diff options
| -rw-r--r-- | AUTHORS | 8 | ||||
| -rw-r--r-- | COPYING | 26 | ||||
| -rw-r--r-- | ChangeLog | 135 | ||||
| -rw-r--r-- | INSTALL | 36 | ||||
| -rw-r--r-- | Makefile.in | 35 | ||||
| -rw-r--r-- | NEWS | 104 | ||||
| -rw-r--r-- | README | 116 | ||||
| -rw-r--r-- | bbexample.c | 373 | ||||
| -rw-r--r-- | carg_parser.c | 26 | ||||
| -rw-r--r-- | carg_parser.h | 66 | ||||
| -rw-r--r-- | cbuffer.c | 26 | ||||
| -rwxr-xr-x | configure | 25 | ||||
| -rw-r--r-- | decoder.c | 42 | ||||
| -rw-r--r-- | decoder.h | 58 | ||||
| -rw-r--r-- | doc/lzlib.info | 1397 | ||||
| -rw-r--r-- | doc/lzlib.texi | 902 | ||||
| -rw-r--r-- | doc/minilzip.1 | 38 | ||||
| -rw-r--r-- | encoder.c | 40 | ||||
| -rw-r--r-- | encoder.h | 26 | ||||
| -rw-r--r-- | encoder_base.c | 55 | ||||
| -rw-r--r-- | encoder_base.h | 65 | ||||
| -rw-r--r-- | fast_encoder.c | 32 | ||||
| -rw-r--r-- | fast_encoder.h | 42 | ||||
| -rw-r--r-- | ffexample.c | 300 | ||||
| -rw-r--r-- | lzcheck.c | 382 | ||||
| -rw-r--r-- | lzip.h | 35 | ||||
| -rw-r--r-- | lzlib.c | 64 | ||||
| -rw-r--r-- | lzlib.h | 38 | ||||
| -rw-r--r-- | main.c | 422 | ||||
| -rwxr-xr-x | testsuite/check.sh | 189 | ||||
| -rw-r--r-- | testsuite/fox.lz | bin | 0 -> 80 bytes | |||
| -rw-r--r-- | testsuite/fox_bcrc.lz | bin | 0 -> 80 bytes | |||
| -rw-r--r-- | testsuite/fox_crc0.lz | bin | 0 -> 80 bytes | |||
| -rw-r--r-- | testsuite/fox_das46.lz | bin | 0 -> 80 bytes | |||
| -rw-r--r-- | testsuite/fox_de20.lz | bin | 0 -> 80 bytes | |||
| -rw-r--r-- | testsuite/fox_lf (renamed from testsuite/test2.txt) | 0 | ||||
| -rw-r--r-- | testsuite/fox_mes81.lz | bin | 0 -> 80 bytes | |||
| -rw-r--r-- | testsuite/fox_s11.lz | bin | 0 -> 80 bytes | |||
| -rw-r--r-- | testsuite/fox_v2.lz | bin | 0 -> 80 bytes | |||
| -rw-r--r-- | testsuite/test_em.txt.lz | bin | 0 -> 14024 bytes |
40 files changed, 3235 insertions, 1868 deletions
@@ -1,7 +1,7 @@ Lzlib was written by Antonio Diaz Diaz. The ideas embodied in lzlib are due to (at least) the following people: -Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for -the definition of Markov chains), G.N.N. Martin (for the definition of -range encoding), Igor Pavlov (for putting all the above together in -LZMA), and Julian Seward (for bzip2's CLI). +Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for the +definition of Markov chains), G.N.N. Martin (for the definition of range +encoding), Igor Pavlov (for putting all the above together in LZMA), and +Julian Seward (for bzip2's CLI). @@ -1,17 +1,17 @@ - Lzlib - Compression library for the lzip format - Copyright (C) Antonio Diaz Diaz. + Lzlib - Compression library for the lzip format + Copyright (C) Antonio Diaz Diaz. - This library is free software. Redistribution and use in source and - binary forms, with or without modification, are permitted provided - that the following conditions are met: + This library is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. @@ -1,16 +1,55 @@ +2021-01-02 Antonio Diaz Diaz <antonio@gnu.org> + + * Version 1.12 released. + * lzlib.h: Define LZ_API_VERSION as 1000 * major + minor. 1.12 = 1012. + This change does not affect the soversion. + * lzlib.h, lzlib.c: New function LZ_api_version. + * LZd_try_verify_trailer: Return 2 if EOF at trailer or EOS marker. + * Decompression speed has been slightly increased. + * decoder.h: Increase 'rd_min_available_bytes' from 8 to 10. + * 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. + 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. + Replace 'decompressed', 'compressed' with 'out', 'in' in output. + Set a valid invocation_name even if argc == 0. + * lzlib.texi: Document the new way of verifying the library version. + Document that 'LZ_(de)compress_close' and 'LZ_(de)compress_errno' + can be called with a null argument. + Document that sync flush marker is not allowed in lzip files. + Document the consequences of not calling 'LZ_decompress_finish'. + Document that 'LZ_decompress_read' returns at least once per member. + Document that 'LZ_(de)compress_read' can be called with a null + buffer pointer argument. + Real code examples for common uses have been added to the tutorial. + * bbexample.c: Don't use 'LZ_(de)compress_write_size'. + * lzcheck.c: New options '-s' (sync) and '-m' (member by member). + Test member by member without 'LZ_decompress_finish'. + * ffexample.c: New file containing example functions for file-to-file + compression/decompression. + * Document extraction from tar.lz in '--help' output and man page. + * Makefile.in: 'install-bin' no longer installs the man page. + New targets 'install-bin-compress' and 'install-bin-strip-compress'. + * testsuite: Add 9 new test files. + 2019-01-02 Antonio Diaz Diaz <antonio@gnu.org> * Version 1.11 released. - * File_* renamed to Lzip_*. + * Rename File_* to Lzip_*. * LZ_decompress_read: Don't return error until all data is read. * decoder.c (LZd_decode_member): Decode truncated data until EOF. + * cbuffer.c (Cb_read_data): Allow a null buffer pointer. * main.c: Don't allow mixing different operations (-d and -t). * main.c: Check return value of close( infd ). * main.c: Compile on DOS with DJGPP. - * lzlib.texi: Improved descriptions of '-0..-9', '-m' and '-s'. + * lzlib.texi: Improve descriptions of '-0..-9', '-m', and '-s'. Document that 'LZ_(de)compress_finish' can be called repeatedly. * configure: Accept appending to CFLAGS, 'CFLAGS+=OPTIONS'. - * Makefile.in: Targets 'install-bin*' renamed to 'install-lib*'. + * Makefile.in: Rename targets 'install-bin*' to 'install-lib*'. * Makefile.in: Targets 'install-bin*' now install minilzip. * INSTALL: Document use of CFLAGS+='-D __USE_MINGW_ANSI_STDIO'. @@ -19,13 +58,13 @@ * Version 1.10 released. * LZ_compress_finish now adjusts dictionary size for each member. * lzlib.c (LZ_decompress_read): Detect corrupt header with HD=3. - * main.c: Added new option '--loose-trailing'. + * main.c: New option '--loose-trailing'. * main.c (main): Option '-S, --volume-size' now keeps input files. - * main.c: Replaced 'bits/byte' with inverse compression ratio. + * 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). - * lzlib.texi: Added chapter 'Invoking minilzip'. + * lzlib.texi: New chapter 'Invoking minilzip'. 2017-04-11 Antonio Diaz Diaz <antonio@gnu.org> @@ -34,14 +73,15 @@ * Compression time of options -1 to -9 has been reduced by 1%. * Decompression time has been reduced by 3%. * main.c: Continue testing if any input file is a terminal. - * License of the library changed to "2-clause BSD". + * Change the license of the library to "2-clause BSD". 2016-05-17 Antonio Diaz Diaz <antonio@gnu.org> * Version 1.8 released. * lzlib.h: Define LZ_API_VERSION to 1. - * decoder.c (LZd_verify_trailer): Removed test of final code. - * main.c: Added new option '-a, --trailing-error'. + * 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. * configure: Avoid warning on some shells when testing for gcc. @@ -52,26 +92,25 @@ 2015-07-08 Antonio Diaz Diaz <antonio@gnu.org> * Version 1.7 released. - * Ported fast encoder and option '-0' from lzip. + * Port fast encoder and option '-0' from lzip. * If open-->write-->finish, produce same dictionary size as lzip. - * Makefile.in: Added new targets 'install*-compress'. + * Makefile.in: New targets 'install*-compress'. 2014-08-27 Antonio Diaz Diaz <antonio@gnu.org> * Version 1.6 released. * Compression ratio of option '-9' has been slightly increased. - * configure: Added new option '--disable-static'. - * configure: Added new option '--disable-ldconfig'. + * configure: New options '--disable-static' and '--disable-ldconfig'. * Makefile.in: Ignore errors from ldconfig. * Makefile.in: Use 'CFLAGS' in every invocation of 'CC'. * main.c (close_and_set_permissions): Behave like 'cp -p'. - * lzlib.texinfo: Renamed to lzlib.texi. - * License changed to "GPL version 2 or later with link exception". + * lzlib.texinfo: Rename to lzlib.texi. + * Change license to "GPL version 2 or later with link exception". 2013-09-15 Antonio Diaz Diaz <antonio@gnu.org> * Version 1.5 released. - * Removed decompression support for version 0 files. + * Remove decompression support for version 0 files. * The LZ_compress_sync_flush mechanism has been fixed (again). * Minor fixes. @@ -82,12 +121,11 @@ * Compression ratio has been slightly increased. * Compression time has been reduced by 8%. * Decompression time has been reduced by 7%. - * lzlib.h: Changed 'long long' values to 'unsigned long long'. + * lzlib.h: Change 'long long' values to 'unsigned long long'. * encoder.c (Mf_init): Reduce minimum buffer size to 64KiB. * lzlib.c (LZ_decompress_read): Tell LZ_header_error from LZ_unexpected_eof the same way as lzip does. - * Makefile.in: Added new target 'install-as-lzip'. - * Makefile.in: Added new target 'install-bin'. + * 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. @@ -95,7 +133,7 @@ * Version 1.3 released. * Translated to C from the C++ source of lzlib 1.2. - * configure: 'datadir' renamed to 'datarootdir'. + * configure: Rename 'datadir' to 'datarootdir'. 2011-10-25 Antonio Diaz Diaz <ant_diaz@teleline.es> @@ -104,10 +142,9 @@ independently of the value of 'pos_state'. This gives better compression for large values of '--match-length' without being slower. - * encoder.h encoder.cc: Optimize pair price calculations. This - reduces compression time for large values of '--match-length' - by up to 6%. - * main.cc: Added new option '-F, --recompress'. + * encoder.h, encoder.cc: Optimize pair price calculations, reducing + compression time for large values of '--match-length' by up to 6%. + * main.cc: New option '-F, --recompress'. * Makefile.in: 'make install' no longer tries to run '/sbin/ldconfig' on systems lacking it. @@ -117,7 +154,7 @@ * Compression time has been reduced by 2%. * All declarations not belonging to the API have been encapsulated in the namespace 'Lzlib'. - * testsuite: 'test1' renamed to 'test.txt'. Added new tests. + * 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. @@ -127,14 +164,11 @@ 2010-05-08 Antonio Diaz Diaz <ant_diaz@teleline.es> * Version 1.0 released. - * Added new function LZ_decompress_member_finished. - * Added new function LZ_decompress_member_version. - * Added new function LZ_decompress_dictionary_size. - * Added new function LZ_decompress_data_crc. - * Variables declared 'extern' have been encapsulated in a - namespace. - * main.cc: Fixed warning about fchown's return value being ignored. - * decoder.h: Input_buffer integrated in Range_decoder. + * New functions LZ_decompress_member_version, LZ_decompress_data_crc, + LZ_decompress_member_finished, and LZ_decompress_dictionary_size. + * Variables declared 'extern' have been encapsulated in a namespace. + * main.cc: Fix warning about fchown's return value being ignored. + * decoder.h: Integrate Input_buffer in Range_decoder. 2010-02-10 Antonio Diaz Diaz <ant_diaz@teleline.es> @@ -145,28 +179,26 @@ 2010-01-17 Antonio Diaz Diaz <ant_diaz@teleline.es> * Version 0.8 released. - * Added new function LZ_decompress_reset. - * Added new function LZ_decompress_sync_to_member. - * Added new function LZ_decompress_write_size. - * Added new function LZ_strerror. - * lzlib.h: API change. Replaced 'enum' with functions for values - of dictionary size limits to make interface names consistent. - * lzlib.h: API change. 'LZ_errno' replaced with 'LZ_Errno'. - * lzlib.h: API change. Replaced 'void *' with 'struct LZ_Encoder *' + * New functions LZ_decompress_reset, LZ_decompress_sync_to_member, + LZ_decompress_write_size, and LZ_strerror. + * lzlib.h: API change. Replace 'enum' with functions for values of + dictionary size limits to make interface names consistent. + * lzlib.h: API change. Rename 'LZ_errno' to 'LZ_Errno'. + * lzlib.h: API change. Replace 'void *' with 'struct LZ_Encoder *' and 'struct LZ_Decoder *' to make interface type safe. - * decoder.cc: Truncated member trailer is now correctly detected. + * decoder.cc: A truncated member trailer is now correctly detected. * encoder.cc: Matchfinder::reset now also clears at_stream_end_, allowing LZ_compress_restart_member to restart a finished stream. * lzlib.cc: Accept only query or close operations after a fatal error has occurred. - * Shared version of lzlib is no longer built by default. + * The shared version of lzlib is no longer built by default. * check.sh: Use 'test1' instead of 'COPYING' for testing. 2009-10-20 Antonio Diaz Diaz <ant_diaz@teleline.es> * Version 0.7 released. * Compression time has been reduced by 4%. - * check.sh: Removed -9 to run in less than 256MiB of RAM. + * check.sh: Remove -9 to run in less than 256MiB of RAM. * lzcheck.cc: Read files of any size up to 2^63 bytes. 2009-09-02 Antonio Diaz Diaz <ant_diaz@teleline.es> @@ -178,15 +210,14 @@ * Version 0.5 released. * Decompression speed has been improved. - * main.cc (signal_handler): Declared as 'extern "C"'. + * main.cc (signal_handler): Declare as 'extern "C"'. 2009-06-03 Antonio Diaz Diaz <ant_diaz@teleline.es> * Version 0.4 released. - * Added new function LZ_compress_sync_flush. - * Added new function LZ_compress_write_size. + * New functions LZ_compress_sync_flush and LZ_compress_write_size. * Decompression speed has been improved. - * Added chapter 'Buffering' to the manual. + * lzlib.texinfo: New chapter 'Buffering'. 2009-05-03 Antonio Diaz Diaz <ant_diaz@teleline.es> @@ -196,16 +227,16 @@ 2009-04-26 Antonio Diaz Diaz <ant_diaz@teleline.es> * Version 0.2 released. - * Fixed a segfault when decompressing trailing garbage. - * Fixed a false positive in LZ_(de)compress_finished. + * Fix a segfault when decompressing trailing garbage. + * Fix a false positive in LZ_(de)compress_finished. 2009-04-21 Antonio Diaz Diaz <ant_diaz@teleline.es> * Version 0.1 released. -Copyright (C) 2009-2019 Antonio Diaz Diaz. +Copyright (C) 2009-2021 Antonio Diaz Diaz. This file is a collection of facts, and thus it is not copyrightable, -but just in case, you have unlimited permission to copy, distribute and +but just in case, you have unlimited permission to copy, distribute, and modify it. @@ -1,13 +1,13 @@ Requirements ------------ -You will need a C compiler. -I use gcc 5.3.0 and 4.1.2, but the code should compile with any standards +You will need a C99 compiler. (gcc 3.3.6 or newer is recommended). +I use gcc 6.1.0 and 4.1.2, but the code should compile with any standards compliant compiler. Gcc is available at http://gcc.gnu.org. The operating system must allow signal handlers read access to objects with static storage duration so that the cleanup handler for Control-C can delete -the partial output file. +the partial output file. (This requirement is for minilzip only). Procedure @@ -41,27 +41,29 @@ the main archive. documentation. (You may need to run ldconfig also). Or type 'make install-compress', which additionally compresses the - info manual after installation. (Installing compressed docs may - become the default in the future). + info manual after installation. + (Installing compressed docs may become the default in the future). - You can install only the library, the info manual or the man page by - typing 'make install-lib', 'make install-info' or 'make install-man' - respectively. + You can install only the library or the info manual by typing + 'make install-lib' or 'make install-info' respectively. - 'make install-bin' installs the minilzip program and its man page. - 'install-bin' installs a shared minilzip if the shared library has + 'make install-bin install-man' installs the program minilzip and its man + page. 'install-bin' installs a shared minilzip if the shared library has been configured. Else it installs a static minilzip. - 'make install-as-lzip' runs 'make install-bin' and then links - minilzip to the name 'lzip'. + 'make install-bin-compress' additionally compresses the man page after + installation. + + 'make install-as-lzip' runs 'make install-bin' and then links minilzip to + the name 'lzip'. Another way ----------- You can also compile lzlib into a separate directory. -To do this, you must use a version of 'make' that supports the 'VPATH' -variable, such as GNU 'make'. 'cd' to the directory where you want the +To do this, you must use a version of 'make' that supports the variable +'VPATH', such as GNU 'make'. 'cd' to the directory where you want the object files and executables to go and run the 'configure' script. -'configure' automatically checks for the source code in '.', in '..' and +'configure' automatically checks for the source code in '.', in '..', and in the directory that 'configure' is in. 'configure' recognizes the option '--srcdir=DIR' to control where to @@ -72,7 +74,7 @@ After running 'configure', you can run 'make' and 'make install' as explained above. -Copyright (C) 2009-2019 Antonio Diaz Diaz. +Copyright (C) 2009-2021 Antonio Diaz Diaz. This file is free documentation: you have unlimited permission to copy, -distribute and modify it. +distribute, and modify it. diff --git a/Makefile.in b/Makefile.in index 1ac03b2..94e3770 100644 --- a/Makefile.in +++ b/Makefile.in @@ -15,8 +15,10 @@ objs = carg_parser.o main.o .PHONY : all install install-bin install-info install-man \ install-strip install-compress install-strip-compress \ install-bin-strip install-info-compress install-man-compress \ - install-lib install-lib-strip uninstall-lib install-as-lzip \ - uninstall uninstall-bin uninstall-info uninstall-man \ + install-bin-compress install-bin-strip-compress \ + install-lib install-lib-strip \ + install-as-lzip \ + uninstall uninstall-bin uninstall-lib uninstall-info uninstall-man \ doc info man check dist clean distclean all : $(progname_static) $(progname_shared) @@ -36,6 +38,9 @@ $(progname)_shared : $(objs) lib$(libname).so.$(pkgversion) bbexample : bbexample.o lib$(libname).a $(CC) $(LDFLAGS) $(CFLAGS) -o $@ bbexample.o lib$(libname).a +ffexample : ffexample.o lib$(libname).a + $(CC) $(LDFLAGS) $(CFLAGS) -o $@ ffexample.o lib$(libname).a + lzcheck : lzcheck.o lib$(libname).a $(CC) $(LDFLAGS) $(CFLAGS) -o $@ lzcheck.o lib$(libname).a @@ -57,6 +62,7 @@ lzlib.o : Makefile $(lzdeps) lzlib_sh.o : Makefile $(lzdeps) main.o : carg_parser.h lzlib.h bbexample.o : Makefile lzlib.h +ffexample.o : Makefile lzlib.h lzcheck.o : Makefile lzlib.h @@ -75,15 +81,17 @@ $(VPATH)/doc/$(progname).1 : $(progname) Makefile : $(VPATH)/configure $(VPATH)/Makefile.in ./config.status -check : $(progname) bbexample lzcheck +check : $(progname) bbexample ffexample lzcheck @$(VPATH)/testsuite/check.sh $(VPATH)/testsuite $(pkgversion) install : install-lib install-info install-strip : install-lib-strip install-info install-compress : install-lib install-info-compress install-strip-compress : install-lib-strip install-info-compress +install-bin-compress : install-bin install-man-compress +install-bin-strip-compress : install-bin-strip install-man-compress -install-bin : all install-man +install-bin : all if [ ! -d "$(DESTDIR)$(bindir)" ] ; then $(INSTALL_DIR) "$(DESTDIR)$(bindir)" ; fi $(INSTALL_PROGRAM) ./$(progname_lzip) "$(DESTDIR)$(bindir)/$(progname)" @@ -119,7 +127,7 @@ install-info : -rm -f "$(DESTDIR)$(infodir)/$(pkgname).info"* $(INSTALL_DATA) $(VPATH)/doc/$(pkgname).info "$(DESTDIR)$(infodir)/$(pkgname).info" -if $(CAN_RUN_INSTALLINFO) ; then \ - install-info --info-dir="$(DESTDIR)$(infodir)" "$(DESTDIR)$(infodir)/$(pkgname).info" ; \ + install-info --info-dir="$(DESTDIR)$(infodir)" "$(DESTDIR)$(infodir)/$(pkgname).info" ; \ fi install-info-compress : install-info @@ -139,7 +147,7 @@ install-as-lzip : install-bin uninstall : uninstall-info uninstall-lib -uninstall-bin : uninstall-man +uninstall-bin : -rm -f "$(DESTDIR)$(bindir)/$(progname)" uninstall-lib : @@ -151,7 +159,7 @@ uninstall-lib : uninstall-info : -if $(CAN_RUN_INSTALLINFO) ; then \ - install-info --info-dir="$(DESTDIR)$(infodir)" --remove "$(DESTDIR)$(infodir)/$(pkgname).info" ; \ + install-info --info-dir="$(DESTDIR)$(infodir)" --remove "$(DESTDIR)$(infodir)/$(pkgname).info" ; \ fi -rm -f "$(DESTDIR)$(infodir)/$(pkgname).info"* @@ -177,16 +185,19 @@ dist : doc $(DISTNAME)/*.c \ $(DISTNAME)/testsuite/check.sh \ $(DISTNAME)/testsuite/test.txt \ - $(DISTNAME)/testsuite/test2.txt \ + $(DISTNAME)/testsuite/fox_lf \ + $(DISTNAME)/testsuite/fox.lz \ + $(DISTNAME)/testsuite/fox_*.lz \ $(DISTNAME)/testsuite/test_sync.lz \ - $(DISTNAME)/testsuite/test.txt.lz + $(DISTNAME)/testsuite/test.txt.lz \ + $(DISTNAME)/testsuite/test_em.txt.lz rm -f $(DISTNAME) lzip -v -9 $(DISTNAME).tar clean : - -rm -f $(progname) $(objs) - -rm -f $(progname)_shared lzlib_sh.o *.so.