1*4e00368fSchristos<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN" 2*4e00368fSchristos "http://www.w3.org/TR/REC-html40/loose.dtd"> 3*4e00368fSchristos<html> 4*4e00368fSchristos<head> 5*4e00368fSchristos<meta http-equiv="Content-Type" content="text/html; charset=ISO-8859-1"> 6*4e00368fSchristos<title>zlib Usage Example</title> 7*4e00368fSchristos<!-- Copyright (c) 2004, 2005 Mark Adler. --> 8*4e00368fSchristos</head> 9*4e00368fSchristos<body bgcolor="#FFFFFF" text="#000000" link="#0000FF" vlink="#00A000"> 10*4e00368fSchristos<h2 align="center"> zlib Usage Example </h2> 11*4e00368fSchristosWe often get questions about how the <tt>deflate()</tt> and <tt>inflate()</tt> functions should be used. 12*4e00368fSchristosUsers wonder when they should provide more input, when they should use more output, 13*4e00368fSchristoswhat to do with a <tt>Z_BUF_ERROR</tt>, how to make sure the process terminates properly, and 14*4e00368fSchristosso on. So for those who have read <tt>zlib.h</tt> (a few times), and 15*4e00368fSchristoswould like further edification, below is an annotated example in C of simple routines to compress and decompress 16*4e00368fSchristosfrom an input file to an output file using <tt>deflate()</tt> and <tt>inflate()</tt> respectively. The 17*4e00368fSchristosannotations are interspersed between lines of the code. So please read between the lines. 18*4e00368fSchristosWe hope this helps explain some of the intricacies of <em>zlib</em>. 19*4e00368fSchristos<p> 20*4e00368fSchristosWithout further adieu, here is the program <a href="zpipe.c"><tt>zpipe.c</tt></a>: 21*4e00368fSchristos<pre><b> 22*4e00368fSchristos/* zpipe.c: example of proper use of zlib's inflate() and deflate() 23*4e00368fSchristos Not copyrighted -- provided to the public domain 24*4e00368fSchristos Version 1.4 11 December 2005 Mark Adler */ 25*4e00368fSchristos 26*4e00368fSchristos/* Version history: 27*4e00368fSchristos 1.0 30 Oct 2004 First version 28*4e00368fSchristos 1.1 8 Nov 2004 Add void casting for unused return values 29*4e00368fSchristos Use switch statement for inflate() return values 30*4e00368fSchristos 1.2 9 Nov 2004 Add assertions to document zlib guarantees 31*4e00368fSchristos 1.3 6 Apr 2005 Remove incorrect assertion in inf() 32*4e00368fSchristos 1.4 11 Dec 2005 Add hack to avoid MSDOS end-of-line conversions 33*4e00368fSchristos Avoid some compiler warnings for input and output buffers 34*4e00368fSchristos */ 35*4e00368fSchristos</b></pre><!-- --> 36*4e00368fSchristosWe now include the header files for the required definitions. From 37*4e00368fSchristos<tt>stdio.h</tt> we use <tt>fopen()</tt>, <tt>fread()</tt>, <tt>fwrite()</tt>, 38*4e00368fSchristos<tt>feof()</tt>, <tt>ferror()</tt>, and <tt>fclose()</tt> for file i/o, and 39*4e00368fSchristos<tt>fputs()</tt> for error messages. From <tt>string.h</tt> we use 40*4e00368fSchristos<tt>strcmp()</tt> for command line argument processing. 41*4e00368fSchristosFrom <tt>assert.h</tt> we use the <tt>assert()</tt> macro. 42*4e00368fSchristosFrom <tt>zlib.h</tt> 43*4e00368fSchristoswe use the basic compression functions <tt>deflateInit()</tt>, 44*4e00368fSchristos<tt>deflate()</tt>, and <tt>deflateEnd()</tt>, and the basic decompression 45*4e00368fSchristosfunctions <tt>inflateInit()</tt>, <tt>inflate()</tt>, and 46*4e00368fSchristos<tt>inflateEnd()</tt>. 47*4e00368fSchristos<pre><b> 48*4e00368fSchristos#include <stdio.h> 49*4e00368fSchristos#include <string.h> 50*4e00368fSchristos#include <assert.h> 51*4e00368fSchristos#include "zlib.h" 52*4e00368fSchristos</b></pre><!-- --> 53*4e00368fSchristosThis is an ugly hack required to avoid corruption of the input and output data on 54*4e00368fSchristosWindows/MS-DOS systems. Without this, those systems would assume that the input and output 55*4e00368fSchristosfiles are text, and try to convert the end-of-line characters from one standard to 56*4e00368fSchristosanother. That would corrupt binary data, and in particular would render the compressed data unusable. 57*4e00368fSchristosThis sets the input and output to binary which suppresses the end-of-line conversions. 58*4e00368fSchristos<tt>SET_BINARY_MODE()</tt> will be used later on <tt>stdin</tt> and <tt>stdout</tt>, at the beginning of <tt>main()</tt>. 59*4e00368fSchristos<pre><b> 60*4e00368fSchristos#if defined(MSDOS) || defined(OS2) || defined(WIN32) || defined(__CYGWIN__) 61*4e00368fSchristos# include <fcntl.h> 62*4e00368fSchristos# include <io.h> 63*4e00368fSchristos# define SET_BINARY_MODE(file) setmode(fileno(file), O_BINARY) 64*4e00368fSchristos#else 65*4e00368fSchristos# define SET_BINARY_MODE(file) 66*4e00368fSchristos#endif 67*4e00368fSchristos</b></pre><!-- --> 68*4e00368fSchristos<tt>CHUNK</tt> is simply the buffer size for feeding data to and pulling data 69*4e00368fSchristosfrom the <em>zlib</em> routines. Larger buffer sizes would be more efficient, 70*4e00368fSchristosespecially for <tt>inflate()</tt>. If the memory is available, buffers sizes 71*4e00368fSchristoson the order of 128K or 256K bytes should be used. 72*4e00368fSchristos<pre><b> 73*4e00368fSchristos#define CHUNK 16384 74*4e00368fSchristos</b></pre><!