diff options
| -rw-r--r-- | ChangeLog | 37 | ||||
| -rw-r--r-- | Makefile.in | 11 | ||||
| -rw-r--r-- | NEWS | 4 | ||||
| -rw-r--r-- | README | 32 | ||||
| -rw-r--r-- | bbexample.c | 22 | ||||
| -rw-r--r-- | carg_parser.c | 42 | ||||
| -rw-r--r-- | carg_parser.h | 34 | ||||
| -rw-r--r-- | cbuffer.c | 28 | ||||
| -rwxr-xr-x | configure | 2 | ||||
| -rw-r--r-- | decoder.c | 6 | ||||
| -rw-r--r-- | decoder.h | 96 | ||||
| -rw-r--r-- | doc/lzlib.info | 291 | ||||
| -rw-r--r-- | doc/lzlib.texi | 225 | ||||
| -rw-r--r-- | doc/minilzip.1 | 31 | ||||
| -rw-r--r-- | encoder.c | 30 | ||||
| -rw-r--r-- | encoder.h | 80 | ||||
| -rw-r--r-- | encoder_base.c | 16 | ||||
| -rw-r--r-- | encoder_base.h | 112 | ||||
| -rw-r--r-- | fast_encoder.c | 4 | ||||
| -rw-r--r-- | fast_encoder.h | 16 | ||||
| -rw-r--r-- | ffexample.c | 22 | ||||
| -rw-r--r-- | lzcheck.c | 24 | ||||
| -rw-r--r-- | lzip.h | 6 | ||||
| -rw-r--r-- | lzlib.c | 110 | ||||
| -rw-r--r-- | lzlib.h | 85 | ||||
| -rw-r--r-- | minilzip.c | 109 | ||||
| -rwxr-xr-x | testsuite/check.sh | 40 |
27 files changed, 724 insertions, 791 deletions
@@ -1,9 +1,18 @@ +2024-11-19 Antonio Diaz Diaz <antonio@gnu.org> + + * Version 1.15-rc1 released. + * lzlib.h: Declare LZ_Errno, LZ_Encoder, and LZ_Decoder as typedef. + * configure, Makefile.in: Use '--soname' conditionally. + * check.sh: Use 'cp' instead of 'cat'. + 2024-10-16 Antonio Diaz Diaz <antonio@gnu.org> * Version 1.15-pre2 released. * decoder.h (Rd_try_reload): Reject a nonzero first LZMA byte. * minilzip.c (do_decompress): Reject empty member in multimember. + (Pp_free): New function. * configure, Makefile.in: Don't use '--soname'; create 'liblz.so.1'. + (Reported by Michael Sullivan). 2024-04-19 Antonio Diaz Diaz <antonio@gnu.org> @@ -49,8 +58,8 @@ * encoder_base.c (LZeb_try_sync_flush): Compensate for the increase in 'rd_min_available_bytes'. * main.c (do_decompress): Fix false report about library stall. - * main.c: New option '--check-lib'. - * main.c (main): Report an error if a file name is empty. + New option '--check-lib'. + (main): Report an error if a file name is empty. Make '-o' behave like '-c', but writing to file instead of stdout. Make '-c' and '-o' check whether the output is a terminal only once. Do not open output if input is a terminal. @@ -99,11 +108,11 @@ (Older versions can adjust dictionary size only once). * lzlib.c (LZ_decompress_read): Detect corrupt header with HD=3. * main.c: New option '--loose-trailing'. - * main.c (main): Option '-S, --volume-size' now keeps input files. - * main.c: Replace 'bits/byte' with inverse compression ratio. - * main.c: Show final diagnostic when testing multiple files. - * main.c: Do not add a second .lz extension to the arg of -o. - * main.c: Show dictionary size at verbosity level 4 (-vvvv). + (main): Make option '-S, --volume-size' keep input files. + Replace 'bits/byte' with inverse compression ratio in output. + (main): Show final diagnostic when testing multiple files. + (set_c_outname): Do not add a second '.lz' to the arg of '-o'. + (do_decompress): Show dictionary size at verbosity level 4 (-vvvv). * lzlib.texi: New chapter 'Invoking minilzip'. 2017-04-11 Antonio Diaz Diaz <antonio@gnu.org> @@ -122,12 +131,11 @@ * lzlib.c (LZ_decompress_sync_to_member): Add skipped size to in_size. * decoder.c (LZd_verify_trailer): Remove test of final code. * main.c: New option '-a, --trailing-error'. - * main.c (main): Delete '--output' file if infd is a terminal. - * main.c (main): Don't use stdin more than once. + (main): Delete '--output' file if infd is a terminal. + (main): Don't use stdin more than once. * configure: Avoid warning on some shells when testing for gcc. * Makefile.in: Detect the existence of install-info. - * check.sh: A POSIX shell is required to run the tests. - * check.sh: Don't check error messages. + * check.sh: Require a POSIX shell. Don't check error messages. 2015-07-08 Antonio Diaz Diaz <antonio@gnu.org> @@ -166,8 +174,8 @@ * lzlib.c (LZ_decompress_read): Tell LZ_header_error from LZ_unexpected_eof the same way as lzip does. * Makefile.in: New targets 'install-as-lzip' and 'install-bin'. - * main.c: Use 'setmode' instead of '_setmode' on Windows and OS/2. * main.c: Define 'strtoull' to 'strtoul' on Windows. + (main): Use 'setmode' instead of '_setmode' on Windows and OS/2. 2012-02-29 Antonio Diaz Diaz <ant_diaz@teleline.es> @@ -195,9 +203,8 @@ * All declarations not belonging to the API have been encapsulated in the namespace 'Lzlib'. * testsuite: Rename 'test1' to 'test.txt'. New tests. - * Match length limits set by options -1 to -9 of minilzip have - been changed to match those of lzip 1.11. - * main.cc: Set stdin/stdout in binary mode on OS2. + * main.cc (main): Set match length limits to same values as lzip 1.11. + (main): Set stdin/stdout in binary mode on OS2. * bbexample.cc: New file containing example functions for buffer-to-buffer compression/decompression. diff --git a/Makefile.in b/Makefile.in index 6d9035c..4f99874 100644 --- a/Makefile.in +++ b/Makefile.in @@ -2,8 +2,9 @@ DISTNAME = $(pkgname)-$(pkgversion) INSTALL = install INSTALL_PROGRAM = $(INSTALL) -m 755 -INSTALL_DATA = $(INSTALL) -m 644 INSTALL_DIR = $(INSTALL) -d -m 755 +INSTALL_DATA = $(INSTALL) -m 644 +INSTALL_SO = $(INSTALL) -m 644 LDCONFIG = /sbin/ldconfig SHELL = /bin/sh CAN_RUN_INSTALLINFO = $(SHELL) -c "install-info --version" > /dev/null 2>&1 @@ -28,6 +29,7 @@ lib$(libname).a : lzlib.o $(AR) $(ARFLAGS) $@ $< lib$(libname).so.$(soversion) : lzlib_sh.o + $(CC) $(CFLAGS) $(LDFLAGS) -fpic -fPIC -shared -Wl,--soname=$@ -o $@ $< || \ $(CC) $(CFLAGS) $(LDFLAGS) -fpic -fPIC -shared -o $@ $< bin : $(progname_static) $(progname_shared) @@ -119,14 +121,15 @@ install-lib : lib fi ; \ rm -f "$(DESTDIR)$(libdir)/lib$(libname).so" ; \ rm -f "$(DESTDIR)$(libdir)/lib$(libname).so.$(soversion)" ; \ - $(INSTALL_PROGRAM) ./lib$(libname).so.$(soversion) "$(DESTDIR)$(libdir)/lib$(libname).so.$(soversion)" ; \ - cd "$(DESTDIR)$(libdir)" && ln -s lib$(libname).so.$(soversion) lib$(libname).so ; \ + $(INSTALL_SO) ./lib$(libname).so.$(soversion) "$(DESTDIR)$(libdir)/lib$(libname).so.$(pkgversion)" ; \ + cd "$(DESTDIR)$(libdir)" && ln -s lib$(libname).so.$(pkgversion) lib$(libname).so ; \ + cd "$(DESTDIR)$(libdir)" && ln -s lib$(libname).so.$(pkgversion) lib$(libname).so.$(soversion) ; \ if [ "${disable_ldconfig}" != yes ] && [ $${run_ldconfig} = yes ] && \ [ -x "$(LDCONFIG)" ] ; then "$(LDCONFIG)" -n "$(DESTDIR)$(libdir)" || true ; fi ; \ fi install-lib-strip : lib - $(MAKE) INSTALL_PROGRAM='$(INSTALL_PROGRAM) -s' install-lib + $(MAKE) INSTALL_SO='$(INSTALL_SO) -s' install-lib install-info : if [ ! -d "$(DESTDIR)$(infodir)" ] ; then $(INSTALL_DIR) "$(DESTDIR)$(infodir)" ; fi @@ -15,5 +15,5 @@ minilzip is no longer built by default. The use of the target 'bin' has been documented in INSTALL. -To improve portability, the linker option '--soname' is no longer used. -Instead the shared library is written to 'liblz.so.1'. +To improve portability, the linker option '--soname' is now used conditionally. +(Reported by Michael Sullivan). @@ -5,30 +5,6 @@ decompression functions, including integrity checking of the decompressed data. The compressed data format used by the library is the lzip format. Lzlib is written in C and is distributed under a 2-clause BSD license. -The lzip file format is designed for data sharing and long-term archiving, -taking into account both data integrity and decoder availability: - - * The lzip format provides very safe integrity checking and some data - recovery means. The program lziprecover can repair bit flip errors - (one of the most common forms of data corruption) in lzip files, and - provides data recovery capabilities, including error-checked merging - of damaged copies of a file. - - * The lzip format is as simple as possible (but not simpler). The lzip - manual provides the source code of a simple decompressor along with a - detailed explanation of how it works, so that with the only help of the - lzip manual it would be possible for a digital archaeologist to extract - the data from a lzip file long after quantum computers eventually - render LZMA obsolete. - - * Additionally the lzip reference implementation is copylefted, which - guarantees that it will remain free forever. - -A nice feature of the lzip format is that a corrupt byte is easier to repair -the nearer it is from the beginning of the file. Therefore, with the help of -lziprecover, losing an entire archive just because of a corrupt byte near -the beginning is a thing of the past. - The functions and variables forming the interface of the compression library are declared in the file 'lzlib.h'. Usage examples of the library are given in the files 'bbexample.c', 'ffexample.c', and 'minilzip.c' from the source @@ -74,10 +50,10 @@ In spite of its name (Lempel-Ziv-Markov chain-Algorithm), LZMA is not a concrete algorithm; it is more like "any algorithm using the LZMA coding scheme". For example, the option '-0' of lzip uses the scheme in almost the simplest way possible; issuing the longest match it can find, or a literal -byte if it can't find a match. Inversely, a much more elaborated way of -finding coding sequences of minimum size than the one currently used by lzip -could be developed, and the resulting sequence could also be coded using the -LZMA coding scheme. +byte if it can't find a match. Inversely, a more elaborate way of finding +coding sequences of minimum size than the one currently used by lzip could +be developed, and the resulting sequence could also be coded using the LZMA +coding scheme. Lzlib currently implements two variants of the LZMA algorithm: fast (used by option '-0' of minilzip) and normal (used by all other compression levels). diff --git a/bbexample.c b/bbexample.c index 50ccf33..d7c324b 100644 --- a/bbexample.c +++ b/bbexample.c @@ -53,7 +53,7 @@ uint8_t * read_file( const char * const name, long * const file_sizep ) fprintf( stderr, "bbexample: %s: Input file is too large.\n", name ); free( buffer ); fclose( f ); return 0; } - buffer_size = ( buffer_size <= LONG_MAX / 2 ) ? 2 * buffer_size : LONG_MAX; + buffer_size = (buffer_size <= LONG_MAX / 2) ? 2 * buffer_size : LONG_MAX; tmp = (uint8_t *)realloc( buffer, buffer_size ); if( !tmp ) { fputs( "bbexample: read_file: Not enough memory.\n", stderr ); @@ -81,14 +81,14 @@ uint8_t * read_file( const char * const name, long * const file_sizep ) uint8_t * bbcompressl( const uint8_t * const inbuf, const long insize, const int level, long * const outlenp ) { - struct Lzma_options + typedef struct Lzma_options { int dictionary_size; /* 4 KiB .. 512 MiB */ int match_len_limit; /* 5 .. 273 */ - }; + } Lzma_options; /* Mapping from gzip/bzip2 style 0..9 compression levels to the corresponding LZMA compression parameters. */ - const struct Lzma_options option_mapping[] = + const Lzma_options option_mapping[] = { { 65535, 16 }, /* -0 (65535,16 chooses fast encoder) */ { 1 << 20, 5 }, /* -1 */ @@ -100,10 +100,10 @@ uint8_t * bbcompressl( const uint8_t * const inbuf, const long insize, { 1 << 24, 68 }, /* -7 */ { 3 << 23, 132 }, /* -8 */ { 1 << 25, 273 } }; /* -9 */ - struct Lzma_options encoder_options; - struct LZ_Encoder * encoder; + Lzma_options encoder_options; + LZ_Encoder * encoder; uint8_t * outbuf; - const long delta_size = ( insize / 4 ) + 64; /* insize may be zero */ + const long delta_size = insize / 4 + 64; /* insize may be zero */ long outsize = delta_size; /* initial outsize */ long inpos = 0; long outpos = 0; @@ -160,7 +160,7 @@ uint8_t * bbcompressl( const uint8_t * const inbuf, const long insize, uint8_t * bbdecompressl( const uint8_t * const inbuf, const long insize, long * const outlenp ) { - struct LZ_Decoder * const decoder = LZ_decompress_open(); + LZ_Decoder * const decoder = LZ_decompress_open(); const long delta_size = insize; /* insize must be > zero */ long outsize = delta_size; /* initial outsize */ uint8_t * outbuf = (uint8_t *)malloc( outsize ); @@ -242,7 +242,7 @@ bool bbcompress( const uint8_t * const inbuf, const int insize, { int inpos = 0, outpos = 0; bool error = false; - struct LZ_Encoder * const encoder = + LZ_Encoder * const encoder = LZ_compress_open( dictionary_size, match_len_limit, INT64_MAX ); if( !encoder || LZ_compress_errno( encoder ) != LZ_ok ) { LZ_compress_close( encoder ); return false; } @@ -278,7 +278,7 @@ bool bbdecompress( const uint8_t * const inbuf, const int insize, { int inpos = 0, outpos = 0; bool error = false; - struct LZ_Decoder * const decoder = LZ_decompress_open(); + LZ_Decoder * const decoder = LZ_decompress_open(); if( !decoder || LZ_decompress_errno( decoder ) != LZ_ok ) { LZ_decompress_close( decoder ); return false; } @@ -340,7 +340,7 @@ int main( const int argc, const char * const argv[] ) { int retval = 0, i; int open_failures = 0; - const bool verbose = ( argc > 2 ); + const bool verbose = argc > 2; if( argc < 2 ) { diff --git a/carg_parser.c b/carg_parser.c index e16fa73..9400342 100644 --- a/carg_parser.c +++ b/carg_parser.c @@ -32,15 +32,15 @@ static void * ap_resize_buffer( void * buf, const int min_size ) } -static char push_back_record( struct Arg_parser * const ap, const int code, +static char push_back_record( Arg_parser * const ap, const int code, const char * const long_name, const char * const argument ) { - struct ap_Record * p; + ap_Record * p; void * tmp = ap_resize_buffer( ap->data, - ( ap->data_size + 1 ) * sizeof (struct ap_Record) ); + ( ap->data_size + 1 ) * sizeof (ap_Record) ); if( !tmp ) return 0; - ap->data = (struct ap_Record *)tmp; + ap->data = (ap_Record *)tmp; p = &(ap->data[ap->data_size]); p->code = code; if( long_name ) @@ -71,7 +71,7 @@ static char push_back_record( struct Arg_parser * const ap, const int code, } -static char add_error( struct Arg_parser * const ap, const char * const msg ) +static char add_error( Arg_parser * const ap, const char * const msg ) { const int len = strlen( msg ); void * tmp = ap_resize_buffer( ap->error, ap->error_size + len + 1 ); @@ -83,7 +83,7 @@ static char add_error( struct Arg_parser * const ap, const char * const msg ) } -static void free_data( struct Arg_parser * const ap ) +static void free_data( Arg_parser * const ap ) { int i; for( i = 0; i < ap->data_size; ++i ) @@ -94,10 +94,9 @@ static void free_data( struct Arg_parser * const ap ) /* Return 0 only if out of memory. */ -static char parse_long_option( struct Arg_parser * const ap, +static char parse_long_option( Arg_parser * const ap, const char * const opt, const char * const arg, - const struct ap_Option options[], - int * const argindp ) + const ap_Option options[], int * const argindp ) { unsigned len; int index = -1, i; @@ -171,10 +170,9 @@ static char parse_long_option( struct Arg_parser * const ap, /* Return 0 only if out of memory. */ -static char parse_short_option( struct Arg_parser * const ap, +static char parse_short_option( Arg_parser * const ap, const char * const opt, const char * const arg, - const struct ap_Option options[], - int * const argindp ) + const ap_Option options[], int * const argindp ) { int cind = 1; /* character index in opt */ @@ -221,9 +219,9 @@ static char parse_short_option( struct Arg_parser * const ap, } -char ap_init( struct Arg_parser * const ap, +char ap_init( Arg_parser * const ap, const int argc, const char * const argv[], - const struct ap_Option options[], const char in_order ) + const ap_Option options[], const char in_order ) { const char ** non_options = 0; /* skipped non-options */ int non_options_size = 0; /* number of skipped non-options */ @@ -282,7 +280,7 @@ out: if( non_options ) free( non_options ); } -void ap_free( struct Arg_parser * const ap ) +void ap_free( Arg_parser * const ap ) { free_data( ap ); if( ap->error ) { free( ap->error ); ap->error = 0; } @@ -290,29 +288,25 @@ void ap_free( struct Arg_parser * const ap ) } -const char * ap_error( const struct Arg_parser * const ap ) - { return ap->error; } +const char * ap_error( const Arg_parser * const ap ) { return ap->error; } +int ap_arguments( const Arg_parser * const ap ) { return ap->data_size; } -int ap_arguments( const struct Arg_parser * const ap ) - { return ap->data_size; } - - -int ap_code( const struct Arg_parser * const ap, const int i ) +int ap_code( const Arg_parser * const ap, const int i ) { if( i < 0 || i >= ap_arguments( ap ) ) return 0; return ap->data[i].code; } -const char * ap_parsed_name( const struct Arg_parser * const ap, const int i ) +const char * ap_parsed_name( const Arg_parser * const ap, const int i ) { if( i < 0 || i >= ap_arguments( ap ) || !ap->data[i].parsed_name ) return ""; return ap->data[i].parsed_name; } -const char * ap_argument( const struct Arg_parser * const ap, const int i ) +const char * ap_argument( const Arg_parser * const ap, const int i ) { if( i < 0 || i >= ap_arguments( ap ) || !ap->data[i].argument ) return ""; return ap->data[i].argument; diff --git a/carg_parser.h b/carg_parser.h index 65a6d7d..294756c 100644 --- a/carg_parser.h +++ b/carg_parser.h @@ -50,52 +50,52 @@ extern "C" { #endif /* ap_yme = yes but maybe empty */ -enum ap_Has_arg { ap_no, ap_yes, ap_maybe, ap_yme }; +typedef enum ap_Has_arg { ap_no, ap_yes, ap_maybe, ap_yme } ap_Has_arg; -struct ap_Option +typedef struct ap_Option { int code; /* Short option letter or code ( code != 0 ) */ const char * long_name; /* Long option name (maybe null) */ - enum ap_Has_arg has_arg; - }; + ap_Has_arg has_arg; + } ap_Option; -struct ap_Record +typedef struct ap_Record { int code; char * parsed_name; char * argument; - }; + } ap_Record; -struct Arg_parser +typedef struct Arg_parser { - struct ap_Record * data; + ap_Record * data; char * error; int data_size; int error_size; - }; + } Arg_parser; -char ap_init( struct Arg_parser * const ap, +char ap_init( Arg_parser * const ap, const int argc, const char * const argv[], - const struct ap_Option options[], const char in_order ); + const ap_Option options[], const char in_order ); -void ap_free( struct Arg_parser * const ap ); +void ap_free( Arg_parser * const ap ); -const char * ap_error( const struct Arg_parser * const ap ); +const char * ap_error( const Arg_parser * const ap ); /* The number of arguments parsed. May be different from argc. */ -int ap_arguments( const struct Arg_parser * const ap ); +int ap_arguments( const Arg_parser * const ap ); /* If ap_code( i ) is 0, ap_argument( i ) is a non-option. Else ap_argument( i ) is the option's argument (or empty). */ -int ap_code( const struct Arg_parser * const ap, const int i ); +int ap_code( const Arg_parser * const ap, const int i ); /* Full name of the option parsed (short or long). */ -const char * ap_parsed_name( const struct Arg_parser * const ap, const int i ); +const char * ap_parsed_name( const Arg_parser * const ap, const int i ); -const char * ap_argument( const struct Arg_parser * const ap, const int i ); +const char * ap_argument( const Arg_parser * const ap, const int i ); #ifdef __cplusplus } @@ -17,15 +17,15 @@ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ -struct Circular_buffer +typedef struct Circular_buffer { uint8_t * buffer; unsigned buffer_size; /* capacity == buffer_size - 1 */ unsigned get; /* buffer is empty when get == put */ unsigned put; - }; + } Circular_buffer; -static inline bool Cb_init( struct Circular_buffer * const cb, +static inline bool Cb_init( Circular_buffer * const cb, const unsigned buf_size ) { cb->buffer_size = buf_size + 1; @@ -36,38 +36,36 @@ static inline bool Cb_init( struct Circular_buffer * const cb, return cb->buffer != 0; } -static inline void Cb_free( struct Circular_buffer * const cb ) +static inline void Cb_free( Circular_buffer * const cb ) { free( cb->buffer ); cb->buffer = 0; } -static inline void Cb_reset( struct Circular_buffer * const cb ) +static inline void Cb_reset( Circular_buffer * const cb ) { cb->get = 0; cb->put = 0; } -static inline unsigned Cb_empty( const struct Circular_buffer * const cb ) +static inline unsigned Cb_empty( const Circular_buffer * const cb ) { return cb->get == cb->put; } -static inline unsigned Cb_used_bytes( const struct Circular_buffer * const cb ) +static inline unsigned Cb_used_bytes( const Circular_buffer * const cb ) { return ( (cb->get <= cb->put) ? 