$(pkgversion) - -rm -f bbexample bbexample.o lzcheck lzcheck.o lzlib.o *.a + -rm -f $(progname) $(objs) lzlib.o lib$(libname).a + -rm -f $(progname)_shared lzlib_sh.o lib$(libname).so* + -rm -f bbexample bbexample.o ffexample ffexample.o lzcheck lzcheck.o distclean : clean -rm -f Makefile config.status *.tar *.tar.lz @@ -1,30 +1,92 @@ -Changes in version 1.11: +Changes in version 1.12: -In case of decompression error caused by corrupt or truncated data, -LZ_decompress_read now does not signal the error immediately to the -application, but waits until all decoded bytes have been read. This -allows tools like tarlz to recover as much data as possible from damaged -members. +The value of the version test macro 'LZ_API_VERSION' is now defined as +1000 * major + minor. For version 1.12 it is 1012. +This change does not affect the soversion. -It is now an error to specify two different operations to minilzip in -the command line (--decompress and --test). +The new function 'LZ_api_version', which returns the LZ_API_VERSION of the +library object code being used, has been added to lzlib. -Minilzip now also checks for errors when closing the input file. +If end of file is found at member trailer or EOS marker, +'LZ_decompress_errno' now returns 'LZ_unexpected_eof' instead of +'LZ_data_error'. -Minilzip now compiles on DOS with DJGPP. (Patch from Robert Riebisch). +Decompression speed has been slightly increased. -The descriptions of '-0..-9', '-m' and '-s' in the manual have been -improved. +A bug has been fixed in minilzip that falsely reported a library stall when +decompressing a file with empty members. -It has been documented that it is safe to call 'LZ_(de)compress_finish' -as many times as needed. Examples in the manual have been simplified. +The new option '--check-lib', which compares the version of lzlib used to +compile minilzip with the version actually being used, has been added to +minilzip. -The configure script now accepts appending options to CFLAGS using the -syntax 'CFLAGS+=OPTIONS'. +Minilzip now reports an error if a file name is empty (minilzip -t ""). -Makefile targets 'install-bin*' have been renamed to 'install-lib*'. New -'install-bin*' targets have been created that install a shared minilzip, -or a static minilzip if the shared library has not been configured. +Option '-o, --output' now behaves like '-c, --stdout', but sending the +output unconditionally to a file instead of to standard output. See the new +description of '-o' in the manual. This change is not backwards compatible. +Therefore commands like: + minilzip -o foo.lz - bar < foo +must now be split into: + minilzip -o foo.lz - < foo + minilzip bar +or rewritten as: + minilzip - bar < foo > foo.lz -It has been documented in INSTALL the use of -CFLAGS+='-D __USE_MINGW_ANSI_STDIO' when compiling on MinGW. +When using '-c' or '-o', minilzip now checks whether the output is a +terminal only once. + +Minilzip now does not even open the output file if the input file is a terminal. + +The words 'decompressed' and 'compressed' have been replaced with the +shorter 'out' and 'in' in the verbose output of minilzip when decompressing +or testing. + +It has been documented in the manual that 'LZ_(de)compress_close' and +'LZ_(de)compress_errno' can be called with a null argument. + +It has been documented in the manual that the LZMA marker '3' ("Sync Flush" +marker) is not allowed in lzip files. Marker '3' 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 +files. + +It has been documented in the manual that not calling 'LZ_decompress_finish' +prevents lzlib from detecting a truncated member. + +It has been documented in the manual that 'LZ_decompress_read' returns at +least once per member so that 'LZ_decompress_member_finished' can be called +(and trailer data retrieved) for each member, even for empty members. +Therefore, 'LZ_decompress_read' returning 0 does not mean that the end of +the stream has been reached. + +It has been documented in the manual that 'LZ_(de)compress_read' can be +called with a null buffer pointer argument. + +Real code examples for the most common uses of the library have been added +to the tutorial. + +'bbexample.c' has been simplified to not use 'LZ_(de)compress_write_size'. + +'lzcheck' now accepts options '-s' (to check LZ_compress_sync_flush) and +'-m' (to check member by member decompression). + +'lzcheck.c' now also tests member by member decompression without +intermediate calls to 'LZ_decompress_finish'. + +The new file 'ffexample.c', containing example functions for file-to-file +compression/decompression, has been added to the distribution. + +The commands needed to extract files from a tar.lz archive have been +documented in the output of 'minilzip --help' and in the man page. + +'make install-bin' no longer installs the minilzip man page. This is to +prevent 'make install-bin install-man-compress' from installing the man page +twice before compressing it. + +The new targets 'install-bin-compress' and 'install-bin-strip-compress', +which install a (stripped) minilzip and a compressed man page, have been +added to the Makefile. + +9 new test files have been added to the testsuite. @@ -1,65 +1,64 @@ Description -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. Lzlib is written in C. +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. +Lzlib is written in C. 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 lziprecover program 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. + 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. +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 'main.c' and 'bbexample.c' from the -source distribution. +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 'main.c' from the source +distribution. Compression/decompression is done by repeatedly calling a couple of -read/write functions until all the data have been processed by the -library. This interface is safer and less error prone than the -traditional zlib interface. +read/write functions until all the data have been processed by the library. +This interface is safer and less error prone than the traditional zlib +interface. Compression/decompression is done when the read function is called. This -means the value returned by the position functions will not be updated -until a read call, even if a lot of data are written. If you want the -data to be compressed in advance, just call the read function with a -size equal to 0. +means the value returned by the position functions will not be updated until +a read call, even if a lot of data are written. If you want the data to be +compressed in advance, just call the read function with a size equal to 0. If all the data to be compressed are written in advance, lzlib will -automatically adjust the header of the compressed data to use the -largest dictionary size that does not exceed neither the data size nor -the limit given to LZ_compress_open. This feature reduces the amount of -memory needed for decompression and allows minilzip to produce identical -compressed output as lzip. +automatically adjust the header of the compressed data to use the largest +dictionary size that does not exceed neither the data size nor the limit +given to 'LZ_compress_open'. This feature reduces the amount of memory +needed for decompression and allows minilzip to produce identical compressed +output as lzip. -Lzlib will correctly decompress a data stream which is the concatenation -of two or more compressed data streams. The result is the concatenation -of the corresponding decompressed data streams. Integrity testing of -concatenated compressed data streams is also supported. +Lzlib will correctly decompress a data stream which is the concatenation of +two or more compressed data streams. The result is the concatenation of the +corresponding decompressed data streams. Integrity testing of concatenated +compressed data streams is also supported. Lzlib is able to compress and decompress streams of unlimited size by -automatically creating multimember output. The members so created are -large, about 2 PiB each. +automatically creating multimember output. The members so created are large, +about 2 PiB each. All the library functions are thread safe. The library does not install any signal handler. The decoder checks the consistency of the compressed @@ -67,16 +66,15 @@ data, so the library should never crash even in case of corrupted input. 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. +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. -Lzlib currently implements two variants of the LZMA algorithm; fast -(used by option '-0' of minilzip) and normal (used by all other -compression levels). +Lzlib currently implements two variants of the LZMA algorithm; fast (used by +option '-0' of minilzip) and normal (used by all other compression levels). The high compression of LZMA comes from combining two basic, well-proven compression ideas: sliding dictionaries (LZ77/78) and markov models (the @@ -85,20 +83,20 @@ similar order-0 entropy coder as its last stage) with segregation of contexts according to what the bits are used for. The ideas embodied in lzlib are due to (at least) the following people: -Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for -the definition of Markov chains), G.N.N. Martin (for the definition of -range encoding), Igor Pavlov (for putting all the above together in -LZMA), and Julian Seward (for bzip2's CLI). +Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for the +definition of Markov chains), G.N.N. Martin (for the definition of range +encoding), Igor Pavlov (for putting all the above together in LZMA), and +Julian Seward (for bzip2's CLI). -LANGUAGE NOTE: Uncompressed = not compressed = plain data; it may never -have been compressed. Decompressed is used to refer to data which have -undergone the process of decompression. +LANGUAGE NOTE: Uncompressed = not compressed = plain data; it may never have +been compressed. Decompressed is used to refer to data which have undergone +the process of decompression. -Copyright (C) 2009-2019 Antonio Diaz Diaz. +Copyright (C) 2009-2021 Antonio Diaz Diaz. This file is free documentation: you have unlimited permission to copy, -distribute and modify it. +distribute, and modify it. The file Makefile.in is a data file used by configure to produce the Makefile. It has the same copyright owner and permissions that configure diff --git a/bbexample.c b/bbexample.c index e953c35..96257ee 100644 --- a/bbexample.c +++ b/bbexample.c @@ -1,21 +1,20 @@ -/* Buffer to buffer example - Test program for the lzlib library - Copyright (C) 2010-2019 Antonio Diaz Diaz. +/* Buffer to buffer example - Test program for the library lzlib + Copyright (C) 2010-2021 Antonio Diaz Diaz. - This program is free software: you have unlimited permission - to copy, distribute and modify it. + This program is free software: you have unlimited permission + to copy, distribute, and modify it. - Usage is: - bbexample filename + Usage: bbexample filename - This program is an example of how buffer-to-buffer - compression/decompression can be implemented using lzlib. + This program is an example of how buffer-to-buffer + compression/decompression can be implemented using lzlib. */ +#define _FILE_OFFSET_BITS 64 + #include <errno.h> #include <limits.h> -#ifndef __cplusplus #include <stdbool.h> -#endif #include <stdint.h> #include <stdio.h> #include <stdlib.h> @@ -24,6 +23,10 @@ #include "lzlib.h" +#ifndef min + #define min(x,y) ((x) <= (y) ? (x) : (y)) +#endif + /* Returns the address of a malloc'd buffer containing the file data and the file size in '*file_sizep'. @@ -35,28 +38,26 @@ uint8_t * read_file( const char * const name, long * const file_sizep ) uint8_t * buffer, * tmp; FILE * const f = fopen( name, "rb" ); if( !f ) - { - fprintf( stderr, "bbexample: Can't open input file '%s': %s\n", - name, strerror( errno ) ); - return 0; - } + { fprintf( stderr, "bbexample: Can't open input file '%s': %s\n", + name, strerror( errno ) ); return 0; } buffer = (uint8_t *)malloc( buffer_size ); if( !buffer ) - { fputs( "bbexample: Not enough memory.\n", stderr ); return 0; } + { fputs( "bbexample: read_file: Not enough memory.\n", stderr ); + fclose( f ); return 0; } file_size = fread( buffer, 1, buffer_size, f ); while( file_size >= buffer_size ) { if( buffer_size >= LONG_MAX ) { fprintf( stderr, "bbexample: Input file '%s' is too large.\n", name ); - free( buffer ); return 0; + free( buffer ); fclose( f ); return 0; } buffer_size = ( buffer_size <= LONG_MAX / 2 ) ? 2 * buffer_size : LONG_MAX; tmp = (uint8_t *)realloc( buffer, buffer_size ); if( !tmp ) - { fputs( "bbexample: Not enough memory.\n", stderr ); - free( buffer ); return 0; } + { fputs( "bbexample: read_file: Not enough memory.\n", stderr ); + free( buffer ); fclose( f ); return 0; } buffer = tmp; file_size += fread( buffer + file_size, 1, buffer_size - file_size, f ); } @@ -64,7 +65,7 @@ uint8_t * read_file( const char * const name, long * const file_sizep ) { fprintf( stderr, "bbexample: Error reading file '%s': %s\n", name, strerror( errno ) ); - free( buffer ); return 0; + free( buffer ); fclose( f ); return 0; } fclose( f ); *file_sizep = file_size; @@ -72,13 +73,13 @@ uint8_t * read_file( const char * const name, long * const file_sizep ) } -/* Compresses 'size' bytes from 'data'. Returns the address of a - malloc'd buffer containing the compressed data and its size in - '*out_sizep'. - In case of error, returns 0 and does not modify '*out_sizep'. +/* Compresses 'insize' bytes from 'inbuf'. + Returns the address of a malloc'd buffer containing the compressed data, + and the size of the data in '*outlenp'. + In case of error, returns 0 and does not modify '*outlenp'. */ -uint8_t * bbcompress( const uint8_t * const data, const long size, - const int level, long * const out_sizep ) +uint8_t * bbcompressl( const uint8_t * const inbuf, const long insize, + const int level, long * const outlenp ) { struct Lzma_options { @@ -100,171 +101,267 @@ uint8_t * bbcompress( const uint8_t * const data, const long size, { 3 << 23, 132 }, /* -8 */ { 1 << 25, 273 } }; /* -9 */ struct Lzma_options encoder_options; - const unsigned long long member_size = 0x7FFFFFFFFFFFFFFFULL; /* INT64_MAX */ struct LZ_Encoder * encoder; - uint8_t * new_data; - const long delta_size = ( size / 4 ) + 64; /* size may be zero */ - long new_data_size = delta_size; /* initial size */ - long new_pos = 0; - long written = 0; + uint8_t * outbuf; + const long delta_size = ( insize / 4 ) + 64; /* insize may be zero */ + long outsize = delta_size; /* initial outsize */ + long inpos = 0; + long outpos = 0; bool error = false; if( level < 0 || level > 9 ) return 0; encoder_options = option_mapping[level]; - if( encoder_options.dictionary_size > size && level != 0 ) - encoder_options.dictionary_size = size; /* saves memory */ + if( encoder_options.dictionary_size > insize && level != 0 ) + encoder_options.dictionary_size = insize; /* saves memory */ if( encoder_options.dictionary_size < LZ_min_dictionary_size() ) encoder_options.dictionary_size = LZ_min_dictionary_size(); encoder = LZ_compress_open( encoder_options.dictionary_size, - encoder_options.match_len_limit, member_size ); - if( !encoder || LZ_compress_errno( encoder ) != LZ_ok ) - { LZ_compress_close( encoder ); return 0; } - - new_data = (uint8_t *)malloc( new_data_size ); - if( !new_data ) - { LZ_compress_close( encoder ); return 0; } + encoder_options.match_len_limit, INT64_MAX ); + outbuf = (uint8_t *)malloc( outsize ); + if( !encoder || LZ_compress_errno( encoder ) != LZ_ok || !outbuf ) + { free( outbuf ); LZ_compress_close( encoder ); return 0; } while( true ) { - int rd; - if( LZ_compress_write_size( encoder ) > 0 ) - { - if( written < size ) - { - const int wr = LZ_compress_write( encoder, data + written, - size - written ); - if( wr < 0 ) { error = true; break; } - written += wr; - } - if( written >= size ) LZ_compress_finish( encoder ); - } - rd = LZ_compress_read( encoder, new_data + new_pos, - new_data_size - new_pos ); - if( rd < 0 ) { error = true; break; } - new_pos += rd; + int ret = LZ_compress_write( encoder, inbuf + inpos, + min( INT_MAX, insize - inpos ) ); + if( ret < 0 ) { error = true; break; } + inpos += ret; + if( inpos >= insize ) LZ_compress_finish( encoder ); + ret = LZ_compress_read( encoder, outbuf + outpos, + min( INT_MAX, outsize - outpos ) ); + if( ret < 0 ) { error = true; break; } + outpos += ret; if( LZ_compress_finished( encoder ) == 1 ) break; - if( new_pos >= new_data_size ) + if( outpos >= outsize ) { uint8_t * tmp; - if( new_data_size > LONG_MAX - delta_size ) { error = true; break; } - new_data_size += delta_size; - tmp = (uint8_t *)realloc( new_data, new_data_size ); + if( outsize > LONG_MAX - delta_size ) { error = true; break; } + outsize += delta_size; + tmp = (uint8_t *)realloc( outbuf, outsize ); if( !tmp ) { error = true; break; } - new_data = tmp; + outbuf = tmp; } } if( LZ_compress_close( encoder ) < 0 ) error = true; - if( error ) { free( new_data ); return 0; } - *out_sizep = new_pos; - return new_data; + if( error ) { free( outbuf ); return 0; } + *outlenp = outpos; + return outbuf; } -/* Decompresses 'size' bytes from 'data'. Returns the address of a - malloc'd buffer containing the decompressed data and its size in - '*out_sizep'. - In case of error, returns 0 and does not modify '*out_sizep'. +/* Decompresses 'insize' bytes from 'inbuf'. + Returns the address of a malloc'd buffer containing the decompressed + data, and the size of the data in '*outlenp'. + In case of error, returns 0 and does not modify '*outlenp'. */ -uint8_t * bbdecompress( const uint8_t * const data, const long size, - long * const out_sizep ) +uint8_t * bbdecompressl( const uint8_t * const inbuf, const long insize, + long * const outlenp ) { struct LZ_Decoder * const decoder = LZ_decompress_open(); - uint8_t * new_data; - const long delta_size = size; /* size must be > zero */ - long new_data_size = delta_size; /* initial size */ - long new_pos = 0; - long written = 0; + const long delta_size = insize; /* insize must be > zero */ + long outsize = delta_size; /* initial outsize */ + uint8_t * outbuf = (uint8_t *)malloc( outsize ); + long inpos = 0; + long outpos = 0; bool error = false; - if( !decoder || LZ_decompress_errno( decoder ) != LZ_ok ) - { LZ_decompress_close( decoder ); return 0; } - - new_data = (uint8_t *)malloc( new_data_size ); - if( !new_data ) - { LZ_decompress_close( decoder ); return 0; } + if( !decoder || LZ_decompress_errno( decoder ) != LZ_ok || !outbuf ) + { free( outbuf ); LZ_decompress_close( decoder ); return 0; } while( true ) { - int rd; - if( LZ_decompress_write_size( decoder ) > 0 ) - { - if( written < size ) - { - const int wr = LZ_decompress_write( decoder, data + written, - size - written ); - if( wr < 0 ) { error = true; break; } - written += wr; - } - if( written >= size ) LZ_decompress_finish( decoder ); - } - rd = LZ_decompress_read( decoder, new_data + new_pos, - new_data_size - new_pos ); - if( rd < 0 ) { error = true; break; } - new_pos += rd; + int ret = LZ_decompress_write( decoder, inbuf + inpos, + min( INT_MAX, insize - inpos ) ); + if( ret < 0 ) { error = true; break; } + inpos += ret; + if( inpos >= insize ) LZ_decompress_finish( decoder ); + ret = LZ_decompress_read( decoder, outbuf + outpos, + min( INT_MAX, outsize - outpos ) ); + if( ret < 0 ) { error = true; break; } + outpos += ret; if( LZ_decompress_finished( decoder ) == 1 ) break; - if( new_pos >= new_data_size ) + if( outpos >= outsize ) { uint8_t * tmp; - if( new_data_size > LONG_MAX - delta_size ) { error = true; break; } - new_data_size += delta_size; - tmp = (uint8_t *)realloc( new_data, new_data_size ); + if( outsize > LONG_MAX - delta_size ) { error = true; break; } + outsize += delta_size; + tmp = (uint8_t *)realloc( outbuf, outsize ); if( !tmp ) { error = true; break; } - new_data = tmp; + outbuf = tmp; } } if( LZ_decompress_close( decoder ) < 0 ) error = true; - if( error ) { free( new_data ); return 0; } - *out_sizep = new_pos; - return new_data; + if( error ) { free( outbuf ); return 0; } + *outlenp = outpos; + return outbuf; } -int main( const int argc, const char * const argv[] ) +/* Test the whole file at all levels. */ +int full_test( const uint8_t * const inbuf, const long insize ) { - uint8_t * in_buffer; - long in_size = 0; int level; + for( level = 0; level <= 9; ++level ) + { + long midsize = 0, outsize = 0; + uint8_t * outbuf; + uint8_t * midbuf = bbcompressl( inbuf, insize, level, &midsize ); + if( !midbuf ) + { fputs( "bbexample: full_test: Not enough memory or compress error.