-- --> 75*4e00368fSchristosThe <tt>def()</tt> routine compresses data from an input file to an output file. The output data 76*4e00368fSchristoswill be in the <em>zlib</em> format, which is different from the <em>gzip</em> or <em>zip</em> 77*4e00368fSchristosformats. The <em>zlib</em> format has a very small header of only two bytes to identify it as 78*4e00368fSchristosa <em>zlib</em> stream and to provide decoding information, and a four-byte trailer with a fast 79*4e00368fSchristoscheck value to verify the integrity of the uncompressed data after decoding. 80*4e00368fSchristos<pre><b> 81*4e00368fSchristos/* Compress from file source to file dest until EOF on source. 82*4e00368fSchristos def() returns Z_OK on success, Z_MEM_ERROR if memory could not be 83*4e00368fSchristos allocated for processing, Z_STREAM_ERROR if an invalid compression 84*4e00368fSchristos level is supplied, Z_VERSION_ERROR if the version of zlib.h and the 85*4e00368fSchristos version of the library linked do not match, or Z_ERRNO if there is 86*4e00368fSchristos an error reading or writing the files. */ 87*4e00368fSchristosint def(FILE *source, FILE *dest, int level) 88*4e00368fSchristos{ 89*4e00368fSchristos</b></pre> 90*4e00368fSchristosHere are the local variables for <tt>def()</tt>. <tt>ret</tt> will be used for <em>zlib</em> 91*4e00368fSchristosreturn codes. <tt>flush</tt> will keep track of the current flushing state for <tt>deflate()</tt>, 92*4e00368fSchristoswhich is either no flushing, or flush to completion after the end of the input file is reached. 93*4e00368fSchristos<tt>have</tt> is the amount of data returned from <tt>deflate()</tt>. The <tt>strm</tt> structure 94*4e00368fSchristosis used to pass information to and from the <em>zlib</em> routines, and to maintain the 95*4e00368fSchristos<tt>deflate()</tt> state. <tt>in</tt> and <tt>out</tt> are the input and output buffers for 96*4e00368fSchristos<tt>deflate()</tt>. 97*4e00368fSchristos<pre><b> 98*4e00368fSchristos int ret, flush; 99*4e00368fSchristos unsigned have; 100*4e00368fSchristos z_stream strm; 101*4e00368fSchristos unsigned char in[CHUNK]; 102*4e00368fSchristos unsigned char out[CHUNK]; 103*4e00368fSchristos</b></pre><!-- --> 104*4e00368fSchristosThe first thing we do is to initialize the <em>zlib</em> state for compression using 105*4e00368fSchristos<tt>deflateInit()</tt>. This must be done before the first use of <tt>deflate()</tt>. 106*4e00368fSchristosThe <tt>zalloc</tt>, <tt>zfree</tt>, and <tt>opaque</tt> fields in the <tt>strm</tt> 107*4e00368fSchristosstructure must be initialized before calling <tt>deflateInit()</tt>. Here they are 108*4e00368fSchristosset to the <em>zlib</em> constant <tt>Z_NULL</tt> to request that <em>zlib</em> use 109*4e00368fSchristosthe default memory allocation routines. An application may also choose to provide 110*4e00368fSchristoscustom memory allocation routines here. <tt>deflateInit()</tt> will allocate on the 111*4e00368fSchristosorder of 256K bytes for the internal state. 112*4e00368fSchristos(See <a href="zlib_tech.html"><em>zlib Technical Details</em></a>.) 113*4e00368fSchristos<p> 114*4e00368fSchristos<tt>deflateInit()</tt> is called with a pointer to the structure to be initialized and 115*4e00368fSchristosthe compression level, which is an integer in the range of -1 to 9. Lower compression 116*4e00368fSchristoslevels result in faster execution, but less compression. Higher levels result in 117*4e00368fSchristosgreater compression, but slower execution. The <em>zlib</em> constant Z_DEFAULT_COMPRESSION, 118*4e00368fSchristosequal to -1, 119*4e00368fSchristosprovides a good compromise between compression and speed and is equivalent to level 6. 120*4e00368fSchristosLevel 0 actually does no compression at all, and in fact expands the data slightly to produce 121*4e00368fSchristosthe <em>zlib</em> format (it is not a byte-for-byte copy of the input). 122*4e00368fSchristosMore advanced applications of <em>zlib</em> 123*4e00368fSchristosmay use <tt>deflateInit2()</tt> here instead. Such an application may want to reduce how 124*4e00368fSchristosmuch memory will be used, at some price in compression. Or it may need to request a 125*4e00368fSchristos<em>gzip</em> header and trailer instead of a <em>zlib</em> header and trailer, or raw 126*4e00368fSchristosencoding with no header or trailer at all. 127*4e00368fSchristos<p> 128*4e00368fSchristosWe must check the return value of <tt>deflateInit()</tt> against the <em>zlib</em> constant 129*4e00368fSchristos<tt>Z_OK</tt> to make sure that it was able to 130*4e00368fSchristosallocate memory for the internal state, and that the provided arguments were valid. 