>zlib</em> header, so <tt>inflate()</tt>
Chris@43 412 requests that that dictionary be provided before it can start to decompress. Without the dictionary,
Chris@43 413 correct decompression is not possible. For this routine, we have no idea what the dictionary is,
Chris@43 414 so the <tt>Z_NEED_DICT</tt> indication is converted to a <tt>Z_DATA_ERROR</tt>.
Chris@43 415 <p>
Chris@43 416 <tt>inflate()</tt> can also return <tt>Z_STREAM_ERROR</tt>, which should not be possible here,
Chris@43 417 but could be checked for as noted above for <tt>def()</tt>. <tt>Z_BUF_ERROR</tt> does not need to be
Chris@43 418 checked for here, for the same reasons noted for <tt>def()</tt>. <tt>Z_STREAM_END</tt> will be
Chris@43 419 checked for later.
Chris@43 420 <pre><b>
Chris@43 421 ret = inflate(&amp;strm, Z_NO_FLUSH);
Chris@43 422 assert(ret != Z_STREAM_ERROR); /* state not clobbered */
Chris@43 423 switch (ret) {
Chris@43 424 case Z_NEED_DICT:
Chris@43 425 ret = Z_DATA_ERROR; /* and fall through */
Chris@43 426 case Z_DATA_ERROR:
Chris@43 427 case Z_MEM_ERROR:
Chris@43 428 (void)inflateEnd(&amp;strm);
Chris@43 429 return ret;
Chris@43 430 }
Chris@43 431 </b></pre>
Chris@43 432 The output of <tt>inflate()</tt> is handled identically to that of <tt>deflate()</tt>.
Chris@43 433 <pre><b>
Chris@43 434 have = CHUNK - strm.avail_out;
Chris@43 435 if (fwrite(out, 1, have, dest) != have || ferror(dest)) {
Chris@43 436 (void)inflateEnd(&amp;strm);
Chris@43 437 return Z_ERRNO;
Chris@43 438 }
Chris@43 439 </b></pre>
Chris@43 440 The inner <tt>do</tt>-loop ends when <tt>inflate()</tt> has no more output as indicated
Chris@43 441 by not filling the output buffer, just as for <tt>deflate()</tt>. In this case, we cannot
Chris@43 442 assert that <tt>strm.avail_in</tt> will be zero, since the deflate stream may end before the file
Chris@43 443 does.
Chris@43 444 <pre><b>
Chris@43 445 } while (strm.avail_out == 0);
Chris@43 446 </b></pre><!-- -->
Chris@43 447 The outer <tt>do</tt>-loop ends when <tt>inflate()</tt> reports that it has reached the
Chris@43 448 end of the input <em>zlib</em> stream, has completed the decompression and integrity
Chris@43 449 check, and has provided all of the output. This is indicated by the <tt>inflate()</tt>
Chris@43 450 return value <tt>Z_STREAM_END</tt>. The inner loop is guaranteed to leave <tt>ret</tt>
Chris@43 451 equal to <tt>Z_STREAM_END</tt> if the last chunk of the input file read contained the end
Chris@43 452 of the <em>zlib</em> stream. So if the return value is not <tt>Z_STREAM_END</tt>, the
Chris@43 453 loop continues to read more input.
Chris@43 454 <pre><b>
Chris@43 455 /* done when inflate() says it's done */
Chris@43 456 } while (ret != Z_STREAM_END);
Chris@43 457 </b></pre><!-- -->
Chris@43 458 At this point, decompression successfully completed, or we broke out of the loop due to no
Chris@43 459 more data being available from the input file. If the last <tt>inflate()</tt> return value
Chris@43 460 is not <tt>Z_STREAM_END</tt>, then the <em>zlib</em> stream was incomplete and a data error
Chris@43 461 is returned. Otherwise, we return with a happy return value. Of course, <tt>inflateEnd()</tt>
Chris@43 462 is called first to avoid a memory leak.
Chris@43 463 <pre><b>
Chris@43 464 /* clean up and return */
Chris@43 465 (void)inflateEnd(&amp;strm);
Chris@43 466 return ret == Z_STREAM_END ? Z_OK : Z_DATA_ERROR;
Chris@43 467 }
Chris@43 468 </b></pre><!-- -->
Chris@43 469 That ends the routines that directly use <em>zlib</em>. The following routines make this
Chris@43 470 a command-line program by running data through the above routines from <tt>stdin</tt> to
Chris@43 471 <tt>stdout</tt>, and handling any errors reported by <tt>def()</tt> or <tt>inf()</tt>.