\n", + stderr ); return 1; } - if( argc < 2 ) + outbuf = bbdecompressl( midbuf, midsize, &outsize ); + free( midbuf ); + if( !outbuf ) + { fputs( "bbexample: full_test: Not enough memory or decompress error.\n", + stderr ); return 1; } + + if( insize != outsize || + ( insize > 0 && memcmp( inbuf, outbuf, insize ) != 0 ) ) + { fputs( "bbexample: full_test: Decompressed data differs from original.\n", + stderr ); free( outbuf ); return 1; } + + free( outbuf ); + } + return 0; + } + + +/* Compresses 'insize' bytes from 'inbuf' to 'outbuf'. + Returns the size of the compressed data in '*outlenp'. + In case of error, or if 'outsize' is too small, returns false and does + not modify '*outlenp'. +*/ +bool bbcompress( const uint8_t * const inbuf, const int insize, + const int dictionary_size, const int match_len_limit, + uint8_t * const outbuf, const int outsize, + int * const outlenp ) + { + int inpos = 0, outpos = 0; + bool error = false; + struct 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; } + + while( true ) { - fputs( "Usage: bbexample filename\n", stderr ); - return 1; + int ret = LZ_compress_write( encoder, inbuf + inpos, insize - inpos ); + if( ret < 0 ) { error = true; break; } + inpos += ret; + if( inpos >= insize ) LZ_compress_finish( encoder ); + ret = LZ_compress_read( encoder, outbuf + outpos, outsize - outpos ); + if( ret < 0 ) { error = true; break; } + outpos += ret; + if( LZ_compress_finished( encoder ) == 1 ) break; + if( outpos >= outsize ) { error = true; break; } } - in_buffer = read_file( argv[1], &in_size ); - if( !in_buffer ) return 1; + if( LZ_compress_close( encoder ) < 0 ) error = true; + if( error ) return false; + *outlenp = outpos; + return true; + } - for( level = 0; level <= 9; ++level ) + +/* Decompresses 'insize' bytes from 'inbuf' to 'outbuf'. + Returns the size of the decompressed data in '*outlenp'. + In case of error, or if 'outsize' is too small, returns false and does + not modify '*outlenp'. +*/ +bool bbdecompress( const uint8_t * const inbuf, const int insize, + uint8_t * const outbuf, const int outsize, + int * const outlenp ) + { + int inpos = 0, outpos = 0; + bool error = false; + struct LZ_Decoder * const decoder = LZ_decompress_open(); + if( !decoder || LZ_decompress_errno( decoder ) != LZ_ok ) + { LZ_decompress_close( decoder ); return false; } + + while( true ) { - uint8_t * mid_buffer, * out_buffer; - long mid_size = 0, out_size = 0; + int ret = LZ_decompress_write( decoder, inbuf + inpos, insize - inpos ); + if( ret < 0 ) { error = true; break; } + inpos += ret; + if( inpos >= insize ) LZ_decompress_finish( decoder ); + ret = LZ_decompress_read( decoder, outbuf + outpos, outsize - outpos ); + if( ret < 0 ) { error = true; break; } + outpos += ret; + if( LZ_decompress_finished( decoder ) == 1 ) break; + if( outpos >= outsize ) { error = true; break; } + } - mid_buffer = bbcompress( in_buffer, in_size, level, &mid_size ); - if( !mid_buffer ) - { - fputs( "bbexample: Not enough memory or compress error.\n", stderr ); - return 1; - } + if( LZ_decompress_close( decoder ) < 0 ) error = true; + if( error ) return false; + *outlenp = outpos; + return true; + } - out_buffer = bbdecompress( mid_buffer, mid_size, &out_size ); - if( !out_buffer ) - { - fputs( "bbexample: Not enough memory or decompress error.\n", stderr ); - return 1; - } - if( in_size != out_size || - ( in_size > 0 && memcmp( in_buffer, out_buffer, in_size ) != 0 ) ) - { - fputs( "bbexample: Decompressed data differs from original.\n", stderr ); - return 1; - } +/* Test at most INT_MAX bytes from the file with buffers of fixed size. */ +int fixed_test( const uint8_t * const inbuf, const int insize ) + { + int dictionary_size = 65535; /* fast encoder */ + int midsize = min( INT_MAX, ( insize / 8 ) * 9LL + 44 ), outsize = insize; + uint8_t * midbuf = (uint8_t *)malloc( midsize ); + uint8_t * outbuf = (uint8_t *)malloc( outsize ); + if( !midbuf || !outbuf ) + { fputs( "bbexample: fixed_test: Not enough memory.\n", stderr ); + free( outbuf ); free( midbuf ); return 1; } + + for( ; dictionary_size <= 8 << 20; dictionary_size += 8323073 ) + { + int midlen, outlen; + if( !bbcompress( inbuf, insize, dictionary_size, 16, midbuf, midsize, &midlen ) ) + { fputs( "bbexample: fixed_test: Not enough memory or compress error.\n", + stderr ); free( outbuf ); free( midbuf ); return 1; } + + if( !bbdecompress( midbuf, midlen, outbuf, outsize, &outlen ) ) + { fputs( "bbexample: fixed_test: Not enough memory or decompress error.\n", + stderr ); free( outbuf ); free( midbuf ); return 1; } + + if( insize != outlen || + ( insize > 0 && memcmp( inbuf, outbuf, insize ) != 0 ) ) + { fputs( "bbexample: fixed_test: Decompressed data differs from original.\n", + stderr ); free( outbuf ); free( midbuf ); return 1; } - free( out_buffer ); - free( mid_buffer ); } - free( in_buffer ); + free( outbuf ); + free( midbuf ); return 0; } + + +int main( const int argc, const char * const argv[] ) + { + int retval = 0, i; + int open_failures = 0; + const bool verbose = ( argc > 2 ); + + if( argc < 2 ) + { + fputs( "Usage: bbexample filename\n", stderr ); + return 1; + } + + for( i = 1; i < argc && retval == 0; ++i ) + { + long insize; + uint8_t * const inbuf = read_file( argv[i], &insize ); + if( !inbuf ) { ++open_failures; continue; } + if( verbose ) fprintf( stderr, " Testing file '%s'\n", argv[i] ); + + retval = full_test( inbuf, insize ); + if( retval == 0 ) retval = fixed_test( inbuf, min( INT_MAX, insize ) ); + free( inbuf ); + } + if( open_failures > 0 && verbose ) + fprintf( stderr, "bbexample: warning: %d %s failed to open.\n", + open_failures, ( open_failures == 1 ) ? "file" : "files" ); + if( retval == 0 && open_failures ) retval = 1; + return retval; + } diff --git a/carg_parser.c b/carg_parser.c index ce01d7b..d0c05d5 100644 --- a/carg_parser.c +++ b/carg_parser.c @@ -1,20 +1,20 @@ -/* Arg_parser - POSIX/GNU command line argument parser. (C version) - Copyright (C) 2006-2019 Antonio Diaz Diaz. +/* Arg_parser - POSIX/GNU command line argument parser. (C version) + Copyright (C) 2006-2021 Antonio Diaz Diaz. - This library is free software. Redistribution and use in source and - binary forms, with or without modification, are permitted provided - that the following conditions are met: + This library is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ #include <stdlib.h> diff --git a/carg_parser.h b/carg_parser.h index dcae2de..c5f2352 100644 --- a/carg_parser.h +++ b/carg_parser.h @@ -1,44 +1,44 @@ -/* Arg_parser - POSIX/GNU command line argument parser. (C version) - Copyright (C) 2006-2019 Antonio Diaz Diaz. +/* Arg_parser - POSIX/GNU command line argument parser. (C version) + Copyright (C) 2006-2021 Antonio Diaz Diaz. - This library is free software. Redistribution and use in source and - binary forms, with or without modification, are permitted provided - that the following conditions are met: + This library is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ -/* Arg_parser reads the arguments in 'argv' and creates a number of - option codes, option arguments and non-option arguments. +/* Arg_parser reads the arguments in 'argv' and creates a number of + option codes, option arguments, and non-option arguments. - In case of error, 'ap_error' returns a non-null pointer to an error - message. + In case of error, 'ap_error' returns a non-null pointer to an error + message. - 'options' is an array of 'struct ap_Option' terminated by an element - containing a code which is zero. A null name means a short-only - option. A code value outside the unsigned char range means a - long-only option. + 'options' is an array of 'struct ap_Option' terminated by an element + containing a code which is zero. A null name means a short-only + option. A code value outside the unsigned char range means a + long-only option. - Arg_parser normally makes it appear as if all the option arguments - were specified before all the non-option arguments for the purposes - of parsing, even if the user of your program intermixed option and - non-option arguments. If you want the arguments in the exact order - the user typed them, call 'ap_init' with 'in_order' = true. + Arg_parser normally makes it appear as if all the option arguments + were specified before all the non-option arguments for the purposes + of parsing, even if the user of your program intermixed option and + non-option arguments. If you want the arguments in the exact order + the user typed them, call 'ap_init' with 'in_order' = true. - The argument '--' terminates all options; any following arguments are - treated as non-option arguments, even if they begin with a hyphen. + The argument '--' terminates all options; any following arguments are + treated as non-option arguments, even if they begin with a hyphen. - The syntax for optional option arguments is '-<short_option><argument>' - (without whitespace), or '--<long_option>=<argument>'. + The syntax for optional option arguments is '-<short_option><argument>' + (without whitespace), or '--<long_option>=<argument>'. */ #ifdef __cplusplus @@ -79,11 +79,11 @@ void ap_free( struct Arg_parser * const ap ); const char * ap_error( const struct Arg_parser * const ap ); - /* The number of arguments parsed (may be different from argc) */ +/* The number of arguments parsed. May be different from argc. */ int ap_arguments( const struct 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). */ +/* 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 ); const char * ap_argument( const struct Arg_parser * const ap, const int i ); @@ -1,20 +1,20 @@ -/* Lzlib - Compression library for the lzip format - Copyright (C) 2009-2019 Antonio Diaz Diaz. +/* Lzlib - Compression library for the lzip format + Copyright (C) 2009-2021 Antonio Diaz Diaz. - This library is free software. Redistribution and use in source and - binary forms, with or without modification, are permitted provided - that the following conditions are met: + This library is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ struct Circular_buffer @@ -1,12 +1,12 @@ #! /bin/sh # configure script for Lzlib - Compression library for the lzip format -# Copyright (C) 2009-2019 Antonio Diaz Diaz. +# Copyright (C) 2009-2021 Antonio Diaz Diaz. # # This configure script is free software: you have unlimited permission -# to copy, distribute and modify it. +# to copy, distribute, and modify it. pkgname=lzlib -pkgversion=1.11 +pkgversion=1.12 soversion=1 progname=minilzip progname_static=${progname} @@ -34,11 +34,7 @@ CFLAGS='-Wall -W -O2' LDFLAGS= # checking whether we are using GNU C. -/bin/sh -c "${CC} --version" > /dev/null 2>&1 || - { - CC=cc - CFLAGS=-O2 - } +/bin/sh -c "${CC} --version" > /dev/null 2>&1 || { CC=cc ; CFLAGS=-O2 ; } # Loop over all args args= @@ -50,11 +46,12 @@ while [ $# != 0 ] ; do shift # Add the argument quoted to args - args="${args} \"${option}\"" + if [ -z "${args}" ] ; then args="\"${option}\"" + else args="${args} \"${option}\"" ; fi # Split out the argument for options that take them case ${option} in - *=*) optarg=`echo ${option} | sed -e 's,^[^=]*=,,;s,/$,,'` ;; + *=*) optarg=`echo "${option}" | sed -e 's,^[^=]*=,,;s,/$,,'` ;; esac # Process the options @@ -151,7 +148,7 @@ if [ -z "${srcdir}" ] ; then if [ ! -r "${srcdir}/${srctrigger}" ] ; then srcdir=.. ; fi if [ ! -r "${srcdir}/${srctrigger}" ] ; then ## the sed command below emulates the dirname command - srcdir=`echo $0 | sed -e 's,[^/]*$,,;s,/$,,;s,^$,.,'` + srcdir=`echo "$0" | sed -e 's,[^/]*$,,;s,/$,,;s,^$,.,'` fi fi @@ -174,7 +171,7 @@ if [ -z "${no_create}" ] ; then # Run this file to recreate the current configuration. # # This script is free software: you have unlimited permission -# to copy, distribute and modify it. +# to copy, distribute, and modify it. exec /bin/sh $0 ${args} --no-create EOF @@ -198,11 +195,11 @@ echo "LDFLAGS = ${LDFLAGS}" rm -f Makefile cat > Makefile << EOF # Makefile for Lzlib - Compression library for the lzip format -# Copyright (C) 2009-2019 Antonio Diaz Diaz. +# Copyright (C) 2009-2021 Antonio Diaz Diaz. # This file was generated automatically by configure. Don't edit. # # This Makefile is free software: you have unlimited permission -# to copy, distribute and modify it. +# to copy, distribute, and modify it. pkgname = ${pkgname} pkgversion = ${pkgversion} @@ -1,27 +1,27 @@ -/* Lzlib - Compression library for the lzip format - Copyright (C) 2009-2019 Antonio Diaz Diaz. +/* Lzlib - Compression library for the lzip format + Copyright (C) 2009-2021 Antonio Diaz Diaz. - This library is free software. Redistribution and use in source and - binary forms, with or without modification, are permitted provided - that the following conditions are met: + This library is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ static int LZd_try_verify_trailer( struct LZ_decoder * const d ) { Lzip_trailer trailer; - if( Rd_available_bytes( d->rdec ) < Lt_size && !d->rdec->at_stream_end ) - return 0; + if( Rd_available_bytes( d->rdec ) < Lt_size ) + { if( !d->rdec->at_stream_end ) return 0; else return 2; } d->verify_trailer_pending = false; d->member_finished = true; @@ -40,10 +40,10 @@ static int LZd_decode_member( struct LZ_decoder * const d ) { struct Range_decoder * const rdec = d->rdec; State * const state = &d->state; - /* unsigned old_mpos = d->rdec->member_position; */ + /* unsigned old_mpos = rdec->member_position; */ if( d->member_finished ) return 0; - if( !Rd_try_reload( rdec, false ) ) + if( !Rd_try_reload( rdec ) ) { if( !rdec->at_stream_end ) return 0; else return 2; } if( d->verify_trailer_pending ) return LZd_try_verify_trailer( d ); @@ -51,7 +51,7 @@ static int LZd_decode_member( struct LZ_decoder * const d ) { int len; const int pos_state = LZd_data_position( d ) & pos_state_mask; - /* const unsigned mpos = d->rdec->member_position; + /* const unsigned mpos = rdec->member_position; if( mpos - old_mpos > rd_min_available_bytes ) return 5; old_mpos = mpos; */ if( !Rd_enough_available_bytes( rdec ) ) /* check unexpected EOF */ @@ -123,6 +123,9 @@ static int LZd_decode_member( struct LZ_decoder * const d ) if( distance == 0xFFFFFFFFU ) /* marker found */ { Rd_normalize( rdec ); + /* const unsigned mpos = rdec->member_position; + if( mpos - old_mpos > rd_min_available_bytes ) return 5; + old_mpos = mpos; */ if( len == min_match_len ) /* End Of Stream marker */ { d->verify_trailer_pending = true; @@ -130,7 +133,8 @@ static int LZd_decode_member( struct LZ_decoder * const d ) } if( len == min_match_len + 1 ) /* Sync Flush marker */ { - if( Rd_try_reload( rdec, true ) ) { /*old_mpos += 5;*/ continue; } + rdec->reload_pending = true; + if( Rd_try_reload( rdec ) ) continue; else { if( !rdec->at_stream_end ) return 0; else break; } } return 4; @@ -1,23 +1,23 @@ -/* Lzlib - Compression library for the lzip format - Copyright (C) 2009-2019 Antonio Diaz Diaz. +/* Lzlib - Compression library for the lzip format + Copyright (C) 2009-2021 Antonio Diaz Diaz. - This library is free software. Redistribution and use in source and - binary forms, with or without modification, are permitted provided - that the following conditions are met: + This library is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ -enum { rd_min_available_bytes = 8 }; +enum { rd_min_available_bytes = 10 }; struct Range_decoder { @@ -133,9 +133,8 @@ static inline bool Rd_unread_data( struct Range_decoder * const rdec, return true; } -static bool Rd_try_reload( struct Range_decoder * const rdec, const bool force ) +static bool Rd_try_reload( struct Range_decoder * const rdec ) { - if( force ) rdec->reload_pending = true; if( rdec->reload_pending && Rd_available_bytes( rdec ) >= 5 ) { int i; @@ -168,7 +167,7 @@ static inline unsigned Rd_decode( struct Range_decoder * const rdec, /* symbol <<= 1; */ /* if( rdec->code >= rdec->range ) { rdec->code -= rdec->range; symbol |= 1; } */ bit = ( rdec->code >= rdec->range ); - symbol = ( symbol << 1 ) + bit; + symbol <<= 1; symbol += bit; rdec->code -= rdec->range & ( 0U - bit ); } return symbol; @@ -182,15 +181,15 @@ static inline unsigned Rd_decode_bit( struct Range_decoder * const rdec, bound = ( rdec->range >> bit_model_total_bits ) * *probability; if( rdec->code < bound ) { - rdec->range = bound; *probability += (bit_model_total - *probability) >> bit_model_move_bits; + rdec->range = bound; return 0; } else { - rdec->range -= bound; - rdec->code -= bound; *probability -= *probability >> bit_model_move_bits; + rdec->code -= bound; + rdec->range -= bound; return 1; } } @@ -198,8 +197,7 @@ static inline unsigned Rd_decode_bit( struct Range_decoder * const rdec, static inline unsigned Rd_decode_tree3( struct Range_decoder * const rdec, Bit_model bm[] ) { - unsigned symbol = 1; - symbol = ( symbol << 1 ) | Rd_decode_bit( rdec, &bm[symbol] ); + unsigned symbol = 2 | Rd_decode_bit( rdec, &bm[1] ); symbol = ( symbol << 1 ) | Rd_decode_bit( rdec, &bm[symbol] ); symbol = ( symbol << 1 ) | Rd_decode_bit( rdec, &bm[symbol] ); return symbol & 7; @@ -208,8 +206,7 @@ static inline unsigned Rd_decode_tree3( struct Range_decoder * const rdec, static inline unsigned Rd_decode_tree6( struct Range_decoder * const rdec, Bit_model bm[] ) { - unsigned symbol = 1; - symbol = ( symbol << 1 ) | Rd_decode_bit( rdec, &bm[symbol] ); + unsigned symbol = 2 | Rd_decode_bit( rdec, &bm[1] ); symbol = ( symbol << 1 ) | Rd_decode_bit( rdec, &bm[symbol] ); symbol = ( symbol << 1 ) | Rd_decode_bit( rdec, &bm[symbol] ); symbol = ( symbol << 1 ) | Rd_decode_bit( rdec, &bm[symbol] ); @@ -238,7 +235,7 @@ Rd_decode_tree_reversed( struct Range_decoder * const rdec, for( i = 0; i < num_bits; ++i ) { const unsigned bit = Rd_decode_bit( rdec, &bm[model] ); - model = ( model << 1 ) + bit; + model <<= 1; model += bit; symbol |= ( bit << i ); } return symbol; @@ -248,12 +245,9 @@ static inline unsigned Rd_decode_tree_reversed4( struct Range_decoder * const rdec, Bit_model bm[] ) { unsigned symbol = Rd_decode_bit( rdec, &bm[1] ); - unsigned model = 2 + symbol; - unsigned bit = Rd_decode_bit( rdec, &bm[model] ); - model = ( model << 1 ) + bit; symbol |= ( bit << 1 ); - bit = Rd_decode_bit( rdec, &bm[model] ); - model = ( model << 1 ) + bit; symbol |= ( bit << 2 ); - symbol |= ( Rd_decode_bit( rdec, &bm[model] ) << 3 ); + symbol += Rd_decode_bit( rdec, &bm[2+symbol] ) << 1; + symbol += Rd_decode_bit( rdec, &bm[4+symbol] ) << 2; + symbol += Rd_decode_bit( rdec, &bm[8+symbol] ) << 3; return symbol; } @@ -266,7 +260,7 @@ static inline unsigned Rd_decode_matched( struct Range_decoder * const rdec, { const unsigned match_bit = ( match_byte <<= 1 ) & mask; const unsigned bit = Rd_decode_bit( rdec, &bm[symbol+match_bit+mask] ); - symbol = ( symbol << 1 ) + bit; + symbol <<= 1; symbol += bit; if( symbol > 0xFF ) return symbol & 0xFF; mask &= ~(match_bit ^ (bit << 8)); /* if( match_bit != bit ) mask = 0; */ } diff --git a/doc/lzlib.info b/doc/lzlib.info index 5a1a573..bef1859 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 1.11, 2 January 2019). +This manual is for Lzlib (version 1.12, 2 January 2021). * Menu: @@ -30,10 +30,10 @@ This manual is for Lzlib (version 1.11, 2 January 2019). * Concept index:: Index of concepts - Copyright (C) 2009-2019 Antonio Diaz Diaz. + Copyright (C) 2009-2021 Antonio Diaz Diaz. - This manual is free documentation: you have unlimited permission to -copy, distribute and modify it. + This manual is free documentation: you have unlimited permission to copy, +distribute, and modify it. File: lzlib.info, Node: Introduction, Next: Library version, Prev: Top, Up: Top @@ -41,102 +41,99 @@ File: lzlib.info, Node: Introduction, Next: Library version, Prev: Top, Up: 1 Introduction ************** -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. Lzlib is written in C. +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. +Lzlib