path: root/lib/libz/compress.3

.\"	$OpenBSD: compress.3,v 1.2 2003/10/03 19:00:14 jmc Exp $
.\"
.\"  Copyright (C) 1995-2002 Jean-loup Gailly and Mark Adler
.\"
.\"  This software is provided 'as-is', without any express or implied
.\"  warranty.  In no event will the authors be held liable for any damages
.\"  arising from the use of this software.
.\"
.\"  Permission is granted to anyone to use this software for any purpose,
.\"  including commercial applications, and to alter it and redistribute it
.\"  freely, subject to the following restrictions:
.\"
.\"  The origin of this software must not be misrepresented; you must not
.\"  claim that you wrote the original software. If you use this software
.\"  in a product, an acknowledgment in the product documentation would be
.\"  appreciated but is not required.
.\"  Altered source versions must be plainly marked as such, and must not be
.\"  misrepresented as being the original software.
.\"  This notice may not be removed or altered from any source distribution.
.\"
.\" Converted to mdoc format for the OpenBSD project
.\" by Jason McIntyre <jmc@openbsd.org>
.\"
.Dd September 17, 2003
.Dt COMPRESS 3
.Os
.Sh NAME
.Nm compress
.Nd zlib general purpose compression library
.Sh SYNOPSIS
.Fd "#include <zlib.h>"
.Pp
Basic functions
.Pp
.Ft const char *
.Fn zlibVersion "void"
.Ft int
.Fn deflateInit "z_streamp strm" "int level"
.Ft int
.Fn deflate "z_streamp strm" "int flush"
.Ft int
.Fn deflateEnd "z_streamp strm"
.Ft int
.Fn inflateInit "z_streamp strm"
.Ft int
.Fn inflate "z_streamp strm" "int flush"
.Ft int
.Fn inflateEnd "z_streamp strm"
.Pp
Advanced functions
.Pp
.Ft int
.Fn deflateInit2 "z_streamp strm" "int level" "int method" "int windowBits" "int memLevel" "int strategy"
.Ft int
.Fn deflateSetDictionary "z_streamp strm" "const Bytef *dictionary" "uInt dictLength"
.Ft int
.Fn deflateCopy "z_streamp dest" "z_streamp source"
.Ft int
.Fn deflateReset "z_streamp strm"
.Ft int
.Fn deflateParams "z_streamp strm" "int level" "int strategy"
.Ft int
.Fn inflateInit2 "z_streamp strm" "int windowBits"
.Ft int
.Fn inflateSetDictionary "z_streamp strm" "const Bytef *dictionary" "uInt dictLength"
.Ft int
.Fn inflateSync "z_streamp strm"
.Ft int
.Fn inflateReset "z_streamp strm"
.Pp
Utility functions
.Pp
.Fd typedef voidp gzFile;
.Pp
.Ft int
.Fn compress "Bytef *dest" "uLongf *destLen" "const Bytef *source" "uLong sourceLen"
.Ft int
.Fn compress2 "Bytef *dest" "uLongf *destLen" "const Bytef *source" "uLong sourceLen" "int level"
.Ft int
.Fn uncompress "Bytef *dest" "uLongf *destLen" "const Bytef *source" "uLong sourceLen"
.Ft gzFile
.Fn gzopen "const char *path" "const char *mode"
.Ft gzFile
.Fn gzdopen "int fd" "const char *mode"
.Ft int
.Fn gzsetparams "gzFile file" "int level" "int strategy"
.Ft int
.Fn gzread "gzFile file" "voidp buf" "unsigned len"
.Ft int
.Fn gzwrite "gzFile file" "const voidp buf" "unsigned len"
.Ft int
.Fn gzprintf "gzFile file" "const char *format" "..."
.Ft int
.Fn gzputs "gzFile file" "const char *s"
.Ft char *
.Fn gzgets "gzFile file" "char *buf" "int len"
.Ft int
.Fn gzputc "gzFile file" "int c"
.Ft int
.Fn gzgetc "gzFile file"
.Ft int
.Fn gzflush "gzFile file" "int flush"
.Ft z_off_t
.Fn gzseek "gzFile file" "z_off_t offset" "int whence"
.Ft int
.Fn gzrewind "gzFile file"
.Ft z_off_t
.Fn gztell "gzFile file"
.Ft int
.Fn gzeof "gzFile file"
.Ft int
.Fn gzclose "gzFile file"
.Ft const char *
.Fn gzerror "gzFile file" "int *errnum"
.Pp
Checksum functions
.Pp
.Ft uLong
.Fn adler32 "uLong adler" "const Bytef *buf" "uInt len"
.Ft uLong
.Fn crc32 "uLong crc" "const Bytef *buf" "uInt len"
.Sh DESCRIPTION
This manual page describes the
.Nm zlib
general purpose compression library, version 1.1.4.
.Pp
The
.Nm zlib
compression library provides in-memory compression and decompression functions,
including integrity checks of the uncompressed data.
This version of the library supports only one compression method
.Pq deflation
but other algorithms will be added later and will have the same
stream interface.
.Pp
Compression can be done in a single step if the buffers are large enough
.Pq for example if an input file is mmap'ed ,
or can be done by repeated calls of the compression function.
In the latter case, the application must provide more input
and/or consume the output
.Pq providing more output space
before each call.
.Pp
The library also supports reading and writing files in
.Xr gzip 1
.Pq .gz
format with an interface similar to that of
.Xr stdio 3 .
.Pp
The library does not install any signal handler.
The decoder checks the consistency of the compressed data,
so the library should never crash even in case of corrupted input.
.Pp
The functions within the library are divided into the following sections:
.Pp
.Bl -dash -offset indent -compact
.It
Basic functions
.It
Advanced functions
.It
Utility functions
.It
Checksum functions
.El
.Sh BASIC FUNCTIONS
.Bl -tag -width Ds
.It Xo
.Fa const char *
.Fn zlibVersion "void" ;
.Xc
.Pp
The application can compare
.Fn zlibVersion
and
.Dv ZLIB_VERSION
for consistency.
If the first character differs, the library code actually used is
not compatible with the
.Aq Pa zlib.h
header file used by the application.
This check is automatically made by
.Fn deflateInit
and
.Fn inflateInit .
.It Xo
.Fa int