131*4e00368fSchristos<tt>deflateInit()</tt> will also check that the version of <em>zlib</em> that the <tt>zlib.h</tt> 132*4e00368fSchristosfile came from matches the version of <em>zlib</em> actually linked with the program. This 133*4e00368fSchristosis especially important for environments in which <em>zlib</em> is a shared library. 134*4e00368fSchristos<p> 135*4e00368fSchristosNote that an application can initialize multiple, independent <em>zlib</em> streams, which can 136*4e00368fSchristosoperate in parallel. The state information maintained in the structure allows the <em>zlib</em> 137*4e00368fSchristosroutines to be reentrant. 138*4e00368fSchristos<pre><b> 139*4e00368fSchristos /* allocate deflate state */ 140*4e00368fSchristos strm.zalloc = Z_NULL; 141*4e00368fSchristos strm.zfree = Z_NULL; 142*4e00368fSchristos strm.opaque = Z_NULL; 143*4e00368fSchristos ret = deflateInit(&strm, level); 144*4e00368fSchristos if (ret != Z_OK) 145*4e00368fSchristos return ret; 146*4e00368fSchristos</b></pre><!-- --> 147*4e00368fSchristosWith the pleasantries out of the way, now we can get down to business. The outer <tt>do</tt>-loop 148*4e00368fSchristosreads all of the input file and exits at the bottom of the loop once end-of-file is reached. 149*4e00368fSchristosThis loop contains the only call of <tt>deflate()</tt>. So we must make sure that all of the 150*4e00368fSchristosinput data has been processed and that all of the output data has been generated and consumed 151*4e00368fSchristosbefore we fall out of the loop at the bottom. 152*4e00368fSchristos<pre><b> 153*4e00368fSchristos /* compress until end of file */ 154*4e00368fSchristos do { 155*4e00368fSchristos</b></pre> 156*4e00368fSchristosWe start off by reading data from the input file. The number of bytes read is put directly 157*4e00368fSchristosinto <tt>avail_in</tt>, and a pointer to those bytes is put into <tt>next_in</tt>. We also 158*4e00368fSchristoscheck 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 159*4e00368fSchristos<em>zlib</em> constant <tt>Z_FINISH</tt>, which is later passed to <tt>deflate()</tt> to 160*4e00368fSchristosindicate that this is the last chunk of input data to compress. We need to use <tt>feof()</tt> 161*4e00368fSchristosto check for end-of-file as opposed to seeing if fewer than <tt>CHUNK</tt> bytes have been read. The 162*4e00368fSchristosreason is that if the input file length is an exact multiple of <tt>CHUNK</tt>, we will miss 163*4e00368fSchristosthe fact that we got to the end-of-file, and not know to tell <tt>deflate()</tt> to finish 164*4e00368fSchristosup the compressed stream. If we are not yet at the end of the input, then the <em>zlib</em> 165*4e00368fSchristosconstant <tt>Z_NO_FLUSH</tt> will be passed to <tt>deflate</tt> to indicate that we are still 166*4e00368fSchristosin the middle of the uncompressed data. 167*4e00368fSchristos<p> 168*4e00368fSchristosIf there is an error in reading from the input file, the process is aborted with 169*4e00368fSchristos<tt>deflateEnd()</tt> being called to free the allocated <em>zlib</em> state before returning 170*4e00368fSchristosthe error. We wouldn't want a memory leak, now would we? <tt>deflateEnd()</tt> can be called 171*4e00368fSchristosat any time after the state has been initialized. Once that's done, <tt>deflateInit()</tt> (or 172*4e00368fSchristos<tt>deflateInit2()</tt>) would have to be called to start a new compression process. There is 173*4e00368fSchristosno point here in checking the <tt>deflateEnd()</tt> return code. The deallocation can't fail. 174*4e00368fSchristos<pre><b> 175*4e00368fSchristos strm.avail_in = fread(in, 1, CHUNK, source); 176*4e00368fSchristos if (ferror(source)) { 177*4e00368fSchristos (void)deflateEnd(&strm); 178*4e00368fSchristos return Z_ERRNO; 179*4e00368fSchristos } 180*4e00368fSchristos flush = feof(source) ? Z_FINISH : Z_NO_FLUSH; 181*4e00368fSchristos strm.next_in = in; 182*4e00368fSchristos</b></pre><!-- --> 183*4e00368fSchristosThe inner <tt>do</tt>-loop passes our chunk of input data to <tt>deflate()</tt>, and then 184*4e00368fSchristoskeeps calling <tt>deflate()</tt> until it is done producing output. Once there is no more 185*4e00368fSchristosnew output, <tt>deflate()</tt> is guaranteed to have consumed all of the input, i.e., 186*4e00368fSchristos<tt>avail_in</tt> will be zero. 187*4e00368fSchristos<pre><b> 188*4e00368fSchristos /* run deflate() on input until output buffer not full, finish 189*4e00368fSchristos compression if all of source has been read in */ 190*4e00368fSchristos do { 191*4e00368fSchristos</b></pre> 192*4e00368fSchristosOutput space is provided to <tt>deflate()</tt> by setting <tt>avail_out</tt> to the number 193*4e00368fSchristosof available output bytes and <tt>next_out</tt> to a pointer to that space. 194*4e00368fSchristos<pre><b> 195*4e00368fSchristos strm.avail_out = CHUNK; 196*4e00368fSchristos strm.next_out = out; 197*4e00368fSchristos</b></pre> 198*4e00368fSchristosNow we call the compression engine itself, <tt>deflate()</tt>. It takes as many of the 199*4e00368fSchristos<tt>avail_in</tt> bytes at <tt>next_in</tt> as it can process, and writes as many as 200*4e00368fSchristos<tt>avail_out</tt> bytes to <tt>next_out</tt>. Those counters and pointers are then 201*4e00368fSchristosupdated past the input data consumed and the output data written. It is the amount of 202*4e00368fSchristosoutput space available that may limit how much input is consumed. 