is written in C. The lzip file format is designed for data sharing and long-term -archiving, taking into account both data integrity and decoder -availability: +archiving, taking into account both data integrity and decoder availability: * The lzip format provides very safe integrity checking and some data - recovery means. The lziprecover program 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. + 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. +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 'main.c' and 'bbexample.c' from the +library are declared in the file 'lzlib.h'. Usage examples of the library +are given in the files 'bbexample.c', 'ffexample.c', and 'main.c' from the source distribution. Compression/decompression is done by repeatedly calling a couple of -read/write functions until all the data have been processed by the -library. This interface is safer and less error prone than the -traditional zlib interface. +read/write functions until all the data have been processed by the library. +This interface is safer and less error prone than the traditional zlib +interface. - Compression/decompression is done when the read function is called. -This means the value returned by the position functions will not be -updated until a read call, even if a lot of data are written. If you -want the data to be compressed in advance, just call the read function -with a SIZE equal to 0. + Compression/decompression is done when the read function is called. This +means the value returned by the position functions will not be updated until +a read call, even if a lot of data are written. If you want the data to be +compressed in advance, just call the read function with a SIZE equal to 0. If all the data to be compressed are written in advance, lzlib will -automatically adjust the header of the compressed data to use the -largest dictionary size that does not exceed neither the data size nor -the limit given to 'LZ_compress_open'. This feature reduces the amount -of memory needed for decompression and allows minilzip to produce -identical compressed output as lzip. +automatically adjust the header of the compressed data to use the largest +dictionary size that does not exceed neither the data size nor the limit +given to 'LZ_compress_open'. This feature reduces the amount of memory +needed for decompression and allows minilzip to produce identical compressed +output as lzip. - Lzlib will correctly decompress a data stream which is the -concatenation of two or more compressed data streams. The result is the -concatenation of the corresponding decompressed data streams. Integrity -testing of concatenated compressed data streams is also supported. + Lzlib will correctly decompress a data stream which is the concatenation +of two or more compressed data streams. The result is the concatenation of +the corresponding decompressed data streams. Integrity testing of +concatenated compressed data streams is also supported. Lzlib is able to compress and decompress streams of unlimited size by -automatically creating multimember output. The members so created are -large, about 2 PiB each. - - All the library functions are thread safe. The library does not -install any signal handler. The decoder checks the consistency of the -compressed data, so the library should never crash even in case of -corrupted input. - - 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. +automatically creating multimember output. The members so created are large, +about 2 PiB each. + + All the library functions are thread safe. The library does not install +any signal handler. The decoder checks the consistency of the compressed +data, so the library should never crash even in case of corrupted input. + + 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. Lzlib currently implements two variants of the LZMA algorithm; fast -(used by option '-0' of minilzip) and normal (used by all other -compression levels). +(used by option '-0' of minilzip) and normal (used by all other compression +levels). - The high compression of LZMA comes from combining two basic, -well-proven compression ideas: sliding dictionaries (LZ77/78) and -markov models (the thing used by every compression algorithm that uses -a range encoder or similar order-0 entropy coder as its last stage) -with segregation of contexts according to what the bits are used for. + The high compression of LZMA comes from combining two basic, well-proven +compression ideas: sliding dictionaries (LZ77/78) and markov models (the +thing used by every compression algorithm that uses a range encoder or +similar order-0 entropy coder as its last stage) with segregation of +contexts according to what the bits are used for. - The ideas embodied in lzlib are due to (at least) the following -people: Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey -Markov (for the definition of Markov chains), G.N.N. Martin (for the -definition of range encoding), Igor Pavlov (for putting all the above -together in LZMA), and Julian Seward (for bzip2's CLI). + The ideas embodied in lzlib are due to (at least) the following people: +Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for the +definition of Markov chains), G.N.N. Martin (for the definition of range +encoding), Igor Pavlov (for putting all the above together in LZMA), and +Julian Seward (for bzip2's CLI). - LANGUAGE NOTE: Uncompressed = not compressed = plain data; it may -never have been compressed. Decompressed is used to refer to data which -have undergone the process of decompression. + LANGUAGE NOTE: Uncompressed = not compressed = plain data; it may never +have been compressed. Decompressed is used to refer to data which have +undergone the process of decompression. File: lzlib.info, Node: Library version, Next: Buffering, Prev: Introduction, Up: Top @@ -144,19 +141,53 @@ File: lzlib.info, Node: Library version, Next: Buffering, Prev: Introduction, 2 Library version ***************** - -- Function: const char * LZ_version ( void ) - Returns the library version as a string. +One goal of lzlib is to keep perfect backward compatibility with older +versions of itself down to 1.0. Any application working with an older lzlib +should work with a newer lzlib. Installing a newer lzlib should not break +anything. This chapter describes the constants and functions that the +application can use to discover the version of the library being used. + + -- Constant: LZ_API_VERSION + This constant is defined in 'lzlib.h' and works as a version test + macro. The application should verify at compile time that + LZ_API_VERSION is greater than or equal to the version required by the + application: + + #if !defined LZ_API_VERSION || LZ_API_VERSION < 1012 + #error "lzlib 1.12 or newer needed." + #endif + + Before version 1.8, lzlib didn't define LZ_API_VERSION. + LZ_API_VERSION was first defined in lzlib 1.8 to 1. + Since lzlib 1.12, LZ_API_VERSION is defined as (major * 1000 + minor). + + NOTE: Version test macros are the library's way of announcing +functionality to the application. They should not be confused with feature +test macros, which allow the application to announce to the library its +desire to have certain symbols and prototypes exposed. + + -- Function: int LZ_api_version ( void ) + If LZ_API_VERSION >= 1012, this function is declared in 'lzlib.h' (else + it doesn't exist). It returns the LZ_API_VERSION of the library object + code being used. The application should verify at run time that the + value returned by 'LZ_api_version' is greater than or equal to the + version required by the application. An application may be dinamically + linked at run time with a different version of lzlib than the one it + was compiled for, and this should not break the program as long as the + library used provides the functionality required by the application. + + #if defined LZ_API_VERSION && LZ_API_VERSION >= 1012 + if( LZ_api_version() < 1012 ) + show_error( "lzlib 1.12 or newer needed." ); + #endif -- Constant: const char * LZ_version_string - This constant is defined in the header file 'lzlib.h'. - - The application should compare LZ_version and LZ_version_string for -consistency. If the first character differs, the library code actually -used may be incompatible with the 'lzlib.h' header file used by the -application. + This string constant is defined in the header file 'lzlib.h' and + represents the version of the library being used at compile time. - if( LZ_version()[0] != LZ_version_string[0] ) - error( "bad library version" ); + -- Function: const char * LZ_version ( void ) + This function returns a string representing the version of the library + being used at run time. File: lzlib.info, Node: Buffering, Next: Parameter limits, Prev: Library version, Up: Top @@ -164,31 +195,31 @@ File: lzlib.info, Node: Buffering, Next: Parameter limits, Prev: Library vers 3 Buffering *********** -Lzlib internal functions need access to a memory chunk at least as large -as the dictionary size (sliding window). For efficiency reasons, the -input buffer for compression is twice or sixteen times as large as the -dictionary size. +Lzlib internal functions need access to a memory chunk at least as large as +the dictionary size (sliding window). For efficiency reasons, the input +buffer for compression is twice or sixteen times as large as the dictionary +size. Finally, for safety reasons, lzlib uses two more internal buffers. - These are the four buffers used by lzlib, and their guaranteed -minimum sizes: + These are the four buffers used by lzlib, and their guaranteed minimum +sizes: - * Input compression buffer. Written to by the 'LZ_compress_write' - function. For the normal variant of LZMA, its size is two times - the dictionary size set with the 'LZ_compress_open' function or - 64 KiB, whichever is larger. For the fast variant, its size is - 1 MiB. + * Input compression buffer. Written to by the function + 'LZ_compress_write'. For the normal variant of LZMA, its size is two + times the dictionary size set with the function 'LZ_compress_open' or + 64 KiB, whichever is larger. For the fast variant, its size is 1 MiB. - * Output compression buffer. Read from by the 'LZ_compress_read' - function. Its size is 64 KiB. + * Output compression buffer. Read from by the function + 'LZ_compress_read'. Its size is 64 KiB. - * Input decompression buffer. Written to by the - 'LZ_decompress_write' function. Its size is 64 KiB. + * Input decompression buffer. Written to by the function + 'LZ_decompress_write'. Its size is 64 KiB. - * Output decompression buffer. Read from by the 'LZ_decompress_read' - function. Its size is the dictionary size set in the header of the - member currently being decompressed or 64 KiB, whichever is larger. + * Output decompression buffer. Read from by the function + 'LZ_decompress_read'. Its size is the dictionary size set in the header + of the member currently being decompressed or 64 KiB, whichever is + larger. File: lzlib.info, Node: Parameter limits, Next: Compression functions, Prev: Buffering, Up: Top @@ -207,8 +238,7 @@ Current values are shown in square brackets. Returns the smallest valid dictionary size [4 KiB]. -- Function: int LZ_max_dictionary_bits ( void ) - Returns the base 2 logarithm of the largest valid dictionary size - [29]. + Returns the base 2 logarithm of the largest valid dictionary size [29]. -- Function: int LZ_max_dictionary_size ( void ) Returns the largest valid dictionary size [512 MiB]. @@ -227,26 +257,26 @@ File: lzlib.info, Node: Compression functions, Next: Decompression functions, These are the functions used to compress data. In case of error, all of them return -1 or 0, for signed and unsigned return values respectively, -except 'LZ_compress_open' whose return value must be verified by -calling 'LZ_compress_errno' before using it. +except 'LZ_compress_open' whose return value must be verified 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 ) + 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 allocated. - - The returned pointer must be verified by calling - 'LZ_compress_errno' before using it. If 'LZ_compress_errno' does - not return 'LZ_ok', the returned pointer must not be used and - should be freed with 'LZ_compress_close' to avoid memory leaks. - - DICTIONARY_SIZE sets the dictionary size to be used, in bytes. - Valid values range from 4 KiB to 512 MiB. Note that dictionary - sizes are quantized. If the specified size does not match one of - the valid sizes, it will be rounded upwards by adding up to + 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 + allocated. + + The returned pointer must be verified by calling 'LZ_compress_errno' + before using it. If 'LZ_compress_errno' does not return 'LZ_ok', the + returned pointer must not be used and should be freed with + 'LZ_compress_close' to avoid memory leaks. + + DICTIONARY_SIZE sets the dictionary size to be used, in bytes. Valid + values range from 4 KiB to 512 MiB. Note that dictionary sizes are + quantized. If the size specified does not match one of the valid + sizes, it will be rounded upwards by adding up to (DICTIONARY_SIZE / 8) to it. MATCH_LEN_LIMIT sets the match length limit in bytes. Valid values @@ -254,95 +284,117 @@ calling 'LZ_compress_errno' before using it. ratios but longer compression times. If DICTIONARY_SIZE is 65535 and MATCH_LEN_LIMIT is 16, the fast - variant of LZMA is chosen, which produces identical compressed - output as 'lzip -0'. (The dictionary size used will be rounded - upwards to 64 KiB). - - MEMBER_SIZE sets the member size limit in bytes. Valid values - range from 100 kB to 2 PiB. Small member size may degrade - compression ratio, so use it only when needed. To produce a - single-member data stream, give MEMBER_SIZE a value larger than - the amount of data to be produced. Values larger than 2 PiB will be - reduced to 2 PiB to prevent the uncompressed size of the member - from overflowing. - - -- Function: int LZ_compress_close ( struct 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. - - -- Function: int LZ_compress_finish ( struct LZ_Encoder * const - ENCODER ) + variant of LZMA is chosen, which produces identical compressed output + as 'lzip -0'. (The dictionary size used will be rounded upwards to + 64 KiB). + + MEMBER_SIZE sets the member size limit in bytes. Valid values range + from 4 KiB to 2 PiB. A small member size may degrade compression + ratio, so use it only when needed. To produce a single-member data + stream, give MEMBER_SIZE a value larger than the amount of data to be + produced. Values larger than 2 PiB will be reduced to 2 PiB to prevent + the uncompressed size of the member from overflowing. + + -- Function: int LZ_compress_close ( struct 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 ) Use this function to tell 'lzlib' that all the data for this member - have already been written (with the 'LZ_compress_write' function). - It is safe to call 'LZ_compress_finish' as many times as needed. - After all the produced compressed data have been read with - 'LZ_compress_read' and '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 ) - 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 'LZ_compress_read' function). + have already been written (with the function 'LZ_compress_write'). It + is safe to call 'LZ_compress_finish' as many times as needed. After + all the compressed data have been read with 'LZ_compress_read' and + '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 ) + 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 ) - Use this function to make available to 'LZ_compress_read' all the - data already written with the 'LZ_compress_write' function. First - call 'LZ_compress_sync_flush'. Then call 'LZ_compress_read' until - it returns 0. + 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 a 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 files. + *Note Data format::. Repeated use of 'LZ_compress_sync_flush' may degrade compression - ratio, so use it only when needed. + ratio, so use it only when needed. If the interval between calls to + 'LZ_compress_sync_flush' is large (comparable to dictionary size), + creating a multimember data stream with 'LZ_compress_restart_member' + may be an alternative. + + Combining multimember stream creation with flushing may be tricky. If + there are more bytes available than those needed to complete + MEMBER_SIZE, 'LZ_compress_restart_member' needs to be called when + '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 ) - The 'LZ_compress_read' function reads up to SIZE bytes from the - stream pointed to by ENCODER, storing the results in BUFFER. - - The return value is the number of bytes actually read. This might - be less than SIZE; for example, if there aren't that many bytes - left in the stream or if more bytes have to be yet written with the - 'LZ_compress_write' function. 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 ) - The 'LZ_compress_write' function writes up to SIZE bytes from - BUFFER to the stream pointed to by ENCODER. + The function 'LZ_compress_read' 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. + + The return value is the number of bytes actually read. This might be + less than SIZE; for example, if there aren't that many bytes left in + the stream or if more bytes have to be yet written with the function + '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 ) + The function 'LZ_compress_write' writes up to SIZE bytes from BUFFER + to the stream pointed to by ENCODER. - The return value is the number of bytes actually written. This - might be less than SIZE. Note that writing less than SIZE bytes is - not an error. + The return value is 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 ) - The 'LZ_compress_write_size' function returns the maximum number of - bytes that can be immediately written through the - 'LZ_compress_write' function. + The function 'LZ_compress_write_size' 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 buffer must be + compressed before a size greater than 0 is returned again. (This is + done to minimize the amount of data that must be copied to the + beginning of the buffer before new data can be accepted). 