0 : cb->buffer_size ) + cb->put - cb->get; } -static inline unsigned Cb_free_bytes( const struct Circular_buffer * const cb ) +static inline unsigned Cb_free_bytes( const Circular_buffer * const cb ) { return ( (cb->get <= cb->put) ? cb->buffer_size : 0 ) - cb->put + cb->get - 1; } -static inline uint8_t Cb_get_byte( struct Circular_buffer * const cb ) +static inline uint8_t Cb_get_byte( Circular_buffer * const cb ) { const uint8_t b = cb->buffer[cb->get]; if( ++cb->get >= cb->buffer_size ) cb->get = 0; return b; } -static inline void Cb_put_byte( struct Circular_buffer * const cb, - const uint8_t b ) +static inline void Cb_put_byte( Circular_buffer * const cb, const uint8_t b ) { cb->buffer[cb->put] = b; if( ++cb->put >= cb->buffer_size ) cb->put = 0; } -static bool Cb_unread_data( struct Circular_buffer * const cb, - const unsigned size ) +static bool Cb_unread_data( Circular_buffer * const cb, const unsigned size ) { if( size > Cb_free_bytes( cb ) ) return false; if( cb->get >= size ) cb->get -= size; @@ -80,7 +78,7 @@ static bool Cb_unread_data( struct Circular_buffer * const cb, If 'out_buffer' is null, the bytes are discarded. Return the number of bytes copied or discarded. */ -static unsigned Cb_read_data( struct Circular_buffer * const cb, +static unsigned Cb_read_data( Circular_buffer * const cb, uint8_t * const out_buffer, const unsigned out_size ) { @@ -113,7 +111,7 @@ static unsigned Cb_read_data( struct Circular_buffer * const cb, /* Copy up to 'in_size' bytes from 'in_buffer' and update 'put'. Return the number of bytes copied. */ -static unsigned Cb_write_data( struct Circular_buffer * const cb, +static unsigned Cb_write_data( Circular_buffer * const cb, const uint8_t * const in_buffer, const unsigned in_size ) { @@ -6,7 +6,7 @@ # to copy, distribute, and modify it. pkgname=lzlib -pkgversion=1.15-pre2 +pkgversion=1.15-rc1 soversion=1 libname=lz libname_static=lib${libname}.a @@ -17,7 +17,7 @@ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ -static int LZd_try_check_trailer( struct LZ_decoder * const d ) +static int LZd_try_check_trailer( LZ_decoder * const d ) { Lzip_trailer trailer; if( Rd_available_bytes( d->rdec ) < Lt_size ) @@ -36,9 +36,9 @@ static int LZd_try_check_trailer( struct LZ_decoder * const d ) /* Return value: 0 = OK, 1 = decoder error, 2 = unexpected EOF, 3 = trailer error, 4 = unknown marker found, 5 = nonzero first LZMA byte found, 6 = library error. */ -static int LZd_decode_member( struct LZ_decoder * const d ) +static int LZd_decode_member( LZ_decoder * const d ) { - struct Range_decoder * const rdec = d->rdec; + Range_decoder * const rdec = d->rdec; State * const state = &d->state; unsigned old_mpos = rdec->member_position; @@ -19,17 +19,17 @@ enum { rd_min_available_bytes = 10 }; -struct Range_decoder +typedef struct Range_decoder { - struct Circular_buffer cb; /* input buffer */ + Circular_buffer cb; /* input buffer */ unsigned long long member_position; uint32_t code; uint32_t range; bool at_stream_end; bool reload_pending; - }; + } Range_decoder; -static inline bool Rd_init( struct Range_decoder * const rdec ) +static inline bool Rd_init( Range_decoder * const rdec ) { if( !Cb_init( &rdec->cb, 65536 + rd_min_available_bytes ) ) return false; rdec->member_position = 0; @@ -40,25 +40,25 @@ static inline bool Rd_init( struct Range_decoder * const rdec ) return true; } -static inline void Rd_free( struct Range_decoder * const rdec ) +static inline void Rd_free( Range_decoder * const rdec ) { Cb_free( &rdec->cb ); } -static inline bool Rd_finished( const struct Range_decoder * const rdec ) +static inline bool Rd_finished( const Range_decoder * const rdec ) { return rdec->at_stream_end && Cb_empty( &rdec->cb ); } -static inline void Rd_finish( struct Range_decoder * const rdec ) +static inline void Rd_finish( Range_decoder * const rdec ) { rdec->at_stream_end = true; } -static inline bool Rd_enough_available_bytes( const struct Range_decoder * const rdec ) +static inline bool Rd_enough_available_bytes( const Range_decoder * const rdec ) { return Cb_used_bytes( &rdec->cb ) >= rd_min_available_bytes; } -static inline unsigned Rd_available_bytes( const struct Range_decoder * const rdec ) +static inline unsigned Rd_available_bytes( const Range_decoder * const rdec ) { return Cb_used_bytes( &rdec->cb ); } -static inline unsigned Rd_free_bytes( const struct Range_decoder * const rdec ) +static inline unsigned Rd_free_bytes( const Range_decoder * const rdec ) { return rdec->at_stream_end ? 0 : Cb_free_bytes( &rdec->cb ); } -static inline unsigned long long Rd_purge( struct Range_decoder * const rdec ) +static inline unsigned long long Rd_purge( Range_decoder * const rdec ) { const unsigned long long size = rdec->member_position + Cb_used_bytes( &rdec->cb ); @@ -67,7 +67,7 @@ static inline unsigned long long Rd_purge( struct Range_decoder * const rdec ) return size; } -static inline void Rd_reset( struct Range_decoder * const rdec ) +static inline void Rd_reset( Range_decoder * const rdec ) { Cb_reset( &rdec->cb ); rdec->member_position = 0; rdec->at_stream_end = false; } @@ -75,7 +75,7 @@ static inline void Rd_reset( struct Range_decoder * const rdec ) /* Seek for a member header and update 'get'. Set '*skippedp' to the number of bytes skipped. Return true if a valid header is found. */ -static bool Rd_find_header( struct Range_decoder * const rdec, +static bool Rd_find_header( Range_decoder * const rdec, unsigned * const skippedp ) { *skippedp = 0; @@ -101,14 +101,14 @@ static bool Rd_find_header( struct Range_decoder * const rdec, } -static inline int Rd_write_data( struct Range_decoder * const rdec, +static inline int Rd_write_data( Range_decoder * const rdec, const uint8_t * const inbuf, const int size ) { if( rdec->at_stream_end || size <= 0 ) return 0; return Cb_write_data( &rdec->cb, inbuf, size ); } -static inline uint8_t Rd_get_byte( struct Range_decoder * const rdec ) +static inline uint8_t Rd_get_byte( Range_decoder * const rdec ) { /* 0xFF avoids decoder error if member is truncated at EOS marker */ if( Rd_finished( rdec ) ) return 0xFF; @@ -116,7 +116,7 @@ static inline uint8_t Rd_get_byte( struct Range_decoder * const rdec ) return Cb_get_byte( &rdec->cb ); } -static inline int Rd_read_data( struct Range_decoder * const rdec, +static inline int Rd_read_data( Range_decoder * const rdec, uint8_t * const outbuf, const int size ) { const int sz = Cb_read_data( &rdec->cb, outbuf, size ); @@ -124,7 +124,7 @@ static inline int Rd_read_data( struct Range_decoder * const rdec, return sz; } -static inline bool Rd_unread_data( struct Range_decoder * const rdec, +static inline bool Rd_unread_data( Range_decoder * const rdec, const unsigned size ) { if( size > rdec->member_position || !Cb_unread_data( &rdec->cb, size ) ) @@ -133,7 +133,7 @@ static inline bool Rd_unread_data( struct Range_decoder * const rdec, return true; } -static int Rd_try_reload( struct Range_decoder * const rdec ) +static int Rd_try_reload( Range_decoder * const rdec ) { if( rdec->reload_pending && Rd_available_bytes( rdec ) >= 5 ) { @@ -149,13 +149,13 @@ static int Rd_try_reload( struct Range_decoder * const rdec ) return !rdec->reload_pending; } -static inline void Rd_normalize( struct Range_decoder * const rdec ) +static inline void Rd_normalize( Range_decoder * const rdec ) { if( rdec->range <= 0x00FFFFFFU ) { rdec->range <<= 8; rdec->code = (rdec->code << 8) | Rd_get_byte( rdec ); } } -static inline unsigned Rd_decode( struct Range_decoder * const rdec, +static inline unsigned Rd_decode( Range_decoder * const rdec, const int num_bits ) { unsigned symbol = 0; @@ -173,7 +173,7 @@ static inline unsigned Rd_decode( struct Range_decoder * const rdec, return symbol; } -static inline unsigned Rd_decode_bit( struct Range_decoder * const rdec, +static inline unsigned Rd_decode_bit( Range_decoder * const rdec, Bit_model * const probability ) { Rd_normalize( rdec ); @@ -193,7 +193,7 @@ static inline unsigned Rd_decode_bit( struct Range_decoder * const rdec, } } -static inline void Rd_decode_symbol_bit( struct Range_decoder * const rdec, +static inline void Rd_decode_symbol_bit( Range_decoder * const rdec, Bit_model * const probability, unsigned * symbol ) { Rd_normalize( rdec ); @@ -213,7 +213,7 @@ static inline void Rd_decode_symbol_bit( struct Range_decoder * const rdec, } } -static inline void Rd_decode_symbol_bit_reversed( struct Range_decoder * const rdec, +static inline void Rd_decode_symbol_bit_reversed( Range_decoder * const rdec, Bit_model * const probability, unsigned * model, unsigned * symbol, const int i ) { @@ -235,7 +235,7 @@ static inline void Rd_decode_symbol_bit_reversed( struct Range_decoder * const r } } -static inline unsigned Rd_decode_tree6( struct Range_decoder * const rdec, +static inline unsigned Rd_decode_tree6( Range_decoder * const rdec, Bit_model bm[] ) { unsigned symbol = 1; @@ -248,7 +248,7 @@ static inline unsigned Rd_decode_tree6( struct Range_decoder * const rdec, return symbol & 0x3F; } -static inline unsigned Rd_decode_tree8( struct Range_decoder * const rdec, +static inline unsigned Rd_decode_tree8( Range_decoder * const rdec, Bit_model bm[] ) { unsigned symbol = 1; @@ -264,7 +264,7 @@ static inline unsigned Rd_decode_tree8( struct Range_decoder * const rdec, } static inline unsigned -Rd_decode_tree_reversed( struct Range_decoder * const rdec, +Rd_decode_tree_reversed( Range_decoder * const rdec, Bit_model bm[], const int num_bits ) { unsigned model = 1; @@ -276,7 +276,7 @@ Rd_decode_tree_reversed( struct Range_decoder * const rdec, } static inline unsigned -Rd_decode_tree_reversed4( struct Range_decoder * const rdec, Bit_model bm[] ) +Rd_decode_tree_reversed4( Range_decoder * const rdec, Bit_model bm[] ) { unsigned model = 1; unsigned symbol = 0; @@ -287,7 +287,7 @@ Rd_decode_tree_reversed4( struct Range_decoder * const rdec, Bit_model bm[] ) return symbol; } -static inline unsigned Rd_decode_matched( struct Range_decoder * const rdec, +static inline unsigned Rd_decode_matched( Range_decoder * const rdec, Bit_model bm[], unsigned match_byte ) { unsigned symbol = 1; @@ -302,8 +302,8 @@ static inline unsigned Rd_decode_matched( struct Range_decoder * const rdec, } } -static inline unsigned Rd_decode_len( struct Range_decoder * const rdec, - struct Len_model * const lm, +static inline unsigned Rd_decode_len( Range_decoder * const rdec, + Len_model * const lm, const int pos_state ) { Bit_model * bm; @@ -329,11 +329,11 @@ len3: enum { lzd_min_free_bytes = max_match_len }; -struct LZ_decoder +typedef struct LZ_decoder { - struct Circular_buffer cb; + Circular_buffer cb; unsigned long long partial_data_pos; - struct Range_decoder * rdec; + Range_decoder * rdec; unsigned dictionary_size; uint32_t crc; bool check_trailer_pending; @@ -356,25 +356,25 @@ struct LZ_decoder Bit_model bm_dis[modeled_distances-end_dis_model+1]; Bit_model bm_align[dis_align_size]; - struct Len_model match_len_model; - struct Len_model rep_len_model; - }; + Len_model match_len_model; + Len_model rep_len_model; + } LZ_decoder; -static inline bool LZd_enough_free_bytes( const struct LZ_decoder * const d ) +static inline bool LZd_enough_free_bytes( const LZ_decoder * const d ) { return Cb_free_bytes( &d->cb ) >= lzd_min_free_bytes; } -static inline uint8_t LZd_peek_prev( const struct LZ_decoder * const d ) +static inline uint8_t LZd_peek_prev( const LZ_decoder * const d ) { return d->cb.buffer[((d->cb.put > 0) ? d->cb.put : d->cb.buffer_size)-1]; } -static inline uint8_t LZd_peek( const struct LZ_decoder * const d, +static inline uint8_t LZd_peek( const LZ_decoder * const d, const unsigned distance ) { - const unsigned i = ( ( d->cb.put > distance ) ? 0 : d->cb.buffer_size ) + + const unsigned i = ( (d->cb.put > distance) ? 0 : d->cb.buffer_size ) + d->cb.put - distance - 1; return d->cb.buffer[i]; } -static inline void LZd_put_byte( struct LZ_decoder * const d, const uint8_t b ) +static inline void LZd_put_byte( LZ_decoder * const d, const uint8_t b ) { CRC32_update_byte( &d->crc, b ); d->cb.buffer[d->cb.put] = b; @@ -382,7 +382,7 @@ static inline void LZd_put_byte( struct LZ_decoder * const d, const uint8_t b ) { d->partial_data_pos += d->cb.put; d->cb.put = 0; d->pos_wrapped = true; } } -static inline void LZd_copy_block( struct LZ_decoder * const d, +static inline void LZd_copy_block( LZ_decoder * const d, const unsigned distance, unsigned len ) { unsigned lpos = d->cb.put, i = lpos - distance - 1; @@ -415,8 +415,7 @@ static inline void LZd_copy_block( struct LZ_decoder * const d, } } -static inline bool LZd_init( struct LZ_decoder * const d, - struct Range_decoder * const rde, +static inline bool LZd_init( LZ_decoder * const d, Range_decoder * const rde, const unsigned dict_size ) { if( !Cb_init( &d->cb, max( 65536, dict_size ) + lzd_min_free_bytes ) ) @@ -451,15 +450,14 @@ static inline bool LZd_init( struct LZ_decoder * const d, return true; } -static inline void LZd_free( struct LZ_decoder * const d ) - { Cb_free( &d->cb ); } +static inline void LZd_free( LZ_decoder * const d ) { Cb_free( &d->cb ); } -static inline bool LZd_member_finished( const struct LZ_decoder * const d ) +static inline bool LZd_member_finished( const LZ_decoder * const d ) { return d->member_finished && Cb_empty( &d->cb ); } -static inline unsigned LZd_crc( const struct LZ_decoder * const d ) +static inline unsigned LZd_crc( const LZ_decoder * const d ) { return d->crc ^ 0xFFFFFFFFU; } static inline unsigned long long -LZd_data_position( const struct LZ_decoder * const d ) +LZd_data_position( const LZ_decoder * const d ) { return d->partial_data_pos + d->cb.put; } diff --git a/doc/lzlib.info b/doc/lzlib.info index 44bf3ce..afe2774 100644 --- a/doc/lzlib.info +++ b/doc/lzlib.info @@ -11,7 +11,7 @@ File: lzlib.info, Node: Top, Next: Introduction, Up: (dir) Lzlib Manual ************ -This manual is for Lzlib (version 1.15-pre2, 16 October 2024). +This manual is for Lzlib (version 1.15-rc1, 19 November 2024). * Menu: @@ -46,31 +46,6 @@ decompression functions, including integrity checking of the decompressed data. The compressed data format used by the library is the lzip format. Lzlib is written in C and is distributed under a 2-clause BSD license. - The lzip file format is designed for data sharing and long-term -archiving, taking into account both data integrity and decoder availability: - - * The lzip format provides very safe integrity checking and some data - recovery means. The program lziprecover can repair bit flip errors - (one of the most common forms of data corruption) in lzip files, and - provides data recovery capabilities, including error-checked merging - of damaged copies of a file. *Note Data safety: (lziprecover)Data - safety. - - * The lzip format is as simple as possible (but not simpler). The lzip - manual provides the source code of a simple decompressor along with a - detailed explanation of how it works, so that with the only help of the - lzip manual it would be possible for a digital archaeologist to extract - the data from a lzip file long after quantum computers eventually - render LZMA obsolete. - - * Additionally the lzip reference implementation is copylefted, which - guarantees that it will remain free forever. - - A nice feature of the lzip format is that a corrupt byte is easier to -repair the nearer it is from the beginning of the file. Therefore, with the -help of lziprecover, losing an entire archive just because of a corrupt -byte near the beginning is a thing of the past. - The functions and variables forming the interface of the compression library are declared in the file 'lzlib.h'. Usage examples of the library are given in the files 'bbexample.c', 'ffexample.c', and 'minilzip.c' from @@ -117,10 +92,10 @@ about 2 PiB each. concrete algorithm; it is more like "any algorithm using the LZMA coding scheme". For example, the option '-0' of lzip uses the scheme in almost the simplest way possible; issuing the longest match it can find, or a literal -byte if it can't find a match. Inversely, a much more elaborated way of -finding coding sequences of minimum size than the one currently used by -lzip could be developed, and the resulting sequence could also be coded -using the LZMA coding scheme. +byte if it can't find a match. Inversely, a more elaborate way of finding +coding sequences of minimum size than the one currently used by lzip could +be developed, and the resulting sequence could also be coded using the LZMA +coding scheme. Lzlib currently implements two variants of the LZMA algorithm: fast (used by option '-0' of minilzip) and normal (used by all other compression @@ -269,9 +244,8 @@ them return -1 or 0, for signed and unsigned return values respectively, except 'LZ_compress_open' whose return value must be checked by calling 'LZ_compress_errno' before using it. - -- Function: struct LZ_Encoder * LZ_compress_open ( const int - DICTIONARY_SIZE, const int MATCH_LEN_LIMIT, const unsigned long - long MEMBER_SIZE ) + -- Function: LZ_Encoder * LZ_compress_open ( const int DICTIONARY_SIZE, + const int MATCH_LEN_LIMIT, const unsigned long long MEMBER_SIZE ) Initializes the internal stream state for compression and returns a pointer that can only be used as the ENCODER argument for the other LZ_compress functions, or a null pointer if the encoder could not be @@ -303,14 +277,14 @@ except 'LZ_compress_open' whose return value must be checked by calling produced. Values larger than 2 PiB are reduced to 2 PiB to prevent the uncompressed size of the member from overflowing. - -- Function: int LZ_compress_close ( struct LZ_Encoder * const ENCODER ) + -- Function: int LZ_compress_close ( LZ_Encoder * const ENCODER ) Frees all dynamically allocated data structures for this stream. This function discards any unprocessed input and does not flush any pending output. After a call to 'LZ_compress_close', ENCODER can no longer be used as an argument to any LZ_compress function. It is safe to call 'LZ_compress_close' with a null argument. - -- Function: int LZ_compress_finish ( struct LZ_Encoder * const ENCODER ) + -- Function: int LZ_compress_finish ( LZ_Encoder * const ENCODER ) Use this function to tell 'lzlib' that all the data for this member have already been written (with the function 'LZ_compress_write'). It is safe to call 'LZ_compress_finish' as many times as needed. After @@ -318,27 +292,26 @@ except 'LZ_compress_open' whose return value must be checked by calling 'LZ_compress_member_finished' returns 1, a new member can be started with 'LZ_compress_restart_member'. - -- Function: int LZ_compress_restart_member ( struct LZ_Encoder * const - ENCODER, const unsigned long long MEMBER_SIZE ) + -- Function: int LZ_compress_restart_member ( LZ_Encoder * const ENCODER, + const unsigned long long MEMBER_SIZE ) Use this function to start a new member in a multimember data stream. Call this function only after 'LZ_compress_member_finished' indicates that the current member has been fully read (with the function 'LZ_compress_read'). *Note member_size::, for a description of MEMBER_SIZE. - -- Function: int LZ_compress_sync_flush ( struct LZ_Encoder * const - ENCODER ) + -- Function: int LZ_compress_sync_flush ( LZ_Encoder * const ENCODER ) Use this function to make available to 'LZ_compress_read' all the data already written with the function 'LZ_compress_write'. First call 'LZ_compress_sync_flush'. Then call 'LZ_compress_read' until it returns 0. - This function writes at least one LZMA marker '3' ("Sync Flush" marker) + This function writes at least one LZMA marker '3' ('Sync Flush' marker) to the compressed output. Note that the sync flush marker is not allowed in lzip files; it is a device for interactive communication between applications using lzlib, but is useless and wasteful in a file, and is excluded from the media type 'application/lzip'. The LZMA - marker '2' ("End Of Stream" marker) is the only marker allowed in lzip + marker '2' ('End Of Stream' marker) is the only marker allowed in lzip files. *Note File format::. Repeated use of 'LZ_compress_sync_flush' may degrade compression @@ -353,8 +326,8 @@ except 'LZ_compress_open' whose return value must be checked by calling 'LZ_compress_member_finished' returns 1, followed by a new call to 'LZ_compress_sync_flush'. - -- Function: int LZ_compress_read ( struct LZ_Encoder * const ENCODER, - uint8_t * const BUFFER, const int SIZE ) + -- Function: int LZ_compress_read ( LZ_Encoder * const ENCODER, uint8_t * + const BUFFER, const int SIZE ) Reads up to SIZE bytes from the stream pointed to by ENCODER, storing the results in BUFFER. If LZ_API_VERSION >= 1012, BUFFER may be a null pointer, in which case the bytes read are discarded. @@ -365,14 +338,13 @@ except 'LZ_compress_open' whose return value must be checked by calling 'LZ_compress_write'. Note that reading less than SIZE bytes is not an error. - -- Function: int LZ_compress_write ( struct LZ_Encoder * const ENCODER, - uint8_t * const BUFFER, const int SIZE ) + -- Function: int LZ_compress_write ( LZ_Encoder * const ENCODER, uint8_t * + const BUFFER, const int SIZE ) Writes up to SIZE bytes from BUFFER to the stream pointed to by ENCODER. Returns the number of bytes actually written. This might be less than SIZE. Note that writing less than SIZE bytes is not an error. - -- Function: int LZ_compress_write_size ( struct LZ_Encoder * const - ENCODER ) + -- Function: int LZ_compress_write_size ( LZ_Encoder * const ENCODER ) Returns the maximum number of bytes that can be immediately written through 'LZ_compress_write'. For efficiency reasons, once the input buffer is full and 'LZ_compress_write_size' returns 0, almost all the @@ -383,39 +355,37 @@ except 'LZ_compress_open' whose return value must be checked by calling It is guaranteed that an immediate call to 'LZ_compress_write' will accept a SIZE up to the returned number of bytes. - -- Function: enum LZ_Errno LZ_compress_errno ( struct LZ_Encoder * const - ENCODER ) + -- Function: LZ_Errno LZ_compress_errno ( LZ_Encoder * const ENCODER ) Returns the current error code for ENCODER. *Note Error codes::. It is safe to call 'LZ_compress_errno' with a null argument, in which case it returns 'LZ_bad_argument'. - -- Function: int LZ_compress_finished ( struct LZ_Encoder * const ENCODER ) + -- Function: int LZ_compress_finished ( LZ_Encoder * const ENCODER ) Returns 1 if all the data have been read and 'LZ_compress_close' can be safely called. Otherwise it returns 0. 