.Fn deflateInit "z_streamp strm" "int level" ;
.Xc
.Pp
The
.Fn deflateInit
function initializes the internal stream state for compression.
The fields
.Fa zalloc ,
.Fa zfree ,
and
.Fa opaque
must be initialized before by the caller.
If
.Fa zalloc
and
.Fa zfree
are set to
.Dv Z_NULL ,
.Fn deflateInit
updates them to use default allocation functions.
.Pp
The compression level must be
.Dv Z_DEFAULT_COMPRESSION ,
or between 0 and 9:
1 gives best speed, 9 gives best compression, 0 gives no compression at all
(the input data is simply copied a block at a time).
.Pp
.Dv Z_DEFAULT_COMPRESSION
requests a default compromise between speed and compression
.Pq currently equivalent to level 6 .
.Pp
.Fn deflateInit
returns
.Dv Z_OK
if successful,
.Dv Z_MEM_ERROR
if there was not enough memory,
.Dv Z_STREAM_ERROR
if level is not a valid compression level,
.Dv Z_VERSION_ERROR
if the
.Nm zlib
library version
.Pq zlib_version
is incompatible with the version assumed by the caller
.Pq ZLIB_VERSION .
.Fa msg
is set to null if there is no error message.
.Fn deflateInit
does not perform any compression: this will be done by
.Fn deflate .
.It Xo
.Fa int
.Fn deflate "z_streamp strm" "int flush" ;
.Xc
.Pp
.Fn deflate
compresses as much data as possible, and stops when the input
buffer becomes empty or the output buffer becomes full.
It may introduce some output latency
.Pq reading input without producing any output
except when forced to flush.
.Pp
The detailed semantics are as follows.
.Fn deflate
performs one or both of the following actions:
.Pp
Compress more input starting at
.Fa next_in
and update
.Fa next_in
and
.Fa avail_in
accordingly.
If not all input can be processed
(because there is not enough room in the output buffer),
.Fa next_in
and
.Fa avail_in
are updated and processing will resume at this point for the next call to
.Fn deflate .
.Pp
Provide more output starting at
.Fa next_out
and update
.Fa next_out
and
.Fa avail_out
accordingly.
This action is forced if the parameter
.Fa flush
is non-zero.
Forcing
.Fa flush
frequently degrades the compression ratio,
so this parameter should be set only when necessary
.Pq in interactive applications .
Some output may be provided even if
.Fa flush
is not set.
.Pp
Before the call to
.Fn deflate ,
the application should ensure that at least
one of the actions is possible, by providing more input and/or consuming
more output, and updating
.Fa avail_in
or
.Fa avail_out
accordingly;
.Fa avail_out
should never be zero before the call.
The application can consume the compressed output when it wants,
for example when the output buffer is full
.Pq avail_out == 0 ,
or after each call to
.Fn deflate .
If
.Fn deflate
returns
.Dv Z_OK
and with zero
.Fa avail_out ,
it must be called again after making room in the
output buffer because there might be more output pending.
.Pp
If the parameter
.Fa flush
is set to
.Dv Z_SYNC_FLUSH ,
all pending output is flushed to the output buffer and the output
is aligned on a byte boundary, so that the decompressor can get all
input data available so far.
(In particular,
.Fa avail_in
is zero after the call if enough output space
has been provided before the call.)
Flushing may degrade compression for some compression algorithms
and so it should be used only when necessary.
.Pp
If
.Fa flush
is set to
.Dv Z_FULL_FLUSH ,
all output is flushed as with
.Dv Z_SYNC_FLUSH ,
and the compression state is reset so that decompression can restart from this
point if previous compressed data has been damaged or if random access
is desired.
Using
.Dv Z_FULL_FLUSH
too often can seriously degrade the compression.
.Pp
If
.Fn deflate
returns with avail_out == 0, this function must be called again
with the same value of the flush parameter and more output space
(updated
.Fa avail_out ) ,
until the flush is complete
.Pf ( Fn deflate
returns with non-zero
.Fa avail_out ) .
.Pp
If the parameter
.Fa flush
is set to
.Dv Z_FINISH ,
pending input is processed, pending output is flushed and
.Fn deflate
returns with
.Dv Z_STREAM_END
if there was enough output space; if
.Fn deflate
returns with
.Dv Z_OK ,
this function must be called again with
.Dv Z_FINISH
and more output space
(updated
.Fa avail_out
but no more input data, until it returns with
.Dv Z_STREAM_END
or an error.
After
.Fn deflate
has returned
.Dv Z_STREAM_END ,
the only possible operations on the stream are
.Fn deflateReset
or
.Fn deflateEnd .
.Pp
.Dv Z_FINISH
can be used immediately after
.Fn deflateInit
if all the compression is to be done in a single step.
In this case,
.Fa avail_out
must be at least 0.1% larger than
.Fa avail_in
plus 12 bytes.
If
.Fn deflate
does not return
.Dv Z_STREAM_END ,
then it must be called again as described above.
.Pp
.Fn deflate
sets strm->adler to the adler32 checksum of all input read so far
(that is,
.Fa total_in
bytes).
.Pp
.Fn deflate
may update
.Fa data_type
if it can make a good guess about the input data type
.Pq Z_ASCII or Z_BINARY .
If in doubt, the data is considered binary.
This field is only for information purposes and does not affect
the compression algorithm in any manner.
.Pp
.Fn deflate
returns
.Dv Z_OK
if some progress has been made
.Pq more input processed or more output produced ,
.Dv Z_STREAM_END
if all input has been consumed and all output has been produced
(only when