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 ) - Returns the current error code for ENCODER. *Note Error codes::. - - -- Function: int LZ_compress_finished ( struct LZ_Encoder * const + -- Function: enum LZ_Errno LZ_compress_errno ( struct 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'. + 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 ) + 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 ) - 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: int LZ_compress_member_finished ( struct 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 ) @@ -351,8 +403,8 @@ calling 'LZ_compress_errno' before using it. -- Function: unsigned long long LZ_compress_member_position ( struct LZ_Encoder * const ENCODER ) - Returns the number of compressed bytes already produced, but - perhaps not yet read, in the current member. + 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 ) @@ -369,130 +421,144 @@ File: lzlib.info, Node: Decompression functions, Next: Error codes, Prev: Com 6 Decompression functions ************************* -These are the functions used to decompress data. In case of error, all -of them return -1 or 0, for signed and unsigned return values -respectively, except 'LZ_decompress_open' whose return value must be -verified by calling 'LZ_decompress_errno' before using it. +These are the functions used to decompress data. In case of error, all of +them return -1 or 0, for signed and unsigned return values respectively, +except 'LZ_decompress_open' whose return value must be verified by calling +'LZ_decompress_errno' before using it. -- Function: struct 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 allocated. - - The returned pointer must be verified by calling - 'LZ_decompress_errno' before using it. If 'LZ_decompress_errno' - does not return 'LZ_ok', the 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 ) - 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. - - -- Function: int LZ_decompress_finish ( struct LZ_Decoder * const - DECODER ) + 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 + allocated. + + The returned pointer must be verified by calling 'LZ_decompress_errno' + before using it. If 'LZ_decompress_errno' does not return 'LZ_ok', the + 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 ) + 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 ) Use this function to tell 'lzlib' that all the data for this stream - have already been written (with the 'LZ_decompress_write' - function). It is safe to call 'LZ_decompress_finish' as many - times as needed. - - -- Function: int LZ_decompress_reset ( struct LZ_Decoder * const + have already been written (with the function 'LZ_decompress_write'). + It is safe to call 'LZ_decompress_finish' as many times as needed. It + is not required to call 'LZ_decompress_finish' if the input stream + 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 ) + 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 is discarded. Position counters are set to 0. + + -- Function: int LZ_decompress_sync_to_member ( struct LZ_Decoder * const DECODER ) - Resets the internal state of DECODER as it was just after opening - it with the 'LZ_decompress_open' function. Data stored in the - internal buffers is discarded. Position counters are set to 0. - - -- Function: int LZ_decompress_sync_to_member ( struct 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 'LZ_decompress_write' will be consumed and - 'LZ_decompress_read' will return 0 until a header is found. - - This function is useful to discard any data preceding the first - member, or to discard the rest of the current member, for example - in 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 ) - The 'LZ_decompress_read' function reads up to SIZE bytes from the - stream pointed to by DECODER, storing the results in BUFFER. - - The return value is the number of bytes actually read. This might - be less than SIZE; for example, if there aren't that many bytes - left in the stream or if more bytes have to be yet written with the - 'LZ_decompress_write' function. Note that reading less than SIZE - bytes is not an error. + 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 + 'LZ_decompress_write' will be consumed and 'LZ_decompress_read' will + return 0 until a header is found. + + This function is useful to discard any data preceding the first member, + or to discard the rest of the current member, for example in 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 ) + The function 'LZ_decompress_read' 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. + + The return value is the number of bytes actually read. This might be + less than SIZE; for example, if there aren't that many bytes left in + the stream or if more bytes have to be yet written with the function + 'LZ_decompress_write'. Note that reading less than SIZE bytes is not + an error. + + 'LZ_decompress_read' returns at least once per member so that + 'LZ_decompress_member_finished' can be called (and trailer data + retrieved) for each member, even for empty members. Therefore, + 'LZ_decompress_read' returning 0 does not mean that the end of the + stream has been reached. The increase in the value returned by + 'LZ_decompress_total_in_size' can be used to tell the end of the stream + from an empty member. In case of decompression error caused by corrupt or truncated data, 'LZ_decompress_read' does not signal the error immediately to the - application, but waits until all decoded bytes have been read. This - allows tools like tarlz to recover as much data as possible from - each damaged member. *Note tarlz manual: (tarlz)Top. + application, but waits until all the bytes decoded have been read. This + 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 ) - The 'LZ_decompress_write' function writes up to SIZE bytes from - BUFFER to the stream pointed to by DECODER. + -- Function: int LZ_decompress_write ( struct LZ_Decoder * const DECODER, + uint8_t * const BUFFER, const int SIZE ) + The function 'LZ_decompress_write' writes up to SIZE bytes from BUFFER + to the stream pointed to by DECODER. - The return value is the number of bytes actually written. This - might be less than SIZE. Note that writing less than SIZE bytes is - not an error. + The return value is 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 ) - The 'LZ_decompress_write_size' function returns the maximum number - of bytes that can be immediately written through the - 'LZ_decompress_write' function. + The function 'LZ_decompress_write_size' 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 1 the value returned. - It is guaranteed that an immediate call to 'LZ_decompress_write' - will accept a SIZE up to the returned number of bytes. + 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 ) - Returns the current error code for DECODER. *Note Error codes::. + -- Function: enum LZ_Errno LZ_decompress_errno ( struct 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 ) - Returns 1 if all the data have been read and 'LZ_decompress_close' - can be safely called. Otherwise it returns 0. - - -- Function: int LZ_decompress_member_finished ( struct LZ_Decoder * - const DECODER ) - Returns 1 if the previous call to 'LZ_decompress_read' finished - reading the current member, indicating that final values for - member are available through 'LZ_decompress_data_crc', - 'LZ_decompress_data_position', and - 'LZ_decompress_member_position'. Otherwise it returns 0. - - -- Function: int LZ_decompress_member_version ( struct 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 + DECODER ) + Returns 1 if the previous call to 'LZ_decompress_read' finished reading + the current member, indicating that final values for member are + available through 'LZ_decompress_data_crc', + 'LZ_decompress_data_position', and 'LZ_decompress_member_position'. + Otherwise it returns 0. + + -- Function: int LZ_decompress_member_version ( struct LZ_Decoder * const + DECODER ) Returns the version of current member from member header. - -- Function: int LZ_decompress_dictionary_size ( struct LZ_Decoder * - const DECODER ) - Returns the dictionary size of current member from member header. + -- Function: int LZ_decompress_dictionary_size ( struct 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 ) - Returns the 32 bit Cyclic Redundancy Check of the data - decompressed from the current member. The returned value is valid - only when 'LZ_decompress_member_finished' returns 1. + -- Function: unsigned LZ_decompress_data_crc ( struct LZ_Decoder * const + DECODER ) + Returns the 32 bit Cyclic Redundancy Check of the data decompressed + from the current member. The returned value is valid only when + 'LZ_decompress_member_finished' returns 1. -- Function: unsigned long long LZ_decompress_data_position ( struct LZ_Decoder * const DECODER ) - Returns the number of decompressed bytes already produced, but - perhaps not yet read, in the current member. + 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 ) - Returns the number of input bytes already decompressed in the - current member. + 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 ) @@ -500,8 +566,8 @@ verified by calling 'LZ_decompress_errno' before using it. -- Function: unsigned long long LZ_decompress_total_out_size ( struct LZ_Decoder * const DECODER ) - Returns the total number of decompressed bytes already produced, - but perhaps not yet read. + Returns the total number of decompressed bytes already produced, but + perhaps not yet read. File: lzlib.info, Node: Error codes, Next: Error messages, Prev: Decompression functions, Up: Top @@ -516,16 +582,15 @@ what kind of error it was, you need to verify the error code by calling Library functions don't change the value returned by 'LZ_(de)compress_errno' when they succeed; thus, the value returned by -'LZ_(de)compress_errno' after a successful call is not necessarily -LZ_ok, 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'. +'LZ_(de)compress_errno' after a successful call is not necessarily LZ_ok, +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'. -- Constant: enum LZ_Errno LZ_ok - The value of this constant is 0 and is used to indicate that there - is no error. + The value of this constant is 0 and is used to indicate that there is + no error. -- Constant: enum LZ_Errno LZ_bad_argument At least one of the arguments passed to the library function was @@ -542,24 +607,22 @@ whether a call failed. If the call failed, then you can examine finished. -- Constant: enum 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. + 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 The end of the data stream was reached in the middle of a member. -- Constant: enum 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 corrupt header in a multimember data - stream, or trailing data too similar to a valid lzip header. - Lziprecover can be used to remove conflicting trailing data from a - file. + 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 corrupt header in a multimember data stream, or + trailing data too similar to a valid lzip header. Lziprecover can be + used to remove conflicting trailing data from a file. -- Constant: enum LZ_Errno LZ_library_error - A bug was detected in the library. Please, report it. *Note - Problems::. + A bug was detected in the library. Please, report it. *Note Problems::. File: lzlib.info, Node: Error messages, Next: Invoking minilzip, Prev: Error codes, Up: Top @@ -568,11 +631,10 @@ File: lzlib.info, Node: Error messages, Next: Invoking minilzip, Prev: Error **************** -- 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. + 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. The value of LZ_ERRNO normally comes from a call to 'LZ_(de)compress_errno'. @@ -583,18 +645,30 @@ File: lzlib.info, Node: Invoking minilzip, Next: Data format, Prev: Error mes 9 Invoking minilzip ******************* -Minilzip is a test program for the lzlib compression library, fully +Minilzip is a test program for the compression library lzlib, fully compatible with lzip 1.4 or newer. + 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, chosen to maximize safety and +interoperability. 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 is better than gzip and bzip2 +from a data recovery perspective. 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. + The format for running minilzip is: minilzip [OPTIONS] [FILES] -'-' used as a FILE argument means standard input. It can be mixed with -other FILES and is read just once, the first time it appears in the -command line. +If no file names are specified, minilzip compresses (or decompresses) from +standard input to standard output. A hyphen '-' used as a FILE argument +means standard input. It can be mixed with other FILES and is read just +once, the first time it appears in the command line. - minilzip supports the following options: + minilzip supports the following options: *Note Argument syntax: +(arg_parser)Argument syntax. '-h' '--help' @@ -602,37 +676,39 @@ command line. '-V' '--version' - Print the version number of minilzip on the standard output and - exit. This version number should be included in all bug reports. + Print the version number of minilzip on the standard output and exit. + This version number should be included in all bug reports. '-a' '--trailing-error' Exit with error status 2 if any remaining input is detected after - decompressing the last member. Such remaining input is usually - trailing garbage that can be safely ignored. + decompressing the last member. Such remaining input is usually trailing + garbage that can be safely ignored. '-b BYTES' '--member-size=BYTES' - When compressing, set the member size limit to BYTES. 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. 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' - Compress or decompress to standard output; keep input files - unchanged. If compressing several files, each file is compressed - independently. This option is needed when reading from a named - pipe (fifo) or from a device. Use it also to recover as much of - the decompressed data as possible when decompressing a corrupt - file. + Compress or decompress to standard output; keep input files unchanged. + If compressing several files, each file is compressed independently. + (The output consists of a sequence of independently compressed + members). This option (or '-o') is needed when reading from a named + pipe (fifo) or from a device. Use it also to recover as much of the + decompressed data as possible when decompressing a corrupt file. '-c' + overrides '-o' and '-S'. '-c' has no effect when testing or listing. '-d' '--decompress' - Decompress the specified files. If a file does not exist or can't - be opened, minilzip continues decompressing the rest of the files. - If a file fails to decompress, or is a terminal, minilzip exits - immediately without decompressing the rest of the files. + Decompress the files specified. If a file does not exist or can't be + opened, minilzip continues decompressing the rest of the files. If a + file fails to decompress, or is a terminal, minilzip exits immediately + without decompressing the rest of the files. '-f' '--force' @@ -640,30 +716,32 @@ command line. '-F' '--recompress' - When compressing, force re-compression of files whose name already - has the '.lz' or '.tlz' suffix. + When compressing, force re-compression of files whose name already has + the '.lz' or '.tlz' suffix. '-k' '--keep' - Keep (don't delete) input files during compression or - decompression. + Keep (don't delete) input files during compression or decompression. '-m BYTES' '--match-length=BYTES' - When compressing, set the match length limit in bytes. After a - match this long is found, the search is finished. Valid values - range from 5 to 273. Larger values usually give better compression - ratios but longer compression times. + When compressing, set the match length limit in bytes. After a match + this long is found, the search is finished. Valid values range from 5 + to 273. Larger values usually give better compression ratios but longer + compression times. '-o FILE' '--output=FILE' - When reading from standard input and '--stdout' has not been - specified, use 'FILE' as the virtual name of the uncompressed - file. This produces a file named 'FILE' when decompressing, or a - file named 'FILE.lz' when compressing. A second '.lz' extension is - not added if 'FILE' already ends in '.lz' or '.tlz'. When - compressing and splitting the output in volumes, several files - named 'FILE00001.lz', 'FILE00002.lz', etc, are created. + If '-c' has not been also specified, write the (de)compressed output to + FILE; keep input files unchanged. If compressing several files, each + file is compressed independently. (The output consists of a sequence of + independently compressed members). This option (or '-c') is needed when + reading from a named pipe (fifo) or from a device. '-o -' is + equivalent to '-c'. '-o' has no effect when testing or listing. + + When compressing and splitting the output in volumes, FILE is used as + a prefix, and several files named 'FILE00001.lz', 'FILE00002.lz', etc, + are created. In this case, only one input file is allowed. '-q' '--quiet' @@ -673,37 +751,36 @@ command line. '--dictionary-size=BYTES' When compressing, set the dictionary size limit in bytes. Minilzip will use for each file the largest dictionary size that does not - exceed neither the file size nor this limit. Valid values range - from 4 KiB to 512 MiB. Values 12 to 29 are interpreted as powers - of two, meaning 2^12 to 2^29 bytes. Dictionary sizes are quantized - so that they can be coded in just one byte (*note - coded-dict-size::). If the specified size does not match one of - the valid sizes, it will be rounded upwards by adding up to - (BYTES / 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. + exceed neither the file size nor this limit. Valid values range from + 4 KiB to 512 MiB. Values 12 to 29 are interpreted as powers of two, + meaning 2^12 to 2^29 bytes. Dictionary sizes are quantized so that + they can be coded in just one byte (*note coded-dict-size::). If the + size specified does not match one of the valid sizes, it will be + rounded upwards by adding up to (BYTES / 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. '-S BYTES' '--volume-size=BYTES' - When compressing, split the compressed output into several volume - files with names 'original_name00001.lz', 'original_name00002.lz', - etc, and set the volume size limit to BYTES. Input files are kept - unchanged. Each volume is a complete, maybe multimember, lzip - file. A small volume size may degrade compression ratio, so use it - only when needed. Valid values range from 100 kB to 4 EiB. + When compressing, and '-c' has not been also specified, split the + compressed output into several volume files with names + 'original_name00001.lz', 'original_name00002.lz', etc, and set the + volume size limit to BYTES. Input files are kept unchanged. Each + volume is a complete, maybe multimember, lzip file. A small volume + size may degrade compression ratio, so use it only when needed. Valid + values range from 100 kB to 4 EiB. '-t' '--test' - Check integrity of the specified files, but don't decompress them. - This really performs a trial decompression and throws away the - result. Use it together with '-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 checking 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. + Check integrity of the files specified, but don't decompress them. This + really performs a trial decompression and throws away the result. Use + it together with '-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 checking 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. '-v' '--verbose' @@ -711,24 +788,24 @@ command line. When compressing, show the compression ratio and size for each file processed. When decompressing or testing, further -v's (up to 4) increase the - verbosity level, showing status, compression ratio, dictionary - size, and trailer contents (CRC, data size, member size). + verbosity level, showing status, compression ratio, dictionary size, + and trailer contents (CRC, data size, member size). '-0 .. -9' - Compression level. Set the compression parameters (dictionary size - and match length limit) as shown in the table below. The default - compression level is '-6', equivalent to '-s8MiB -m36'. Note that - '-9' can be much slower than '-0'. These options have no effect - when decompressing or testing. + Compression level. Set the compression parameters (dictionary size and + match length limit) as shown in the table below. The default + compression level is '-6', equivalent to '-s8MiB -m36'. Note that '-9' + can be much slower than '-0'. These options have no effect when + decompressing or testing. - The bidimensional parameter space of LZMA can't be mapped to a - linear scale optimal for all files. If your files are large, very - repetitive, etc, you may need to use the '--dictionary-size' and - '--match-length' options directly to achieve optimal performance. + The bidimensional parameter space of LZMA can't be mapped to a linear + scale optimal for all files. If your files are large, very repetitive, + etc, you may need to use the options '--dictionary-size' and + '--match-length' directly to achieve optimal performance. - If several compression levels or '-s' or '-m' options are given, - the last setting is used. For example '-9 -s64MiB' is equivalent - to '-s64MiB -m273' + If several compression levels or '-s' or '-m' options are given, the + last setting is used. For example '-9 -s64MiB' is equivalent to + '-s64MiB -m273' Level Dictionary size (-s) Match length limit (-m) -0 64 KiB 16 bytes @@ -747,11 +824,19 @@ command line. Aliases for GNU gzip compatibility. '--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. + 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. + +'--check-lib' + Compare the version of lzlib used to compile minilzip with the version + actually being used and exit. Report any differences found. Exit with + error status 1 if differences are found. A mismatch may indicate that + lzlib is not correctly installed or that a different version of lzlib + has been installed after compiling the shared version of minilzip. + 'minilzip -v --check-lib' shows the version of lzlib being used and + the value of 'LZ_API_VERSION' (if defined). *Note Library version::. Numbers given as arguments to options may be followed by a multiplier @@ -770,10 +855,10 @@ Z zettabyte (10^21) | Zi zebibyte (2^70) Y yottabyte (10^24) | Yi yobibyte (2^80) - Exit status: 0 for a normal exit, 1 for environmental problems (file -not found, invalid flags, I/O errors, etc), 2 to indicate a corrupt or -invalid input file, 3 for an internal consistency error (eg, bug) which -caused minilzip to panic. + Exit status: 0 for a normal exit, 1 for environmental problems (file not +found, invalid flags, I/O errors, etc), 2 to indicate a corrupt or invalid +input file, 3 for an internal consistency error (eg, bug) which caused +minilzip to panic. File: lzlib.info, Node: Data format, Next: Examples, Prev: Invoking minilzip, Up: Top @@ -787,11 +872,13 @@ when there is no longer anything to take away. In the diagram below, a box like this: + +---+ | | <-- the vertical bars might be missing +---+ represents one byte; a box like this: + +==============+ | | +==============+ @@ -799,11 +886,12 @@ when there is no longer anything to take away. represents a variable number of bytes. - A lzip data stream consists of a series of "members" (compressed -data sets). The members simply appear one after another in the data -stream, with no additional information before, between, or after them. + A lzip data stream consists of a series of "members" (compressed data +sets). The members simply appear one after another in the data stream, with +no additional information before, between, or after them. Each member has the following structure: + +--+--+--+--+----+----+=============+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | ID string | VN | DS | LZMA stream | CRC32 | Data size | Member size | +--+--+--+--+----+----+=============+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ @@ -811,17 +899,16 @@ stream, with no additional information before, between, or after them. All multibyte values are stored in little endian order. 