'LZ_compress_finished' implies 'LZ_compress_member_finished'. - -- Function: int LZ_compress_member_finished ( struct LZ_Encoder * const - ENCODER ) + -- Function: int LZ_compress_member_finished ( LZ_Encoder * const ENCODER ) Returns 1 if the current member, in a multimember data stream, has been fully read and 'LZ_compress_restart_member' can be safely called. Otherwise it returns 0. - -- Function: unsigned long long LZ_compress_data_position ( struct - LZ_Encoder * const ENCODER ) + -- Function: unsigned long long LZ_compress_data_position ( LZ_Encoder * + const ENCODER ) Returns the number of input bytes already compressed in the current member. - -- Function: unsigned long long LZ_compress_member_position ( struct - LZ_Encoder * const ENCODER ) + -- Function: unsigned long long LZ_compress_member_position ( LZ_Encoder * + const ENCODER ) Returns the number of compressed bytes already produced, but perhaps not yet read, in the current member. - -- Function: unsigned long long LZ_compress_total_in_size ( struct - LZ_Encoder * const ENCODER ) + -- Function: unsigned long long LZ_compress_total_in_size ( LZ_Encoder * + const ENCODER ) Returns the total number of input bytes already compressed. - -- Function: unsigned long long LZ_compress_total_out_size ( struct - LZ_Encoder * const ENCODER ) + -- Function: unsigned long long LZ_compress_total_out_size ( LZ_Encoder * + const ENCODER ) Returns the total number of compressed bytes already produced, but perhaps not yet read. @@ -430,7 +400,7 @@ them return -1 or 0, for signed and unsigned return values respectively, except 'LZ_decompress_open' whose return value must be checked by calling 'LZ_decompress_errno' before using it. - -- Function: struct LZ_Decoder * LZ_decompress_open ( void ) + -- Function: LZ_Decoder * LZ_decompress_open ( void ) Initializes the internal stream state for decompression and returns a pointer that can only be used as the DECODER argument for the other LZ_decompress functions, or a null pointer if the decoder could not be @@ -441,14 +411,14 @@ except 'LZ_decompress_open' whose return value must be checked by calling returned pointer must not be used and should be freed with 'LZ_decompress_close' to avoid memory leaks. - -- Function: int LZ_decompress_close ( struct LZ_Decoder * const DECODER ) + -- Function: int LZ_decompress_close ( LZ_Decoder * const DECODER ) Frees all dynamically allocated data structures for this stream. This function discards any unprocessed input and does not flush any pending output. After a call to 'LZ_decompress_close', DECODER can no longer be used as an argument to any LZ_decompress function. It is safe to call 'LZ_decompress_close' with a null argument. - -- Function: int LZ_decompress_finish ( struct LZ_Decoder * const DECODER ) + -- Function: int LZ_decompress_finish ( LZ_Decoder * const DECODER ) Use this function to tell 'lzlib' that all the data for this stream have already been written (with the function 'LZ_decompress_write'). It is safe to call 'LZ_decompress_finish' as many times as needed. It @@ -456,13 +426,13 @@ except 'LZ_decompress_open' whose return value must be checked by calling only contains whole members, but not calling it prevents lzlib from detecting a truncated member. - -- Function: int LZ_decompress_reset ( struct LZ_Decoder * const DECODER ) + -- Function: int LZ_decompress_reset ( LZ_Decoder * const DECODER ) Resets the internal state of DECODER as it was just after opening it with the function 'LZ_decompress_open'. Data stored in the internal buffers are discarded. Position counters are set to 0. - -- Function: int LZ_decompress_sync_to_member ( struct LZ_Decoder * const - DECODER ) + -- Function: int LZ_decompress_sync_to_member ( LZ_Decoder * const DECODER + ) Resets the error state of DECODER and enters a search state that lasts until a new member header (or the end of the stream) is found. After a successful call to 'LZ_decompress_sync_to_member', data written with @@ -474,8 +444,8 @@ except 'LZ_decompress_open' whose return value must be checked by calling case of a data error. If the decoder is already at the beginning of a member, this function does nothing. - -- Function: int LZ_decompress_read ( struct LZ_Decoder * const DECODER, - uint8_t * const BUFFER, const int SIZE ) + -- Function: int LZ_decompress_read ( LZ_Decoder * const DECODER, uint8_t + * const BUFFER, const int SIZE ) Reads up to SIZE bytes from the stream pointed to by DECODER, storing the results in BUFFER. If LZ_API_VERSION >= 1012, BUFFER may be a null pointer, in which case the bytes read are discarded. @@ -500,14 +470,13 @@ except 'LZ_decompress_open' whose return value must be checked by calling allows tools like tarlz to recover as much data as possible from each damaged member. *Note tarlz manual: (tarlz)Top. - -- Function: int LZ_decompress_write ( struct LZ_Decoder * const DECODER, - uint8_t * const BUFFER, const int SIZE ) + -- Function: int LZ_decompress_write ( LZ_Decoder * const DECODER, uint8_t + * const BUFFER, const int SIZE ) Writes up to SIZE bytes from BUFFER to the stream pointed to by DECODER. Returns the number of bytes actually written. This might be less than SIZE. Note that writing less than SIZE bytes is not an error. - -- Function: int LZ_decompress_write_size ( struct LZ_Decoder * const - DECODER ) + -- Function: int LZ_decompress_write_size ( LZ_Decoder * const DECODER ) Returns the maximum number of bytes that can be immediately written through 'LZ_decompress_write'. This number varies smoothly; each compressed byte consumed may be overwritten immediately, increasing by @@ -516,19 +485,17 @@ except 'LZ_decompress_open' whose return value must be checked by calling It is guaranteed that an immediate call to 'LZ_decompress_write' will accept a SIZE up to the returned number of bytes. - -- Function: enum LZ_Errno LZ_decompress_errno ( struct LZ_Decoder * const - DECODER ) + -- Function: LZ_Errno LZ_decompress_errno ( LZ_Decoder * const DECODER ) Returns the current error code for DECODER. *Note Error codes::. It is safe to call 'LZ_decompress_errno' with a null argument, in which case it returns 'LZ_bad_argument'. - -- Function: int LZ_decompress_finished ( struct LZ_Decoder * const - DECODER ) + -- Function: int LZ_decompress_finished ( LZ_Decoder * const DECODER ) Returns 1 if all the data have been read and 'LZ_decompress_close' can be safely called. Otherwise it returns 0. 'LZ_decompress_finished' does not imply 'LZ_decompress_member_finished'. - -- Function: int LZ_decompress_member_finished ( struct LZ_Decoder * const + -- Function: int LZ_decompress_member_finished ( LZ_Decoder * const DECODER ) Returns 1 if the previous call to 'LZ_decompress_read' finished reading the current member, indicating that final values for the member are @@ -536,37 +503,36 @@ except 'LZ_decompress_open' whose return value must be checked by calling 'LZ_decompress_data_position', and 'LZ_decompress_member_position'. Otherwise it returns 0. - -- Function: int LZ_decompress_member_version ( struct LZ_Decoder * const - DECODER ) + -- Function: int LZ_decompress_member_version ( LZ_Decoder * const DECODER + ) Returns the version of the current member, read from the member header. - -- Function: int LZ_decompress_dictionary_size ( struct LZ_Decoder * const + -- Function: int LZ_decompress_dictionary_size ( LZ_Decoder * const DECODER ) Returns the dictionary size of the current member, read from the member header. - -- Function: unsigned LZ_decompress_data_crc ( struct LZ_Decoder * const - DECODER ) + -- Function: unsigned LZ_decompress_data_crc ( LZ_Decoder * const DECODER ) Returns the 32 bit Cyclic Redundancy Check of the data decompressed from the current member. The value returned is valid only when 'LZ_decompress_member_finished' returns 1. - -- Function: unsigned long long LZ_decompress_data_position ( struct - LZ_Decoder * const DECODER ) + -- Function: unsigned long long LZ_decompress_data_position ( LZ_Decoder * + const DECODER ) Returns the number of decompressed bytes already produced, but perhaps not yet read, in the current member. - -- Function: unsigned long long LZ_decompress_member_position ( struct - LZ_Decoder * const DECODER ) + -- Function: unsigned long long LZ_decompress_member_position ( LZ_Decoder + * const DECODER ) Returns the number of input bytes already decompressed in the current member. - -- Function: unsigned long long LZ_decompress_total_in_size ( struct - LZ_Decoder * const DECODER ) + -- Function: unsigned long long LZ_decompress_total_in_size ( LZ_Decoder * + const DECODER ) Returns the total number of input bytes already decompressed. - -- Function: unsigned long long LZ_decompress_total_out_size ( struct - LZ_Decoder * const DECODER ) + -- Function: unsigned long long LZ_decompress_total_out_size ( LZ_Decoder + * const DECODER ) Returns the total number of decompressed bytes already produced, but perhaps not yet read. @@ -587,35 +553,36 @@ what kind of error it was, you need to check the error code by calling and you should not use 'LZ_(de)compress_errno' to determine whether a call failed. If the call failed, then you can examine 'LZ_(de)compress_errno'. - The error codes are defined in the header file 'lzlib.h'. + The error codes are defined in the header file 'lzlib.h'. 'LZ_Errno' is +an enum type: - -- Constant: enum LZ_Errno LZ_ok + -- Constant: LZ_Errno LZ_ok The value of this constant is 0 and is used to indicate that there is no error. - -- Constant: enum LZ_Errno LZ_bad_argument + -- Constant: LZ_Errno LZ_bad_argument At least one of the arguments passed to the library function was invalid. - -- Constant: enum LZ_Errno LZ_mem_error + -- Constant: LZ_Errno LZ_mem_error No memory available. The system cannot allocate more virtual memory because its capacity is full. - -- Constant: enum LZ_Errno LZ_sequence_error + -- Constant: LZ_Errno LZ_sequence_error A library function was called in the wrong order. For example 'LZ_compress_restart_member' was called before - 'LZ_compress_member_finished' indicates that the current member is + 'LZ_compress_member_finished' indicated that the current member is finished. - -- Constant: enum LZ_Errno LZ_header_error + -- Constant: LZ_Errno LZ_header_error An invalid member header (one with the wrong magic bytes) was read. If this happens at the end of the data stream it may indicate trailing data. - -- Constant: enum LZ_Errno LZ_unexpected_eof + -- Constant: LZ_Errno LZ_unexpected_eof The end of the data stream was reached in the middle of a member. - -- Constant: enum LZ_Errno LZ_data_error + -- Constant: LZ_Errno LZ_data_error The data stream is corrupt. If 'LZ_decompress_member_position' is 6 or less, it indicates either a format version not supported, an invalid dictionary size, a nonzero first LZMA byte, a corrupt header in a @@ -623,7 +590,7 @@ failed. If the call failed, then you can examine 'LZ_(de)compress_errno'. header. Lziprecover can be used to repair some of these errors and to remove conflicting trailing data from a file. - -- Constant: enum LZ_Errno LZ_library_error + -- Constant: LZ_Errno LZ_library_error A bug was detected in the library. Please, report it. *Note Problems::. @@ -632,11 +599,11 @@ File: lzlib.info, Node: Error messages, Next: Invoking minilzip, Prev: Error 8 Error messages **************** - -- Function: const char * LZ_strerror ( const enum LZ_Errno LZ_ERRNO ) - Returns the standard error message for a given error code. The messages - are fairly short; there are no multi-line messages or embedded - newlines. This function makes it easy for your program to report - informative error messages about the failure of a library call. + -- Function: const char * LZ_strerror ( const LZ_Errno LZ_ERRNO ) + Returns the error message corresponding to the error code LZ_ERRNO. + The messages are fairly short; there are no multi-line messages or + embedded newlines. This function makes it easy for your program to + report informative error messages about the failure of a library call. The value of LZ_ERRNO normally comes from a call to 'LZ_(de)compress_errno'. @@ -647,22 +614,22 @@ File: lzlib.info, Node: Invoking minilzip, Next: File format, Prev: Error mes 9 Invoking minilzip ******************* -Minilzip is a test program for the compression library lzlib, compatible -(interoperable) with lzip 1.4 or newer. Minilzip is not intended to be -installed because lzip has more features, but minilzip is well tested and -you can use it as your main compressor if so you wish. +Minilzip is a test program for the compression library lzlib. Minilzip is +not intended to be installed because lzip has more features, but minilzip is +well tested and you can use it as your main compressor if so you wish. +*Note lzip: (lzip)Top. Lzip is a lossless data compressor with a user interface similar to the -one of gzip or bzip2. Lzip uses a simplified form of the 'Lempel-Ziv-Markov -chain-Algorithm' (LZMA) stream format to maximize interoperability. The -maximum dictionary size is 512 MiB so that any lzip file can be decompressed -on 32-bit machines. Lzip provides accurate and robust 3-factor integrity -checking. Lzip can compress about as fast as gzip (lzip -0) or compress most -files more than bzip2 (lzip -9). Decompression speed is intermediate between -gzip and bzip2. Lzip provides better data recovery capabilities than gzip -and bzip2. Lzip has been designed, written, and tested with great care to -replace gzip and bzip2 as the standard general-purpose compressed format for -Unix-like systems. +one of gzip or bzip2. Lzip uses a simplified form of LZMA (Lempel-Ziv-Markov +chain-Algorithm) designed to achieve complete interoperability between +implementations. The maximum dictionary size is 512 MiB so that any lzip +file can be decompressed on 32-bit machines. Lzip provides accurate and +robust 3-factor integrity checking. 'lzip -0' compresses about as fast as +gzip, while 'lzip -9' compresses most files more than bzip2. Decompression +speed is intermediate between gzip and bzip2. Lzip provides better data +recovery capabilities than gzip and bzip2. Lzip has been designed, written, +and tested with great care to replace gzip and bzip2 as general-purpose +compressed format for Unix-like systems. The format for running minilzip is: @@ -675,7 +642,7 @@ once, the first time it appears in the command line. Remember to prepend './' to any file name beginning with a hyphen, or use '--'. minilzip supports the following options: *Note Argument syntax: -(arg_parser)Argument syntax. +(plzip)Argument syntax. '-h' '--help' @@ -694,11 +661,12 @@ minilzip supports the following options: *Note Argument syntax: '-b BYTES' '--member-size=BYTES' - When compressing, set the member size limit to BYTES. It is advisable - to keep members smaller than RAM size so that they can be repaired with - lziprecover in case of corruption. A small member size may degrade - compression ratio, so use it only when needed. Valid values range from - 100 kB to 2 PiB. Defaults to 2 PiB. + When compressing, set the member size limit to BYTES. If BYTES is + smaller than the compressed size, a multimember file is produced. It is + advisable to keep members smaller than RAM size so that they can be + repaired with lziprecover in case of corruption. A small member size + may degrade compression ratio, so use it only when needed. Valid + values range from 100 kB to 2 PiB. Defaults to 2 PiB. '-c' '--stdout' @@ -719,7 +687,8 @@ minilzip supports the following options: *Note Argument syntax: status 1. If a file fails to decompress, or is a terminal, minilzip exits immediately with error status 2 without decompressing the rest of the files. A terminal is considered an uncompressed file, and - therefore invalid. + therefore invalid. A multimember file with one or more empty members + is accepted if redirected to standard input. '-f' '--force' @@ -772,6 +741,8 @@ minilzip supports the following options: *Note Argument syntax: For maximum compression you should use a dictionary size limit as large as possible, but keep in mind that the decompression memory requirement is affected at compression time by the choice of dictionary size limit. + The dictionary size used for decompression is the same dictionary size + used for compression. '-S BYTES' '--volume-size=BYTES' @@ -791,7 +762,8 @@ minilzip supports the following options: *Note Argument syntax: fails the test, does not exist, can't be opened, or is a terminal, minilzip continues testing the rest of the files. A final diagnostic is shown at verbosity level 1 or higher if any file fails the test - when testing multiple files. + when testing multiple files. A multimember file with one or more empty + members is accepted if redirected to standard input. '-v' '--verbose' @@ -839,7 +811,7 @@ minilzip supports the following options: *Note Argument syntax: When decompressing or testing, allow trailing data whose first bytes are so similar to the magic bytes of a lzip header that they can be confused with a corrupt header. Use this option if a file triggers a - "corrupt header" error and the cause is not indeed a corrupt header. + 'corrupt header' error and the cause is not indeed a corrupt header. '--check-lib' Compare the version of lzlib used to compile minilzip with the version @@ -935,11 +907,11 @@ not allowed in multimember files. Valid values for dictionary size range from 4 KiB to 512 MiB. 'LZMA stream' - The LZMA stream, terminated by an "End Of Stream" marker. Uses default + The LZMA stream, terminated by an 'End Of Stream' marker. Uses default values for encoder properties. *Note Stream format: (lzip)Stream format, for a complete description. - Lzip only uses the LZMA marker '2' ("End Of Stream" marker). Lzlib - also uses the LZMA marker '3' ("Sync Flush" marker). *Note + Lzip only uses the LZMA marker '2' ('End Of Stream' marker). Lzlib + also uses the LZMA marker '3' ('Sync Flush' marker). *Note sync_flush::. 'CRC32 (4 bytes)' @@ -999,7 +971,7 @@ bool bbcompress( const uint8_t * const inbuf, const int insize, { int inpos = 0, outpos = 0; bool error = false; - struct LZ_Encoder * const encoder = + LZ_Encoder * const encoder = LZ_compress_open( dictionary_size, match_len_limit, INT64_MAX ); if( !encoder || LZ_compress_errno( encoder ) != LZ_ok ) { LZ_compress_close( encoder ); return false; } @@ -1042,7 +1014,7 @@ bool bbdecompress( const uint8_t * const inbuf, const int insize, { int inpos = 0, outpos = 0; bool error = false; - struct LZ_Decoder * const decoder = LZ_decompress_open(); + LZ_Decoder * const decoder = LZ_decompress_open(); if( !decoder || LZ_decompress_errno( decoder ) != LZ_ok ) { LZ_decompress_close( decoder ); return false; } @@ -1073,7 +1045,7 @@ File: lzlib.info, Node: File compression, Next: File decompression, Prev: Buf File-to-file compression using LZ_compress_write_size. -int ffcompress( struct LZ_Encoder * const encoder, +int ffcompress( LZ_Encoder * const encoder, FILE * const infile, FILE * const outfile ) { enum { buffer_size = 16384 }; @@ -1106,7 +1078,7 @@ File: lzlib.info, Node: File decompression, Next: File compression mm, Prev: File-to-file decompression using LZ_decompress_write_size. -int ffdecompress( struct LZ_Decoder * const decoder, +int ffdecompress( LZ_Decoder * const decoder, FILE * const infile, FILE * const outfile ) { enum { buffer_size = 16384 }; @@ -1145,8 +1117,7 @@ int ffmmcompress( FILE * const infile, FILE * const outfile ) enum { buffer_size = 16384, member_size = 4096 }; uint8_t buffer[buffer_size]; bool done = false; - struct LZ_Encoder * const encoder = - LZ_compress_open( 65535, 16, member_size ); + LZ_Encoder * const encoder = LZ_compress_open( 65535, 16, member_size ); if( !encoder || LZ_compress_errno( encoder ) != LZ_ok ) { fputs( "ffexample: Not enough memory.