.Fa flush
is set to
.Dv Z_FINISH ) ,
.Dv Z_STREAM_ERROR
if the stream state was inconsistent
(for example, if
.Fa next_in
or
.Fa next_out
was
.Dv NULL ) ,
.Dv Z_BUF_ERROR
if no progress is possible
(for example,
.Fa avail_in
or
.Fa avail_out
was zero).
.It Xo
.Fa int
.Fn deflateEnd "z_streamp strm" ;
.Xc
.Pp
All dynamically allocated data structures for this stream are freed.
This function discards any unprocessed input and does not flush any
pending output.
.Pp
.Fn deflateEnd
returns
.Dv Z_OK
if successful,
.Dv Z_STREAM_ERROR
if the stream state was inconsistent,
.Dv Z_DATA_ERROR
if the stream was freed prematurely
.Pq some input or output was discarded .
In the error case,
.Fa msg
may be set but then points to a static string
.Pq which must not be deallocated .
.It Xo
.Fa int
.Fn inflateInit "z_streamp strm" ;
.Xc
The
.Fn inflateInit
function initializes the internal stream state for decompression.
The fields
.Fa next_in ,
.Fa avail_in ,
.Fa zalloc ,
.Fa zfree ,
and
.Fa opaque
must be initialized before by the caller.
If
.Fa next_in
is not
.Dv Z_NULL
and
.Fa avail_in
is large enough
.Pq the exact value depends on the compression method ,
.Fn inflateInit
determines the compression method from the
.Nm zlib
header and allocates all data structures accordingly;
otherwise the allocation will be deferred to the first call to
.Fn inflate .
If
.Fa zalloc
and
.Fa zfree
are set to
.Dv Z_NULL ,
.Fn inflateInit
updates them to use default allocation functions.
.Pp
.Fn inflateInit
returns
.Dv Z_OK
if successful,
.Dv Z_MEM_ERROR
if there was not enough memory,
.Dv Z_VERSION_ERROR
if the
.Nm zlib
library version is incompatible with the version assumed by the caller.
.Fa msg
is set to null if there is no error message.
.Fn inflateInit
does not perform any decompression apart from reading the
.Nm zlib
header if present: this will be done by
.Fn inflate .
(So
.Fa next_in
and
.Fa avail_in
may be modified,
but
.Fa next_out
and
.Fa avail_out
are unchanged.)
.It Xo
.Fa int
.Fn inflate "z_streamp strm" "int flush" ;
.Xc
.Fn inflate
decompresses as much data as possible, and stops when the input
buffer becomes empty or the output buffer becomes full.
It may introduce some output latency
.Pq reading input without producing any output
except when forced to flush.
.Pp
The detailed semantics are as follows.
.Fn inflate
performs one or both of the following actions:
.Pp
Decompress more input starting at
.Fa next_in
and update
.Fa next_in
and
.Fa avail_in
accordingly.
If not all input can be processed
(because there is not enough room in the output buffer),
.Fa next_in
is updated and processing will resume at this point for the next call to
.Fn inflate .
.Pp
Provide more output starting at
.Fa next_out
and update
.Fa next_out
and
.Fa avail_out
accordingly.
.Fn inflate
provides as much output as possible,
until there is no more input data or no more space in the output buffer
.Pq see below about the flush parameter .
.Pp
Before the call to
.Fn inflate ,
the application should ensure that at least one of the actions is possible,
by providing more input and/or consuming more output,
and updating the next_* and avail_* values accordingly.
The application can consume the uncompressed output when it wants,
for example when the output buffer is full (avail_out == 0),
or after each call to
.Fn inflate .
If
.Fn inflate
returns
.Dv Z_OK
and with zero
.Fa avail_out ,
it must be called again after making room
in the output buffer because there might be more output pending.
.Pp
If the parameter
.Fa flush
is set to
.Dv Z_SYNC_FLUSH ,
.Fn inflate
flushes as much output as possible to the output buffer.
The flushing behavior of
.Fn inflate
is not specified for values of the flush parameter other than
.Dv Z_SYNC_FLUSH
and
.Dv Z_FINISH ,
but the current implementation actually flushes as much output
as possible anyway.
.Pp
.Fn inflate
should normally be called until it returns
.Dv Z_STREAM_END
or an error.
However if all decompression is to be performed in a single step
.Pq a single call to inflate ,
the parameter
.Fa flush
should be set to
.Dv Z_FINISH .
In this case all pending input is processed and all pending output is flushed;
.Fa avail_out
must be large enough to hold all the uncompressed data.
(The size of the uncompressed data may have been saved
by the compressor for this purpose.)
The next operation on this stream must be
.Fn inflateEnd
to deallocate the decompression state.
The use of
.Dv Z_FINISH
is never required, but can be used to inform
.Fn inflate
that a faster routine may be used for the single
.Fn inflate
call.
.Pp
If a preset dictionary is needed at this point (see
.Fn inflateSetDictionary
below),
.Fn inflate
sets strm->adler to the adler32 checksum of the dictionary
chosen by the compressor and returns
.Dv Z_NEED_DICT ;
otherwise it sets strm->adler to the adler32 checksum of all output produced
so far
(that is,
.Fa total_out
bytes)
and returns
.Dv Z_OK ,
.Dv Z_STREAM_END ,
or an error code as described below.
At the end of the stream,
.Fn inflate
checks that its computed adler32 checksum is equal to that saved by the
compressor and returns
.Dv Z_STREAM_END
only if the checksum is correct.
.Pp
.Fn inflate
returns
.Dv Z_OK
if some progress has been made
.Pq more input processed or more output produced ,
.Dv Z_STREAM_END