'ID string (the "magic" bytes)' - A four byte string, identifying the lzip format, with the value - "LZIP" (0x4C, 0x5A, 0x49, 0x50). + A four byte string, identifying the lzip format, with the value "LZIP" + (0x4C, 0x5A, 0x49, 0x50). 'VN (version number, 1 byte)' - Just in case something needs to be modified in the future. 1 for - now. + Just in case something needs to be modified in the future. 1 for now. 'DS (coded dictionary size, 1 byte)' The dictionary size is calculated by taking a power of 2 (the base - size) and subtracting from it a fraction between 0/16 and 7/16 of - the base size. + size) and subtracting from it a fraction between 0/16 and 7/16 of the + base size. Bits 4-0 contain the base 2 logarithm of the base size (12 to 29). Bits 7-5 contain the numerator of the fraction (0 to 7) to subtract from the base size to obtain the dictionary size. @@ -830,22 +917,23 @@ stream, with no additional information before, between, or after them. 'LZMA stream' The LZMA stream, finished by an end of stream marker. Uses default - values for encoder properties. *Note Stream format: (lzip)Stream + 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). + also uses the LZMA marker '3' ("Sync Flush" marker). *Note + sync_flush::. 'CRC32 (4 bytes)' - CRC of the uncompressed original data. + Cyclic Redundancy Check (CRC) of the uncompressed original data. 'Data size (8 bytes)' Size of the uncompressed original data. 'Member size (8 bytes)' - Total size of the member, including header and trailer. This field - acts as a distributed index, allows the verification of stream - integrity, and facilitates safe recovery of undamaged members from - multimember files. + Total size of the member, including header and trailer. This field acts + as a distributed index, allows the verification of stream integrity, + and facilitates safe recovery of undamaged members from multimember + files. @@ -854,115 +942,303 @@ File: lzlib.info, Node: Examples, Next: Problems, Prev: Data format, Up: Top 11 A small tutorial with examples ********************************* -This chapter shows the order in which the library functions should be -called depending on what kind of data stream you want to compress or -decompress. See the file 'bbexample.c' in the source distribution for -an example of how buffer-to-buffer compression/decompression can be -implemented using lzlib. - - Note that lzlib's interface is symmetrical. That is, the code for -normal compression and decompression is identical except because one -calls LZ_compress* functions while the other calls LZ_decompress* -functions. - - -Example 1: Normal compression (MEMBER_SIZE > total output). - - 1) LZ_compress_open - 2) LZ_compress_write - 3) if no more data to write, call LZ_compress_finish - 4) LZ_compress_read - 5) go back to step 2 until LZ_compress_finished returns 1 - 6) LZ_compress_close - - -Example 2: Normal compression using LZ_compress_write_size. - - 1) LZ_compress_open - 2) go to step 5 if LZ_compress_write_size returns 0 - 3) LZ_compress_write - 4) if no more data to write, call LZ_compress_finish - 5) LZ_compress_read - 6) go back to step 2 until LZ_compress_finished returns 1 - 7) LZ_compress_close +This chapter provides real code examples for the most common uses of the +library. See these examples in context in the files 'bbexample.c' and +'ffexample.c' from the source distribution of lzlib. + Note that the interface of lzlib is symmetrical. That is, the code for +normal compression and decompression is identical except because one calls +LZ_compress* functions while the other calls LZ_decompress* functions. -Example 3: Decompression. - - 1) LZ_decompress_open - 2) LZ_decompress_write - 3) if no more data to write, call LZ_decompress_finish - 4) LZ_decompress_read - 5) go back to step 2 until LZ_decompress_finished returns 1 - 6) LZ_decompress_close - - -Example 4: Decompression using LZ_decompress_write_size. - - 1) LZ_decompress_open - 2) go to step 5 if LZ_decompress_write_size returns 0 - 3) LZ_decompress_write - 4) if no more data to write, call LZ_decompress_finish - 5) LZ_decompress_read - 5a) optionally, if LZ_decompress_member_finished returns 1, read - final values for member with LZ_decompress_data_crc, etc. - 6) go back to step 2 until LZ_decompress_finished returns 1 - 7) LZ_decompress_close - - -Example 5: Multimember compression (MEMBER_SIZE < total output). - - 1) LZ_compress_open - 2) go to step 5 if LZ_compress_write_size returns 0 - 3) LZ_compress_write - 4) if no more data to write, call LZ_compress_finish - 5) LZ_compress_read - 6) go back to step 2 until LZ_compress_member_finished returns 1 - 7) go to step 10 if LZ_compress_finished() returns 1 - 8) LZ_compress_restart_member - 9) go back to step 2 - 10) LZ_compress_close - +* Menu: -Example 6: Multimember compression (user-restarted members). +* Buffer compression:: Buffer-to-buffer single-member compression +* Buffer decompression:: Buffer-to-buffer decompression +* File compression:: File-to-file single-member compression +* File decompression:: File-to-file decompression +* File compression mm:: File-to-file multimember compression +* Skipping data errors:: Decompression with automatic resynchronization - 1) LZ_compress_open (with MEMBER_SIZE > largest member). - 2) LZ_compress_write - 3) LZ_compress_read - 4) go back to step 2 until member termination is desired - 5) LZ_compress_finish - 6) LZ_compress_read - 7) go back to step 6 until LZ_compress_member_finished returns 1 - 9) go to step 12 if all input data have been written and - LZ_compress_finished returns 1 - 10) LZ_compress_restart_member - 11) go back to step 2 - 12) LZ_compress_close + +File: lzlib.info, Node: Buffer compression, Next: Buffer decompression, Up: Examples + +11.1 Buffer compression +======================= + +Buffer-to-buffer single-member compression (MEMBER_SIZE > total output). + +/* Compresses 'insize' bytes from 'inbuf' to 'outbuf'. + Returns the size of the compressed data in '*outlenp'. + In case of error, or if 'outsize' is too small, returns false and does + not modify '*outlenp'. +*/ +bool bbcompress( const uint8_t * const inbuf, const int insize, + const int dictionary_size, const int match_len_limit, + uint8_t * const outbuf, const int outsize, + int * const outlenp ) + { + int inpos = 0, outpos = 0; + bool error = false; + struct 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; } + + while( true ) + { + int ret = LZ_compress_write( encoder, inbuf + inpos, insize - inpos ); + if( ret < 0 ) { error = true; break; } + inpos += ret; + if( inpos >= insize ) LZ_compress_finish( encoder ); + ret = LZ_compress_read( encoder, outbuf + outpos, outsize - outpos ); + if( ret < 0 ) { error = true; break; } + outpos += ret; + if( LZ_compress_finished( encoder ) == 1 ) break; + if( outpos >= outsize ) { error = true; break; } + } + + if( LZ_compress_close( encoder ) < 0 ) error = true; + if( error ) return false; + *outlenp = outpos; + return true; + } + +File: lzlib.info, Node: Buffer decompression, Next: File compression, Prev: Buffer compression, Up: Examples + +11.2 Buffer decompression +========================= + +Buffer-to-buffer decompression. + +/* Decompresses 'insize' bytes from 'inbuf' to 'outbuf'. + Returns the size of the decompressed data in '*outlenp'. + In case of error, or if 'outsize' is too small, returns false and does + not modify '*outlenp'. +*/ +bool bbdecompress( const uint8_t * const inbuf, const int insize, + uint8_t * const outbuf, const int outsize, + int * const outlenp ) + { + int inpos = 0, outpos = 0; + bool error = false; + struct LZ_Decoder * const decoder = LZ_decompress_open(); + if( !decoder || LZ_decompress_errno( decoder ) != LZ_ok ) + { LZ_decompress_close( decoder ); return false; } + + while( true ) + { + int ret = LZ_decompress_write( decoder, inbuf + inpos, insize - inpos ); + if( ret < 0 ) { error = true; break; } + inpos += ret; + if( inpos >= insize ) LZ_decompress_finish( decoder ); + ret = LZ_decompress_read( decoder, outbuf + outpos, outsize - outpos ); + if( ret < 0 ) { error = true; break; } + outpos += ret; + if( LZ_decompress_finished( decoder ) == 1 ) break; + if( outpos >= outsize ) { error = true; break; } + } + + if( LZ_decompress_close( decoder ) < 0 ) error = true; + if( error ) return false; + *outlenp = outpos; + return true; + } -Example 7: Decompression with automatic removal of leading data. + +File: lzlib.info, Node: File compression, Next: File decompression, Prev: Buffer decompression, Up: Examples + +11.3 File compression +===================== + +File-to-file compression using LZ_compress_write_size. + +int ffcompress( struct LZ_Encoder * const encoder, + FILE * const infile, FILE * const outfile ) + { + enum { buffer_size = 16384 }; + uint8_t buffer[buffer_size]; + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_compress_write_size( encoder ) ); + if( size > 0 ) + { + len = fread( buffer, 1, size, infile ); + ret = LZ_compress_write( encoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) ) LZ_compress_finish( encoder ); + } + ret = LZ_compress_read( encoder, buffer, buffer_size ); + if( ret < 0 ) break; + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_compress_finished( encoder ) == 1 ) return 0; + } + return 1; + } - 1) LZ_decompress_open - 2) LZ_decompress_sync_to_member - 3) go to step 6 if LZ_decompress_write_size returns 0 - 4) LZ_decompress_write - 5) if no more data to write, call LZ_decompress_finish - 6) LZ_decompress_read - 7) go back to step 3 until LZ_decompress_finished returns 1 - 8) LZ_decompress_close + +File: lzlib.info, Node: File decompression, Next: File compression mm, Prev: File compression, Up: Examples + +11.4 File decompression +======================= + +File-to-file decompression using LZ_decompress_write_size. + +int ffdecompress( struct LZ_Decoder * const decoder, + FILE * const infile, FILE * const outfile ) + { + enum { buffer_size = 16384 }; + uint8_t buffer[buffer_size]; + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_decompress_write_size( decoder ) ); + if( size > 0 ) + { + len = fread( buffer, 1, size, infile ); + ret = LZ_decompress_write( decoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) ) LZ_decompress_finish( decoder ); + } + ret = LZ_decompress_read( decoder, buffer, buffer_size ); + if( ret < 0 ) break; + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_decompress_finished( decoder ) == 1 ) return 0; + } + return 1; + } + +File: lzlib.info, Node: File compression mm, Next: Skipping data errors, Prev: File decompression, Up: Examples + +11.5 File-to-file multimember compression +========================================= + +Example 1: Multimember compression with members of fixed size +(MEMBER_SIZE < total output). + +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 ); + if( !encoder || LZ_compress_errno( encoder ) != LZ_ok ) + { fputs( "ffexample: Not enough memory.\n", stderr ); + LZ_compress_close( encoder ); return 1; } + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_compress_write_size( encoder ) ); + if( size > 0 ) + { + len = fread( buffer, 1, size, infile ); + ret = LZ_compress_write( encoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) ) LZ_compress_finish( encoder ); + } + ret = LZ_compress_read( encoder, buffer, buffer_size ); + if( ret < 0 ) break; + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_compress_member_finished( encoder ) == 1 ) + { + if( LZ_compress_finished( encoder ) == 1 ) { done = true; break; } + if( LZ_compress_restart_member( encoder, member_size ) < 0 ) break; + } + } + if( LZ_compress_close( encoder ) < 0 ) done = false; + return done; + } + + +Example 2: Multimember compression (user-restarted members). (Call +LZ_compress_open with MEMBER_SIZE > largest member). + +/* Compresses 'infile' to 'outfile' as a multimember stream with one member + for each line of text terminated by a newline character or by EOF. + Returns 0 if success, 1 if error. +*/ +int fflfcompress( struct LZ_Encoder * const encoder, + FILE * const infile, FILE * const outfile ) + { + enum { buffer_size = 16384 }; + uint8_t buffer[buffer_size]; + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_compress_write_size( encoder ) ); + if( size > 0 ) + { + for( len = 0; len < size; ) + { + int ch = getc( infile ); + if( ch == EOF || ( buffer[len++] = ch ) == '\n' ) break; + } + /* avoid writing an empty member to outfile */ + if( len == 0 && LZ_compress_data_position( encoder ) == 0 ) return 0; + ret = LZ_compress_write( encoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) || buffer[len-1] == '\n' ) + LZ_compress_finish( encoder ); + } + ret = LZ_compress_read( encoder, buffer, buffer_size ); + if( ret < 0 ) break; + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_compress_member_finished( encoder ) == 1 ) + { + if( feof( infile ) && LZ_compress_finished( encoder ) == 1 ) return 0; + if( LZ_compress_restart_member( encoder, INT64_MAX ) < 0 ) break; + } + } + return 1; + } -Example 8: Streamed decompression with automatic resynchronization to -next member in case of data error. - - 1) LZ_decompress_open - 2) go to step 5 if LZ_decompress_write_size returns 0 - 3) LZ_decompress_write - 4) if no more data to write, call LZ_decompress_finish - 5) if LZ_decompress_read produces LZ_header_error or LZ_data_error, - call LZ_decompress_sync_to_member - 6) go back to step 2 until LZ_decompress_finished returns 1 - 7) LZ_decompress_close + +File: lzlib.info, Node: Skipping data errors, Prev: File compression mm, Up: Examples + +11.6 Skipping data errors +========================= + +/* Decompresses 'infile' to 'outfile' with automatic resynchronization to + next member in case of data error, including the automatic removal of + leading garbage. +*/ +int ffrsdecompress( struct LZ_Decoder * const decoder, + FILE * const infile, FILE * const outfile ) + { + enum { buffer_size = 16384 }; + uint8_t buffer[buffer_size]; + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_decompress_write_size( decoder ) ); + if( size > 0 ) + { + len = fread( buffer, 1, size, infile ); + ret = LZ_decompress_write( decoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) ) LZ_decompress_finish( decoder ); + } + ret = LZ_decompress_read( decoder, buffer, buffer_size ); + if( ret < 0 ) + { + if( LZ_decompress_errno( decoder ) == LZ_header_error || + LZ_decompress_errno( decoder ) == LZ_data_error ) + { LZ_decompress_sync_to_member( decoder ); continue; } + else break; + } + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_decompress_finished( decoder ) == 1 ) return 0; + } + return 1; + } File: lzlib.info, Node: Problems, Next: Concept index, Prev: Examples, Up: Top @@ -970,15 +1246,14 @@ File: lzlib.info, Node: Problems, Next: Concept index, Prev: Examples, Up: T 12 Reporting bugs ***************** -There are probably bugs in lzlib. There are certainly errors and -omissions in this manual. If you report them, they will get fixed. If -you don't, no one will ever know about them and they will remain unfixed -for all eternity, if not longer. +There are probably bugs in lzlib. There are certainly errors and omissions +in this manual. If you report them, they will get fixed. If you don't, no +one will ever know about them and they will remain unfixed for all +eternity, if not longer. If you find a bug in lzlib, please send electronic mail to -<lzip-bug@nongnu.org>. Include the version number, which you can find -by running 'minilzip --version' or in 'LZ_version_string' from -'lzlib.h'. +<lzip-bug@nongnu.org>. Include the version number, which you can find by +running 'minilzip --version' or in 'LZ_version_string' from 'lzlib.h'. File: lzlib.info, Node: Concept index, Prev: Problems, Up: Top @@ -989,41 +1264,53 @@ Concept index * Menu: -* buffering: Buffering. (line 6) -* bugs: Problems. (line 6) -* compression functions: Compression functions. (line 6) -* data format: Data format. (line 6) -* decompression functions: Decompression functions. - (line 6) -* error codes: Error codes. (line 6) -* error messages: Error messages. (line 6) -* examples: Examples. (line 6) -* getting help: Problems. (line 6) -* introduction: Introduction. (line 6) -* invoking: Invoking minilzip. (line 6) -* library version: Library version. (line 6) -* options: Invoking minilzip. (line 6) -* parameter limits: Parameter limits. (line 6) +* buffer compression: Buffer compression. (line 6) +* buffer decompression: Buffer decompression. (line 6) +* buffering: Buffering. (line 6) +* bugs: Problems. (line 6) +* compression functions: Compression functions. (line 6) +* data format: Data format. (line 6) +* decompression functions: Decompression functions. (line 6) +* error codes: Error codes. (line 6) +* error messages: Error messages. (line 6) +* examples: Examples. (line 6) +* file compression: File compression. (line 6) +* file decompression: File decompression. (line 6) +* getting help: Problems. (line 6) +* introduction: Introduction. (line 6) +* invoking: Invoking minilzip. (line 6) +* library version: Library version. (line 6) +* multimember compression: File compression mm. (line 6) +* options: Invoking minilzip. (line 6) +* parameter limits: Parameter limits. (line 6) +* skipping data errors: Skipping data errors. (line 6) Tag Table: Node: Top220 -Node: Introduction1341 -Node: Library version6397 -Node: Buffering7042 -Node: Parameter limits8267 -Node: Compression functions9226 -Node: Decompression functions16010 -Node: Error codes22599 -Node: Error messages24900 -Node: Invoking minilzip25485 -Ref: --trailing-error26243 -Node: Data format33248 -Ref: coded-dict-size34557 -Node: Examples35817 -Node: Problems39813 -Node: Concept index40385 +Node: Introduction1342 +Node: Library version6413 +Node: Buffering8918 +Node: Parameter limits10143 +Node: Compression functions11097 +Ref: member_size12907 +Ref: sync_flush14673 +Node: Decompression functions19493 +Node: Error codes27187 +Node: Error messages29478 +Node: Invoking minilzip30057 +Node: Data format39651 +Ref: coded-dict-size40957 +Node: Examples42267 +Node: Buffer compression43228 +Node: Buffer decompression44754 +Node: File compression46174 +Node: File decompression47157 +Node: File compression mm48161 +Node: Skipping data errors51193 +Node: Problems52505 +Node: Concept index53077 End Tag Table diff --git a/doc/lzlib.texi