203*4e00368fSchristosHence the inner loop to make sure that 204*4e00368fSchristosall of the input is consumed by providing more output space each time. Since <tt>avail_in</tt> 205*4e00368fSchristosand <tt>next_in</tt> are updated by <tt>deflate()</tt>, we don't have to mess with those 206*4e00368fSchristosbetween <tt>deflate()</tt> calls until it's all used up. 207*4e00368fSchristos<p> 208*4e00368fSchristosThe parameters to <tt>deflate()</tt> are a pointer to the <tt>strm</tt> structure containing 209*4e00368fSchristosthe input and output information and the internal compression engine state, and a parameter 210*4e00368fSchristosindicating whether and how to flush data to the output. Normally <tt>deflate</tt> will consume 211*4e00368fSchristosseveral K bytes of input data before producing any output (except for the header), in order 212*4e00368fSchristosto accumulate statistics on the data for optimum compression. It will then put out a burst of 213*4e00368fSchristoscompressed data, and proceed to consume more input before the next burst. Eventually, 214*4e00368fSchristos<tt>deflate()</tt> 215*4e00368fSchristosmust be told to terminate the stream, complete the compression with provided input data, and 216*4e00368fSchristoswrite out the trailer check value. <tt>deflate()</tt> will continue to compress normally as long 217*4e00368fSchristosas the flush parameter is <tt>Z_NO_FLUSH</tt>. Once the <tt>Z_FINISH</tt> parameter is provided, 218*4e00368fSchristos<tt>deflate()</tt> will begin to complete the compressed output stream. However depending on how 219*4e00368fSchristosmuch output space is provided, <tt>deflate()</tt> may have to be called several times until it 220*4e00368fSchristoshas provided the complete compressed stream, even after it has consumed all of the input. The flush 221*4e00368fSchristosparameter must continue to be <tt>Z_FINISH</tt> for those subsequent calls. 222*4e00368fSchristos<p> 223*4e00368fSchristosThere are other values of the flush parameter that are used in more advanced applications. You can 224*4e00368fSchristosforce <tt>deflate()</tt> to produce a burst of output that encodes all of the input data provided 225*4e00368fSchristosso far, even if it wouldn't have otherwise, for example to control data latency on a link with 226*4e00368fSchristoscompressed data. You can also ask that <tt>deflate()</tt> do that as well as erase any history up to 227*4e00368fSchristosthat point so that what follows can be decompressed independently, for example for random access 228*4e00368fSchristosapplications. Both requests will degrade compression by an amount depending on how often such 229*4e00368fSchristosrequests are made. 230*4e00368fSchristos<p> 231*4e00368fSchristos<tt>deflate()</tt> has a return value that can indicate errors, yet we do not check it here. Why 232*4e00368fSchristosnot? Well, it turns out that <tt>deflate()</tt> can do no wrong here. Let's go through 233*4e00368fSchristos<tt>deflate()</tt>'s return values and dispense with them one by one. The possible values are 234*4e00368fSchristos<tt>Z_OK</tt>, <tt>Z_STREAM_END</tt>, <tt>Z_STREAM_ERROR</tt>, or <tt>Z_BUF_ERROR</tt>. <tt>Z_OK</tt> 235*4e00368fSchristosis, well, ok. <tt>Z_STREAM_END</tt> is also ok and will be returned for the last call of 236*4e00368fSchristos<tt>deflate()</tt>. This is already guaranteed by calling <tt>deflate()</tt> with <tt>Z_FINISH</tt> 237*4e00368fSchristosuntil it has no more output. <tt>Z_STREAM_ERROR</tt> is only possible if the stream is not 238*4e00368fSchristosinitialized properly, but we did initialize it properly. There is no harm in checking for 239*4e00368fSchristos<tt>Z_STREAM_ERROR</tt> here, for example to check for the possibility that some 240*4e00368fSchristosother part of the application inadvertently clobbered the memory containing the <em>zlib</em> state. 241*4e00368fSchristos<tt>Z_BUF_ERROR</tt> will be explained further below, but 242*4e00368fSchristossuffice it to say that this is simply an indication that <tt>deflate()</tt> could not consume 243*4e00368fSchristosmore input or produce more output. <tt>deflate()</tt> can be called again with more output space 244*4e00368fSchristosor more available input, which it will be in this code. 245*4e00368fSchristos<pre><b> 246*4e00368fSchristos ret = deflate(&strm, flush); /* no bad return value */ 247*4e00368fSchristos assert(ret != Z_STREAM_ERROR); /* state not clobbered */ 248*4e00368fSchristos</b></pre> 249*4e00368fSchristosNow we compute how much output <tt>deflate()</tt> provided on the last call, which is the 250*4e00368fSchristosdifference between how much space was provided before the call, and how