if the end of the compressed data has been reached and all uncompressed output
has been produced,
.Dv Z_NEED_DICT
if a preset dictionary is needed at this point,
.Dv Z_DATA_ERROR
if the input data was corrupted (input stream not conforming to the
.Nm zlib
format or incorrect adler32 checksum),
.Dv Z_STREAM_ERROR
if the stream structure was inconsistent
(for example, if
.Fa next_in
or
.Fa next_out
was
.Dv NULL ) ,
.Dv Z_MEM_ERROR
if there was not enough memory,
.Dv Z_BUF_ERROR
if no progress is possible or if there was not enough room in the output buffer
when
.Dv Z_FINISH
is used.
In the
.Dv Z_DATA_ERROR
case, the application may then call
.Fn inflateSync
to look for a good compression block.
.It Xo
.Fa int
.Fn inflateEnd "z_streamp strm" ;
.Xc
All dynamically allocated data structures for this stream are freed.
This function discards any unprocessed input and does not flush any
pending output.
.Pp
.Fn inflateEnd
returns
.Dv Z_OK
if successful, or
.Dv Z_STREAM_ERROR
if the stream state was inconsistent.
In the error case,
.Fa msg
may be set but then points to a static string
.Pq which must not be deallocated .
.El
.Sh ADVANCED FUNCTIONS
The following functions are needed only in some special applications.
.Bl -tag -width Ds
.It Xo
.Fa int
.Fn deflateInit2 "z_streamp strm" "int level" "int method" "int windowBits" "int memLevel" "int strategy" ;
.Xc
.Pp
This is another version of
.Fn deflateInit
with more compression options.
The fields
.Fa next_in ,
.Fa zalloc ,
.Fa zfree ,
and
.Fa opaque
must be initialized before by the caller.
.Pp
The
.Fa method
parameter is the compression method.
It must be
.Dv Z_DEFLATED
in this version of the library.
.Pp
The
.Fa windowBits
parameter is the base two logarithm of the window size
.Pq the size of the history buffer .
It should be in the range 8..15 for this version of the library.
Larger values of this parameter result in better compression
at the expense of memory usage.
The default value is 15 if
.Fn deflateInit
is used instead.
.Pp
The
.Fa memLevel
parameter specifies how much memory should be allocated
for the internal compression state.
memLevel=1 uses minimum memory but is slow and reduces compression ratio;
memLevel=9 uses maximum memory for optimal speed.
The default value is 8.
See
.Aq Pa zconf.h
for total memory usage as a function of
.Fa windowBits
and
.Fa memLevel .
.Pp
The
.Fa strategy
parameter is used to tune the compression algorithm.
Use the value
.Dv Z_DEFAULT_STRATEGY
for normal data;
.Dv Z_FILTERED
for data produced by a filter
.Pq or predictor ;
or
.Dv Z_HUFFMAN_ONLY
to force Huffman encoding only
.Pq no string match .
Filtered data consists mostly of small values with a
somewhat random distribution.
In this case, the compression algorithm is tuned to compress them better.
The effect of
.Dv Z_FILTERED
is to force more Huffman coding and less string matching;
it is somewhat intermediate between
.Dv Z_DEFAULT
and
.Dv Z_HUFFMAN_ONLY .
The
.Fa strategy
parameter only affects the compression ratio but not the correctness of the
compressed output, even if it is not set appropriately.
.Pp
.Fn deflateInit2
returns
.Dv Z_OK
if successful,
.Dv Z_MEM_ERROR
if there was not enough memory,
.Dv Z_STREAM_ERROR
if a parameter is invalid
.Pq such as an invalid method .
.Fa msg
is set to null if there is no error message.
.Fn deflateInit2
does not perform any compression: this will be done by
.Fn deflate .
.It Xo
.Fa int
.Fn deflateSetDictionary "z_streamp strm" "const Bytef *dictionary" "uInt dictLength" ;
.Xc
.Pp
Initializes the compression dictionary from the given byte sequence
without producing any compressed output.
This function must be called immediately after
.Fn deflateInit ,
.Fn deflateInit2 ,
or
.Fn deflateReset ,
before any call to
.Fn deflate .
The compressor and decompressor must use exactly the same dictionary
(see
.Fn inflateSetDictionary ) .
.Pp
The dictionary should consist of strings
.Pq byte sequences
that are likely to be encountered later in the data to be compressed,
with the most commonly used strings preferably put towards
the end of the dictionary.
Using a dictionary is most useful when the data to be compressed is short
and can be predicted with good accuracy;
the data can then be compressed better than with the default empty dictionary.
.Pp
Depending on the size of the compression data structures selected by
.Fn deflateInit
or
.Fn deflateInit2 ,
a part of the dictionary may in effect be discarded,
for example if the dictionary is larger than the window size in
.Fn deflate
or
.Fn deflate2 .
Thus the strings most likely to be useful should be
put at the end of the dictionary, not at the front.
.Pp
Upon return of this function, strm->adler is set to the Adler32 value
of the dictionary; the decompressor may later use this value to determine
which dictionary has been used by the compressor.
(The Adler32 value applies to the whole dictionary even if only a subset
of the dictionary is actually used by the compressor.)
.Pp
.Fn deflateSetDictionary
returns
.Dv Z_OK
if successful,
or
.Dv Z_STREAM_ERROR
if a parameter is invalid
.Pq such as NULL dictionary
or the stream state is inconsistent
(for example if
.Fn deflate
has already been called for this stream or if the compression method is bsort).
.Fn deflateSetDictionary