\n", stderr ); LZ_compress_close( encoder ); return 1; } @@ -1183,7 +1154,7 @@ LZ_compress_open with MEMBER_SIZE > largest member). for each line of text terminated by a newline character or by EOF. Return 0 if success, 1 if error. */ -int fflfcompress( struct LZ_Encoder * const encoder, +int fflfcompress( LZ_Encoder * const encoder, FILE * const infile, FILE * const outfile ) { enum { buffer_size = 16384 }; @@ -1229,7 +1200,7 @@ File: lzlib.info, Node: Skipping data errors, Prev: File compression mm, Up: next member in case of data error, including the automatic removal of leading garbage. */ -int ffrsdecompress( struct LZ_Decoder * const decoder, +int ffrsdecompress( LZ_Decoder * const decoder, FILE * const infile, FILE * const outfile ) { enum { buffer_size = 16384 }; @@ -1310,27 +1281,27 @@ Concept index Tag Table: Node: Top215 Node: Introduction1343 -Node: Library version6831 -Node: Buffering9382 -Node: Parameter limits10607 -Node: Compression functions11561 -Ref: member_size13354 -Ref: sync_flush15116 -Node: Decompression functions19804 -Node: Error codes27362 -Node: Error messages29719 -Node: Invoking minilzip30298 -Node: File format41020 -Ref: coded-dict-size42518 -Node: Examples43925 -Node: Buffer compression44886 -Node: Buffer decompression46406 -Node: File compression47820 -Node: File decompression48803 -Node: File compression mm49807 -Node: Skipping data errors52836 -Node: Problems54141 -Node: Concept index54702 +Node: Library version5506 +Node: Buffering8057 +Node: Parameter limits9282 +Node: Compression functions10236 +Ref: member_size12012 +Ref: sync_flush13753 +Node: Decompression functions18319 +Node: Error codes25706 +Node: Error messages28051 +Node: Invoking minilzip28634 +Node: File format39716 +Ref: coded-dict-size41214 +Node: Examples42621 +Node: Buffer compression43582 +Node: Buffer decompression45095 +Node: File compression46502 +Node: File decompression47478 +Node: File compression mm48475 +Node: Skipping data errors51486 +Node: Problems52784 +Node: Concept index53345 End Tag Table diff --git a/doc/lzlib.texi b/doc/lzlib.texi index 23bbbfb..4600e0d 100644 --- a/doc/lzlib.texi +++ b/doc/lzlib.texi @@ -6,8 +6,8 @@ @finalout @c %**end of header -@set UPDATED 16 October 2024 -@set VERSION 1.15-pre2 +@set UPDATED 19 November 2024 +@set VERSION 1.15-rc1 @dircategory Compression @direntry @@ -66,42 +66,10 @@ distribute, and modify it. @uref{http://www.nongnu.org/lzip/lzlib.html,,Lzlib} is a data compression library providing in-memory LZMA compression and decompression functions, including integrity checking of the decompressed -data. The compressed data format used by the library is the lzip format. +data. The compressed data format used by the library is the +@uref{http://www.nongnu.org/lzip/lzip.html,,lzip} format. Lzlib is written in C and is distributed under a 2-clause BSD license. -The lzip file format is designed for data sharing and long-term archiving, -taking into account both data integrity and decoder availability: - -@itemize @bullet -@item -The lzip format provides very safe integrity checking and some data -recovery means. The program -@uref{http://www.nongnu.org/lzip/manual/lziprecover_manual.html#Data-safety,,lziprecover} -can repair bit flip errors (one of the most common forms of data corruption) -in lzip files, and provides data recovery capabilities, including -error-checked merging of damaged copies of a file. -@ifnothtml -@xref{Data safety,,,lziprecover}. -@end ifnothtml - -@item -The lzip format is as simple as possible (but not simpler). The lzip -manual provides the source code of a simple decompressor along with a -detailed explanation of how it works, so that with the only help of the -lzip manual it would be possible for a digital archaeologist to extract -the data from a lzip file long after quantum computers eventually -render LZMA obsolete. - -@item -Additionally the lzip reference implementation is copylefted, which -guarantees that it will remain free forever. -@end itemize - -A nice feature of the lzip format is that a corrupt byte is easier to repair -the nearer it is from the beginning of the file. Therefore, with the help of -lziprecover, losing an entire archive just because of a corrupt byte near -the beginning is a thing of the past. - The functions and variables forming the interface of the compression library are declared in the file @file{lzlib.h}. Usage examples of the library are given in the files @file{bbexample.c}, @file{ffexample.c}, and @@ -149,10 +117,10 @@ In spite of its name (Lempel-Ziv-Markov chain-Algorithm), LZMA is not a concrete algorithm; it is more like "any algorithm using the LZMA coding scheme". For example, the option @option{-0} of lzip uses the scheme in almost the simplest way possible; issuing the longest match it can find, or -a literal byte if it can't find a match. Inversely, a much more elaborated -way of finding coding sequences of minimum size than the one currently used -by lzip could be developed, and the resulting sequence could also be coded -using the LZMA coding scheme. +a literal byte if it can't find a match. Inversely, a more elaborate way of +finding coding sequences of minimum size than the one currently used by lzip +could be developed, and the resulting sequence could also be coded using the +LZMA coding scheme. Lzlib currently implements two variants of the LZMA algorithm: fast (used by option @option{-0} of minilzip) and normal (used by all other compression levels). @@ -308,7 +276,7 @@ except @samp{LZ_compress_open} whose return value must be checked by calling @samp{LZ_compress_errno} before using it. -@deftypefun {struct LZ_Encoder *} LZ_compress_open ( const int @var{dictionary_size}, const int @var{match_len_limit}, const unsigned long long @var{member_size} ) +@deftypefun {LZ_Encoder *} LZ_compress_open ( const int @var{dictionary_size}, const int @var{match_len_limit}, const unsigned long long @var{member_size} ) Initializes the internal stream state for compression and returns a pointer that can only be used as the @var{encoder} argument for the other LZ_compress functions, or a null pointer if the encoder could not @@ -344,7 +312,7 @@ the uncompressed size of the member from overflowing. @end deftypefun -@deftypefun int LZ_compress_close ( struct LZ_Encoder * const @var{encoder} ) +@deftypefun int LZ_compress_close ( LZ_Encoder * const @var{encoder} ) Frees all dynamically allocated data structures for this stream. This function discards any unprocessed input and does not flush any pending output. After a call to @samp{LZ_compress_close}, @var{encoder} can no @@ -353,7 +321,7 @@ It is safe to call @samp{LZ_compress_close} with a null argument. @end deftypefun -@deftypefun int LZ_compress_finish ( struct LZ_Encoder * const @var{encoder} ) +@deftypefun int LZ_compress_finish ( LZ_Encoder * const @var{encoder} ) Use this function to tell @samp{lzlib} that all the data for this member have already been written (with the function @samp{LZ_compress_write}). It is safe to call @samp{LZ_compress_finish} as many times as needed. @@ -363,7 +331,7 @@ started with @samp{LZ_compress_restart_member}. @end deftypefun -@deftypefun int LZ_compress_restart_member ( struct LZ_Encoder * const @var{encoder}, const unsigned long long @var{member_size} ) +@deftypefun int LZ_compress_restart_member ( LZ_Encoder * const @var{encoder}, const unsigned long long @var{member_size} ) Use this function to start a new member in a multimember data stream. Call this function only after @samp{LZ_compress_member_finished} indicates that the current member has been fully read (with the function @@ -373,18 +341,18 @@ the current member has been fully read (with the function @anchor{sync_flush} -@deftypefun int LZ_compress_sync_flush ( struct LZ_Encoder * const @var{encoder} ) +@deftypefun int LZ_compress_sync_flush ( LZ_Encoder * const @var{encoder} ) Use this function to make available to @samp{LZ_compress_read} all the data already written with the function @samp{LZ_compress_write}. First call @samp{LZ_compress_sync_flush}. Then call @samp{LZ_compress_read} until it returns 0. -This function writes at least one LZMA marker @samp{3} ("Sync Flush" marker) +This function writes at least one LZMA marker @samp{3} ('Sync Flush' marker) to the compressed output. Note that the sync flush marker is not allowed in lzip files; it is a device for interactive communication between applications using lzlib, but is useless and wasteful in a file, and is excluded from the media type @samp{application/lzip}. The LZMA marker -@samp{2} ("End Of Stream" marker) is the only marker allowed in lzip files. +@samp{2} ('End Of Stream' marker) is the only marker allowed in lzip files. @xref{File format}. Repeated use of @samp{LZ_compress_sync_flush} may degrade compression @@ -401,7 +369,7 @@ are more bytes available than those needed to complete @var{member_size}, @end deftypefun -@deftypefun int LZ_compress_read ( struct LZ_Encoder * const @var{encoder}, uint8_t * const @var{buffer}, const int @var{size} ) +@deftypefun int LZ_compress_read ( LZ_Encoder * const @var{encoder}, uint8_t * const @var{buffer}, const int @var{size} ) Reads up to @var{size} bytes from the stream pointed to by @var{encoder}, storing the results in @var{buffer}. If @w{LZ_API_VERSION >= 1012}, @var{buffer} may be a null pointer, in which case the bytes read are @@ -415,7 +383,7 @@ not an error. @end deftypefun -@deftypefun int LZ_compress_write ( struct LZ_Encoder * const @var{encoder}, uint8_t * const @var{buffer}, const int @var{size} ) +@deftypefun int LZ_compress_write ( LZ_Encoder * const @var{encoder}, uint8_t * const @var{buffer}, const int @var{size} ) Writes up to @var{size} bytes from @var{buffer} to the stream pointed to by @var{encoder}. Returns the number of bytes actually written. This might be less than @var{size}. Note that writing less than @var{size} bytes is not an @@ -423,7 +391,7 @@ error. @end deftypefun -@deftypefun int LZ_compress_write_size ( struct LZ_Encoder * const @var{encoder} ) +@deftypefun int LZ_compress_write_size ( LZ_Encoder * const @var{encoder} ) Returns the maximum number of bytes that can be immediately written through @samp{LZ_compress_write}. For efficiency reasons, once the input buffer is full and @samp{LZ_compress_write_size} returns 0, almost all the buffer must @@ -436,44 +404,44 @@ accept a @var{size} up to the returned number of bytes. @end deftypefun -@deftypefun {enum LZ_Errno} LZ_compress_errno ( struct LZ_Encoder * const @var{encoder} ) +@deftypefun {LZ_Errno} LZ_compress_errno ( LZ_Encoder * const @var{encoder} ) Returns the current error code for @var{encoder}. @xref{Error codes}. It is safe to call @samp{LZ_compress_errno} with a null argument, in which case it returns @samp{LZ_bad_argument}. @end deftypefun -@deftypefun int LZ_compress_finished ( struct LZ_Encoder * const @var{encoder} ) +@deftypefun int LZ_compress_finished ( LZ_Encoder * const @var{encoder} ) Returns 1 if all the data have been read and @samp{LZ_compress_close} can be safely called. Otherwise it returns 0. @samp{LZ_compress_finished} implies @samp{LZ_compress_member_finished}. @end deftypefun -@deftypefun int LZ_compress_member_finished ( struct LZ_Encoder * const @var{encoder} ) +@deftypefun int LZ_compress_member_finished ( LZ_Encoder * const @var{encoder} ) Returns 1 if the current member, in a multimember data stream, has been fully read and @samp{LZ_compress_restart_member} can be safely called. Otherwise it returns 0. @end deftypefun -@deftypefun {unsigned long long} LZ_compress_data_position ( struct LZ_Encoder * const @var{encoder} ) +@deftypefun {unsigned long long} LZ_compress_data_position ( LZ_Encoder * const @var{encoder} ) Returns the number of input bytes already compressed in the current member. @end deftypefun -@deftypefun {unsigned long long} LZ_compress_member_position ( struct LZ_Encoder * const @var{encoder} ) +@deftypefun {unsigned long long} LZ_compress_member_position ( LZ_Encoder * const @var{encoder} ) Returns the number of compressed bytes already produced, but perhaps not yet read, in the current member. @end deftypefun -@deftypefun {unsigned long long} LZ_compress_total_in_size ( struct LZ_Encoder * const @var{encoder} ) +@deftypefun {unsigned long long} LZ_compress_total_in_size ( LZ_Encoder * const @var{encoder} ) Returns the total number of input bytes already compressed. @end deftypefun -@deftypefun {unsigned long long} LZ_compress_total_out_size ( struct LZ_Encoder * const @var{encoder} ) +@deftypefun {unsigned long long} LZ_compress_total_out_size ( LZ_Encoder * const @var{encoder} ) Returns the total number of compressed bytes already produced, but perhaps not yet read. @end deftypefun @@ -489,7 +457,7 @@ except @samp{LZ_decompress_open} whose return value must be checked by calling @samp{LZ_decompress_errno} before using it. -@deftypefun {struct LZ_Decoder *} LZ_decompress_open ( void ) +@deftypefun {LZ_Decoder *} LZ_decompress_open ( void ) Initializes the internal stream state for decompression and returns a pointer that can only be used as the @var{decoder} argument for the other LZ_decompress functions, or a null pointer if the decoder could not be @@ -502,7 +470,7 @@ the returned pointer must not be used and should be freed with @end deftypefun -@deftypefun int LZ_decompress_close ( struct LZ_Decoder * const @var{decoder} ) +@deftypefun int LZ_decompress_close ( LZ_Decoder * const @var{decoder} ) Frees all dynamically allocated data structures for this stream. This function discards any unprocessed input and does not flush any pending output. After a call to @samp{LZ_decompress_close}, @var{decoder} can no @@ -511,7 +479,7 @@ It is safe to call @samp{LZ_decompress_close} with a null argument. @end deftypefun -@deftypefun int LZ_decompress_finish ( struct LZ_Decoder * const @var{decoder} ) +@deftypefun int LZ_decompress_finish ( LZ_Decoder * const @var{decoder} ) Use this function to tell @samp{lzlib} that all the data for this stream have already been written (with the function @samp{LZ_decompress_write}). It is safe to call @samp{LZ_decompress_finish} as many times as needed. @@ -521,14 +489,14 @@ detecting a truncated member. @end deftypefun -@deftypefun int LZ_decompress_reset ( struct LZ_Decoder * const @var{decoder} ) +@deftypefun int LZ_decompress_reset ( LZ_Decoder * const @var{decoder} ) Resets the internal state of @var{decoder} as it was just after opening it with the function @samp{LZ_decompress_open}. Data stored in the internal buffers are discarded. Position counters are set to 0. @end deftypefun -@deftypefun int LZ_decompress_sync_to_member ( struct LZ_Decoder * const @var{decoder} ) +@deftypefun int LZ_decompress_sync_to_member ( LZ_Decoder * const @var{decoder} ) Resets the error state of @var{decoder} and enters a search state that lasts until a new member header (or the end of the stream) is found. After a successful call to @samp{LZ_decompress_sync_to_member}, data written with @@ -542,7 +510,7 @@ does nothing. @end deftypefun -@deftypefun int LZ_decompress_read ( struct LZ_Decoder * const @var{decoder}, uint8_t * const @var{buffer}, const int @var{size} ) +@deftypefun int LZ_decompress_read ( LZ_Decoder * const @var{decoder}, uint8_t * const @var{buffer}, const int @var{size} ) Reads up to @var{size} bytes from the stream pointed to by @var{decoder}, storing the results in @var{buffer}. If @w{LZ_API_VERSION >= 1012}, @var{buffer} may be a null pointer, in which case the bytes read are @@ -574,7 +542,7 @@ recover as much data as possible from each damaged member. @end deftypefun -@deftypefun int LZ_decompress_write ( struct LZ_Decoder * const @var{decoder}, uint8_t * const @var{buffer}, const int @var{size} ) +@deftypefun int LZ_decompress_write ( LZ_Decoder * const @var{decoder}, uint8_t * const @var{buffer}, const int @var{size} ) Writes up to @var{size} bytes from @var{buffer} to the stream pointed to by @var{decoder}. Returns the number of bytes actually written. This might be less than @var{size}. Note that writing less than @var{size} bytes is not an @@ -582,7 +550,7 @@ error. @end deftypefun -@deftypefun int LZ_decompress_write_size ( struct LZ_Decoder * const @var{decoder} ) +@deftypefun int LZ_decompress_write_size ( LZ_Decoder * const @var{decoder} ) Returns the maximum number of bytes that can be immediately written through @samp{LZ_decompress_write}. This number varies smoothly; each compressed byte consumed may be overwritten immediately, increasing by 1 the value @@ -593,21 +561,21 @@ accept a @var{size} up to the returned number of bytes. @end deftypefun -@deftypefun {enum LZ_Errno} LZ_decompress_errno ( struct LZ_Decoder * const @var{decoder} ) +@deftypefun {LZ_Errno} LZ_decompress_errno ( LZ_Decoder * const @var{decoder} ) Returns the current error code for @var{decoder}. @xref{Error codes}. It is safe to call @samp{LZ_decompress_errno} with a null argument, in which case it returns @samp{LZ_bad_argument}. @end deftypefun -@deftypefun int LZ_decompress_finished ( struct LZ_Decoder * const @var{decoder} ) +@deftypefun int LZ_decompress_finished ( LZ_Decoder * const @var{decoder} ) Returns 1 if all the data have been read and @samp{LZ_decompress_close} can be safely called. Otherwise it returns 0. @samp{LZ_decompress_finished} does not imply @samp{LZ_decompress_member_finished}. @end deftypefun -@deftypefun int LZ_decompress_member_finished ( struct LZ_Decoder * const @var{decoder} ) +@deftypefun int LZ_decompress_member_finished ( LZ_Decoder * const @var{decoder} ) Returns 1 if the previous call to @samp{LZ_decompress_read} finished reading the current member, indicating that final values for the member are available through @samp{LZ_decompress_data_crc}, @samp{LZ_decompress_data_position}, @@ -615,40 +583,40 @@ and @samp{LZ_decompress_member_position}. Otherwise it returns 0. @end deftypefun -@deftypefun int LZ_decompress_member_version ( struct LZ_Decoder * const @var{decoder} ) +@deftypefun int LZ_decompress_member_version ( LZ_Decoder * const @var{decoder} ) Returns the version of the current member, read from the member header. @end deftypefun -@deftypefun int LZ_decompress_dictionary_size ( struct LZ_Decoder * const @var{decoder} ) +@deftypefun int LZ_decompress_dictionary_size ( LZ_Decoder * const @var{decoder} ) Returns the dictionary size of the current member, read from the member header. @end deftypefun -@deftypefun {unsigned} LZ_decompress_data_crc ( struct LZ_Decoder * const @var{decoder} ) +@deftypefun {unsigned} LZ_decompress_data_crc ( LZ_Decoder * const @var{decoder} ) Returns the 32 bit Cyclic Redundancy Check of the data decompressed from the current member. The value returned is valid only when @samp{LZ_decompress_member_finished} returns 1. @end deftypefun -@deftypefun {unsigned long long} LZ_decompress_data_position ( struct LZ_Decoder * const @var{decoder} ) +@deftypefun {unsigned long long} LZ_decompress_data_position ( LZ_Decoder * const @var{decoder} ) Returns the number of decompressed bytes already produced, but perhaps not yet read, in the current member. @end deftypefun -@deftypefun {unsigned long long} LZ_decompress_member_position ( struct LZ_Decoder * const @var{decoder} ) +@deftypefun {unsigned long long} LZ_decompress_member_position ( LZ_Decoder * const @var{decoder} ) Returns the number of input bytes already decompressed in the current member. @end deftypefun -@deftypefun {unsigned long long} LZ_decompress_total_in_size ( struct LZ_Decoder * const @var{decoder} ) +@deftypefun {unsigned long long} LZ_decompress_total_in_size ( LZ_Decoder * const @var{decoder} ) Returns the total number of input bytes already decompressed. @end deftypefun -@deftypefun {unsigned long long} LZ_decompress_total_out_size ( struct LZ_Decoder * const @var{decoder} ) +@deftypefun {unsigned long long} LZ_decompress_total_out_size ( LZ_Decoder * const @var{decoder} ) Returns the total number of decompressed bytes already produced, but perhaps not yet read. @end deftypefun @@ -671,37 +639,38 @@ to determine whether a call failed. If the call failed, then you can examine @samp{LZ_(de)compress_errno}. The error codes are defined in the header file @file{lzlib.h}. +@samp{LZ_Errno} is an enum type: -@deftypevr Constant {enum LZ_Errno} LZ_ok +@deftypevr Constant {LZ_Errno} LZ_ok The value of this constant is 0 and is used to indicate that there is no error. @end deftypevr -@deftypevr Constant {enum LZ_Errno} LZ_bad_argument +@deftypevr Constant {LZ_Errno} LZ_bad_argument At least one of the arguments passed to the library function was invalid. @end deftypevr -@deftypevr Constant {enum LZ_Errno} LZ_mem_error -No memory available. The system cannot allocate more virtual memory -because its capacity is full. +@deftypevr Constant {LZ_Errno} LZ_mem_error +No memory available. The system cannot allocate more virtual memory because +its capacity is full. @end deftypevr -@deftypevr Constant {enum LZ_Errno} LZ_sequence_error +@deftypevr Constant {LZ_Errno} LZ_sequence_error A library function was called in the wrong order. For example @samp{LZ_compress_restart_member} was called before -@samp{LZ_compress_member_finished} indicates that the current member is +@samp{LZ_compress_member_finished} indicated that the current member is finished. @end deftypevr -@deftypevr Constant {enum LZ_Errno} LZ_header_error +@deftypevr Constant {LZ_Errno} LZ_header_error An invalid member header (one with the wrong magic bytes) was read. If this happens at the end of the data stream it may indicate trailing data. @end deftypevr -@deftypevr Constant {enum LZ_Errno} LZ_unexpected_eof +@deftypevr Constant {LZ_Errno} LZ_unexpected_eof The end of the data stream was reached in the middle of a member. @end deftypevr -@deftypevr Constant {enum LZ_Errno} LZ_data_error +@deftypevr Constant {LZ_Errno} LZ_data_error The data stream is corrupt. If @samp{LZ_decompress_member_position} is 6 or less, it indicates either a format version not supported, an invalid dictionary size, a nonzero first LZMA byte, a corrupt header in a multimember @@ -710,7 +679,7 @@ Lziprecover can be used to repair some of these errors and to remove conflicting trailing data from a file. @end deftypevr -@deftypevr Constant {enum LZ_Errno} LZ_library_error +@deftypevr Constant {LZ_Errno} LZ_library_error A bug was detected in the library. Please, report it. @xref{Problems}. @end deftypevr @@ -719,11 +688,11 @@ A bug was detected in the library. Please, report it. @xref{Problems}. @chapter Error messages @cindex error messages -@deftypefun {const char *} LZ_strerror ( const enum LZ_Errno @var{lz_errno} ) -Returns the standard error message for a given error code. The messages -are fairly short; there are no multi-line messages or embedded newlines. -This function makes it easy for your program to report informative error -messages about the failure of a library call. +@deftypefun {const char *} LZ_strerror ( const LZ_Errno @var{lz_errno} ) +Returns the error message corresponding to the error code @var{lz_errno}. +The messages are fairly short; there are no multi-line messages or embedded +newlines. This function makes it easy for your program to report informative +error messages about the failure of a library call. The value of @var{lz_errno} normally comes from a call to @samp{LZ_(de)compress_errno}. @@ -735,23 +704,25 @@ The value of @var{lz_errno} normally comes from a call to @cindex