much output space 251*4e00368fSchristosis still available after the call. Then that data, if any, is written to the output file. 252*4e00368fSchristosWe can then reuse the output buffer for the next call of <tt>deflate()</tt>. Again if there 253*4e00368fSchristosis a file i/o error, we call <tt>deflateEnd()</tt> before returning to avoid a memory leak. 254*4e00368fSchristos<pre><b> 255*4e00368fSchristos have = CHUNK - strm.avail_out; 256*4e00368fSchristos if (fwrite(out, 1, have, dest) != have || ferror(dest)) { 257*4e00368fSchristos (void)deflateEnd(&strm); 258*4e00368fSchristos return Z_ERRNO; 259*4e00368fSchristos } 260*4e00368fSchristos</b></pre> 261*4e00368fSchristosThe inner <tt>do</tt>-loop is repeated until the last <tt>deflate()</tt> call fails to fill the 262*4e00368fSchristosprovided output buffer. Then we know that <tt>deflate()</tt> has done as much as it can with 263*4e00368fSchristosthe provided input, and that all of that input has been consumed. We can then fall out of this 264*4e00368fSchristosloop and reuse the input buffer. 265*4e00368fSchristos<p> 266*4e00368fSchristosThe way we tell that <tt>deflate()</tt> has no more output is by seeing that it did not fill 267*4e00368fSchristosthe output buffer, leaving <tt>avail_out</tt> greater than zero. However suppose that 268*4e00368fSchristos<tt>deflate()</tt> has no more output, but just so happened to exactly fill the output buffer! 269*4e00368fSchristos<tt>avail_out</tt> is zero, and we can't tell that <tt>deflate()</tt> has done all it can. 270*4e00368fSchristosAs far as we know, <tt>deflate()</tt> 271*4e00368fSchristoshas more output for us. So we call it again. But now <tt>deflate()</tt> produces no output 272*4e00368fSchristosat all, and <tt>avail_out</tt> remains unchanged as <tt>CHUNK</tt>. That <tt>deflate()</tt> call 273*4e00368fSchristoswasn't able to do anything, either consume input or produce output, and so it returns 274*4e00368fSchristos<tt>Z_BUF_ERROR</tt>. (See, I told you I'd cover this later.) However this is not a problem at 275*4e00368fSchristosall. Now we finally have the desired indication that <tt>deflate()</tt> is really done, 276*4e00368fSchristosand so we drop out of the inner loop to provide more input to <tt>deflate()</tt>. 277*4e00368fSchristos<p> 278*4e00368fSchristosWith <tt>flush</tt> set to <tt>Z_FINISH</tt>, this final set of <tt>deflate()</tt> calls will 279*4e00368fSchristoscomplete the output stream. Once that is done, subsequent calls of <tt>deflate()</tt> would return 280*4e00368fSchristos<tt>Z_STREAM_ERROR</tt> if the flush parameter is not <tt>Z_FINISH</tt>, and do no more processing 281*4e00368fSchristosuntil the state is reinitialized. 282*4e00368fSchristos<p> 283*4e00368fSchristosSome applications of <em>zlib</em> have two loops that call <tt>deflate()</tt> 284*4e00368fSchristosinstead of the single inner loop we have here. The first loop would call 285*4e00368fSchristoswithout flushing and feed all of the data to <tt>deflate()</tt>. The second loop would call 286*4e00368fSchristos<tt>deflate()</tt> with no more 287*4e00368fSchristosdata and the <tt>Z_FINISH</tt> parameter to complete the process. As you can see from this 288*4e00368fSchristosexample, that can be avoided by simply keeping track of the current flush state. 289*4e00368fSchristos<pre><b> 290*4e00368fSchristos } while (strm.avail_out == 0); 291*4e00368fSchristos assert(strm.avail_in == 0); /* all input will be used */ 292*4e00368fSchristos</b></pre><!-- --> 293*4e00368fSchristosNow we check to see if we have already processed all of the input file. That information was 294*4e00368fSchristossaved in the <tt>flush</tt> variable, so we see if that was set to <tt>Z_FINISH</tt>. If so, 295*4e00368fSchristosthen we're done and we fall out of the outer loop. We're guaranteed to get <tt>Z_STREAM_END</tt> 296*4e00368fSchristosfrom the last <tt>deflate()</tt> call, since we ran it until the last chunk of input was 297*4e00368fSchristosconsumed and all of the output was generated. 298*4e00368fSchristos<pre><b> 299*4e00368fSchristos /* done when last data in file processed */ 300*4e00368fSchristos } while (flush != Z_FINISH); 301*4e00368fSchristos assert(ret == Z_STREAM_END); /* stream will be complete */ 302*4e00368fSchristos</b></pre><!-- --> 303*4e00368fSchristosThe process is complete, but we still need to deallocate the state to avoid a memory leak 304*4e00368fSchristos(or rather more like a memory hemorrhage if you didn't do this). Then 305*4e00368fSchristosfinally we can return with a happy return value. 306*4e00368fSchristos<pre><b> 307*4e00368fSchristos /* clean up and return */ 308*4e00368fSchristos (void)deflateEnd(&strm); 309*4e00368fSchristos return Z_OK; 310*4e00368fSchristos} 311*4e00368fSchristos</b></pre><!