does not perform any compression: this will be done by
.Fn deflate .
.It Xo
.Fa int
.Fn deflateCopy "z_streamp dest" "z_streamp source" ;
.Xc
.Pp
The
.Fn deflateCopy
function sets the destination stream as a complete copy of the source stream.
.Pp
This function can be useful when several compression strategies will be
tried, for example when there are several ways of pre-processing the input
data with a filter.
The streams that will be discarded should then be freed by calling
.Fn deflateEnd .
Note that
.Fn deflateCopy
duplicates the internal compression state which can be quite large,
so this strategy is slow and can consume lots of memory.
.Pp
.Fn deflateCopy
returns
.Dv Z_OK
if successful,
.Dv Z_MEM_ERROR
if there was not enough memory,
.Dv Z_STREAM_ERROR
if the source stream state was inconsistent
(such as
.Fa zalloc
being NULL).
.Fa msg
is left unchanged in both source and destination.
.It Xo
.Fa int
.Fn deflateReset "z_streamp strm" ;
.Xc
.Pp
This function is equivalent to
.Fn deflateEnd
followed by
.Fn deflateInit ,
but does not free and reallocate all the internal compression state.
The stream will keep the same compression level and any other attributes
that may have been set by
.Fn deflateInit2 .
.Pp
.Fn deflateReset
returns
.Dv Z_OK
if successful, or
.Dv Z_STREAM_ERROR
if the source stream state was inconsistent
(such as
.Fa zalloc
or
.Fa state
being NULL).
.It Xo
.Fa int
.Fn deflateParams "z_streamp strm" "int level" "int strategy" ;
.Xc
.Pp
The
.Fn deflateParams
function dynamically updates the compression level and compression strategy.
The interpretation of level and strategy is as in
.Fn deflateInit2 .
This can be used to switch between compression and straight copy
of the input data, or to switch to a different kind of input data
requiring a different strategy.
If the compression level is changed, the input available so far
is compressed with the old level
.Pq and may be flushed ;
the new level will take effect only at the next call to
.Fn deflate .
.Pp
Before the call to
.Fn deflateParams ,
the stream state must be set as for a call to
.Fn deflate ,
since the currently available input may have to be compressed and flushed.
In particular, strm->avail_out must be non-zero.
.Pp
.Fn deflateParams
returns
.Dv Z_OK
if successful,
.Dv Z_STREAM_ERROR
if the source stream state was inconsistent or if a parameter was invalid, or
.Dv Z_BUF_ERROR
if strm->avail_out was zero.
.It Xo
.Fa int
.Fn inflateInit2 "z_streamp strm" "int windowBits" ;
.Xc
.Pp
This is another version of
.Fn inflateInit
with an extra parameter.
The fields
.Fa next_in ,
.Fa avail_in ,
.Fa zalloc ,
.Fa zfree ,
and
.Fa opaque
must be initialized before by the caller.
.Pp
The
.Fa windowBits
parameter is the base two logarithm of the maximum window size
.Pq the size of the history buffer .
It should be in the range 8..15 for this version of the library.
The default value is 15 if
.Fn inflateInit
is used instead.
If a compressed stream with a larger window size is given as input,
.Fn inflate
will return with the error code
.Dv Z_DATA_ERROR
instead of trying to allocate a larger window.
.Pp
.Fn inflateInit2
returns
.Dv Z_OK
if successful,
.Dv Z_MEM_ERROR
if there was not enough memory,
.Dv Z_STREAM_ERROR
if a parameter is invalid
(such as a negative
.Fa memLevel ) .
.Fa msg
is set to null if there is no error message.
.Fn inflateInit2
does not perform any decompression apart from reading the
.Nm zlib
header if present: this will be done by
.Fn inflate .
(So
.Fa next_in
and
.Fa avail_in
may be modified, but
.Fa next_out
and
.Fa avail_out
are unchanged.)
.It Xo
.Fa int
.Fn inflateSetDictionary "z_streamp strm" "const Bytef *dictionary" "uInt dictLength" ;
.Xc
.Pp
Initializes the decompression dictionary from the given uncompressed byte
sequence.
This function must be called immediately after a call to
.Fn inflate
if this call returned
.Dv Z_NEED_DICT .
The dictionary chosen by the compressor can be determined from the
Adler32 value returned by this call to
.Fn inflate .
The compressor and decompressor must use exactly the same dictionary
(see
.Fn deflateSetDictionary ) .
.Pp
.Fn inflateSetDictionary
returns
.Dv Z_OK
if successful,
.Dv Z_STREAM_ERROR
if a parameter is invalid
.Pq such as NULL dictionary
or the stream state is inconsistent,
.Dv Z_DATA_ERROR
if the given dictionary doesn't match the expected one
.Pq incorrect Adler32 value .
.Fn inflateSetDictionary
does not perform any decompression: this will be done by subsequent calls of
.Fn inflate .
.It Xo
.Fa int
.Fn inflateSync "z_streamp strm" ;
.Xc
.Pp
Skips invalid compressed data until a full flush point
(see above the description of
.Fn deflate
with
.Dv Z_FULL_FLUSH )
can be found, or until all available input is skipped.
No output is provided.
.Pp
.Fn inflateSync
returns
.Dv Z_OK
if a full flush point has been found,
.Dv Z_BUF_ERROR
if no more input was provided,
.Dv Z_DATA_ERROR
if no flush point has been found, or
.Dv Z_STREAM_ERROR
if the stream structure was inconsistent.
In the success case, the application may save the current value of
.Fa total_in
which indicates where valid compressed data was found.
In the error case, the application may repeatedly call
.Fn inflateSync ,
providing more input each time, until success or end of the input data.
.It Xo
.Fa int