invoking @cindex options -Minilzip is a test program for the compression library lzlib, compatible -(interoperable) with lzip 1.4 or newer. Minilzip is not intended to be -installed because lzip has more features, but minilzip is well tested and -you can use it as your main compressor if so you wish. +Minilzip is a test program for the compression library lzlib. Minilzip is +not intended to be installed because lzip has more features, but minilzip is +well tested and you can use it as your main compressor if so you wish. +@ifnothtml +@xref{Top,lzip,,lzip}. +@end ifnothtml @uref{http://www.nongnu.org/lzip/lzip.html,,Lzip} is a lossless data compressor with a user interface similar to the one -of gzip or bzip2. Lzip uses a simplified form of the 'Lempel-Ziv-Markov -chain-Algorithm' (LZMA) stream format to maximize interoperability. The -maximum dictionary size is 512 MiB so that any lzip file can be decompressed -on 32-bit machines. Lzip provides accurate and robust 3-factor integrity -checking. Lzip can compress about as fast as gzip @w{(lzip -0)} or compress most -files more than bzip2 @w{(lzip -9)}. Decompression speed is intermediate between -gzip and bzip2. Lzip provides better data recovery capabilities than gzip -and bzip2. Lzip has been designed, written, and tested with great care to -replace gzip and bzip2 as the standard general-purpose compressed format for -Unix-like systems. +of gzip or bzip2. Lzip uses a simplified form of LZMA (Lempel-Ziv-Markov +chain-Algorithm) designed to achieve complete interoperability between +implementations. The maximum dictionary size is 512 MiB so that any lzip +file can be decompressed on 32-bit machines. Lzip provides accurate and +robust 3-factor integrity checking. @w{@samp{lzip -0}} compresses about as fast as +gzip, while @w{@samp{lzip -9}} compresses most files more than bzip2. Decompression +speed is intermediate between gzip and bzip2. Lzip provides better data +recovery capabilities than gzip and bzip2. Lzip has been designed, written, +and tested with great care to replace gzip and bzip2 as general-purpose +compressed format for Unix-like systems. @noindent The format for running minilzip is: @@ -769,9 +740,9 @@ prepend @file{./} to any file name beginning with a hyphen, or use @samp{--}. @noindent minilzip supports the following -@uref{http://www.nongnu.org/arg-parser/manual/arg_parser_manual.html#Argument-syntax,,options}: +@uref{http://www.nongnu.org/lzip/manual/plzip_manual.html#Argument-syntax,,options}: @ifnothtml -@xref{Argument syntax,,,arg_parser}. +@xref{Argument syntax,,,plzip}. @end ifnothtml @table @code @@ -792,9 +763,10 @@ garbage that can be safely ignored. @item -b @var{bytes} @itemx --member-size=@var{bytes} -When compressing, set the member size limit to @var{bytes}. It is advisable -to keep members smaller than RAM size so that they can be repaired with -lziprecover in case of corruption. A small member size may degrade +When compressing, set the member size limit to @var{bytes}. If @var{bytes} +is smaller than the compressed size, a multimember file is produced. It is +advisable to keep members smaller than RAM size so that they can be repaired +with lziprecover in case of corruption. A small member size may degrade compression ratio, so use it only when needed. Valid values range from @w{100 kB} to @w{2 PiB}. Defaults to @w{2 PiB}. @@ -816,7 +788,8 @@ already exists and @option{--force} has not been specified, minilzip continues decompressing the rest of the files and exits with error status 1. If a file fails to decompress, or is a terminal, minilzip exits immediately with error status 2 without decompressing the rest of the files. A terminal is -considered an uncompressed file, and therefore invalid. +considered an uncompressed file, and therefore invalid. A multimember file +with one or more empty members is accepted if redirected to standard input. @item -f @itemx --force @@ -870,6 +843,8 @@ to it. For maximum compression you should use a dictionary size limit as large as possible, but keep in mind that the decompression memory requirement is affected at compression time by the choice of dictionary size limit. +The dictionary size used for decompression is the same dictionary size used +for compression. @item -S @var{bytes} @itemx --volume-size=@var{bytes} @@ -889,7 +864,8 @@ together with @option{-v} to see information about the files. If a file fails the test, does not exist, can't be opened, or is a terminal, minilzip continues testing the rest of the files. A final diagnostic is shown at verbosity level 1 or higher if any file fails the test when testing multiple -files. +files. A multimember file with one or more empty members is accepted if +redirected to standard input. @item -v @itemx --verbose @@ -936,8 +912,8 @@ Aliases for GNU gzip compatibility. @item --loose-trailing When decompressing or testing, allow trailing data whose first bytes are so similar to the magic bytes of a lzip header that they can be confused -with a corrupt header. Use this option if a file triggers a "corrupt -header" error and the cause is not indeed a corrupt header. +with a corrupt header. Use this option if a file triggers a 'corrupt +header' error and the cause is not indeed a corrupt header. @item --check-lib Compare the @uref{#Library-version,,version of lzlib} used to compile @@ -1044,7 +1020,7 @@ Example: 0xD3 = 2^19 - 6 * 2^15 = 512 KiB - 6 * 32 KiB = 320 KiB@* Valid values for dictionary size range from 4 KiB to 512 MiB. @item LZMA stream -The LZMA stream, terminated by an "End Of Stream" marker. Uses default values +The LZMA stream, terminated by an 'End Of Stream' marker. Uses default values for encoder properties. @ifnothtml @xref{Stream format,,,lzip}, @@ -1054,8 +1030,8 @@ See @uref{http://www.nongnu.org/lzip/manual/lzip_manual.html#Stream-format,,Stream format} @end ifhtml for a complete description.@* -Lzip only uses the LZMA marker @samp{2} ("End Of Stream" marker). Lzlib -also uses the LZMA marker @samp{3} ("Sync Flush" marker). @xref{sync_flush}. +Lzip only uses the LZMA marker @samp{2} ('End Of Stream' marker). Lzlib +also uses the LZMA marker @samp{3} ('Sync Flush' marker). @xref{sync_flush}. @item CRC32 (4 bytes) Cyclic Redundancy Check (CRC) of the original uncompressed data. @@ -1115,7 +1091,7 @@ bool bbcompress( const uint8_t * const inbuf, const int insize, { int inpos = 0, outpos = 0; bool error = false; - struct LZ_Encoder * const encoder = + LZ_Encoder * const encoder = LZ_compress_open( dictionary_size, match_len_limit, INT64_MAX ); if( !encoder || LZ_compress_errno( encoder ) != LZ_ok ) { LZ_compress_close( encoder ); return false; } @@ -1159,7 +1135,7 @@ bool bbdecompress( const uint8_t * const inbuf, const int insize, { int inpos = 0, outpos = 0; bool error = false; - struct LZ_Decoder * const decoder = LZ_decompress_open(); + LZ_Decoder * const decoder = LZ_decompress_open(); if( !decoder || LZ_decompress_errno( decoder ) != LZ_ok ) { LZ_decompress_close( decoder ); return false; } @@ -1191,7 +1167,7 @@ bool bbdecompress( const uint8_t * const inbuf, const int insize, File-to-file compression using LZ_compress_write_size. @verbatim -int ffcompress( struct LZ_Encoder * const encoder, +int ffcompress( LZ_Encoder * const encoder, FILE * const infile, FILE * const outfile ) { enum { buffer_size = 16384 }; @@ -1225,7 +1201,7 @@ int ffcompress( struct LZ_Encoder * const encoder, File-to-file decompression using LZ_decompress_write_size. @verbatim -int ffdecompress( struct LZ_Decoder * const decoder, +int ffdecompress( LZ_Decoder * const decoder, FILE * const infile, FILE * const outfile ) { enum { buffer_size = 16384 }; @@ -1265,8 +1241,7 @@ int ffmmcompress( FILE * const infile, FILE * const outfile ) enum { buffer_size = 16384, member_size = 4096 }; uint8_t buffer[buffer_size]; bool done = false; - struct LZ_Encoder * const encoder = - LZ_compress_open( 65535, 16, member_size ); + LZ_Encoder * const encoder = LZ_compress_open( 65535, 16, member_size ); if( !encoder || LZ_compress_errno( encoder ) != LZ_ok ) { fputs( "ffexample: Not enough memory.\n", stderr ); LZ_compress_close( encoder ); return 1; } @@ -1306,7 +1281,7 @@ Example 2: Multimember compression (user-restarted members). for each line of text terminated by a newline character or by EOF. Return 0 if success, 1 if error. */ -int fflfcompress( struct LZ_Encoder * const encoder, +int fflfcompress( LZ_Encoder * const encoder, FILE * const infile, FILE * const outfile ) { enum { buffer_size = 16384 }; @@ -1353,7 +1328,7 @@ int fflfcompress( struct LZ_Encoder * const encoder, next member in case of data error, including the automatic removal of leading garbage. */ -int ffrsdecompress( struct LZ_Decoder * const decoder, +int ffrsdecompress( LZ_Decoder * const decoder, FILE * const infile, FILE * const outfile ) { enum { buffer_size = 16384 }; diff --git a/doc/minilzip.1 b/doc/minilzip.1 index c60e528..8d35853 100644 --- a/doc/minilzip.1 +++ b/doc/minilzip.1 @@ -1,25 +1,26 @@ .\" DO NOT MODIFY THIS FILE! It was generated by help2man 1.49.2. -.TH MINILZIP "1" "October 2024" "minilzip 1.15-pre2" "User Commands" +.TH MINILZIP "1" "November 2024" "minilzip 1.15-rc1" "User Commands" .SH NAME minilzip \- reduces the size of files .SH SYNOPSIS .B minilzip [\fI\,options\/\fR] [\fI\,files\/\fR] .SH DESCRIPTION -Minilzip is a test program for the compression library lzlib, compatible -(interoperable) with lzip 1.4 or newer. +Minilzip is a test program for the compression library lzlib. Minilzip is +not intended to be installed because lzip has more features, but minilzip is +well tested and you can use it as your main compressor if so you wish. .PP Lzip is a lossless data compressor with a user interface similar to the one -of gzip or bzip2. Lzip uses a simplified form of the 'Lempel\-Ziv\-Markov -chain\-Algorithm' (LZMA) stream format to maximize interoperability. The -maximum dictionary size is 512 MiB so that any lzip file can be decompressed -on 32\-bit machines. Lzip provides accurate and robust 3\-factor integrity -checking. Lzip can compress about as fast as gzip (lzip \fB\-0\fR) or compress most -files more than bzip2 (lzip \fB\-9\fR). Decompression speed is intermediate between -gzip and bzip2. Lzip provides better data recovery capabilities than gzip -and bzip2. Lzip has been designed, written, and tested with great care to -replace gzip and bzip2 as the standard general\-purpose compressed format for -Unix\-like systems. +of gzip or bzip2. Lzip uses a simplified form of LZMA (Lempel\-Ziv\-Markov +chain\-Algorithm) designed to achieve complete interoperability between +implementations. The maximum dictionary size is 512 MiB so that any lzip +file can be decompressed on 32\-bit machines. Lzip provides accurate and +robust 3\-factor integrity checking. 'lzip \fB\-0\fR' compresses about as fast as +gzip, while 'lzip \fB\-9\fR' compresses most files more than bzip2. Decompression +speed is intermediate between gzip and bzip2. Lzip provides better data +recovery capabilities than gzip and bzip2. Lzip has been designed, written, +and tested with great care to replace gzip and bzip2 as general\-purpose +compressed format for Unix\-like systems. .SH OPTIONS .TP \fB\-h\fR, \fB\-\-help\fR @@ -32,7 +33,7 @@ output version information and exit exit with error status if trailing data .TP \fB\-b\fR, \fB\-\-member\-size=\fR<bytes> -set member size limit in bytes +set member size limit of multimember files .TP \fB\-c\fR, \fB\-\-stdout\fR write to standard output, keep input files @@ -120,7 +121,7 @@ License GPLv2+: GNU GPL version 2 or later <http://gnu.org/licenses/gpl.html> .br This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. -Using lzlib 1.15\-pre2 +Using lzlib 1.15\-rc1 Using LZ_API_VERSION = 1015 .SH "SEE ALSO" The full documentation for @@ -17,7 +17,7 @@ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ -static int LZe_get_match_pairs( struct LZ_encoder * const e, struct Pair * pairs ) +static int LZe_get_match_pairs( LZ_encoder * const e, Pair * pairs ) { int32_t * ptr0 = e->eb.mb.pos_array + ( e->eb.mb.cyclic_pos << 1 ); int32_t * ptr1 = ptr0 + 1; @@ -31,8 +31,8 @@ static int LZe_get_match_pairs( struct LZ_encoder * const e, struct Pair * pairs int maxlen = 3; /* only used if pairs != 0 */ int num_pairs = 0; - const int min_pos = ( e->eb.mb.pos > e->eb.mb.dictionary_size ) ? - e->eb.mb.pos - e->eb.mb.dictionary_size : 0; + const int min_pos = (e->eb.mb.pos > e->eb.mb.dictionary_size) ? + e->eb.mb.pos - e->eb.mb.dictionary_size : 0; const uint8_t * const data = Mb_ptr_to_current_pos( &e->eb.mb ); unsigned tmp = crc32[data[0]] ^ data[1]; @@ -121,7 +121,7 @@ static int LZe_get_match_pairs( struct LZ_encoder * const e, struct Pair * pairs } -static void LZe_update_distance_prices( struct LZ_encoder * const e ) +static void LZe_update_distance_prices( LZ_encoder * const e ) { int dis, len_state; for( dis = start_dis_model; dis < modeled_distances; ++dis ) @@ -160,7 +160,7 @@ static void LZe_update_distance_prices( struct LZ_encoder * const e ) ( trials[0].dis4 == -1 ) means literal. A match/rep longer or equal than match_len_limit finishes the sequence. */ -static int LZe_sequence_optimizer( struct LZ_encoder * const e, +static int LZe_sequence_optimizer( LZ_encoder * const e, const int reps[num_rep_distances], const State state ) { @@ -174,7 +174,7 @@ static int LZe_sequence_optimizer( struct LZ_encoder * const e, } else num_pairs = LZe_read_match_distances( e ); - const int main_len = ( num_pairs > 0 ) ? e->pairs[num_pairs-1].len : 0; + const int main_len = (num_pairs > 0) ? e->pairs[num_pairs-1].len : 0; int replens[num_rep_distances]; int rep_index = 0; @@ -270,7 +270,7 @@ static int LZe_sequence_optimizer( struct LZ_encoder * const e, } const int num_pairs = LZe_read_match_distances( e ); - const int newlen = ( num_pairs > 0 ) ? e->pairs[num_pairs-1].len : 0; + const int newlen = (num_pairs > 0) ? e->pairs[num_pairs-1].len : 0; if( newlen >= e->match_len_limit ) { e->pending_num_pairs = num_pairs; @@ -279,7 +279,7 @@ static int LZe_sequence_optimizer( struct LZ_encoder * const e, } /* give final values to current trial */ - struct Trial * cur_trial = &e->trials[cur]; + Trial * cur_trial = &e->trials[cur]; State cur_state; { const int dis4 = cur_trial->dis4; @@ -324,7 +324,7 @@ static int LZe_sequence_optimizer( struct LZ_encoder * const e, next_price += LZeb_price_matched( &e->eb, prev_byte, cur_byte, match_byte ); /* try last updates to next trial */ - struct Trial * next_trial = &e->trials[cur+1]; + Trial * next_trial = &e->trials[cur+1]; Tr_update( next_trial, next_price, -1, cur ); /* literal */ @@ -466,12 +466,12 @@ static int LZe_sequence_optimizer( struct LZ_encoder * const e, } -static bool LZe_encode_member( struct LZ_encoder * const e ) +static bool LZe_encode_member( LZ_encoder * const e ) { - const bool best = ( e->match_len_limit > 12 ); + const bool best = e->match_len_limit > 12; const int dis_price_count = best ? 1 : 512; const int align_price_count = best ? 1 : dis_align_size; - const int price_count = ( e->match_len_limit > 36 ) ? 1013 : 4093; + const int price_count = (e->match_len_limit > 36) ? 1013 : 4093; int i; State * const state = &e->eb.state; @@ -522,7 +522,7 @@ static bool LZe_encode_member( struct LZ_encoder * const e ) const int len = e->trials[i].price; int dis = e->trials[i].dis4; - bool bit = ( dis < 0 ); + bool bit = dis < 0; Re_encode_bit( &e->eb.renc, &e->eb.bm_match[*state][pos_state], !bit ); if( bit ) /* literal byte */ { @@ -541,11 +541,11 @@ static bool LZe_encode_member( struct LZ_encoder * const e ) { CRC32_update_buf( &e->eb.crc, Mb_ptr_to_current_pos( &e->eb.mb ) - ahead, len ); mtf_reps( dis, e->eb.reps ); - bit = ( dis < num_rep_distances ); + bit = dis < num_rep_distances; Re_encode_bit( &e->eb.renc, &e->eb.bm_rep[*state], bit ); if( bit ) /* repeated match */ { - bit = ( dis == 0 ); + bit = dis == 0; Re_encode_bit( &e->eb.renc, &e->eb.bm_rep0[*state], !bit ); if( bit ) Re_encode_bit( &e->eb.renc, &e->eb.bm_len[*state][pos_state], len > 1 ); @@ -17,16 +17,16 @@ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ -struct Len_prices +typedef struct Len_prices { - const struct Len_model * lm; + const Len_model * lm; int len_symbols; int count; int prices[pos_states][max_len_symbols]; int counters[pos_states]; /* may decrement below 0 */ - }; + } Len_prices; -static inline void Lp_update_low_mid_prices( struct Len_prices * const lp, +static inline void Lp_update_low_mid_prices( Len_prices * const lp, const int pos_state ) { int * const pps = lp->prices[pos_state]; @@ -41,7 +41,7 @@ static inline void Lp_update_low_mid_prices( struct Len_prices * const lp, price_symbol3( lp->lm->bm_mid[pos_state], len - len_low_symbols ); } -static inline void Lp_update_high_prices( struct Len_prices * const lp ) +static inline void Lp_update_high_prices( Len_prices * const lp ) { const int tmp = price1( lp->lm->choice1 ) + price1( lp->lm->choice2 ); int len; @@ -52,24 +52,23 @@ static inline void Lp_update_high_prices( struct Len_prices * const lp ) price_symbol8( lp->lm->bm_high, len - len_low_symbols - len_mid_symbols ); } -static inline void Lp_reset( struct Len_prices * const lp ) +static inline void Lp_reset( Len_prices * const lp ) { int i; for( i = 0; i < pos_states; ++i ) lp->counters[i] = 0; } -static inline void Lp_init( struct Len_prices * const lp, - const struct Len_model * const lm, +static inline void Lp_init( Len_prices * const lp, const Len_model * const lm, const int match_len_limit ) { lp->lm = lm; lp->len_symbols = match_len_limit + 1 - min_match_len; - lp->count = ( match_len_limit > 12 ) ? 1 : lp->len_symbols; + lp->count = (match_len_limit > 12) ? 1 : lp->len_symbols; Lp_reset( lp ); } -static inline void Lp_decrement_counter( struct Len_prices * const lp, +static inline void Lp_decrement_counter( Len_prices * const lp, const int pos_state ) { --lp->counters[pos_state]; } -static inline void Lp_update_prices( struct Len_prices * const lp ) +static inline void Lp_update_prices( Len_prices * const lp ) { int pos_state; bool high_pending = false; @@ -81,23 +80,23 @@ static inline void Lp_update_prices( struct Len_prices * const lp ) Lp_update_high_prices( lp ); } -static inline int Lp_price( const struct Len_prices * const lp, +static inline int Lp_price( const Len_prices * const lp, const int len, const int pos_state ) { return lp->prices[pos_state][len - min_match_len]; } -struct Pair /* distance-length pair */ +typedef struct Pair /* distance-length pair */ { int dis; int len; - }; + } Pair; enum { infinite_price = 0x0FFFFFFF, max_num_trials = 1 << 13, single_step_trial = -2, dual_step_trial = -1 }; -struct Trial +typedef struct Trial { State state; int price; /* dual use var; cumulative price, match length */ @@ -107,9 +106,9 @@ struct Trial /* -1 literal + rep0 */ /* >= 0 ( rep or match ) + literal + rep0 */ int reps[num_rep_distances]; - }; + } Trial; -static inline void Tr_update( struct Trial * const trial, const int pr, +static inline void Tr_update( Trial * const trial, const int pr, const int distance4, const int p_i ) { if( pr < trial->price ) @@ -117,7 +116,7 @@ static inline void Tr_update( struct Trial * const trial, const int pr, trial->prev_index2 = single_step_trial; } } -static inline void Tr_update2( struct Trial * const trial, const int pr, +static inline void Tr_update2( Trial * const trial, const int pr, const int p_i ) { if( pr < trial->price ) @@ -125,7 +124,7 @@ static inline void Tr_update2( struct Trial * const trial, const int pr, trial->prev_index2 = dual_step_trial; } } -static inline void Tr_update3( struct Trial * const trial, const int pr, +static inline void Tr_update3( Trial * const trial, const int pr, const int distance4, const int p_i, const int p_i2 ) { @@ -135,16 +134,16 @@ static inline void Tr_update3( struct Trial * const trial, const int pr, } -struct LZ_encoder +typedef struct LZ_encoder { - struct LZ_encoder_base eb; + LZ_encoder_base eb; int cycles; int match_len_limit; - struct Len_prices match_len_prices; - struct Len_prices rep_len_prices; + Len_prices match_len_prices; + Len_prices rep_len_prices; int pending_num_pairs; - struct Pair pairs[max_match_len+1]; - struct Trial trials[max_num_trials]; + Pair pairs[max_match_len+1]; + Trial trials[max_num_trials]; int dis_slot_prices[len_states][2*max_dictionary_bits]; int dis_prices[len_states][modeled_distances]; @@ -154,10 +153,9 @@ struct LZ_encoder int dis_price_counter; int align_price_counter; bool been_flushed; - }; + } LZ_encoder; -static inline bool Mb_dec_pos( struct Matchfinder_base * const mb, - const int ahead ) +static inline bool Mb_dec_pos( Matchfinder_base * const mb, const int ahead ) { if( ahead < 0 || mb->pos < ahead ) return false; mb->pos -= ahead; @@ -166,7 +164,7 @@ static inline bool Mb_dec_pos( struct Matchfinder_base * const mb, return true; } -static int LZe_get_match_pairs( struct LZ_encoder * const e, struct Pair * pairs ); +static int LZe_get_match_pairs( LZ_encoder * const e, Pair * pairs ); /* move-to-front dis in/into reps; do nothing if( dis4 <= 0 ) */ static inline void mtf_reps( const int dis4, int reps[num_rep_distances] ) @@ -184,13 +182,13 @@ static inline void mtf_reps( const int dis4, int reps[num_rep_distances] ) } } -static inline int LZeb_price_shortrep( const struct LZ_encoder_base * const eb, +static inline int LZeb_price_shortrep( const LZ_encoder_base * const eb, const State state, const int pos_state ) { return price0( eb->bm_rep0[state] ) + price0( eb->bm_len[state][pos_state] ); } -static inline int LZeb_price_rep( const struct LZ_encoder_base * const eb, +static inline int LZeb_price_rep( const LZ_encoder_base * const eb, const int rep, const State state, const int pos_state ) { @@ -207,7 +205,7 @@ static inline int LZeb_price_rep( const struct LZ_encoder_base * const eb, return price; } -static inline int LZe_price_rep0_len( const struct LZ_encoder * const e, +static inline int LZe_price_rep0_len( const LZ_encoder * const e, const int len, const State state, const int pos_state ) { @@ -215,7 +213,7 @@ static inline int LZe_price_rep0_len( const struct LZ_encoder * const e, Lp_price( &e->rep_len_prices, len, pos_state ); } -static inline int LZe_price_pair( const struct LZ_encoder * const e, +static inline int LZe_price_pair( const LZ_encoder * const e, const int dis, const int len, const int pos_state ) { @@ -228,7 +226,7 @@ static inline int LZe_price_pair( const struct LZ_encoder * const e, e->align_prices[dis & (dis_align_size - 1)]; } -static inline int LZe_read_match_distances( struct LZ_encoder * const e ) +static inline int LZe_read_match_distances( LZ_encoder * const e ) { const int num_pairs = LZe_get_match_pairs( e, e->pairs ); if( num_pairs > 0 ) @@ -241,7 +239,7 @@ static inline int LZe_read_match_distances( struct LZ_encoder * const e ) return num_pairs; } -static inline bool LZe_move_and_update( struct LZ_encoder * const e, int n ) +static inline bool LZe_move_and_update( LZ_encoder * const e, int n ) { while( true ) { @@ -252,13 +250,13 @@ static inline bool LZe_move_and_update( struct LZ_encoder * const e, int n ) return true; } -static inline void LZe_backward( struct LZ_encoder * const e, int cur ) +static inline void LZe_backward( LZ_encoder * const e, int cur ) { int dis4 = e->trials[cur].dis4; while( cur > 0 ) { const int prev_index = e->trials[cur].prev_index; - struct Trial * const prev_trial = &e->trials[prev_index]; + Trial * const prev_trial = &e->trials[prev_index]; if( e->trials[cur].prev_index2 != single_step_trial ) { @@ -267,7 +265,7 @@ static inline void LZe_backward( struct LZ_encoder * const e, int cur ) prev_trial->prev_index2 = single_step_trial; if( e->trials[cur].prev_index2 >= 0 ) { - struct Trial * const prev_trial2 = &e->trials[prev_index-1]; + Trial * const prev_trial2 = &e->trials[prev_index-1]; prev_trial2->dis4 = dis4; dis4 = 0; /* rep0 */ prev_trial2->prev_index = e->trials[cur].prev_index2; prev_trial2->prev_index2 = single_step_trial; @@ -282,7 +280,7 @@ static inline void LZe_backward( struct LZ_encoder * const e, int cur ) enum { num_prev_positions3 = 1 << 16, num_prev_positions2 = 1 << 10 }; -static inline bool LZe_init( struct LZ_encoder * const e, +static inline bool LZe_init( LZ_encoder * const e, const int dict_size, const int len_limit, const unsigned long long member_size ) { @@ -297,7 +295,7 @@ static inline bool LZe_init( struct LZ_encoder * const e, if( !LZeb_init( &e->eb, before_size, dict_size, after_size, dict_factor, num_prev_positions23, pos_array_factor, min_free_bytes, member_size ) ) return false; - e->cycles = ( len_limit < max_match_len ) ? 