-- --> 312*4e00368fSchristosNow we do the same thing for decompression in the <tt>inf()</tt> routine. <tt>inf()</tt> 313*4e00368fSchristosdecompresses what is hopefully a valid <em>zlib</em> stream from the input file and writes the 314*4e00368fSchristosuncompressed data to the output file. Much of the discussion above for <tt>def()</tt> 315*4e00368fSchristosapplies to <tt>inf()</tt> as well, so the discussion here will focus on the differences between 316*4e00368fSchristosthe two. 317*4e00368fSchristos<pre><b> 318*4e00368fSchristos/* Decompress from file source to file dest until stream ends or EOF. 319*4e00368fSchristos inf() returns Z_OK on success, Z_MEM_ERROR if memory could not be 320*4e00368fSchristos allocated for processing, Z_DATA_ERROR if the deflate data is 321*4e00368fSchristos invalid or incomplete, Z_VERSION_ERROR if the version of zlib.h and 322*4e00368fSchristos the version of the library linked do not match, or Z_ERRNO if there 323*4e00368fSchristos is an error reading or writing the files. */ 324*4e00368fSchristosint inf(FILE *source, FILE *dest) 325*4e00368fSchristos{ 326*4e00368fSchristos</b></pre> 327*4e00368fSchristosThe local variables have the same functionality as they do for <tt>def()</tt>. The 328*4e00368fSchristosonly difference is that there is no <tt>flush</tt> variable, since <tt>inflate()</tt> 329*4e00368fSchristoscan tell from the <em>zlib</em> stream itself when the stream is complete. 330*4e00368fSchristos<pre><b> 331*4e00368fSchristos int ret; 332*4e00368fSchristos unsigned have; 333*4e00368fSchristos z_stream strm; 334*4e00368fSchristos unsigned char in[CHUNK]; 335*4e00368fSchristos unsigned char out[CHUNK]; 336*4e00368fSchristos</b></pre><!-- --> 337*4e00368fSchristosThe initialization of the state is the same, except that there is no compression level, 338*4e00368fSchristosof course, and two more elements of the structure are initialized. <tt>avail_in</tt> 339*4e00368fSchristosand <tt>next_in</tt> must be initialized before calling <tt>inflateInit()</tt>. This 340*4e00368fSchristosis because the application has the option to provide the start of the zlib stream in 341*4e00368fSchristosorder for <tt>inflateInit()</tt> to have access to information about the compression 342*4e00368fSchristosmethod to aid in memory allocation. In the current implementation of <em>zlib</em> 343*4e00368fSchristos(up through versions 1.2.x), the method-dependent memory allocations are deferred to the first call of 344*4e00368fSchristos<tt>inflate()</tt> anyway. However those fields must be initialized since later versions 345*4e00368fSchristosof <em>zlib</em> that provide more compression methods may take advantage of this interface. 346*4e00368fSchristosIn any case, no decompression is performed by <tt>inflateInit()</tt>, so the 347*4e00368fSchristos<tt>avail_out</tt> and <tt>next_out</tt> fields do not need to be initialized before calling. 348*4e00368fSchristos<p> 349*4e00368fSchristosHere <tt>avail_in</tt> is set to zero and <tt>next_in</tt> is set to <tt>Z_NULL</tt> to 350*4e00368fSchristosindicate that no input data is being provided. 351*4e00368fSchristos<pre><b> 352*4e00368fSchristos /* allocate inflate state */ 353*4e00368fSchristos strm.zalloc = Z_NULL; 354*4e00368fSchristos strm.zfree = Z_NULL; 355*4e00368fSchristos strm.opaque = Z_NULL; 356*4e00368fSchristos strm.avail_in = 0; 357*4e00368fSchristos strm.next_in = Z_NULL; 358*4e00368fSchristos ret = inflateInit(&strm); 359*4e00368fSchristos if (ret != Z_OK) 360*4e00368fSchristos return ret; 361*4e00368fSchristos</b></pre><!-- --> 362*4e00368fSchristosThe outer <tt>do</tt>-loop decompresses input until <tt>inflate()</tt> indicates 363*4e00368fSchristosthat it has reached the end of the compressed data and has produced all of the uncompressed 364*4e00368fSchristosoutput. This is in contrast to <tt>def()</tt> which processes all of the input file. 365*4e00368fSchristosIf end-of-file is reached before the compressed data self-terminates, then the compressed 366*4e00368fSchristosdata is incomplete and an error is returned. 367*4e00368fSchristos<pre><b> 368*4e00368fSchristos /* decompress until deflate stream ends or end of file */ 369*4e00368fSchristos do { 370*4e00368fSchristos</b></pre> 371*4e00368fSchristosWe read input data and set the <tt>strm</tt> structure accordingly. If we've reached the 372*4e00368fSchristosend of the input file, then we leave the outer loop and report an error, since the 373*4e00368fSchristoscompressed data is incomplete. Note that we may read more data than is eventually consumed 374*4e00368fSchristosby <tt>inflate()</tt>, if the input file continues past the <em>zlib</em> stream. 375*4e00368fSchristosFor applications where <em>zlib</em> streams are embedded in other data, this routine would 376*4e00368fSchristosneed to be modified to return the unused data, or at least indicate how much of the input 377*4e00368fSchristosdata was not used, so the application would know where to pick up after the <em>zlib</em> stream. 