.Fn inflateReset "z_streamp strm" ;
.Xc
.Pp
This function is equivalent to
.Fn inflateEnd
followed by
.Fn inflateInit ,
but does not free and reallocate all the internal decompression state.
The stream will keep attributes that may have been set by
.Fn inflateInit2 .
.Pp
.Fn inflateReset
returns
.Dv Z_OK
if successful, or
.Dv Z_STREAM_ERROR
if the source stream state was inconsistent
(such as
.Fa zalloc
or
.Fa state
being NULL).
.El
.Sh UTILITY FUNCTIONS
The following utility functions are implemented on top of the
basic stream-oriented functions.
To simplify the interface,
some default options are assumed (compression level and memory usage,
standard memory allocation functions).
The source code of these utility functions can easily be modified
if you need special options.
.Bl -tag -width Ds
.It Xo
.Fa int
.Fn compress "Bytef *dest" "uLongf *destLen" "const Bytef *source" "uLong sourceLen" ;
.Xc
.Pp
The
.Fn compress
function compresses the source buffer into the destination buffer.
.Fa sourceLen
is the byte length of the source buffer.
Upon entry,
.Fa destLen
is the total size of the destination buffer,
which must be at least 0.1% larger than
.Fa sourceLen
plus 12 bytes.
Upon exit,
.Fa destLen
is the actual size of the compressed buffer.
This function can be used to compress a whole file at once if the
input file is mmap'ed.
.Pp
.Fn compress
returns
.Dv Z_OK
if successful,
.Dv Z_MEM_ERROR
if there was not enough memory, or
.Dv Z_BUF_ERROR
if there was not enough room in the output buffer.
.It Xo
.Fa int
.Fn compress2 "Bytef *dest" "uLongf *destLen" "const Bytef *source" "uLong sourceLen" "int level" ;
.Xc
.Pp
The
.Fn compress2
function compresses the source buffer into the destination buffer.
The
.Fa level
parameter has the same meaning as in
.Fn deflateInit .
.Fa sourceLen
is the byte length of the source buffer.
Upon entry,
.Fa destLen
is the total size of the destination buffer,
which must be at least 0.1% larger than
.Fa sourceLen
plus 12 bytes.
Upon exit,
.Fa destLen
is the actual size of the compressed buffer.
.Pp
.Fn compress2
returns
.Dv Z_OK
if successful,
.Dv Z_MEM_ERROR
if there was not enough memory,
.Dv Z_BUF_ERROR
if there was not enough room in the output buffer, or
.Dv Z_STREAM_ERROR
if the level parameter is invalid.
.It Xo
.Fa int
.Fn uncompress "Bytef *dest" "uLongf *destLen" "const Bytef *source" "uLong sourceLen" ;
.Xc
.Pp
The
.Fn uncompress
function decompresses the source buffer into the destination buffer.
.Fa sourceLen
is the byte length of the source buffer.
Upon entry,
.Fa destLen
is the total size of the destination buffer,
which must be large enough to hold the entire uncompressed data.
(The size of the uncompressed data must have been saved previously
by the compressor and transmitted to the decompressor
by some mechanism outside the scope of this compression library.)
Upon exit,
.Fa destLen
is the actual size of the compressed buffer.
This function can be used to decompress a whole file at once if the
input file is mmap'ed.
.Pp
.Fn uncompress
returns
.Dv Z_OK
if successful,
.Dv Z_MEM_ERROR
if there was not enough memory,
.Dv Z_BUF_ERROR
if there was not enough room in the output buffer, or
.Dv Z_DATA_ERROR
if the input data was corrupted.
.It Xo
.Fa gzFile
.Fn gzopen "const char *path" "const char *mode" ;
.Xc
.Pp
The
.Fn gzopen
function opens a gzip
.Pq .gz
file for reading or writing.
The mode parameter is as in
.Xr fopen 3
.Pq \&"rb\&" or \&"wb\&"
but can also include a compression level
.Pq "wb9"
or a strategy:
.Sq f
for filtered data, as in
.Qq wb6f ;
.Sq h
for Huffman only compression, as in
.Qq wb1h .
(See the description of
.Fn deflateInit2
for more information about the strategy parameter.)
.Pp
.Fn gzopen
can be used to read a file which is not in gzip format;
in this case
.Fn gzread
will directly read from the file without decompression.
.Pp
.Fn gzopen
returns
.Dv NULL
if the file could not be opened or if there was
insufficient memory to allocate the (de)compression state;
errno can be checked to distinguish the two cases (if errno is zero, the
.Nm zlib
error is
.Dv Z_MEM_ERROR ) .
.It Xo
.Fa gzFile
.Fn gzdopen "int fd" "const char *mode" ;
.Xc
.Pp
The
.Fn gzdopen
function associates a gzFile with the file descriptor
.Fa fd .
File descriptors are obtained from calls like
.Xr open 2 ,
.Xr dup 2 ,
.Xr creat 3 ,
.Xr pipe 2 ,
or
.Xr fileno 3
(if the file has been previously opened with
.Xr fopen 3 ) .
The
.Fa mode
parameter is as in
.Fn gzopen .
.Pp
The next call to
.Fn gzclose
on the returned gzFile will also close the file descriptor fd,
just like fclose(fdopen(fd), mode) closes the file descriptor fd.
If you want to keep fd open, use gzdopen(dup(fd), mode).
.Pp
.Fn gzdopen
returns
.Dv NULL
if there was insufficient memory to allocate the (de)compression state.
.It Xo
.Fa int
.Fn gzsetparams "gzFile file" "int level" "int strategy" ;
.Xc
.Pp
The
.Fn gzsetparams
function dynamically updates the compression level or strategy.
See the description of
.Fn deflateInit2
for the meaning of these parameters.
.Pp
.Fn gzsetparams
returns
.Dv Z_OK
if successful, or
.Dv Z_STREAM_ERROR
if the file was not opened for writing.
.It Xo
.Fa int
.Fn gzread "gzFile file" "voidp buf" "unsigned len" ;
.Xc
.Pp
The
.Fn gzread
function reads the given number of uncompressed bytes from the compressed file.