16 + ( len_limit / 2 ) : 256; + e->cycles = (len_limit < max_match_len) ? 16 + ( len_limit / 2 ) : 256; e->match_len_limit = len_limit; Lp_init( &e->match_len_prices, &e->eb.match_len_model, e->match_len_limit ); Lp_init( &e->rep_len_prices, &e->eb.rep_len_model, e->match_len_limit ); @@ -312,7 +310,7 @@ static inline bool LZe_init( struct LZ_encoder * const e, return true; } -static inline void LZe_reset( struct LZ_encoder * const e, +static inline void LZe_reset( LZ_encoder * const e, const unsigned long long member_size ) { LZeb_reset( &e->eb, member_size ); diff --git a/encoder_base.c b/encoder_base.c index cd790fb..2773887 100644 --- a/encoder_base.c +++ b/encoder_base.c @@ -17,7 +17,7 @@ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ -static bool Mb_normalize_pos( struct Matchfinder_base * const mb ) +static bool Mb_normalize_pos( Matchfinder_base * const mb ) { if( mb->pos > mb->stream_pos ) { mb->pos = mb->stream_pos; return false; } @@ -40,7 +40,7 @@ static bool Mb_normalize_pos( struct Matchfinder_base * const mb ) } -static bool Mb_init( struct Matchfinder_base * const mb, const int before_size, +static bool Mb_init( Matchfinder_base * const mb, const int before_size, const int dict_size, const int after_size, const int dict_factor, const int num_prev_positions23, const int pos_array_factor ) @@ -83,7 +83,7 @@ static bool Mb_init( struct Matchfinder_base * const mb, const int before_size, } -static void Mb_adjust_array( struct Matchfinder_base * const mb ) +static void Mb_adjust_array( Matchfinder_base * const mb ) { int size = 1 << max( 16, real_bits( mb->dictionary_size - 1 ) - 2 ); if( mb->dictionary_size > 1 << 26 ) size >>= 1; /* 64 MiB */ @@ -94,7 +94,7 @@ static void Mb_adjust_array( struct Matchfinder_base * const mb ) } -static void Mb_adjust_dictionary_size( struct Matchfinder_base * const mb ) +static void Mb_adjust_dictionary_size( Matchfinder_base * const mb ) { if( mb->stream_pos < mb->dictionary_size ) { @@ -105,7 +105,7 @@ static void Mb_adjust_dictionary_size( struct Matchfinder_base * const mb ) } -static void Mb_reset( struct Matchfinder_base * const mb ) +static void Mb_reset( Matchfinder_base * const mb ) { int i; if( mb->stream_pos > mb->pos ) @@ -124,7 +124,7 @@ static void Mb_reset( struct Matchfinder_base * const mb ) /* End Of Stream marker => (dis == 0xFFFFFFFFU, len == min_match_len) */ -static void LZeb_try_full_flush( struct LZ_encoder_base * const eb ) +static void LZeb_try_full_flush( LZ_encoder_base * const eb ) { if( eb->member_finished || Cb_free_bytes( &eb->renc.cb ) < max_marker_size + eb->renc.ff_count + Lt_size ) return; @@ -144,7 +144,7 @@ static void LZeb_try_full_flush( struct LZ_encoder_base * const eb ) /* Sync Flush marker => (dis == 0xFFFFFFFFU, len == min_match_len + 1) */ -static void LZeb_try_sync_flush( struct LZ_encoder_base * const eb ) +static void LZeb_try_sync_flush( LZ_encoder_base * const eb ) { const unsigned min_size = eb->renc.ff_count + max_marker_size; if( eb->member_finished || @@ -163,7 +163,7 @@ static void LZeb_try_sync_flush( struct LZ_encoder_base * const eb ) } -static void LZeb_reset( struct LZ_encoder_base * const eb, +static void LZeb_reset( LZ_encoder_base * const eb, const unsigned long long member_size ) { const unsigned long long min_member_size = min_dictionary_size; diff --git a/encoder_base.h b/encoder_base.h index 094f679..56165d4 100644 --- a/encoder_base.h +++ b/encoder_base.h @@ -217,7 +217,7 @@ static inline int price_matched( const Bit_model bm[], unsigned symbol, } -struct Matchfinder_base +typedef struct Matchfinder_base { unsigned long long partial_data_pos; uint8_t * buffer; /* input buffer */ @@ -238,52 +238,52 @@ struct Matchfinder_base int saved_dictionary_size; /* dictionary_size restored by Mb_reset */ bool at_stream_end; /* stream_pos shows real end of file */ bool sync_flush_pending; - }; + } Matchfinder_base; -static bool Mb_normalize_pos( struct Matchfinder_base * const mb ); +static bool Mb_normalize_pos( Matchfinder_base * const mb ); -static bool Mb_init( struct Matchfinder_base * const mb, const int before_size, +static bool Mb_init( Matchfinder_base * const mb, const int before_size, const int dict_size, const int after_size, const int dict_factor, const int num_prev_positions23, const int pos_array_factor ); -static inline void Mb_free( struct Matchfinder_base * const mb ) +static inline void Mb_free( Matchfinder_base * const mb ) { free( mb->prev_positions ); free( mb->buffer ); } -static inline uint8_t Mb_peek( const struct Matchfinder_base * const mb, +static inline uint8_t Mb_peek( const Matchfinder_base * const mb, const int distance ) { return mb->buffer[mb->pos-distance]; } -static inline int Mb_available_bytes( const struct Matchfinder_base * const mb ) +static inline int Mb_available_bytes( const Matchfinder_base * const mb ) { return mb->stream_pos - mb->pos; } static inline unsigned long long -Mb_data_position( const struct Matchfinder_base * const mb ) +Mb_data_position( const Matchfinder_base * const mb ) { return mb->partial_data_pos + mb->pos; } -static inline void Mb_finish( struct Matchfinder_base * const mb ) +static inline void Mb_finish( Matchfinder_base * const mb ) { mb->at_stream_end = true; mb->sync_flush_pending = false; } -static inline bool Mb_data_finished( const struct Matchfinder_base * const mb ) +static inline bool Mb_data_finished( const Matchfinder_base * const mb ) { return mb->at_stream_end && mb->pos >= mb->stream_pos; } -static inline bool Mb_flushing_or_end( const struct Matchfinder_base * const mb ) +static inline bool Mb_flushing_or_end( const Matchfinder_base * const mb ) { return mb->at_stream_end || mb->sync_flush_pending; } -static inline int Mb_free_bytes( const struct Matchfinder_base * const mb ) +static inline int Mb_free_bytes( const Matchfinder_base * const mb ) { if( Mb_flushing_or_end( mb ) ) return 0; return mb->buffer_size - mb->stream_pos; } static inline bool -Mb_enough_available_bytes( const struct Matchfinder_base * const mb ) +Mb_enough_available_bytes( const Matchfinder_base * const mb ) { return mb->pos + mb->after_size <= mb->stream_pos || ( Mb_flushing_or_end( mb ) && mb->pos < mb->stream_pos ); } static inline const uint8_t * -Mb_ptr_to_current_pos( const struct Matchfinder_base * const mb ) +Mb_ptr_to_current_pos( const Matchfinder_base * const mb ) { return mb->buffer + mb->pos; } -static int Mb_write_data( struct Matchfinder_base * const mb, +static int Mb_write_data( Matchfinder_base * const mb, const uint8_t * const inbuf, const int size ) { const int sz = min( mb->buffer_size - mb->stream_pos, size ); @@ -293,7 +293,7 @@ static int Mb_write_data( struct Matchfinder_base * const mb, return sz; } -static inline int Mb_true_match_len( const struct Matchfinder_base * const mb, +static inline int Mb_true_match_len( const Matchfinder_base * const mb, const int index, const int distance ) { const uint8_t * const data = mb->buffer + mb->pos; @@ -303,7 +303,7 @@ static inline int Mb_true_match_len( const struct Matchfinder_base * const mb, return i; } -static inline bool Mb_move_pos( struct Matchfinder_base * const mb ) +static inline bool Mb_move_pos( Matchfinder_base * const mb ) { if( ++mb->cyclic_pos > mb->dictionary_size ) mb->cyclic_pos = 0; if( ++mb->pos >= mb->pos_limit ) return Mb_normalize_pos( mb ); @@ -311,9 +311,9 @@ static inline bool Mb_move_pos( struct Matchfinder_base * const mb ) } -struct Range_encoder +typedef struct Range_encoder { - struct Circular_buffer cb; + Circular_buffer cb; unsigned min_free_bytes; uint64_t low; unsigned long long partial_member_pos; @@ -321,13 +321,13 @@ struct Range_encoder unsigned ff_count; uint8_t cache; Lzip_header header; - }; + } Range_encoder; -static inline void Re_shift_low( struct Range_encoder * const renc ) +static inline void Re_shift_low( Range_encoder * const renc ) { if( renc->low >> 24 != 0xFF ) { - const bool carry = ( renc->low > 0xFFFFFFFFU ); + const bool carry = renc->low > 0xFFFFFFFFU; Cb_put_byte( &renc->cb, renc->cache + carry ); for( ; renc->ff_count > 0; --renc->ff_count ) Cb_put_byte( &renc->cb, 0xFF + carry ); @@ -337,7 +337,7 @@ static inline void Re_shift_low( struct Range_encoder * const renc ) renc->low = ( renc->low & 0x00FFFFFFU ) << 8; } -static inline void Re_reset( struct Range_encoder * const renc, +static inline void Re_reset( Range_encoder * const renc, const unsigned dictionary_size ) { Cb_reset( &renc->cb ); @@ -350,7 +350,7 @@ static inline void Re_reset( struct Range_encoder * const renc, int i; for( i = 0; i < Lh_size; ++i ) Cb_put_byte( &renc->cb, renc->header[i] ); } -static inline bool Re_init( struct Range_encoder * const renc, +static inline bool Re_init( Range_encoder * const renc, const unsigned dictionary_size, const unsigned min_free_bytes ) { @@ -361,17 +361,17 @@ static inline bool Re_init( struct Range_encoder * const renc, return true; } -static inline void Re_free( struct Range_encoder * const renc ) +static inline void Re_free( Range_encoder * const renc ) { Cb_free( &renc->cb ); } static inline unsigned long long -Re_member_position( const struct Range_encoder * const renc ) +Re_member_position( const Range_encoder * const renc ) { return renc->partial_member_pos + Cb_used_bytes( &renc->cb ) + renc->ff_count; } -static inline bool Re_enough_free_bytes( const struct Range_encoder * const renc ) +static inline bool Re_enough_free_bytes( const Range_encoder * const renc ) { return Cb_free_bytes( &renc->cb ) >= renc->min_free_bytes + renc->ff_count; } -static inline int Re_read_data( struct Range_encoder * const renc, +static inline int Re_read_data( Range_encoder * const renc, uint8_t * const out_buffer, const int out_size ) { const int size = Cb_read_data( &renc->cb, out_buffer, out_size ); @@ -379,7 +379,7 @@ static inline int Re_read_data( struct Range_encoder * const renc, return size; } -static inline void Re_flush( struct Range_encoder * const renc ) +static inline void Re_flush( Range_encoder * const renc ) { int i; for( i = 0; i < 5; ++i ) Re_shift_low( renc ); renc->low = 0; @@ -388,7 +388,7 @@ static inline void Re_flush( struct Range_encoder * const renc ) renc->cache = 0; } -static inline void Re_encode( struct Range_encoder * const renc, +static inline void Re_encode( Range_encoder * const renc, const int symbol, const int num_bits ) { unsigned mask; @@ -400,7 +400,7 @@ static inline void Re_encode( struct Range_encoder * const renc, } } -static inline void Re_encode_bit( struct Range_encoder * const renc, +static inline void Re_encode_bit( Range_encoder * const renc, Bit_model * const probability, const bool bit ) { const uint32_t bound = ( renc->range >> bit_model_total_bits ) * *probability; @@ -418,7 +418,7 @@ static inline void Re_encode_bit( struct Range_encoder * const renc, if( renc->range <= 0x00FFFFFFU ) { renc->range <<= 8; Re_shift_low( renc ); } } -static inline void Re_encode_tree3( struct Range_encoder * const renc, +static inline void Re_encode_tree3( Range_encoder * const renc, Bit_model bm[], const int symbol ) { bool bit = ( symbol >> 2 ) & 1; @@ -429,7 +429,7 @@ static inline void Re_encode_tree3( struct Range_encoder * const renc, Re_encode_bit( renc, &bm[model], symbol & 1 ); } -static inline void Re_encode_tree6( struct Range_encoder * const renc, +static inline void Re_encode_tree6( Range_encoder * const renc, Bit_model bm[], const unsigned symbol ) { bool bit = ( symbol >> 5 ) & 1; @@ -446,7 +446,7 @@ static inline void Re_encode_tree6( struct Range_encoder * const renc, Re_encode_bit( renc, &bm[model], symbol & 1 ); } -static inline void Re_encode_tree8( struct Range_encoder * const renc, +static inline void Re_encode_tree8( Range_encoder * const renc, Bit_model bm[], const int symbol ) { int model = 1; @@ -459,7 +459,7 @@ static inline void Re_encode_tree8( struct Range_encoder * const renc, } } -static inline void Re_encode_tree_reversed( struct Range_encoder * const renc, +static inline void Re_encode_tree_reversed( Range_encoder * const renc, Bit_model bm[], int symbol, const int num_bits ) { int model = 1; @@ -473,7 +473,7 @@ static inline void Re_encode_tree_reversed( struct Range_encoder * const renc, } } -static inline void Re_encode_matched( struct Range_encoder * const renc, +static inline void Re_encode_matched( Range_encoder * const renc, Bit_model bm[], unsigned symbol, unsigned match_byte ) { @@ -489,17 +489,17 @@ static inline void Re_encode_matched( struct Range_encoder * const renc, } } -static inline void Re_encode_len( struct Range_encoder * const renc, - struct Len_model * const lm, +static inline void Re_encode_len( Range_encoder * const renc, + Len_model * const lm, int symbol, const int pos_state ) { - bool bit = ( ( symbol -= min_match_len ) >= len_low_symbols ); + bool bit = ( symbol -= min_match_len ) >= len_low_symbols; Re_encode_bit( renc, &lm->choice1, bit ); if( !bit ) Re_encode_tree3( renc, lm->bm_low[pos_state], symbol ); else { - bit = ( ( symbol -= len_low_symbols ) >= len_mid_symbols ); + bit = ( symbol -= len_low_symbols ) >= len_mid_symbols; Re_encode_bit( renc, &lm->choice2, bit ); if( !bit ) Re_encode_tree3( renc, lm->bm_mid[pos_state], symbol ); @@ -512,9 +512,9 @@ static inline void Re_encode_len( struct Range_encoder * const renc, enum { max_marker_size = 16, num_rep_distances = 4 }; /* must be 4 */ -struct LZ_encoder_base +typedef struct LZ_encoder_base { - struct Matchfinder_base mb; + Matchfinder_base mb; unsigned long long member_size_limit; uint32_t crc; @@ -528,18 +528,18 @@ struct LZ_encoder_base Bit_model bm_dis_slot[len_states][1<<dis_slot_bits]; Bit_model bm_dis[modeled_distances-end_dis_model+1]; Bit_model bm_align[dis_align_size]; - struct Len_model match_len_model; - struct Len_model rep_len_model; - struct Range_encoder renc; + Len_model match_len_model; + Len_model rep_len_model; + Range_encoder renc; int reps[num_rep_distances]; State state; bool member_finished; - }; + } LZ_encoder_base; -static void LZeb_reset( struct LZ_encoder_base * const eb, +static void LZeb_reset( LZ_encoder_base * const eb, const unsigned long long member_size ); -static inline bool LZeb_init( struct LZ_encoder_base * const eb, +static inline bool LZeb_init( LZ_encoder_base * const eb, const int before_size, const int dict_size, const int after_size, const int dict_factor, const int num_prev_positions23, @@ -555,34 +555,34 @@ static inline bool LZeb_init( struct LZ_encoder_base * const eb, return true; } -static inline bool LZeb_member_finished( const struct LZ_encoder_base * const eb ) +static inline bool LZeb_member_finished( const LZ_encoder_base * const eb ) { return eb->member_finished && Cb_empty( &eb->renc.cb ); } -static inline void LZeb_free( struct LZ_encoder_base * const eb ) +static inline void LZeb_free( LZ_encoder_base * const eb ) { Re_free( &eb->renc ); Mb_free( &eb->mb ); } -static inline unsigned LZeb_crc( const struct LZ_encoder_base * const eb ) +static inline unsigned LZeb_crc( const LZ_encoder_base * const eb ) { return eb->crc ^ 0xFFFFFFFFU; } -static inline int LZeb_price_literal( const struct LZ_encoder_base * const eb, +static inline int LZeb_price_literal( const LZ_encoder_base * const eb, const uint8_t prev_byte, const uint8_t symbol ) { return price_symbol8( eb->bm_literal[get_lit_state(prev_byte)], symbol ); } -static inline int LZeb_price_matched( const struct LZ_encoder_base * const eb, +static inline int LZeb_price_matched( const LZ_encoder_base * const eb, const uint8_t prev_byte, const uint8_t symbol, const uint8_t match_byte ) { return price_matched( eb->bm_literal[get_lit_state(prev_byte)], symbol, match_byte ); } -static inline void LZeb_encode_literal( struct LZ_encoder_base * const eb, +static inline void LZeb_encode_literal( LZ_encoder_base * const eb, const uint8_t prev_byte, const uint8_t symbol ) { Re_encode_tree8( &eb->renc, eb->bm_literal[get_lit_state(prev_byte)], symbol ); } -static inline void LZeb_encode_matched( struct LZ_encoder_base * const eb, +static inline void LZeb_encode_matched( LZ_encoder_base * const eb, const uint8_t prev_byte, const uint8_t symbol, const uint8_t match_byte ) { Re_encode_matched( &eb->renc, eb->bm_literal[get_lit_state(prev_byte)], symbol, match_byte ); } -static inline void LZeb_encode_pair( struct LZ_encoder_base * const eb, +static inline void LZeb_encode_pair( LZ_encoder_base * const eb, const unsigned dis, const int len, const int pos_state ) { diff --git a/fast_encoder.c b/fast_encoder.c index bb6363a..8b0db25 100644 --- a/fast_encoder.c +++ b/fast_encoder.c @@ -17,7 +17,7 @@ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ -static int FLZe_longest_match_len( struct FLZ_encoder * const fe, int * const distance ) +static int FLZe_longest_match_len( FLZ_encoder * const fe, int * const distance ) { enum { len_limit = 16 }; int32_t * ptr0 = fe->eb.mb.pos_array + fe->eb.mb.cyclic_pos; @@ -58,7 +58,7 @@ static int FLZe_longest_match_len( struct FLZ_encoder * const fe, int * const di } -static bool FLZe_encode_member( struct FLZ_encoder * const fe ) +static bool FLZe_encode_member( FLZ_encoder * const fe ) { int rep = 0, i; State * const state = &fe->eb.state; diff --git a/fast_encoder.h b/fast_encoder.h index b9421f4..f4b2d3c 100644 --- a/fast_encoder.h +++ b/fast_encoder.h @@ -17,13 +17,13 @@ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ -struct FLZ_encoder +typedef struct FLZ_encoder { - struct LZ_encoder_base eb; + LZ_encoder_base eb; unsigned key4; /* key made from latest 4 bytes */ - }; + } FLZ_encoder; -static inline void FLZe_reset_key4( struct FLZ_encoder * const fe ) +static inline void FLZe_reset_key4( FLZ_encoder * const fe ) { int i; fe->key4 = 0; @@ -31,9 +31,9 @@ static inline void FLZe_reset_key4( struct FLZ_encoder * const fe ) fe->key4 = ( fe->key4 << 4 ) ^ fe->eb.mb.buffer[i]; } -static inline bool FLZe_update_and_move( struct FLZ_encoder * const fe, int n ) +static inline bool FLZe_update_and_move( FLZ_encoder * const fe, int n ) { - struct Matchfinder_base * const mb = &fe->eb.mb; + Matchfinder_base * const mb = &fe->eb.mb; while( --n >= 0 ) { if( Mb_available_bytes( mb ) >= 4 ) @@ -48,7 +48,7 @@ static inline bool FLZe_update_and_move( struct FLZ_encoder * const fe, int n ) return true; } -static inline bool FLZe_init( struct FLZ_encoder * const fe, +static inline bool FLZe_init( FLZ_encoder * const fe, const unsigned long long member_size ) { enum { before_size = 0, @@ -65,6 +65,6 @@ static inline bool FLZe_init( struct FLZ_encoder * const fe, member_size ); } -static inline void FLZe_reset( struct FLZ_encoder * const fe, +static inline void FLZe_reset( FLZ_encoder * const fe, const unsigned long long member_size ) { LZeb_reset( &fe->eb, member_size ); } diff --git a/ffexample.c b/ffexample.c index 826abcd..0c87046 100644 --- a/ffexample.c +++ b/ffexample.c @@ -53,7 +53,7 @@ static void show_help( void ) } -int ffcompress( struct LZ_Encoder * const encoder, +int ffcompress( LZ_Encoder * const encoder, FILE * const infile, FILE * const outfile ) { enum { buffer_size = 16384 }; @@ -79,7 +79,7 @@ int ffcompress( struct LZ_Encoder * const encoder, } -int ffdecompress( struct LZ_Decoder * const decoder, +int ffdecompress( LZ_Decoder * const decoder, FILE * const infile, FILE * const outfile ) { enum { buffer_size = 16384 }; @@ -105,8 +105,7 @@ int ffdecompress( struct LZ_Decoder * const decoder, } -int ffboth( struct LZ_Encoder * const encoder, - struct LZ_Decoder * const decoder, +int ffboth( LZ_Encoder * const encoder, LZ_Decoder * const decoder, FILE * const infile, FILE * const outfile ) { enum { buffer_size = 16384 }; @@ -147,8 +146,7 @@ int ffmmcompress( FILE * const infile, FILE * const outfile ) enum { buffer_size = 16384, member_size = 4096 }; uint8_t buffer[buffer_size]; bool done = false; - struct LZ_Encoder * const encoder = - LZ_compress_open( 65535, 16, member_size ); + LZ_Encoder * const encoder = LZ_compress_open( 65535, 16, member_size ); if( !encoder || LZ_compress_errno( encoder ) != LZ_ok ) { fputs( "ffexample: Not enough memory.