378*4e00368fSchristos<pre><b> 379*4e00368fSchristos strm.avail_in = fread(in, 1, CHUNK, source); 380*4e00368fSchristos if (ferror(source)) { 381*4e00368fSchristos (void)inflateEnd(&strm); 382*4e00368fSchristos return Z_ERRNO; 383*4e00368fSchristos } 384*4e00368fSchristos if (strm.avail_in == 0) 385*4e00368fSchristos break; 386*4e00368fSchristos strm.next_in = in; 387*4e00368fSchristos</b></pre><!-- --> 388*4e00368fSchristosThe inner <tt>do</tt>-loop has the same function it did in <tt>def()</tt>, which is to 389*4e00368fSchristoskeep calling <tt>inflate()</tt> until has generated all of the output it can with the 390*4e00368fSchristosprovided input. 391*4e00368fSchristos<pre><b> 392*4e00368fSchristos /* run inflate() on input until output buffer not full */ 393*4e00368fSchristos do { 394*4e00368fSchristos</b></pre> 395*4e00368fSchristosJust like in <tt>def()</tt>, the same output space is provided for each call of <tt>inflate()</tt>. 396*4e00368fSchristos<pre><b> 397*4e00368fSchristos strm.avail_out = CHUNK; 398*4e00368fSchristos strm.next_out = out; 399*4e00368fSchristos</b></pre> 400*4e00368fSchristosNow we run the decompression engine itself. There is no need to adjust the flush parameter, since 401*4e00368fSchristosthe <em>zlib</em> format is self-terminating. The main difference here is that there are 402*4e00368fSchristosreturn values that we need to pay attention to. <tt>Z_DATA_ERROR</tt> 403*4e00368fSchristosindicates that <tt>inflate()</tt> detected an error in the <em>zlib</em> compressed data format, 404*4e00368fSchristoswhich means that either the data is not a <em>zlib</em> stream to begin with, or that the data was 405*4e00368fSchristoscorrupted somewhere along the way since it was compressed. The other error to be processed is 406*4e00368fSchristos<tt>Z_MEM_ERROR</tt>, which can occur since memory allocation is deferred until <tt>inflate()</tt> 407*4e00368fSchristosneeds it, unlike <tt>deflate()</tt>, whose memory is allocated at the start by <tt>deflateInit()</tt>. 408*4e00368fSchristos<p> 409*4e00368fSchristosAdvanced applications may use 410*4e00368fSchristos<tt>deflateSetDictionary()</tt> to prime <tt>deflate()</tt> with a set of likely data to improve the 411*4e00368fSchristosfirst 32K or so of compression. This is noted in the <em>zlib</em> header, so <tt>inflate()</tt> 412*4e00368fSchristosrequests that that dictionary be provided before it can start to decompress. Without the dictionary, 413*4e00368fSchristoscorrect decompression is not possible. For this routine, we have no idea what the dictionary is, 414*4e00368fSchristosso the <tt>Z_NEED_DICT</tt> indication is converted to a <tt>Z_DATA_ERROR</tt>. 415*4e00368fSchristos<p> 416*4e00368fSchristos<tt>inflate()</tt> can also return <tt>Z_STREAM_ERROR</tt>, which should not be possible here, 417*4e00368fSchristosbut could be checked for as noted above for <tt>def()</tt>. <tt>Z_BUF_ERROR</tt> does not need to be 418*4e00368fSchristoschecked for here, for the same reasons noted for <tt>def()</tt>. <tt>Z_STREAM_END</tt> will be 419*4e00368fSchristoschecked for later. 420*4e00368fSchristos<pre><b> 421*4e00368fSchristos ret = inflate(&strm, Z_NO_FLUSH); 422*4e00368fSchristos assert(ret != Z_STREAM_ERROR); /* state not clobbered */ 423*4e00368fSchristos switch (ret) { 424*4e00368fSchristos case Z_NEED_DICT: 425*4e00368fSchristos ret = Z_DATA_ERROR; /* and fall through */ 426*4e00368fSchristos case Z_DATA_ERROR: 427*4e00368fSchristos case Z_MEM_ERROR: 428*4e00368fSchristos (void)inflateEnd(&strm); 429*4e00368fSchristos return ret; 430*4e00368fSchristos } 431*4e00368fSchristos</b></pre> 432*4e00368fSchristosThe output of <tt>inflate()</tt> is handled identically to that of <tt>deflate()</tt>. 433*4e00368fSchristos<pre><b> 434*4e00368fSchristos have = CHUNK - strm.avail_out; 435*4e00368fSchristos if (fwrite(out, 1, have, dest) != have || ferror(dest)) { 436*4e00368fSchristos (void)inflateEnd(&strm); 437*4e00368fSchristos return Z_ERRNO; 438*4e00368fSchristos } 439*4e00368fSchristos</b></pre> 440*4e00368fSchristosThe inner <tt>do</tt>-loop ends when <tt>inflate()</tt> has no more output as indicated 441*4e00368fSchristosby not filling the output buffer, just as for <tt>deflate()</tt>. In this case, we cannot 442*4e00368fSchristosassert that <tt>strm.avail_in</tt> will be zero, since the deflate stream may end before the file 443*4e00368fSchristosdoes. 444*4e00368fSchristos<pre><b> 445*4e00368fSchristos } while (strm.avail_out == 0); 446*4e00368fSchristos</b></pre><!-- --> 447*4e00368fSchristosThe outer <tt>do</tt>-loop ends when <tt>inflate()</tt> reports that it has reached the 448*4e00368fSchristosend of the input <em>zlib</em> stream, has completed the decompression and integrity 449*4e00368fSchristoscheck, and has provided all of the output. This is indicated by the <tt>inflate()</tt> 450*4e00368fSchristosreturn value <tt>Z_STREAM_END</tt>. The inner loop is guaranteed to leave <tt>ret</tt> 451*4e00368fSchristosequal to <tt>Z_STREAM_END</tt> if the last chunk of the input file read contained the end 452*4e00368fSchristosof the <em>zlib</em> stream. So if the return value is not <tt>Z_STREAM_END</tt>, the 453*4e00368fSchristosloop continues to read more input. 