If the input file was not in gzip format,
.Fn gzread
copies the given number of bytes into the buffer.
.Pp
.Fn gzread
returns the number of uncompressed bytes actually read
(0 for end of file, \-1 for error).
.It Xo
.Fa int
.Fn gzwrite "gzFile file" "const voidp buf" "unsigned len" ;
.Xc
.Pp
The
.Fn gzwrite
function writes the given number of uncompressed bytes into the compressed file.
.Fn gzwrite
returns the number of uncompressed bytes actually written
.Pq 0 in case of error .
.It Xo
.Fa int
.Fn gzprintf "gzFile file" "const char *format" "..." ;
.Xc
.Pp
The
.Fn gzprintf
function converts, formats, and writes the args to the compressed file
under control of the format string, as in
.Xr fprintf 3 .
.Fn gzprintf
returns the number of uncompressed bytes actually written
.Pq 0 in case of error .
.It Xo
.Fa int
.Fn gzputs "gzFile file" "const char *s" ;
.Xc
.Pp
The
.Fn gzputs
function writes the given null-terminated string to the compressed file,
excluding the terminating null character.
.Pp
.Fn gzputs
returns the number of characters written, or \-1 in case of error.
.It Xo
.Fa char *
.Fn gzgets "gzFile file" "char *buf" "int len" ;
.Xc
.Pp
The
.Fn gzgets
function reads bytes from the compressed file until len\-1 characters are read,
or a newline character is read and transferred to
.Fa buf ,
or an end-of-file condition is encountered.
The string is then terminated with a null character.
.Pp
.Fn gzgets
returns
.Fa buf ,
or
.Dv Z_NULL
in case of error.
.It Xo
.Fa int
.Fn gzputc "gzFile file" "int c" ;
.Xc
.Pp
The
.Fn gzputc
function writes
.Fa c ,
converted to an unsigned char, into the compressed file.
.Fn gzputc
returns the value that was written, or \-1 in case of error.
.It Xo
.Fa int
.Fn gzgetc "gzFile file" ;
.Xc
.Pp
The
.Fn gzgetc
function reads one byte from the compressed file.
.Fn gzgetc
returns this byte or \-1 in case of end of file or error.
.It Xo
.Fa int
.Fn gzflush "gzFile file" "int flush" ;
.Xc
.Pp
The
.Fn gzflush
function flushes all pending output into the compressed file.
The parameter
.Fa flush
is as in the
.Fn deflate
function.
The return value is the
.Nm zlib
error number (see function
.Fn gzerror
below).
.Fn gzflush
returns
.Dv Z_OK
if the flush parameter is
.Dv Z_FINISH
and all output could be flushed.
.Pp
.Fn gzflush
should be called only when strictly necessary because it can
degrade compression.
.It Xo
.Fa z_off_t
.Fn gzseek "gzFile file" "z_off_t offset" "int whence" ;
.Xc
.Pp
Sets the starting position for the next
.Fn gzread
or
.Fn gzwrite
on the given compressed file.
The offset represents a number of bytes in the uncompressed data stream.
The whence parameter is defined as in
.Xr lseek 2 ;
the value
.Dv SEEK_END
is not supported.
.Pp
If the file is opened for reading, this function is emulated but can be
extremely slow.
If the file is opened for writing, only forward seeks are supported;
.Fn gzseek
then compresses a sequence of zeroes up to the new starting position.
.Pp
.Fn gzseek
returns the resulting offset location as measured in bytes from
the beginning of the uncompressed stream, or \-1 in case of error,
in particular if the file is opened for writing and the new starting position
would be before the current position.
.It Xo
.Fa int
.Fn gzrewind "gzFile file" ;
.Xc
.Pp
The
.Fn gzrewind
function rewinds the given
.Fa file .
This function is supported only for reading.
.Pp
gzrewind(file) is equivalent to (int)gzseek(file, 0L, SEEK_SET).
.It Xo
.Fa z_off_t
.Fn gztell "gzFile file" ;
.Xc
.Pp
The
.Fn gztell
function returns the starting position for the next
.Fn gzread
or
.Fn gzwrite
on the given compressed file.
This position represents a number of bytes in the uncompressed data stream.
.Pp
gztell(file) is equivalent to gzseek(file, 0L, SEEK_CUR).
.It Xo
.Fa int
.Fn gzeof "gzFile file" ;
.Xc
.Pp
The
.Fn gzeof
function returns 1 when
.Dv EOF
has previously been detected reading the given input stream, otherwise zero.
.It Xo
.Fa int
.Fn gzclose "gzFile file" ;
.Xc
.Pp
The
.Fn gzclose
function flushes all pending output if necessary, closes the compressed file
and deallocates all the (de)compression state.
The return value is the
.Nm zlib
error number (see function
.Fn gzerror
below).
.It Xo
.Fa const char *
.Fn gzerror "gzFile file" "int *errnum" ;
.Xc
.Pp
The
.Fn gzerror
function returns the error message for the last error which occurred on the
given compressed
.Fa file .
.Fa errnum
is set to the
.Nm zlib
error number.
If an error occurred in the file system and not in the compression library,
.Fa errnum
is set to
.Dv Z_ERRNO
and the application may consult errno to get the exact error code.
.El
.Sh CHECKSUM FUNCTIONS
These functions are not related to compression but are exported
anyway because they might be useful in applications using the
compression library.
.Bl -tag -width Ds
.It Xo
.Fa uLong
.Fn adler32 "uLong adler" "const Bytef *buf" "uInt len" ;
.Xc
The
.Fn adler32
function updates a running Adler-32 checksum with the bytes buf[0..len-1]
and returns the updated checksum.
If
.Fa buf
is
.Dv NULL ,
this function returns the required initial value for the checksum.
.Pp
An Adler-32 checksum is almost as reliable as a CRC32 but can be computed
much faster.
Usage example:
.Bd -unfilled -offset indent
uLong adler = adler32(0L, Z_NULL, 0);