Chris@43 472 <p>
Chris@43 473 <tt>zerr()</tt> is used to interpret the possible error codes from <tt>def()</tt>
Chris@43 474 and <tt>inf()</tt>, as detailed in their comments above, and print out an error message.
Chris@43 475 Note that these are only a subset of the possible return values from <tt>deflate()</tt>
Chris@43 476 and <tt>inflate()</tt>.
Chris@43 477 <pre><b>
Chris@43 478 /* report a zlib or i/o error */
Chris@43 479 void zerr(int ret)
Chris@43 480 {
Chris@43 481 fputs("zpipe: ", stderr);
Chris@43 482 switch (ret) {
Chris@43 483 case Z_ERRNO:
Chris@43 484 if (ferror(stdin))
Chris@43 485 fputs("error reading stdin\n", stderr);
Chris@43 486 if (ferror(stdout))
Chris@43 487 fputs("error writing stdout\n", stderr);
Chris@43 488 break;
Chris@43 489 case Z_STREAM_ERROR:
Chris@43 490 fputs("invalid compression level\n", stderr);
Chris@43 491 break;
Chris@43 492 case Z_DATA_ERROR:
Chris@43 493 fputs("invalid or incomplete deflate data\n", stderr);
Chris@43 494 break;
Chris@43 495 case Z_MEM_ERROR:
Chris@43 496 fputs("out of memory\n", stderr);
Chris@43 497 break;
Chris@43 498 case Z_VERSION_ERROR:
Chris@43 499 fputs("zlib version mismatch!\n", stderr);
Chris@43 500 }
Chris@43 501 }
Chris@43 502 </b></pre><!-- -->
Chris@43 503 Here is the <tt>main()</tt> routine used to test <tt>def()</tt> and <tt>inf()</tt>. The
Chris@43 504 <tt>zpipe</tt> command is simply a compression pipe from <tt>stdin</tt> to <tt>stdout</tt>, if
Chris@43 505 no arguments are given, or it is a decompression pipe if <tt>zpipe -d</tt> is used. If any other
Chris@43 506 arguments are provided, no compression or decompression is performed. Instead a usage
Chris@43 507 message is displayed. Examples are <tt>zpipe < foo.txt > foo.txt.z</tt> to compress, and
Chris@43 508 <tt>zpipe -d < foo.txt.z > foo.txt</tt> to decompress.
Chris@43 509 <pre><b>
Chris@43 510 /* compress or decompress from stdin to stdout */
Chris@43 511 int main(int argc, char **argv)
Chris@43 512 {
Chris@43 513 int ret;
Chris@43 514
Chris@43 515 /* avoid end-of-line conversions */
Chris@43 516 SET_BINARY_MODE(stdin);
Chris@43 517 SET_BINARY_MODE(stdout);
Chris@43 518
Chris@43 519 /* do compression if no arguments */
Chris@43 520 if (argc == 1) {
Chris@43 521 ret = def(stdin, stdout, Z_DEFAULT_COMPRESSION);
Chris@43 522 if (ret != Z_OK)
Chris@43 523 zerr(ret);
Chris@43 524 return ret;
Chris@43 525 }
Chris@43 526
Chris@43 527 /* do decompression if -d specified */
Chris@43 528 else if (argc == 2 &amp;&amp; strcmp(argv[1], "-d") == 0) {
Chris@43 529 ret = inf(stdin, stdout);
Chris@43 530 if (ret != Z_OK)
Chris@43 531 zerr(ret);
Chris@43 532 return ret;
Chris@43 533 }
Chris@43 534
Chris@43 535 /* otherwise, report usage */
Chris@43 536 else {
Chris@43 537 fputs("zpipe usage: zpipe [-d] &lt; source &gt; dest\n", stderr);
Chris@43 538 return 1;
Chris@43 539 }
Chris@43 540 }
Chris@43 541 </b></pre>
Chris@43 542 <hr>
Chris@43 543 <i>Copyright (c) 2004, 2005 by Mark Adler<br>Last modified 11 December 2005</i>
Chris@43 544 </body>
Chris@43 545 </html>