454*4e00368fSchristos<pre><b> 455*4e00368fSchristos /* done when inflate() says it's done */ 456*4e00368fSchristos } while (ret != Z_STREAM_END); 457*4e00368fSchristos</b></pre><!-- --> 458*4e00368fSchristosAt this point, decompression successfully completed, or we broke out of the loop due to no 459*4e00368fSchristosmore data being available from the input file. If the last <tt>inflate()</tt> return value 460*4e00368fSchristosis not <tt>Z_STREAM_END</tt>, then the <em>zlib</em> stream was incomplete and a data error 461*4e00368fSchristosis returned. Otherwise, we return with a happy return value. Of course, <tt>inflateEnd()</tt> 462*4e00368fSchristosis called first to avoid a memory leak. 463*4e00368fSchristos<pre><b> 464*4e00368fSchristos /* clean up and return */ 465*4e00368fSchristos (void)inflateEnd(&strm); 466*4e00368fSchristos return ret == Z_STREAM_END ? Z_OK : Z_DATA_ERROR; 467*4e00368fSchristos} 468*4e00368fSchristos</b></pre><!-- --> 469*4e00368fSchristosThat ends the routines that directly use <em>zlib</em>. The following routines make this 470*4e00368fSchristosa command-line program by running data through the above routines from <tt>stdin</tt> to 471*4e00368fSchristos<tt>stdout</tt>, and handling any errors reported by <tt>def()</tt> or <tt>inf()</tt>. 472*4e00368fSchristos<p> 473*4e00368fSchristos<tt>zerr()</tt> is used to interpret the possible error codes from <tt>def()</tt> 474*4e00368fSchristosand <tt>inf()</tt>, as detailed in their comments above, and print out an error message. 475*4e00368fSchristosNote that these are only a subset of the possible return values from <tt>deflate()</tt> 476*4e00368fSchristosand <tt>inflate()</tt>. 477*4e00368fSchristos<pre><b> 478*4e00368fSchristos/* report a zlib or i/o error */ 479*4e00368fSchristosvoid zerr(int ret) 480*4e00368fSchristos{ 481*4e00368fSchristos fputs("zpipe: ", stderr); 482*4e00368fSchristos switch (ret) { 483*4e00368fSchristos case Z_ERRNO: 484*4e00368fSchristos if (ferror(stdin)) 485*4e00368fSchristos fputs("error reading stdin\n", stderr); 486*4e00368fSchristos if (ferror(stdout)) 487*4e00368fSchristos fputs("error writing stdout\n", stderr); 488*4e00368fSchristos break; 489*4e00368fSchristos case Z_STREAM_ERROR: 490*4e00368fSchristos fputs("invalid compression level\n", stderr); 491*4e00368fSchristos break; 492*4e00368fSchristos case Z_DATA_ERROR: 493*4e00368fSchristos fputs("invalid or incomplete deflate data\n", stderr); 494*4e00368fSchristos break; 495*4e00368fSchristos case Z_MEM_ERROR: 496*4e00368fSchristos fputs("out of memory\n", stderr); 497*4e00368fSchristos break; 498*4e00368fSchristos case Z_VERSION_ERROR: 499*4e00368fSchristos fputs("zlib version mismatch!\n", stderr); 500*4e00368fSchristos } 501*4e00368fSchristos} 502*4e00368fSchristos</b></pre><!-- --> 503*4e00368fSchristosHere is the <tt>main()</tt> routine used to test <tt>def()</tt> and <tt>inf()</tt>. The 504*4e00368fSchristos<tt>zpipe</tt> command is simply a compression pipe from <tt>stdin</tt> to <tt>stdout</tt>, if 505*4e00368fSchristosno arguments are given, or it is a decompression pipe if <tt>zpipe -d</tt> is used. If any other 506*4e00368fSchristosarguments are provided, no compression or decompression is performed. Instead a usage 507*4e00368fSchristosmessage is displayed. Examples are <tt>zpipe < foo.txt > foo.txt.z</tt> to compress, and 508*4e00368fSchristos<tt>zpipe -d < foo.txt.z > foo.txt</tt> to decompress. 509*4e00368fSchristos<pre><b> 510*4e00368fSchristos/* compress or decompress from stdin to stdout */ 511*4e00368fSchristosint main(int argc, char **argv) 512*4e00368fSchristos{ 513*4e00368fSchristos int ret; 514*4e00368fSchristos 515*4e00368fSchristos /* avoid end-of-line conversions */ 516*4e00368fSchristos SET_BINARY_MODE(stdin); 517*4e00368fSchristos SET_BINARY_MODE(stdout); 518*4e00368fSchristos 519*4e00368fSchristos /* do compression if no arguments */ 520*4e00368fSchristos if (argc == 1) { 521*4e00368fSchristos ret = def(stdin, stdout, Z_DEFAULT_COMPRESSION); 522*4e00368fSchristos if (ret != Z_OK) 523*4e00368fSchristos zerr(ret); 524*4e00368fSchristos return ret; 525*4e00368fSchristos } 526*4e00368fSchristos 527*4e00368fSchristos /* do decompression if -d specified */ 528*4e00368fSchristos else if (argc == 2 && strcmp(argv[1], "-d") == 0) { 529*4e00368fSchristos ret = inf(stdin, stdout); 530*4e00368fSchristos if (ret != Z_OK) 531*4e00368fSchristos zerr(ret); 532*4e00368fSchristos return ret; 533*4e00368fSchristos } 534*4e00368fSchristos 535*4e00368fSchristos /* otherwise, report usage */ 536*4e00368fSchristos else { 537*4e00368fSchristos fputs("zpipe usage: zpipe [-d] < source > dest\n", stderr); 538*4e00368fSchristos return 1; 539*4e00368fSchristos } 540*4e00368fSchristos} 541*4e00368fSchristos</b></pre> 542*4e00368fSchristos<hr> 543*4e00368fSchristos<i>Copyright (c) 2004, 2005 by Mark Adler<br>Last modified 11 December 2005</i> 544*4e00368fSchristos</body> 545*4e00368fSchristos</html> 546