while (read_buffer(buffer, length) != EOF) {
adler = adler32(adler, buffer, length);
}
if (adler != original_adler) error();
.Ed
.It Xo
.Fa uLong
.Fn crc32 "uLong crc" "const Bytef *buf" "uInt len" ;
.Xc
The
.Fn crc32
function updates a running CRC with the bytes buf[0..len-1]
and returns the updated CRC.
If
.Fa buf
is
.Dv NULL ,
this function returns the required initial value for the CRC.
Pre- and post-conditioning
.Pq one's complement
is performed within this function so it shouldn't be done by the application.
Usage example:
.Bd -unfilled -offset indent
uLong crc = crc32(0L, Z_NULL, 0);

while (read_buffer(buffer, length) != EOF) {
crc = crc32(crc, buffer, length);
}
if (crc != original_crc) error();
.Ed
.El
.Sh STRUCTURES
.Bd -unfilled
struct internal_state;

typedef struct z_stream_s {
    Bytef    *next_in;  /* next input byte */
    uInt     avail_in;  /* number of bytes available at next_in */
    uLong    total_in;  /* total nb of input bytes read so far */

    Bytef    *next_out; /* next output byte should be put there */
    uInt     avail_out; /* remaining free space at next_out */
    uLong    total_out; /* total nb of bytes output so far */

    char     *msg;      /* last error message, NULL if no error */
    struct internal_state FAR *state; /* not visible by applications */

    alloc_func zalloc;  /* used to allocate the internal state */
    free_func  zfree;   /* used to free the internal state */
    voidpf     opaque;  /* private data object passed to zalloc and zfree*/

    int     data_type;  /*best guess about the data type: ascii or binary*/
    uLong   adler;      /* adler32 value of the uncompressed data */
    uLong   reserved;   /* reserved for future use */
} z_stream;

typedef z_stream FAR * z_streamp;
.Ed
.Pp
The application must update
.Fa next_in
and
.Fa avail_in
when
.Fa avail_in
has dropped to zero.
It must update
.Fa next_out
and
.Fa avail_out
when
.Fa avail_out
has dropped to zero.
The application must initialize
.Fa zalloc ,
.Fa zfree ,
and
.Fa opaque
before calling the init function.
All other fields are set by the compression library
and must not be updated by the application.
.Pp
The
.Fa opaque
value provided by the application will be passed as the first
parameter for calls to
.Fn zalloc
and
.Fn zfree .
This can be useful for custom memory management.
The compression library attaches no meaning to the
.Fa opaque
value.
.Pp
.Fa zalloc
must return
.Dv Z_NULL
if there is not enough memory for the object.
If
.Nm zlib
is used in a multi-threaded application,
.Fa zalloc
and
.Fa zfree
must be thread safe.
.Pp
On 16-bit systems, the functions
.Fa zalloc
and
.Fa zfree
must be able to allocate exactly 65536 bytes,
but will not be required to allocate more than this if the symbol MAXSEG_64K
is defined (see
.Aq Pa zconf.h ) .
.Pp
WARNING: On MSDOS, pointers returned by
.Fa zalloc
for objects of exactly 65536 bytes *must* have their offset normalized to zero.
The default allocation function provided by this library ensures this (see
.Pa zutil.c ) .
To reduce memory requirements and avoid any allocation of 64K objects,
at the expense of compression ratio,
compile the library with -DMAX_WBITS=14 (see
.Aq Pa zconf.h ) .
.Pp
The fields
.Fa total_in
and
.Fa total_out
can be used for statistics or progress reports.
After compression,
.Fa total_in
holds the total size of the uncompressed data and may be saved for use
in the decompressor
(particularly if the decompressor wants to decompress everything
in a single step).
.Sh CONSTANTS
.Bd -unfilled
#define Z_NO_FLUSH      0
#define Z_PARTIAL_FLUSH 1 /* will be removed, use Z_SYNC_FLUSH instead */
#define Z_SYNC_FLUSH    2
#define Z_FULL_FLUSH    3
#define Z_FINISH        4
/* Allowed flush values; see deflate() below for details */

#define Z_OK            0
#define Z_STREAM_END    1
#define Z_NEED_DICT     2
#define Z_ERRNO        (-1)
#define Z_STREAM_ERROR (-2)
#define Z_DATA_ERROR   (-3)
#define Z_MEM_ERROR    (-4)
#define Z_BUF_ERROR    (-5)
#define Z_VERSION_ERROR (-6)
/* Return codes for the compression/decompression functions.
 * Negative values are errors,
 * positive values are used for special but normal events.
 */

#define Z_NO_COMPRESSION         0
#define Z_BEST_SPEED             1
#define Z_BEST_COMPRESSION       9
#define Z_DEFAULT_COMPRESSION  (-1)
/* compression levels */

#define Z_FILTERED            1
#define Z_HUFFMAN_ONLY        2
#define Z_DEFAULT_STRATEGY    0
/* compression strategy; see deflateInit2() below for details */

#define Z_BINARY   0
#define Z_ASCII    1
#define Z_UNKNOWN  2
/* Possible values of the data_type field */

#define Z_DEFLATED   8
/* The deflate compression method
 * (the only one supported in this version)
*/

#define Z_NULL  0  /* for initializing zalloc, zfree, opaque */

#define zlib_version zlibVersion()
/* for compatibility with versions < 1.0.2 */
.Ed
.Sh VARIOUS HACKS
deflateInit and inflateInit are macros to allow checking the
.Nm zlib
version and the compiler's view of
.Fa z_stream .
.Bl -tag -width Ds
.It Xo
.Fa int
.Fn deflateInit_ "z_stream strm" "int level" "const char *version" "int stream_size" ;
.Xc
.It Xo
.Fa int
.Fn inflateInit_ "z_stream strm" "const char *version" "int stream_size" ;
.Xc
.It Xo
.Fa int
.Fo deflateInit2_
.Fa "z_stream strm"
.Fa "int level"
.Fa "int method"
.Fa "int windowBits"
.Fa "int memLevel"
.Fa "int strategy"
.Fa "const char *version"
.Fa "int stream_size"
.Fc
.Xc
.It Xo
.Fa int
.Fn inflateInit2_ "z_stream strm" "int windowBits" "const char *version" "int stream_size" ;
.Xc
.It Xo
.Fa const char *
.Fn zError "int err" ;
.Xc
.It Xo
.Fa int
.Fn inflateSyncPoint "z_streamp z" ;
.Xc
.It Xo
.Fa const uLongf *
.Fn "get_crc_table" "void" ;
.Xc
.El
.Sh SEE ALSO
.Xr compress 1
.Pp
.Bl -tag -width 12n -compact
.It RFC 1950
ZLIB Compressed Data Format Specification.
.It RFC 1951
DEFLATE Compressed Data Format Specification.
.It RFC 1952
GZIP File Format Specification.
.El
.Pp
.Pa http://www.gzip.org/zlib/
.Sh HISTORY
This manual page is based on an HTML version of
.Aq Pa zlib.h
converted by
.An piaip Aq piaip@csie.ntu.edu.tw
and was converted to mdoc format by the OpenBSD project.
.Sh AUTHORS
.An Jean-loup Gailly Aq jloup@gzip.org
.An Mark Adler Aq madler@alumni.caltech.edu