\n", stderr ); LZ_compress_close( encoder ); return 1; } @@ -182,7 +180,7 @@ int ffmmcompress( FILE * const infile, FILE * const outfile ) for each line of text terminated by a newline character or by EOF. Return 0 if success, 1 if error. */ -int fflfcompress( struct LZ_Encoder * const encoder, +int fflfcompress( LZ_Encoder * const encoder, FILE * const infile, FILE * const outfile ) { enum { buffer_size = 16384 }; @@ -223,7 +221,7 @@ int fflfcompress( struct LZ_Encoder * const encoder, next member in case of data error, including the automatic removal of leading garbage. */ -int ffrsdecompress( struct LZ_Decoder * const decoder, +int ffrsdecompress( LZ_Decoder * const decoder, FILE * const infile, FILE * const outfile ) { enum { buffer_size = 16384 }; @@ -262,11 +260,11 @@ int main( const int argc, const char * const argv[] ) setmode( STDOUT_FILENO, O_BINARY ); #endif - struct LZ_Encoder * const encoder = LZ_compress_open( 65535, 16, INT64_MAX ); - struct LZ_Decoder * const decoder = LZ_decompress_open(); - FILE * const infile = ( argc >= 3 && strcmp( argv[2], "-" ) != 0 ) ? + LZ_Encoder * const encoder = LZ_compress_open( 65535, 16, INT64_MAX ); + LZ_Decoder * const decoder = LZ_decompress_open(); + FILE * const infile = (argc >= 3 && strcmp( argv[2], "-" ) != 0) ? fopen( argv[2], "rb" ) : stdin; - FILE * const outfile = ( argc >= 4 && strcmp( argv[3], "-" ) != 0 ) ? + FILE * const outfile = (argc >= 4 && strcmp( argv[3], "-" ) != 0) ? fopen( argv[3], "wb" ) : stdout; int retval; @@ -41,10 +41,10 @@ static void show_line( const uint8_t * const buffer, const int size ) } -static struct LZ_Encoder * xopen_encoder( const int dictionary_size ) +static LZ_Encoder * xopen_encoder( const int dictionary_size ) { const int match_len_limit = 16; - struct LZ_Encoder * const encoder = + LZ_Encoder * const encoder = LZ_compress_open( dictionary_size, match_len_limit, member_size ); if( !encoder || LZ_compress_errno( encoder ) != LZ_ok ) { @@ -63,9 +63,9 @@ static struct LZ_Encoder * xopen_encoder( const int dictionary_size ) } -static struct LZ_Decoder * xopen_decoder( void ) +static LZ_Decoder * xopen_decoder( void ) { - struct LZ_Decoder * const decoder = LZ_decompress_open(); + LZ_Decoder * const decoder = LZ_decompress_open(); if( !decoder || LZ_decompress_errno( decoder ) != LZ_ok ) { LZ_decompress_close( decoder ); @@ -76,8 +76,7 @@ static struct LZ_Decoder * xopen_decoder( void ) } -static void xclose_encoder( struct LZ_Encoder * const encoder, - const bool finish ) +static void xclose_encoder( LZ_Encoder * const encoder, const bool finish ) { if( finish ) { @@ -105,8 +104,7 @@ static void xclose_encoder( struct LZ_Encoder * const encoder, } -static void xclose_decoder( struct LZ_Decoder * const decoder, - const bool finish ) +static void xclose_decoder( LZ_Decoder * const decoder, const bool finish ) { if( finish ) { @@ -160,8 +158,8 @@ static const uint8_t * next_line( FILE * const file, int * const sizep ) static int check_sync_flush( FILE * const file, const int dictionary_size ) { - struct LZ_Encoder * const encoder = xopen_encoder( dictionary_size ); - struct LZ_Decoder * const decoder = xopen_decoder(); + LZ_Encoder * const encoder = xopen_encoder( dictionary_size ); + LZ_Decoder * const decoder = xopen_decoder(); int retval = 0; while( retval <= 1 ) /* test LZ_compress_sync_flush */ @@ -267,8 +265,8 @@ static int check_sync_flush( FILE * const file, const int dictionary_size ) */ static int check_members( FILE * const file, const int dictionary_size ) { - struct LZ_Encoder * const encoder = xopen_encoder( dictionary_size ); - struct LZ_Decoder * const decoder = xopen_decoder(); + LZ_Encoder * const encoder = xopen_encoder( dictionary_size ); + LZ_Decoder * const decoder = xopen_decoder(); int retval = 0; while( retval <= 1 ) /* test LZ_compress_restart_member */ @@ -362,7 +360,7 @@ int main( const int argc, const char * const argv[] ) ( strcmp( argv[1], "-m" ) == 0 || strcmp( argv[1], "-s" ) == 0 ) ) ? argv[1][1] : 0; const int first = opt ? 2 : 1; - const bool verbose = ( opt != 0 || argc > first + 1 ); + const bool verbose = opt != 0 || argc > first + 1; if( argc < 2 ) { @@ -94,16 +94,16 @@ static inline void Bm_init( Bit_model * const probability ) static inline void Bm_array_init( Bit_model bm[], const int size ) { int i; for( i = 0; i < size; ++i ) Bm_init( &bm[i] ); } -struct Len_model +typedef struct Len_model { Bit_model choice1; Bit_model choice2; Bit_model bm_low[pos_states][len_low_symbols]; Bit_model bm_mid[pos_states][len_mid_symbols]; Bit_model bm_high[len_high_symbols]; - }; + } Len_model; -static inline void Lm_init( struct Len_model * const lm ) +static inline void Lm_init( Len_model * const lm ) { Bm_init( &lm->choice1 ); Bm_init( &lm->choice2 ); @@ -39,14 +39,14 @@ struct LZ_Encoder { unsigned long long partial_in_size; unsigned long long partial_out_size; - struct LZ_encoder_base * lz_encoder_base; /* these 3 pointers make a */ - struct LZ_encoder * lz_encoder; /* polymorphic encoder */ - struct FLZ_encoder * flz_encoder; - enum LZ_Errno lz_errno; + LZ_encoder_base * lz_encoder_base; /* these 3 pointers make a */ + LZ_encoder * lz_encoder; /* polymorphic encoder */ + FLZ_encoder * flz_encoder; + LZ_Errno lz_errno; bool fatal; }; -static void LZ_Encoder_init( struct LZ_Encoder * const e ) +static void LZ_Encoder_init( LZ_Encoder * const e ) { e->partial_in_size = 0; e->partial_out_size = 0; @@ -62,16 +62,16 @@ struct LZ_Decoder { unsigned long long partial_in_size; unsigned long long partial_out_size; - struct Range_decoder * rdec; - struct LZ_decoder * lz_decoder; - enum LZ_Errno lz_errno; + Range_decoder * rdec; + LZ_decoder * lz_decoder; + LZ_Errno lz_errno; Lzip_header member_header; /* header of current member */ bool fatal; bool first_header; /* true until first header is read */ bool seeking; }; -static void LZ_Decoder_init( struct LZ_Decoder * const d ) +static void LZ_Decoder_init( LZ_Decoder * const d ) { int i; d->partial_in_size = 0; @@ -86,7 +86,7 @@ static void LZ_Decoder_init( struct LZ_Decoder * const d ) } -static bool check_encoder( struct LZ_Encoder * const e ) +static bool check_encoder( LZ_Encoder * const e ) { if( !e ) return false; if( !e->lz_encoder_base || ( !e->lz_encoder && !e->flz_encoder ) || @@ -96,7 +96,7 @@ static bool check_encoder( struct LZ_Encoder * const e ) } -static bool check_decoder( struct LZ_Decoder * const d ) +static bool check_decoder( LZ_Decoder * const d ) { if( !d ) return false; if( !d->rdec ) @@ -111,7 +111,7 @@ int LZ_api_version( void ) { return LZ_API_VERSION; } const char * LZ_version( void ) { return LZ_version_string; } -const char * LZ_strerror( const enum LZ_Errno lz_errno ) +const char * LZ_strerror( const LZ_Errno lz_errno ) { switch( lz_errno ) { @@ -138,13 +138,12 @@ int LZ_max_match_len_limit( void ) { return max_match_len; } /* --------------------- Compression Functions --------------------- */ -struct LZ_Encoder * LZ_compress_open( const int dictionary_size, - const int match_len_limit, - const unsigned long long member_size ) +LZ_Encoder * LZ_compress_open( const int dictionary_size, + const int match_len_limit, + const unsigned long long member_size ) { Lzip_header header; - struct LZ_Encoder * const e = - (struct LZ_Encoder *)malloc( sizeof (struct LZ_Encoder) ); + LZ_Encoder * const e = (LZ_Encoder *)malloc( sizeof (LZ_Encoder) ); if( !e ) return 0; LZ_Encoder_init( e ); if( !Lh_set_dictionary_size( header, dictionary_size ) || @@ -156,14 +155,14 @@ struct LZ_Encoder * LZ_compress_open( const int dictionary_size, { if( dictionary_size == 65535 && match_len_limit == 16 ) { - e->flz_encoder = (struct FLZ_encoder *)malloc( sizeof (struct FLZ_encoder) ); + e->flz_encoder = (FLZ_encoder *)malloc( sizeof (FLZ_encoder) ); if( e->flz_encoder && FLZe_init( e->flz_encoder, member_size ) ) { e->lz_encoder_base = &e->flz_encoder->eb; return e; } free( e->flz_encoder ); e->flz_encoder = 0; } else { - e->lz_encoder = (struct LZ_encoder *)malloc( sizeof (struct LZ_encoder) ); + e->lz_encoder = (LZ_encoder *)malloc( sizeof (LZ_encoder) ); if( e->lz_encoder && LZe_init( e->lz_encoder, Lh_get_dictionary_size( header ), match_len_limit, member_size ) ) { e->lz_encoder_base = &e->lz_encoder->eb; return e; } @@ -176,7 +175,7 @@ struct LZ_Encoder * LZ_compress_open( const int dictionary_size, } -int LZ_compress_close( struct LZ_Encoder * const e ) +int LZ_compress_close( LZ_Encoder * const e ) { if( !e ) return -1; if( e->lz_encoder_base ) @@ -187,7 +186,7 @@ int LZ_compress_close( struct LZ_Encoder * const e ) } -int LZ_compress_finish( struct LZ_Encoder * const e ) +int LZ_compress_finish( LZ_Encoder * const e ) { if( !check_encoder( e ) || e->fatal ) return -1; Mb_finish( &e->lz_encoder_base->mb ); @@ -205,7 +204,7 @@ int LZ_compress_finish( struct LZ_Encoder * const e ) } -int LZ_compress_restart_member( struct LZ_Encoder * const e, +int LZ_compress_restart_member( LZ_Encoder * const e, const unsigned long long member_size ) { if( !check_encoder( e ) || e->fatal ) return -1; @@ -224,7 +223,7 @@ int LZ_compress_restart_member( struct LZ_Encoder * const e, } -int LZ_compress_sync_flush( struct LZ_Encoder * const e ) +int LZ_compress_sync_flush( LZ_Encoder * const e ) { if( !check_encoder( e ) || e->fatal ) return -1; if( !e->lz_encoder_base->mb.at_stream_end ) @@ -233,13 +232,13 @@ int LZ_compress_sync_flush( struct LZ_Encoder * const e ) } -int LZ_compress_read( struct LZ_Encoder * const e, +int LZ_compress_read( LZ_Encoder * const e, uint8_t * const buffer, const int size ) { if( !check_encoder( e ) || e->fatal ) return -1; if( size < 0 ) return 0; - { struct LZ_encoder_base * const eb = e->lz_encoder_base; + { LZ_encoder_base * const eb = e->lz_encoder_base; int out_size = Re_read_data( &eb->renc, buffer, size ); /* minimize number of calls to encode_member */ if( out_size < size || size == 0 ) @@ -255,7 +254,7 @@ int LZ_compress_read( struct LZ_Encoder * const e, } -int LZ_compress_write( struct LZ_Encoder * const e, +int LZ_compress_write( LZ_Encoder * const e, const uint8_t * const buffer, const int size ) { if( !check_encoder( e ) || e->fatal ) return -1; @@ -263,21 +262,21 @@ int LZ_compress_write( struct LZ_Encoder * const e, } -int LZ_compress_write_size( struct LZ_Encoder * const e ) +int LZ_compress_write_size( LZ_Encoder * const e ) { if( !check_encoder( e ) || e->fatal ) return -1; return Mb_free_bytes( &e->lz_encoder_base->mb ); } -enum LZ_Errno LZ_compress_errno( struct LZ_Encoder * const e ) +LZ_Errno LZ_compress_errno( LZ_Encoder * const e ) { if( !e ) return LZ_bad_argument; return e->lz_errno; } -int LZ_compress_finished( struct LZ_Encoder * const e ) +int LZ_compress_finished( LZ_Encoder * const e ) { if( !check_encoder( e ) ) return -1; return Mb_data_finished( &e->lz_encoder_base->mb ) && @@ -285,35 +284,35 @@ int LZ_compress_finished( struct LZ_Encoder * const e ) } -int LZ_compress_member_finished( struct LZ_Encoder * const e ) +int LZ_compress_member_finished( LZ_Encoder * const e ) { if( !check_encoder( e ) ) return -1; return LZeb_member_finished( e->lz_encoder_base ); } -unsigned long long LZ_compress_data_position( struct LZ_Encoder * const e ) +unsigned long long LZ_compress_data_position( LZ_Encoder * const e ) { if( !check_encoder( e ) ) return 0; return Mb_data_position( &e->lz_encoder_base->mb ); } -unsigned long long LZ_compress_member_position( struct LZ_Encoder * const e ) +unsigned long long LZ_compress_member_position( LZ_Encoder * const e ) { if( !check_encoder( e ) ) return 0; return Re_member_position( &e->lz_encoder_base->renc ); } -unsigned long long LZ_compress_total_in_size( struct LZ_Encoder * const e ) +unsigned long long LZ_compress_total_in_size( LZ_Encoder * const e ) { if( !check_encoder( e ) ) return 0; return e->partial_in_size + Mb_data_position( &e->lz_encoder_base->mb ); } -unsigned long long LZ_compress_total_out_size( struct LZ_Encoder * const e ) +unsigned long long LZ_compress_total_out_size( LZ_Encoder * const e ) { if( !check_encoder( e ) ) return 0; return e->partial_out_size + Re_member_position( &e->lz_encoder_base->renc ); @@ -322,14 +321,13 @@ unsigned long long LZ_compress_total_out_size( struct LZ_Encoder * const e ) /* -------------------- Decompression Functions -------------------- */ -struct LZ_Decoder * LZ_decompress_open( void ) +LZ_Decoder * LZ_decompress_open( void ) { - struct LZ_Decoder * const d = - (struct LZ_Decoder *)malloc( sizeof (struct LZ_Decoder) ); + LZ_Decoder * const d = (LZ_Decoder *)malloc( sizeof (LZ_Decoder) ); if( !d ) return 0; LZ_Decoder_init( d ); - d->rdec = (struct Range_decoder *)malloc( sizeof (struct Range_decoder) ); + d->rdec = (Range_decoder *)malloc( sizeof (Range_decoder) ); if( !d->rdec || !Rd_init( d->rdec ) ) { if( d->rdec ) { Rd_free( d->rdec ); free( d->rdec ); d->rdec = 0; } @@ -339,7 +337,7 @@ struct LZ_Decoder * LZ_decompress_open( void ) } -int LZ_decompress_close( struct LZ_Decoder * const d ) +int LZ_decompress_close( LZ_Decoder * const d ) { if( !d ) return -1; if( d->lz_decoder ) @@ -350,7 +348,7 @@ int LZ_decompress_close( struct LZ_Decoder * const d ) } -int LZ_decompress_finish( struct LZ_Decoder * const d ) +int LZ_decompress_finish( LZ_Decoder * const d ) { if( !check_decoder( d ) || d->fatal ) return -1; if( d->seeking ) @@ -360,7 +358,7 @@ int LZ_decompress_finish( struct LZ_Decoder * const d ) } -int LZ_decompress_reset( struct LZ_Decoder * const d ) +int LZ_decompress_reset( LZ_Decoder * const d ) { if( !check_decoder( d ) ) return -1; if( d->lz_decoder ) @@ -376,7 +374,7 @@ int LZ_decompress_reset( struct LZ_Decoder * const d ) } -int LZ_decompress_sync_to_member( struct LZ_Decoder * const d ) +int LZ_decompress_sync_to_member( LZ_Decoder * const d ) { unsigned skipped = 0; if( !check_decoder( d ) ) return -1; @@ -395,7 +393,7 @@ int LZ_decompress_sync_to_member( struct LZ_Decoder * const d ) } -int LZ_decompress_read( struct LZ_Decoder * const d, +int LZ_decompress_read( LZ_Decoder * const d, uint8_t * const buffer, const int size ) { int result; @@ -467,7 +465,7 @@ int LZ_decompress_read( struct LZ_Decoder * const d, d->fatal = true; return -1; } - d->lz_decoder = (struct LZ_decoder *)malloc( sizeof (struct LZ_decoder) ); + d->lz_decoder = (LZ_decoder *)malloc( sizeof (LZ_decoder) ); if( !d->lz_decoder || !LZd_init( d->lz_decoder, d->rdec, Lh_get_dictionary_size( d->member_header ) ) ) { /* not enough free memory */ @@ -496,7 +494,7 @@ get_data: } -int LZ_decompress_write( struct LZ_Decoder * const d, +int LZ_decompress_write( LZ_Decoder * const d, const uint8_t * const buffer, const int size ) { int result; @@ -519,21 +517,21 @@ int LZ_decompress_write( struct LZ_Decoder * const d, } -int LZ_decompress_write_size( struct LZ_Decoder * const d ) +int LZ_decompress_write_size( LZ_Decoder * const d ) { if( !check_decoder( d ) || d->fatal ) return -1; return Rd_free_bytes( d->rdec ); } -enum LZ_Errno LZ_decompress_errno( struct LZ_Decoder * const d ) +LZ_Errno LZ_decompress_errno( LZ_Decoder * const d ) { if( !d ) return LZ_bad_argument; return d->lz_errno; } -int LZ_decompress_finished( struct LZ_Decoder * const d ) +int LZ_decompress_finished( LZ_Decoder * const d ) { if( !check_decoder( d ) || d->fatal ) return -1; return Rd_finished( d->rdec ) && @@ -541,28 +539,28 @@ int LZ_decompress_finished( struct LZ_Decoder * const d ) } -int LZ_decompress_member_finished( struct LZ_Decoder * const d ) +int LZ_decompress_member_finished( LZ_Decoder * const d ) { if( !check_decoder( d ) || d->fatal ) return -1; return d->lz_decoder && LZd_member_finished( d->lz_decoder ); } -int LZ_decompress_member_version( struct LZ_Decoder * const d ) +int LZ_decompress_member_version( LZ_Decoder * const d ) { if( !check_decoder( d ) ) return -1; return Lh_version( d->member_header ); } -int LZ_decompress_dictionary_size( struct LZ_Decoder * const d ) +int LZ_decompress_dictionary_size( LZ_Decoder * const d ) { if( !check_decoder( d ) ) return -1; return Lh_get_dictionary_size( d->member_header ); } -unsigned LZ_decompress_data_crc( struct LZ_Decoder * const d ) +unsigned LZ_decompress_data_crc( LZ_Decoder * const d ) { if( check_decoder( d ) && d->lz_decoder ) return LZd_crc( d->lz_decoder ); @@ -570,7 +568,7 @@ unsigned LZ_decompress_data_crc( struct LZ_Decoder * const d ) } -unsigned long long LZ_decompress_data_position( struct LZ_Decoder * const d ) +unsigned long long LZ_decompress_data_position( LZ_Decoder * const d ) { if( check_decoder( d ) && d->lz_decoder ) return LZd_data_position( d->lz_decoder ); @@ -578,21 +576,21 @@ unsigned long long LZ_decompress_data_position( struct LZ_Decoder * const d ) } -unsigned long long LZ_decompress_member_position( struct LZ_Decoder * const d ) +unsigned long long LZ_decompress_member_position( LZ_Decoder * const d ) { if( !check_decoder( d ) ) return 0; return d->rdec->member_position; } -unsigned long long LZ_decompress_total_in_size( struct LZ_Decoder * const d ) +unsigned long long LZ_decompress_total_in_size( LZ_Decoder * const d ) { if( !check_decoder( d ) ) return 0; return d->partial_in_size + d->rdec->member_position; } -unsigned long long LZ_decompress_total_out_size( struct LZ_Decoder * const d ) +unsigned long long LZ_decompress_total_out_size( LZ_Decoder * const d ) { if( !check_decoder( d ) ) return 0; if( d->lz_decoder ) @@ -26,16 +26,17 @@ extern "C" { #define LZ_API_VERSION 1015 -static const char * const LZ_version_string = "1.15-pre2"; +static const char * const LZ_version_string = "1.15-rc1"; -enum LZ_Errno { LZ_ok = 0, LZ_bad_argument, LZ_mem_error, - LZ_sequence_error, LZ_header_error, LZ_unexpected_eof, - LZ_data_error, LZ_library_error }; +typedef enum LZ_Errno + { LZ_ok = 0, LZ_bad_argument, LZ_mem_error, + LZ_sequence_error, LZ_header_error, LZ_unexpected_eof, + LZ_data_error, LZ_library_error } LZ_Errno; int LZ_api_version( void ); /* new in 1.12 */ const char * LZ_version( void ); -const char * LZ_strerror( const enum LZ_Errno lz_errno ); +const char * LZ_strerror( const LZ_Errno lz_errno ); int LZ_min_dictionary_bits( void ); int LZ_min_dictionary_size( void ); @@ -47,63 +48,63 @@ int LZ_max_match_len_limit( void ); /* --------------------- Compression Functions --------------------- */ -struct LZ_Encoder; +typedef struct LZ_Encoder LZ_Encoder; -struct LZ_Encoder * LZ_compress_open( const int dictionary_size, - const int match_len_limit, - const unsigned long long member_size ); -int LZ_compress_close( struct LZ_Encoder * const encoder ); +LZ_Encoder * LZ_compress_open( const int dictionary_size, + const int match_len_limit, + const unsigned long long member_size ); +int LZ_compress_close( LZ_Encoder * const encoder ); -int LZ_compress_finish( struct LZ_Encoder * const encoder ); -int LZ_compress_restart_member( struct LZ_Encoder * const encoder, +int LZ_compress_finish( LZ_Encoder * const encoder ); +int LZ_compress_restart_member( LZ_Encoder * const encoder, const unsigned long long member_size ); -int LZ_compress_sync_flush( struct LZ_Encoder * const encoder ); +int LZ_compress_sync_flush( LZ_Encoder * const encoder ); -int LZ_compress_read( struct LZ_Encoder * const encoder, +int LZ_compress_read( LZ_Encoder * const encoder, uint8_t * const buffer, const int size ); -int LZ_compress_write( struct LZ_Encoder * const encoder, +int LZ_compress_write( LZ_Encoder * const encoder, const uint8_t * const buffer, const int size ); -int LZ_compress_write_size( struct LZ_Encoder * const encoder ); +int LZ_compress_write_size( LZ_Encoder * const encoder ); -enum LZ_Errno LZ_compress_errno( struct LZ_Encoder * const encoder ); -int LZ_compress_finished( struct LZ_Encoder * const encoder ); -int LZ_compress_member_finished( struct LZ_Encoder * const encoder ); +LZ_Errno LZ_compress_errno( LZ_Encoder * const encoder ); +int LZ_compress_finished( LZ_Encoder * const encoder ); +int LZ_compress_member_finished( LZ_Encoder * const encoder ); -unsigned long long LZ_compress_data_position( struct LZ_Encoder * const encoder ); -unsigned long long LZ_compress_member_position( struct LZ_Encoder * const encoder ); -unsigned long long LZ_compress_total_in_size( struct LZ_Encoder * const encoder ); -unsigned long long LZ_compress_total_out_size( struct LZ_Encoder * const encoder ); +unsigned long long LZ_compress_data_position( LZ_Encoder * const encoder ); +unsigned long long LZ_compress_member_position( LZ_Encoder * const encoder ); +unsigned long long LZ_compress_total_in_size( LZ_Encoder * const encoder ); +unsigned long long LZ_compress_total_out_size( LZ_Encoder * const encoder ); /* -------------------- Decompression Functions -------------------- */ -struct LZ_Decoder; +typedef struct LZ_Decoder LZ_Decoder; -struct LZ_Decoder * LZ_decompress_open( void ); -int LZ_decompress_close( struct LZ_Decoder * const decoder ); +LZ_Decoder * LZ_decompress_open( void ); +int LZ_decompress_close( LZ_Decoder * const decoder ); -int LZ_decompress_finish( struct LZ_Decoder * const decoder ); -int LZ_decompress_reset( struct LZ_Decoder * const decoder ); -int LZ_decompress_sync_to_member( struct LZ_Decoder * const decoder ); +int LZ_decompress_finish( LZ_Decoder * const decoder ); +int LZ_decompress_reset( LZ_Decoder * const decoder ); +int LZ_decompress_sync_to_member( LZ_Decoder * const decoder ); -int LZ_decompress_read( struct LZ_Decoder * const decoder, +int LZ_decompress_read( LZ_Decoder * const decoder, uint8_t * const buffer, const int size ); -int LZ_decompress_write( struct LZ_Decoder * const decoder, +int LZ_decompress_write( LZ_Decoder * const decoder, const uint8_t * const buffer, const int size ); -int LZ_decompress_write_size( struct LZ_Decoder * const decoder ); +int LZ_decompress_write_size( LZ_Decoder * const decoder ); -enum LZ_Errno LZ_decompress_errno( struct LZ_Decoder * const decoder ); -int LZ_decompress_finished( struct LZ_Decoder * const decoder ); -int LZ_decompress_member_finished( struct LZ_Decoder * const decoder ); +LZ_Errno LZ_decompress_errno( LZ_Decoder * const decoder ); +int LZ_decompress_finished( LZ_Decoder * const decoder ); +int LZ_decompress_member_finished( LZ_Decoder * const decoder ); -int LZ_decompress_member_version( struct LZ_Decoder * const decoder ); -int LZ_decompress_dictionary_size( struct LZ_Decoder * const decoder ); -unsigned LZ_decompress_data_crc( struct LZ_Decoder * const decoder ); +int LZ_decompress_member_version( LZ_Decoder * const decoder ); +int LZ_decompress_dictionary_size( LZ_Decoder * const decoder ); +unsigned LZ_decompress_data_crc( LZ_Decoder * const decoder ); -unsigned long long LZ_decompress_data_position( struct LZ_Decoder * const decoder ); -unsigned long long LZ_decompress_member_position( struct LZ_Decoder * const decoder ); -unsigned long long LZ_decompress_total_in_size( struct LZ_Decoder * const decoder ); -unsigned long long LZ_decompress_total_out_size( struct LZ_Decoder * const decoder ); +unsigned long long LZ_decompress_data_position( LZ_Decoder * const decoder ); +unsigned long long LZ_decompress_member_position( LZ_Decoder * const decoder ); +unsigned long long LZ_decompress_total_in_size( LZ_Decoder * const decoder ); +unsigned long long LZ_decompress_total_out_size( LZ_Decoder * const decoder ); #ifdef __cplusplus } @@ -99,13 +99,13 @@ static const struct { const char * from; const char * to; } known_extensions[] = { ".tlz", ".tar" }, { 0, 0 } }; -struct Lzma_options +typedef struct Lzma_options { int dictionary_size; /* 4 KiB .. 512 MiB */ int match_len_limit; /* 5 .. 273 */ - }; + } Lzma_options; -enum Mode { m_compress, m_decompress, m_test }; +typedef enum Mode { m_compress, m_decompress, m_test } Mode; /* Variables used in signal handler context. They are not declared volatile because the handler never returns. */ @@ -116,25 +116,26 @@ static bool delete_output_on_interrupt = false; static void show_help( void ) { - printf( "Minilzip is a test program for the compression library lzlib, compatible\n" - "(interoperable) with lzip 1.4 or newer.\n" + printf( "Minilzip is a test program for the compression library lzlib. Minilzip is\n" + "not intended to be installed because lzip has more features, but minilzip is\n" + "well tested and you can use it as your main compressor if so you wish.\n" "\nLzip is a lossless data compressor with a user interface similar to the one\n" - "of gzip or bzip2. Lzip uses a simplified form of the 'Lempel-Ziv-Markov\n" - "chain-Algorithm' (LZMA) stream format to maximize interoperability. The\n" - "maximum dictionary size is 512 MiB so that any lzip file can be decompressed\n" - "on 32-bit machines. Lzip provides accurate and robust 3-factor integrity\n" - "checking. Lzip can compress about as fast as gzip (lzip -0) or compress most\n" - "files more than bzip2 (lzip -9). Decompression speed is intermediate between\n" - "gzip and bzip2. Lzip provides better data recovery capabilities than gzip\n" - "and bzip2. Lzip has been designed, written, and tested with great care to\n" - "replace gzip and bzip2 as the standard general-purpose compressed format for\n" - "Unix-like systems.\n" + "of gzip or bzip2. Lzip uses a simplified form of LZMA (Lempel-Ziv-Markov\n" + "chain-Algorithm) designed to achieve complete interoperability between\n" + "implementations. The maximum dictionary size is 512 MiB so that any lzip\n" + "file can be decompressed on 32-bit machines. Lzip provides accurate and\n" + "robust 3-factor integrity checking. 'lzip -0' compresses about as fast as\n" + "gzip, while 'lzip -9' compresses most files more than bzip2. Decompression\n" + "speed is intermediate between gzip and bzip2. Lzip provides better data\n" + "recovery capabilities than gzip and bzip2. Lzip has been designed, written,\n" + "and tested with great care to replace gzip and bzip2 as general-purpose\n" + "compressed format for Unix-like systems.\n" "\nUsage: %s [options] [files]\n", invocation_name ); printf( "\nOptions:\n" " -h, --help display this help and exit\n" " -V, --version output version information and exit\n" " -a, --trailing-error exit with error status if trailing data\n" - " -b, --member-size=<bytes> set member size limit in bytes\n" + " -b, --member-size=<bytes> set member size limit of multimember files\n" " -c, --stdout write to standard output, keep input files\n" " -d, --decompress decompress, test compressed file integrity\n" " -f, --force overwrite existing output files\n" @@ -264,16 +265,16 @@ static void * resize_buffer( void * buf, const unsigned min_size ) } -struct Pretty_print +typedef struct Pretty_print { const char * name; char * padded_name; const char * stdin_name; unsigned longest_name; bool first_post; - }; + } Pretty_print; -static void Pp_init( struct Pretty_print * const pp, +static void Pp_init( Pretty_print * const pp, const char * const filenames[], const int num_filenames ) { pp->name = 0; @@ -294,11 +295,10 @@ static void Pp_init( struct Pretty_print * const pp, if( pp->longest_name == 0 ) pp->longest_name = stdin_name_len; } -void Pp_free( struct Pretty_print * const pp ) +void Pp_free( Pretty_print * const pp ) { if( pp->padded_name ) { free( pp->padded_name ); pp->padded_name = 0; } } -static void Pp_set_name( struct Pretty_print * const pp, - const char * const filename ) +static void Pp_set_name( Pretty_print * const pp, const char * const filename ) { unsigned name_len, padded_name_len, i = 0; @@ -316,10 +316,10 @@ static void Pp_set_name( struct Pretty_print * const pp, pp->first_post = true; } -static void Pp_reset( struct Pretty_print * const pp ) +static void Pp_reset( Pretty_print * const pp ) { if( pp->name && pp->name[0] ) pp->first_post = true; } -static void Pp_show_msg( struct Pretty_print * const pp, const char * const msg ) +static void Pp_show_msg( Pretty_print * const pp, const char * const msg ) { if( verbosity < 0 ) return; if( pp->first_post ) @@ -339,7 +339,7 @@ static void show_header( const unsigned dictionary_size ) const char * p = ""; const char * np = " "; unsigned num = dictionary_size; - bool exact = ( num % factor == 0 ); + bool exact = num % factor == 0; int i; for( i = 0; i < n && ( num > 9999 || ( exact && num >= factor ) ); ++i ) { num /= factor; if( num % factor != 0 ) exact = false; @@ -406,7 +406,7 @@ static unsigned long long getnum( const char * const arg, if( !errno && tail[0] ) { - const unsigned factor = ( tail[1] == 'i' ) ? 1024 : 1000; + const unsigned factor = (tail[1] == 'i') ? 1024 : 1000; int exponent = 0; /* 0 = bad multiplier */ int i; switch( tail[0] ) @@ -459,7 +459,7 @@ static int get_dict_size( const char * const arg, const char * const option_name } -static void set_mode( enum Mode * const program_modep, const enum Mode new_mode ) +static void set_mode( Mode * const program_modep, const Mode new_mode ) { if( *program_modep != m_compress && *program_modep != new_mode ) { @@ -524,7 +524,7 @@ static void set_d_outname( const char * const name, const int eindex ) static int open_instream( const char * const name, struct stat * const in_statsp, - const enum Mode program_mode, const int eindex, + const Mode program_mode, const int eindex, const bool one_to_one, const bool recompress ) { if( program_mode == m_compress && !recompress && eindex >= 0 ) @@ -541,9 +541,9 @@ static int open_instream( const char * const name, struct stat * const in_statsp { const int i = fstat( infd, in_statsp ); const mode_t mode = in_statsp->st_mode; - const bool can_read = ( i == 0 && - ( S_ISBLK( mode ) || S_ISCHR( mode ) || - S_ISFIFO( mode ) || S_ISSOCK( mode ) ) ); + const bool can_read = i == 0 && + ( S_ISBLK( mode ) || S_ISCHR( mode ) || + S_ISFIFO( mode ) || S_ISSOCK( mode ) ); if( i != 0 || ( !S_ISREG( mode ) && ( !can_read || one_to_one ) ) ) { if( verbosity >= 0 ) @@ -609,7 +609,7 @@ static void signal_handler( int sig ) static bool check_tty_in( const char * const input_filename, const int infd, - const enum Mode program_mode, int * const retval ) + const Mode program_mode, int * const retval ) { if( ( program_mode == m_decompress || program_mode == m_test ) && isatty( infd ) ) /* for example /dev/tty */ @@ -621,7 +621,7 @@ static bool check_tty_in( const char * const input_filename, const int infd, return true; } -static bool check_tty_out( const enum Mode program_mode ) +static bool check_tty_out( const Mode program_mode ) { if( program_mode == m_compress && isatty( outfd ) ) { show_file_error( output_filename[0] ? @@ -717,10 +717,10 @@ static bool next_filename( void ) } -static int do_compress( struct LZ_Encoder * const encoder, +static int do_compress( LZ_Encoder * const encoder, const unsigned long long member_size, const unsigned long long volume_size, const int infd, - struct Pretty_print * const pp, + Pretty_print * const pp, const struct stat * const in_statsp ) { unsigned long long partial_volume_size = 0; @@ -819,11 +819,11 @@ static int do_compress( struct LZ_Encoder * const encoder, static int compress( const unsigned long long member_size, const unsigned long long volume_size, const int infd, - const struct Lzma_options * const encoder_options, - struct Pretty_print * const pp, + const Lzma_options * const encoder_options, + Pretty_print * const pp, const struct stat * const in_statsp ) { - struct LZ_Encoder * const encoder = + LZ_Encoder * const encoder = LZ_compress_open( encoder_options->dictionary_size, encoder_options->match_len_limit, ( volume_size > 0 ) ? min( member_size, volume_size ) : member_size ); @@ -844,8 +844,8 @@ static int compress( const unsigned long long member_size, } -static int do_decompress( struct LZ_Decoder * const decoder, const int infd, - struct Pretty_print * const pp, const bool from_stdin, +static int do_decompress( LZ_Decoder * const decoder, const int infd, + Pretty_print * const pp, const bool from_stdin, const bool ignore_trailing, const bool loose_trailing, const bool testing ) { @@ -853,7 +853,7 @@ static int do_decompress( struct LZ_Decoder * const decoder, const int infd, uint8_t buffer[buffer_size]; /* read/write buffer */ unsigned long long total_in = 0; /* to detect library stall */ bool first_member; - bool empty = false, nonempty = false; + bool empty = false, multi = false; for( first_member = true; ; ) { @@ -894,10 +894,11 @@ static int do_decompress( struct LZ_Decoder * const decoder, const int infd, { const unsigned long long data_size = LZ_decompress_data_position( decoder ); if( !from_stdin ) - { if( data_size == 0 ) empty = true; else nonempty = true; } + { multi = !first_member; if( data_size == 0 ) empty = true; } if( verbosity >= 1 ) { - const unsigned long long member_size = LZ_decompress_member_position( decoder ); + const unsigned long long member_size = + LZ_decompress_member_position( decoder ); if( verbosity >= 2 || ( verbosity == 1 && first_member ) ) Pp_show_msg( pp, 0 ); if( verbosity >= 2 ) @@ -925,7 +926,7 @@ static int do_decompress( struct LZ_Decoder * const decoder, const int infd, if( out_size < 0 || ( first_member && out_size == 0 ) ) { const unsigned long long member_pos = LZ_decompress_member_position( decoder ); - const enum LZ_Errno lz_errno = LZ_decompress_errno( decoder ); + const LZ_Errno lz_errno = LZ_decompress_errno( decoder ); if( lz_errno == LZ_library_error ) internal_error( "library error (LZ_decompress_read)." ); if( member_pos <= 6 ) @@ -988,17 +989,17 @@ static int do_decompress( struct LZ_Decoder * const decoder, const int infd, } } if( verbosity == 1 ) fputs( testing ? "ok\n" : "done\n", stderr ); - if( empty && nonempty ) + if( empty && multi ) { show_file_error( pp->name, "Empty member not allowed.", 0 ); return 2; } return 0; } -static int decompress( const int infd, struct Pretty_print * const pp, +static int decompress( const int infd, Pretty_print * const pp, const bool from_stdin, const bool ignore_trailing, const bool loose_trailing, const bool testing ) { - struct LZ_Decoder * const decoder = LZ_decompress_open(); + LZ_Decoder * const decoder = LZ_decompress_open(); int retval; if( !decoder || LZ_decompress_errno( decoder ) != LZ_ok ) @@ -1046,7 +1047,7 @@ int main( const int argc, const char * const argv[] ) { /* Mapping from gzip/bzip2 style 0..9 compression levels to the corresponding LZMA compression parameters. */ - const struct Lzma_options option_mapping[] = + const Lzma_options option_mapping[] = { { 65535, 16 }, /* -0 (65535,16 chooses fast encoder) */ { 1 << 20, 5 }, /* -1 */ @@ -1058,14 +1059,13 @@ int main( const int argc, const char * const argv[] ) { 1 << 24, 68 }, /* -7 */ { 3 << 23, 132 }, /* -8 */ { 1 << 25, 273 } }; /* -9 */ - struct Lzma_options encoder_options = option_mapping[6]; /* default = "-6" */ + Lzma_options encoder_options = option_mapping[6]; /* default = "-6" */ const unsigned long long max_member_size = 0x0008000000000000ULL; /* 2 PiB */ const unsigned long long max_volume_size = 0x4000000000000000ULL; /* 4 EiB */ unsigned long long member_size = max_member_size; unsigned long long volume_size = 0; const char * default_output_filename = ""; - enum Mode program_mode = m_compress; - int i; + Mode program_mode = m_compress; bool force = false; bool ignore_trailing = true; bool keep_input_files = false; @@ -1075,7 +1075,7 @@ int main( const int argc, const char * const argv[] ) if( argc > 0 ) invocation_name = argv[0]; enum { opt_chk = 256, opt_lt }; - const struct ap_Option options[] = + const ap_Option options[] = { { '0', "fast", ap_no }, { '1', 0, ap_no }, @@ -1109,7 +1109,7 @@ int main( const int argc, const char * const argv[] ) { 0, 0, ap_no } }; /* static because valgrind complains and memory management in C sucks */ - static struct Arg_parser parser; + static Arg_parser parser; if( !ap_init( &parser, argc, argv, options, 0 ) ) { show_error( mem_msg, 0, false ); return 1; } if( ap_error( &parser ) ) /* bad option */ @@ -1179,6 +1179,7 @@ int main( const int argc, const char * const argv[] ) filenames = resize_buffer( filenames, num_filenames * sizeof filenames[0] ); filenames[0] = "-"; + int i; bool filenames_given = false; for( i = 0; argind + i < ap_arguments( &parser ); ++i ) { @@ -1208,7 +1209,7 @@ int main( const int argc, const char * const argv[] ) if( !to_stdout && program_mode != m_test && ( filenames_given || to_file ) ) set_signals( signal_handler ); - static struct Pretty_print pp; + static Pretty_print pp; Pp_init( &pp, filenames, num_filenames ); int failed_tests = 0; diff --git a/testsuite/check.sh b/testsuite/check.sh index 21f6def..c52cceb 100755 --- a/testsuite/check.sh +++ b/testsuite/check.sh @@ -73,7 +73,7 @@ done "${LZIP}" -q -o out.lz nx_file [ $? = 1 ] || test_failed $LINENO [ ! -e out.lz ] || test_failed $LINENO -"${LZIP}" -qf -S100k -o out in in +"${LZIP}" -qf -S100k -o out in in # only one file with -o and -S [ $? = 1 ] || test_failed $LINENO { [ ! -e out ] && [ ! -e out.lz ] ; } || test_failed $LINENO # these are for code coverage @@ -143,7 +143,6 @@ cmp in out || test_failed $LINENO printf "to be overwritten" > out || framework_failure "${LZIP}" -df -o out < "${in_lz}" || test_failed $LINENO cmp in out || test_failed $LINENO -rm -f out || framework_failure "${LZIP}" -d -o ./- "${in_lz}" || test_failed $LINENO cmp in ./- || test_failed $LINENO rm -f ./- || framework_failure @@ -156,7 +155,7 @@ cp "${in_lz}" anyothername || framework_failure test_failed $LINENO cmp in out || test_failed $LINENO cmp in anyothername.out || test_failed $LINENO -rm -f out anyothername.out || framework_failure +rm -f anyothername.out || framework_failure "${LZIP}" -tq in "${in_lz}" [ $? = 2 ] || test_failed $LINENO @@ -222,7 +221,9 @@ touch empty em || framework_failure "${LZIP}" -0 em || test_failed $LINENO "${LZIP}" -dk em.lz || test_failed $LINENO cmp empty em || test_failed $LINENO -rm -f empty em || framework_failure +cat em.lz em.lz | "${LZIP}" -t || test_failed $LINENO +cat em.lz em.lz | "${LZIP}" -d > em || test_failed $LINENO +cmp empty em || test_failed $LINENO cat em.lz "${in_lz}" | "${LZIP}" -t || test_failed $LINENO cat em.lz "${in_lz}" | "${LZIP}" -d > out || test_failed $LINENO cmp in out || test_failed $LINENO @@ -352,6 +353,19 @@ rm -f in8 || framework_failure printf "\ntesting bad input..." +cat em.lz em.lz > ee.lz || framework_failure +"${LZIP}" -t < ee.lz || test_failed $LINENO +"${LZIP}" -d < ee.lz > em || test_failed $LINENO +cmp empty em || test_failed $LINENO +"${LZIP}" -tq ee.lz +[ $? = 2 ] || test_failed $LINENO +"${LZIP}" -dq ee.lz +[ $? = 2 ] || test_failed $LINENO +[ ! -e ee ] || test_failed $LINENO +"${LZIP}" -cdq ee.lz > em +[ $? = 2 ] || test_failed $LINENO +cmp empty em || test_failed $LINENO +rm -f empty em || framework_failure cat "${in_lz}" em.lz "${in_lz}" > inein.lz || framework_failure "${LZIP}" -t < inein.lz || test_failed $LINENO "${LZIP}" -d < inein.lz > out2 || test_failed $LINENO @@ -364,13 +378,13 @@ cmp in2 out2 || test_failed $LINENO "${LZIP}" -cdq inein.lz > out2 [ $? = 2 ] || test_failed $LINENO cmp in2 out2 || test_failed $LINENO -rm -f out2 inein.lz || framework_failure +rm -f out2 inein.lz em.lz || framework_failure headers='LZIp LZiP LZip LzIP LzIp LziP lZIP lZIp lZiP lzIP' body='\001\014\000\000\101\376\367\377\377\340\000\200\000\215\357\002\322\001\000\000\000\000\000\000\000\045\000\000\000\000\000\000\000' cp "${in_lz}" int.lz || framework_failure printf "LZIP${body}" >> int.lz || framework_failure -if "${LZIP}" -tq int.lz ; then +if "${LZIP}" -t int.lz ; then for header in ${headers} ; do printf "${header}${body}" > int.lz || framework_failure "${LZIP}" -tq int.lz # first member @@ -407,7 +421,7 @@ if "${LZIP}" -tq int.lz ; then [ $? = 2 ] || test_failed $LINENO ${header} done else - printf "\nwarning: skipping header test: 'printf' does not work on your system." + printf "warning: skipping header test: 'printf' does not work on your system." fi rm -f int.lz || framework_failure @@ -425,12 +439,12 @@ for i in fox_bcrc.lz fox_crc0.lz fox_das46.lz fox_mes81.lz ; do [ $? = 2 ] || test_failed $LINENO $i cmp fox out || test_failed $LINENO $i done -rm -f fox out || framework_failure +rm -f fox || framework_failure cat "${in_lz}" "${in_lz}" > in2.lz || framework_failure cat "${in_lz}" "${in_lz}" "${in_lz}" > in3.lz || framework_failure if dd if=in3.lz of=trunc.lz bs=14682 count=1 2> /dev/null && - [ -e trunc.lz ] && cmp in2.lz trunc.lz > /dev/null 2>&1 ; then + [ -e trunc.lz ] && cmp in2.lz trunc.lz ; then for i in 6 20 14664 14683 14684 14685 14686 14687 14688 ; do dd if=in3.lz of=trunc.lz bs=$i count=1 2> /dev/null "${LZIP}" -tq trunc.lz @@ -443,7 +457,7 @@ if dd if=in3.lz of=trunc.lz bs=14682 count=1 2> /dev/null && [ $? = 2 ] || test_failed $LINENO $i done else - printf "\nwarning: skipping truncation test: 'dd' does not work on your system." + printf "warning: skipping truncation test: 'dd' does not work on your system." fi rm -f in2.lz in3.lz trunc.lz || framework_failure @@ -454,10 +468,12 @@ cat "${in_lz}" >> ingin.lz || framework_failure [ $? = 2 ] || test_failed $LINENO "${LZIP}" -atq < ingin.lz [ $? = 2 ] || test_failed $LINENO -"${LZIP}" -acdq ingin.lz > /dev/null +"${LZIP}" -acdq ingin.lz > out [ $? = 2 ] || test_failed $LINENO -"${LZIP}" -adq < ingin.lz > /dev/null +cmp in out || test_failed $LINENO +"${LZIP}" -adq < ingin.lz > out [ $? = 2 ] || test_failed $LINENO +cmp in out || test_failed $LINENO "${LZIP}" -t ingin.lz || test_failed $LINENO "${LZIP}" -t < ingin.lz || test_failed $LINENO "${LZIP}" -dk ingin.lz || test_failed $LINENO |