b/doc/lzlib.texi index 55e80c2..644a3d7 100644 --- a/doc/lzlib.texi +++ b/doc/lzlib.texi @@ -6,8 +6,8 @@ @finalout @c %**end of header -@set UPDATED 2 January 2019 -@set VERSION 1.11 +@set UPDATED 2 January 2021 +@set VERSION 1.12 @dircategory Data Compression @direntry @@ -29,6 +29,7 @@ @contents @end ifnothtml +@ifnottex @node Top @top @@ -51,20 +52,22 @@ This manual is for Lzlib (version @value{VERSION}, @value{UPDATED}). @end menu @sp 1 -Copyright @copyright{} 2009-2019 Antonio Diaz Diaz. +Copyright @copyright{} 2009-2021 Antonio Diaz Diaz. -This manual is free documentation: you have unlimited permission -to copy, distribute and modify it. +This manual is free documentation: you have unlimited permission to copy, +distribute, and modify it. +@end ifnottex @node Introduction @chapter Introduction @cindex introduction -@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. Lzlib is written in C. +@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. +Lzlib is written in C. The lzip file format is designed for data sharing and long-term archiving, taking into account both data integrity and decoder availability: @@ -72,9 +75,9 @@ 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 +recovery means. The program @uref{http://www.nongnu.org/lzip/manual/lziprecover_manual.html#Data-safety,,lziprecover} -program can repair bit flip errors (one of the most common forms of data +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 @@ -94,42 +97,42 @@ 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. +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 @samp{lzlib.h}. Usage examples of the -library are given in the files @samp{main.c} and @samp{bbexample.c} from -the source distribution. +The functions and variables forming the interface of the compression library +are declared in the file @samp{lzlib.h}. Usage examples of the library are +given in the files @samp{bbexample.c}, @samp{ffexample.c}, and @samp{main.c} +from the source distribution. Compression/decompression is done by repeatedly calling a couple of -read/write functions until all the data have been processed by the -library. This interface is safer and less error prone than the -traditional zlib interface. +read/write functions until all the data have been processed by the library. +This interface is safer and less error prone than the traditional zlib +interface. Compression/decompression is done when the read function is called. This -means the value returned by the position functions will not be updated -until a read call, even if a lot of data are written. If you want the -data to be compressed in advance, just call the read function with a -@var{size} equal to 0. +means the value returned by the position functions will not be updated until +a read call, even if a lot of data are written. If you want the data to be +compressed in advance, just call the read function with a @var{size} equal +to 0. If all the data to be compressed are written in advance, lzlib will -automatically adjust the header of the compressed data to use the -largest dictionary size that does not exceed neither the data size nor -the limit given to @samp{LZ_compress_open}. This feature reduces the -amount of memory needed for decompression and allows minilzip to produce -identical compressed output as lzip. +automatically adjust the header of the compressed data to use the largest +dictionary size that does not exceed neither the data size nor the limit +given to @samp{LZ_compress_open}. This feature reduces the amount of memory +needed for decompression and allows minilzip to produce identical compressed +output as lzip. -Lzlib will correctly decompress a data stream which is the concatenation -of two or more compressed data streams. The result is the concatenation -of the corresponding decompressed data streams. Integrity testing of -concatenated compressed data streams is also supported. +Lzlib will correctly decompress a data stream which is the concatenation of +two or more compressed data streams. The result is the concatenation of the +corresponding decompressed data streams. Integrity testing of concatenated +compressed data streams is also supported. Lzlib is able to compress and decompress streams of unlimited size by -automatically creating multimember output. The members so created are -large, about @w{2 PiB} each. +automatically creating multimember output. The members so created are large, +about @w{2 PiB} each. All the library functions are thread safe. The library does not install any signal handler. The decoder checks the consistency of the compressed @@ -139,14 +142,14 @@ 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 @samp{-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. +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. -Lzlib currently implements two variants of the LZMA algorithm; fast -(used by option @samp{-0} of minilzip) and normal (used by all other -compression levels). +Lzlib currently implements two variants of the LZMA algorithm; fast (used by +option @samp{-0} of minilzip) and normal (used by all other compression +levels). The high compression of LZMA comes from combining two basic, well-proven compression ideas: sliding dictionaries (LZ77/78) and markov models (the @@ -155,37 +158,74 @@ similar order-0 entropy coder as its last stage) with segregation of contexts according to what the bits are used for. The ideas embodied in lzlib are due to (at least) the following people: -Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for -the definition of Markov chains), G.N.N. Martin (for the definition of -range encoding), Igor Pavlov (for putting all the above together in -LZMA), and Julian Seward (for bzip2's CLI). +Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for the +definition of Markov chains), G.N.N. Martin (for the definition of range +encoding), Igor Pavlov (for putting all the above together in LZMA), and +Julian Seward (for bzip2's CLI). -LANGUAGE NOTE: Uncompressed = not compressed = plain data; it may never -have been compressed. Decompressed is used to refer to data which have -undergone the process of decompression. +LANGUAGE NOTE: Uncompressed = not compressed = plain data; it may never have +been compressed. Decompressed is used to refer to data which have undergone +the process of decompression. @node Library version @chapter Library version @cindex library version -@deftypefun {const char *} LZ_version ( void ) -Returns the library version as a string. -@end deftypefun +One goal of lzlib is to keep perfect backward compatibility with older +versions of itself down to 1.0. Any application working with an older lzlib +should work with a newer lzlib. Installing a newer lzlib should not break +anything. This chapter describes the constants and functions that the +application can use to discover the version of the library being used. -@deftypevr Constant {const char *} LZ_version_string -This constant is defined in the header file @samp{lzlib.h}. -@end deftypevr +@defvr Constant LZ_API_VERSION +This constant is defined in @samp{lzlib.h} and works as a version test +macro. The application should verify at compile time that LZ_API_VERSION is +greater than or equal to the version required by the application: -The application should compare LZ_version and LZ_version_string for -consistency. If the first character differs, the library code actually -used may be incompatible with the @samp{lzlib.h} header file used by the -application. +@example +#if !defined LZ_API_VERSION || LZ_API_VERSION < 1012 +#error "lzlib 1.12 or newer needed." +#endif +@end example + +Before version 1.8, lzlib didn't define LZ_API_VERSION.@* +LZ_API_VERSION was first defined in lzlib 1.8 to 1.@* +Since lzlib 1.12, LZ_API_VERSION is defined as (major * 1000 + minor). +@end defvr + +NOTE: Version test macros are the library's way of announcing functionality +to the application. They should not be confused with feature test macros, +which allow the application to announce to the library its desire to have +certain symbols and prototypes exposed. + +@deftypefun int LZ_api_version ( void ) +If LZ_API_VERSION >= 1012, this function is declared in @samp{lzlib.h} (else +it doesn't exist). It returns the LZ_API_VERSION of the library object code +being used. The application should verify at run time that the value +returned by @code{LZ_api_version} is greater than or equal to the version +required by the application. An application may be dinamically linked at run +time with a different version of lzlib than the one it was compiled for, and +this should not break the program as long as the library used provides the +functionality required by the application. @example -if( LZ_version()[0] != LZ_version_string[0] ) - error( "bad library version" ); +#if defined LZ_API_VERSION && LZ_API_VERSION >= 1012 + if( LZ_api_version() < 1012 ) + show_error( "lzlib 1.12 or newer needed." ); +#endif @end example +@end deftypefun + +@deftypevr Constant {const char *} LZ_version_string +This string constant is defined in the header file @samp{lzlib.h} and +represents the version of the library being used at compile time. +@end deftypevr + +@deftypefun {const char *} LZ_version ( void ) +This function returns a string representing the version of the library being +used at run time. +@end deftypefun @node Buffering @@ -199,26 +239,23 @@ dictionary size. Finally, for safety reasons, lzlib uses two more internal buffers. -These are the four buffers used by lzlib, and their guaranteed minimum -sizes: +These are the four buffers used by lzlib, and their guaranteed minimum sizes: @itemize @bullet -@item Input compression buffer. Written to by the -@samp{LZ_compress_write} function. For the normal variant of LZMA, its -size is two times the dictionary size set with the -@samp{LZ_compress_open} function or @w{64 KiB}, whichever is larger. For -the fast variant, its size is @w{1 MiB}. - -@item Output compression buffer. Read from by the -@samp{LZ_compress_read} function. Its size is @w{64 KiB}. - -@item Input decompression buffer. Written to by the -@samp{LZ_decompress_write} function. Its size is @w{64 KiB}. - -@item Output decompression buffer. Read from by the -@samp{LZ_decompress_read} function. Its size is the dictionary size set -in the header of the member currently being decompressed or @w{64 KiB}, -whichever is larger. +@item Input compression buffer. Written to by the function +@samp{LZ_compress_write}. For the normal variant of LZMA, its size is two +times the dictionary size set with the function @samp{LZ_compress_open} or +@w{64 KiB}, whichever is larger. For the fast variant, its size is @w{1 MiB}. + +@item Output compression buffer. Read from by the function +@samp{LZ_compress_read}. Its size is @w{64 KiB}. + +@item Input decompression buffer. Written to by the function +@samp{LZ_decompress_write}. Its size is @w{64 KiB}. + +@item Output decompression buffer. Read from by the function +@samp{LZ_decompress_read}. Its size is the dictionary size set in the header +of the member currently being decompressed or @w{64 KiB}, whichever is larger. @end itemize @@ -277,26 +314,26 @@ should be freed with @samp{LZ_compress_close} to avoid memory leaks. @var{dictionary_size} sets the dictionary size to be used, in bytes. Valid values range from @w{4 KiB} to @w{512 MiB}. Note that dictionary -sizes are quantized. If the specified size does not match one of the +sizes are quantized. If the size specified does not match one of the valid sizes, it will be rounded upwards by adding up to @w{(@var{dictionary_size} / 8)} to it. @var{match_len_limit} sets the match length limit in bytes. Valid values -range from 5 to 273. Larger values usually give better compression -ratios but longer compression times. - -If @var{dictionary_size} is 65535 and @var{match_len_limit} is 16, the -fast variant of LZMA is chosen, which produces identical compressed -output as @code{lzip -0}. (The dictionary size used will be rounded -upwards to @w{64 KiB}). - -@var{member_size} sets the member size limit in bytes. Valid values -range from @w{100 kB} to @w{2 PiB}. Small member size may degrade -compression ratio, so use it only when needed. To produce a -single-member data stream, give @var{member_size} a value larger than -the amount of data to be produced. Values larger than @w{2 PiB} will be -reduced to @w{2 PiB} to prevent the uncompressed size of the member from -overflowing. +range from 5 to 273. Larger values usually give better compression ratios +but longer compression times. + +If @var{dictionary_size} is 65535 and @var{match_len_limit} is 16, the fast +variant of LZMA is chosen, which produces identical compressed output as +@w{@samp{lzip -0}}. (The dictionary size used will be rounded upwards to +@w{64 KiB}). + +@anchor{member_size} +@var{member_size} sets the member size limit in bytes. Valid values range +from @w{4 KiB} to @w{2 PiB}. A small member size may degrade compression +ratio, so use it only when needed. To produce a single-member data stream, +give @var{member_size} a value larger than the amount of data to be +produced. Values larger than @w{2 PiB} will be reduced to @w{2 PiB} to +prevent the uncompressed size of the member from overflowing. @end deftypefun @@ -305,53 +342,73 @@ 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 longer be used as an argument to any LZ_compress function. +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} ) Use this function to tell @samp{lzlib} that all the data for this member -have already been written (with the @samp{LZ_compress_write} function). +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. -After all the produced compressed data have been read with -@samp{LZ_compress_read} and @samp{LZ_compress_member_finished} returns -1, a new member can be started with @samp{LZ_compress_restart_member}. +After all the compressed data have been read with @samp{LZ_compress_read} +and @samp{LZ_compress_member_finished} returns 1, a new member can be +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} ) -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 -@samp{LZ_compress_read} function). +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 +@samp{LZ_compress_read}). @xref{member_size}, for a description of +@var{member_size}. @end deftypefun +@anchor{sync_flush} @deftypefun int LZ_compress_sync_flush ( struct LZ_Encoder * const @var{encoder} ) -Use this function to make available to @samp{LZ_compress_read} all the -data already written with the @samp{LZ_compress_write} function. First -call @samp{LZ_compress_sync_flush}. Then call @samp{LZ_compress_read} -until it returns 0. +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 a 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. @xref{Data format}. Repeated use of @samp{LZ_compress_sync_flush} may degrade compression -ratio, so use it only when needed. +ratio, so use it only when needed. If the interval between calls to +@samp{LZ_compress_sync_flush} is large (comparable to dictionary size), +creating a multimember data stream with @samp{LZ_compress_restart_member} +may be an alternative. + +Combining multimember stream creation with flushing may be tricky. If there +are more bytes available than those needed to complete @var{member_size}, +@samp{LZ_compress_restart_member} needs to be called when +@samp{LZ_compress_member_finished} returns 1, followed by a new call to +@samp{LZ_compress_sync_flush}. @end deftypefun @deftypefun int LZ_compress_read ( struct LZ_Encoder * const @var{encoder}, uint8_t * const @var{buffer}, const int @var{size} ) -The @samp{LZ_compress_read} function reads up to @var{size} bytes from -the stream pointed to by @var{encoder}, storing the results in -@var{buffer}. - -The return value is the number of bytes actually read. This might be -less than @var{size}; for example, if there aren't that many bytes left -in the stream or if more bytes have to be yet written with the -@samp{LZ_compress_write} function. Note that reading less than -@var{size} bytes is not an error. +The function @samp{LZ_compress_read} 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 discarded. + +The return value is the number of bytes actually read. This might be less +than @var{size}; for example, if there aren't that many bytes left in the +stream or if more bytes have to be yet written with the function +@samp{LZ_compress_write}. Note that reading less than @var{size} bytes is +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} ) -The @samp{LZ_compress_write} function writes up to @var{size} bytes from +The function @samp{LZ_compress_write} writes up to @var{size} bytes from @var{buffer} to the stream pointed to by @var{encoder}. The return value is the number of bytes actually written. This might be @@ -361,9 +418,13 @@ not an error. @deftypefun int LZ_compress_write_size ( struct LZ_Encoder * const @var{encoder} ) -The @samp{LZ_compress_write_size} function returns the maximum number of -bytes that can be immediately written through the @samp{LZ_compress_write} -function. +The function @samp{LZ_compress_write_size} 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 be +compressed before a size greater than 0 is returned again. (This is done to +minimize the amount of data that must be copied to the beginning of the +buffer before new data can be accepted). It is guaranteed that an immediate call to @samp{LZ_compress_write} will accept a @var{size} up to the returned number of bytes. @@ -372,6 +433,8 @@ accept a @var{size} up to the returned number of bytes. @deftypefun {enum LZ_Errno} LZ_compress_errno ( struct 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 @@ -390,8 +453,7 @@ Otherwise it returns 0. @deftypefun {unsigned long long} LZ_compress_data_position ( struct LZ_Encoder * const @var{encoder} ) -Returns the number of input bytes already compressed in the current -member. +Returns the number of input bytes already compressed in the current member. @end deftypefun @@ -441,19 +503,23 @@ 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 longer be used as an argument to any LZ_decompress function. +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} ) Use this function to tell @samp{lzlib} that all the data for this stream -have already been written (with the @samp{LZ_decompress_write} function). +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. +It is not required to call @samp{LZ_decompress_finish} if the input stream +only contains whole members, but not calling it prevents lzlib from +detecting a truncated member. @end deftypefun @deftypefun int LZ_decompress_reset ( struct LZ_Decoder * const @var{decoder} ) Resets the internal state of @var{decoder} as it was just after opening -it with the @samp{LZ_decompress_open} function. Data stored in the +it with the function @samp{LZ_decompress_open}. Data stored in the internal buffers is discarded. Position counters are set to 0. @end deftypefun @@ -473,19 +539,28 @@ function does nothing. @deftypefun int LZ_decompress_read ( struct LZ_Decoder * const @var{decoder}, uint8_t * const @var{buffer}, const int @var{size} ) -The @samp{LZ_decompress_read} function reads up to @var{size} bytes from -the stream pointed to by @var{decoder}, storing the results in -@var{buffer}. +The function @samp{LZ_decompress_read} 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 discarded. + +The return value is the number of bytes actually read. This might be less +than @var{size}; for example, if there aren't that many bytes left in the +stream or if more bytes have to be yet written with the function +@samp{LZ_decompress_write}. Note that reading less than @var{size} bytes is +not an error. -The return value is the number of bytes actually read. This might be -less than @var{size}; for example, if there aren't that many bytes left -in the stream or if more bytes have to be yet written with the -@samp{LZ_decompress_write} function. Note that reading less than -@var{size} bytes is not an error. +@samp{LZ_decompress_read} returns at least once per member so that +@samp{LZ_decompress_member_finished} can be called (and trailer data +retrieved) for each member, even for empty members. Therefore, +@samp{LZ_decompress_read} returning 0 does not mean that the end of the +stream has been reached. The increase in the value returned by +@samp{LZ_decompress_total_in_size} can be used to tell the end of the stream +from an empty member. In case of decompression error caused by corrupt or truncated data, @samp{LZ_decompress_read} does not signal the error immediately to the -application, but waits until all decoded bytes have been read. This +application, but waits until all the bytes decoded have been read. This allows tools like @uref{http://www.nongnu.org/lzip/manual/tarlz_manual.html,,tarlz} to recover as much data as possible from each damaged member. @@ -496,7 +571,7 @@ recover as much data as possible from each damaged member. @deftypefun int LZ_decompress_write ( struct LZ_Decoder * const @var{decoder}, uint8_t * const @var{buffer}, const int @var{size} ) -The @samp{LZ_decompress_write} function writes up to @var{size} bytes from +The function @samp{LZ_decompress_write} writes up to @var{size} bytes from @var{buffer} to the stream pointed to by @var{decoder}. The return value is the number of bytes actually written. This might be @@ -506,32 +581,35 @@ not an error. @deftypefun int LZ_decompress_write_size ( struct LZ_Decoder * const @var{decoder} ) -The @samp{LZ_decompress_write_size} function returns the maximum number -of bytes that can be immediately written through the -@samp{LZ_decompress_write} function. +The function @samp{LZ_decompress_write_size} 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 returned. -It is guaranteed that an immediate call to @samp{LZ_decompress_write} -will accept a @var{size} up to the returned number of bytes. +It is guaranteed that an immediate call to @samp{LZ_decompress_write} will +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} ) 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} ) Returns 1 if all the data have been read and @samp{LZ_decompress_close} -can be safely called. Otherwise it returns 0. +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} ) -Returns 1 if the previous call to @samp{LZ_decompress_read} finished -reading the current member, indicating that final values for member are -available through @samp{LZ_decompress_data_crc}, -@samp{LZ_decompress_data_position}, and -@samp{LZ_decompress_member_position}. Otherwise it returns 0. +Returns 1 if the previous call to @samp{LZ_decompress_read} finished reading +the current member, indicating that final values for member are available +through @samp{LZ_decompress_data_crc}, @samp{LZ_decompress_data_position}, +and @samp{LZ_decompress_member_position}. Otherwise it returns 0. @end deftypefun @@ -541,7 +619,8 @@ Returns the version of current member from member header. @deftypefun int LZ_decompress_dictionary_size ( struct LZ_Decoder * const @var{decoder} ) -Returns the dictionary size of current member from member header. +Returns the dictionary size of the current member, read from the member +header. @end deftypefun @@ -559,8 +638,7 @@ not yet read, in the current member. @deftypefun {unsigned long long} LZ_decompress_member_position ( struct LZ_Decoder * const @var{decoder} ) -Returns the number of input bytes already decompressed in the current -member. +Returns the number of input bytes already decompressed in the current member. @end deftypefun @@ -599,8 +677,7 @@ error. @end deftypevr @deftypevr Constant {enum LZ_Errno} LZ_bad_argument -At least one of the arguments passed to the library function was -invalid. +At least one of the arguments passed to the library function was invalid. @end deftypevr @deftypevr Constant {enum LZ_Errno} LZ_mem_error @@ -617,8 +694,7 @@ finished. @deftypevr Constant {enum 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. +this happens at the end of the data stream it may indicate trailing data. @end deftypevr @deftypevr Constant {enum LZ_Errno} LZ_unexpected_eof @@ -658,9 +734,20 @@ The value of @var{lz_errno} normally comes from a call to @cindex invoking @cindex options -Minilzip is a test program for the lzlib compression library, fully +Minilzip is a test program for the compression library lzlib, fully compatible with lzip 1.4 or newer. +@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, chosen to maximize safety and interoperability. 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 is better than gzip and bzip2 from a data recovery +perspective. 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. + @noindent The format for running minilzip is: @@ -669,11 +756,16 @@ minilzip [@var{options}] [@var{files}] @end example @noindent -@samp{-} used as a @var{file} argument means standard input. It can be -mixed with other @var{files} and is read just once, the first time it -appears in the command line. +If no file names are specified, minilzip compresses (or decompresses) from +standard input to standard output. A hyphen @samp{-} used as a @var{file} +argument means standard input. It can be mixed with other @var{files} and is +read just once, the first time it appears in the command line. -minilzip supports the following options: +minilzip supports the following +@uref{http://www.nongnu.org/arg-parser/manual/arg_parser_manual.html#Argument-syntax,,options}: +@ifnothtml +@xref{Argument syntax,,,arg_parser}. +@end ifnothtml @table @code @item -h @@ -685,7 +777,6 @@ Print an informative help message describing the options and exit. Print the version number of minilzip on the standard output and exit. This version number should be included in all bug reports. -@anchor{--trailing-error} @item -a @itemx --trailing-error Exit with error status 2 if any remaining input is detected after @@ -694,21 +785,25 @@ 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}. 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}. +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 +compression ratio, so use it only when needed. Valid values range from +@w{100 kB} to @w{2 PiB}. Defaults to @w{2 PiB}. @item -c @itemx --stdout -Compress or decompress to standard output; keep input files unchanged. -If compressing several files, each file is compressed independently. -This option is needed when reading from a named pipe (fifo) or from a -device. Use it also to recover as much of the decompressed data as -possible when decompressing a corrupt file. +Compress or decompress to standard output; keep input files unchanged. If +compressing several files, each file is compressed independently. (The +output consists of a sequence of independently compressed members). This +option (or @samp{-o}) is needed when reading from a named pipe (fifo) or +from a device. Use it also to recover as much of the decompressed data as +possible when decompressing a corrupt file. @samp{-c} overrides @samp{-o} +and @samp{-S}. @samp{-c} has no effect when testing or listing. @item -d @itemx --decompress -Decompress the specified files. If a file does not exist or can't be +Decompress the files specified. If a file does not exist or can't be opened, minilzip continues decompressing the rest of the files. If a file fails to decompress, or is a terminal, minilzip exits immediately without decompressing the rest of the files. @@ -735,14 +830,17 @@ compression times. @item -o @var{file} @itemx --output=@var{file} -When reading from standard input and @samp{--stdout} has not been -specified, use @samp{@var{file}} as the virtual name of the uncompressed -file. This produces a file named @samp{@var{file}} when decompressing, -or a file named @samp{@var{file}.lz} when compressing. A second -@samp{.lz} extension is not added if @samp{@var{file}} already ends in -@samp{.lz} or @samp{.tlz}. When compressing and splitting the output in -volumes, several files named @samp{@var{file}00001.lz}, -@samp{@var{file}00002.lz}, etc, are created. +If @samp{-c} has not been also specified, write the (de)compressed output to +@var{file}; keep input files unchanged. If compressing several files, each +file is compressed independently. (The output consists of a sequence of +independently compressed members). This option (or @samp{-c}) is needed when +reading from a named pipe (fifo) or from a device. @w{@samp{-o -}} is +equivalent to @samp{-c}. @samp{-o} has no effect when testing or listing. + +When compressing and splitting the output in volumes, @var{file} is used as +a prefix, and several files named @samp{@var{file}00001.lz}, +@samp{@var{file}00002.lz}, etc, are created. In this case, only one input +file is allowed. @item -q @itemx --quiet @@ -755,7 +853,7 @@ for each file the largest dictionary size that does not exceed neither the file size nor this limit. Valid values range from @w{4 KiB} to @w{512 MiB}. Values 12 to 29 are interpreted as powers of two, meaning 2^12 to 2^29 bytes. Dictionary sizes are quantized so that they can be -coded in just one byte (@pxref{coded-dict-size}). If the specified size +coded in just one byte (@pxref{coded-dict-size}). If the size specified does not match one of the valid sizes, it will be rounded upwards by adding up to @w{(@var{bytes} / 8)} to it. @@ -765,16 +863,17 @@ is affected at compression time by the choice of dictionary size limit. @item -S @var{bytes} @itemx --volume-size=@var{bytes} -When compressing, split the compressed output into several volume files -with names @samp{original_name00001.lz}, @samp{original_name00002.lz}, -etc, and set the volume size limit to @var{bytes}. Input files are kept -unchanged. Each volume is a complete, maybe multimember, lzip file. A -small volume size may degrade compression ratio, so use it only when -needed. Valid values range from @w{100 kB} to @w{4 EiB}. +When compressing, and @samp{-c} has not been also specified, split the +compressed output into several volume files with names +@samp{original_name00001.lz}, @samp{original_name00002.lz}, etc, and set the +volume size limit to @var{bytes}. Input files are kept unchanged. Each +volume is a complete, maybe multimember, lzip file. A small volume size may +degrade compression ratio, so use it only when needed. Valid values range +from @w{100 kB} to @w{4 EiB}. @item -t @itemx --test -Check integrity of the specified files, but don't decompress them. This +Check integrity of the files specified, but don't decompress them. This really performs a trial decompression and throws away the result. Use it together with @samp{-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 @@ -800,8 +899,8 @@ effect when decompressing or testing. The bidimensional parameter space of LZMA can't be mapped to a linear scale optimal for all files. If your files are large, very repetitive, -etc, you may need to use the @samp{--dictionary-size} and -@samp{--match-length} options directly to achieve optimal performance. +etc, you may need to use the options @samp{--dictionary-size} and +@samp{--match-length} directly to achieve optimal performance. If several compression levels or @samp{-s} or @samp{-m} options are given, the last setting is used. For example @w{@samp{-9 -s64MiB}} is @@ -831,6 +930,18 @@ 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. +@item --check-lib +Compare the @uref{#Library-version,,version of lzlib} used to compile +minilzip with the version actually being used and exit. Report any +differences found. Exit with error status 1 if differences are found. A +mismatch may indicate that lzlib is not correctly installed or that a +different version of lzlib has been installed after compiling the shared +version of minilzip. @w{@samp{minilzip -v --check-lib}} shows the version of +lzlib being used and the value of @samp{LZ_API_VERSION} (if defined). +@ifnothtml +@xref{Library version}. +@end ifnothtml + @end table Numbers given as arguments to options may be followed by a multiplier @@ -867,6 +978,7 @@ when there is no longer anything to take away.@* @sp 1 In the diagram below, a box like this: + @verbatim +---+ | | <-- the vertical bars might be missing @@ -874,6 +986,7 @@ In the diagram below, a box like this: @end verbatim represents one byte; a box like this: + @verbatim +==============+ | | @@ -888,6 +1001,7 @@ The members simply appear one after another in the data stream, with no additional information before, between, or after them. Each member has the following structure: + @verbatim +--+--+--+--+----+----+=============+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | ID string | VN | DS | LZMA stream | CRC32 | Data size | Member size | @@ -907,8 +1021,7 @@ Just in case something needs to be modified in the future. 1 for now. @anchor{coded-dict-size} @item DS (coded dictionary size, 1 byte) The dictionary size is calculated by taking a power of 2 (the base size) -and subtracting from it a fraction between 0/16 and 7/16 of the base -size.@* +and subtracting from it a fraction between 0/16 and 7/16 of the base size.@* Bits 4-0 contain the base 2 logarithm of the base size (12 to 29).@* Bits 7-5 contain the numerator of the fraction (0 to 7) to subtract from the base size to obtain the dictionary size.@* @@ -916,8 +1029,8 @@ 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, finished by an end of stream marker. Uses default -values for encoder properties. +The LZMA stream, finished by an end of stream marker. Uses default values +for encoder properties. @ifnothtml @xref{Stream format,,,lzip}, @end ifnothtml @@ -927,10 +1040,10 @@ See @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). +also uses the LZMA marker @samp{3} ("Sync Flush" marker). @xref{sync_flush}. @item CRC32 (4 bytes) -CRC of the uncompressed original data. +Cyclic Redundancy Check (CRC) of the uncompressed original data. @item Data size (8 bytes) Size of the uncompressed original data. @@ -947,138 +1060,313 @@ facilitates safe recovery of undamaged members from multimember files. @chapter A small tutorial with examples @cindex examples -This chapter shows the order in which the library functions should be -called depending on what kind of data stream you want to compress or -decompress. See the file @samp{bbexample.c} in the source distribution -for an example of how buffer-to-buffer compression/decompression can be -implemented using lzlib. +This chapter provides real code examples for the most common uses of the +library. See these examples in context in the files @samp{bbexample.c} and +@samp{ffexample.c} from the source distribution of lzlib. -Note that lzlib's interface is symmetrical. That is, the code for normal -compression and decompression is identical except because one calls +Note that the interface of lzlib is symmetrical. That is, the code for +normal compression and decompression is identical except because one calls LZ_compress* functions while the other calls LZ_decompress* functions. -@sp 1 -@noindent -Example 1: Normal compression (@var{member_size} > total output). +@menu +* Buffer compression:: Buffer-to-buffer single-member compression +* Buffer decompression:: Buffer-to-buffer decompression +* File compression:: File-to-file single-member compression +* File decompression:: File-to-file decompression +* File compression mm:: File-to-file multimember compression +* Skipping data errors:: Decompression with automatic resynchronization +@end menu -@example -1) LZ_compress_open -2) LZ_compress_write -3) if no more data to write, call LZ_compress_finish -4) LZ_compress_read -5) go back to step 2 until LZ_compress_finished returns 1 -6) LZ_compress_close -@end example -@sp 1 -@noindent -Example 2: Normal compression using LZ_compress_write_size. +@node Buffer compression +@section Buffer compression +@cindex buffer compression -@example -1) LZ_compress_open -2) go to step 5 if LZ_compress_write_size returns 0 -3) LZ_compress_write -4) if no more data to write, call LZ_compress_finish -5) LZ_compress_read -6) go back to step 2 until LZ_compress_finished returns 1 -7) LZ_compress_close -@end example +Buffer-to-buffer single-member compression +@w{(@var{member_size} > total output)}. -@sp 1 -@noindent -Example 3: Decompression. +@verbatim +/* Compresses 'insize' bytes from 'inbuf' to 'outbuf'. + Returns the size of the compressed data in '*outlenp'. + In case of error, or if 'outsize' is too small, returns false and does + not modify '*outlenp'. +*/ +bool bbcompress( const uint8_t * const inbuf, const int insize, + const int dictionary_size, const int match_len_limit, + uint8_t * const outbuf, const int outsize, + int * const outlenp ) + { + int inpos = 0, outpos = 0; + bool error = false; + struct 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; } + + while( true ) + { + int ret = LZ_compress_write( encoder, inbuf + inpos, insize - inpos ); + if( ret < 0 ) { error = true; break; } + inpos += ret; + if( inpos >= insize ) LZ_compress_finish( encoder ); + ret = LZ_compress_read( encoder, outbuf + outpos, outsize - outpos ); + if( ret < 0 ) { error = true; break; } + outpos += ret; + if( LZ_compress_finished( encoder ) == 1 ) break; + if( outpos >= outsize ) { error = true; break; } + } + + if( LZ_compress_close( encoder ) < 0 ) error = true; + if( error ) return false; + *outlenp = outpos; + return true; + } +@end verbatim -@example -1) LZ_decompress_open -2) LZ_decompress_write -3) if no more data to write, call LZ_decompress_finish -4) LZ_decompress_read -5) go back to step 2 until LZ_decompress_finished returns 1 -6) LZ_decompress_close -@end example -@sp 1 -@noindent -Example 4: Decompression using LZ_decompress_write_size. +@node Buffer decompression +@section Buffer decompression +@cindex buffer decompression -@example -1) LZ_decompress_open -2) go to step 5 if LZ_decompress_write_size returns 0 -3) LZ_decompress_write -4) if no more data to write, call LZ_decompress_finish -5) LZ_decompress_read -5a) optionally, if LZ_decompress_member_finished returns 1, read - final values for member with LZ_decompress_data_crc, etc. -6) go back to step 2 until LZ_decompress_finished returns 1 -7) LZ_decompress_close -@end example +Buffer-to-buffer decompression. -@sp 1 -@noindent -Example 5: Multimember compression (@var{member_size} < total output). +@verbatim +/* Decompresses 'insize' bytes from 'inbuf' to 'outbuf'. + Returns the size of the decompressed data in '*outlenp'. + In case of error, or if 'outsize' is too small, returns false and does + not modify '*outlenp'. +*/ +bool bbdecompress( const uint8_t * const inbuf, const int insize, + uint8_t * const outbuf, const int outsize, + int * const outlenp ) + { + int inpos = 0, outpos = 0; + bool error = false; + struct LZ_Decoder * const decoder = LZ_decompress_open(); + if( !decoder || LZ_decompress_errno( decoder ) != LZ_ok ) + { LZ_decompress_close( decoder ); return false; } + + while( true ) + { + int ret = LZ_decompress_write( decoder, inbuf + inpos, insize - inpos ); + if( ret < 0 ) { error = true; break; } + inpos += ret; + if( inpos >= insize ) LZ_decompress_finish( decoder ); + ret = LZ_decompress_read( decoder, outbuf + outpos, outsize - outpos ); + if( ret < 0 ) { error = true; break; } + outpos += ret; + if( LZ_decompress_finished( decoder ) == 1 ) break; + if( outpos >= outsize ) { error = true; break; } + } + + if( LZ_decompress_close( decoder ) < 0 ) error = true; + if( error ) return false; + *outlenp = outpos; + return true; + } +@end verbatim -@example - 1) LZ_compress_open - 2) go to step 5 if LZ_compress_write_size returns 0 - 3) LZ_compress_write - 4) if no more data to write, call LZ_compress_finish - 5) LZ_compress_read - 6) go back to step 2 until LZ_compress_member_finished returns 1 - 7) go to step 10 if LZ_compress_finished() returns 1 - 8) LZ_compress_restart_member - 9) go back to step 2 -10) LZ_compress_close -@end example -@sp 1 -@noindent -Example 6: Multimember compression (user-restarted members). +@node File compression +@section File compression +@cindex file compression -@example - 1) LZ_compress_open (with @var{member_size} > largest member). - 2) LZ_compress_write - 3) LZ_compress_read - 4) go back to step 2 until member termination is desired - 5) LZ_compress_finish - 6) LZ_compress_read - 7) go back to step 6 until LZ_compress_member_finished returns 1 - 9) go to step 12 if all input data have been written and - LZ_compress_finished returns 1 -10) LZ_compress_restart_member -11) go back to step 2 -12) LZ_compress_close -@end example +File-to-file compression using LZ_compress_write_size. -@sp 1 -@noindent -Example 7: Decompression with automatic removal of leading data. +@verbatim +int ffcompress( struct LZ_Encoder * const encoder, + FILE * const infile, FILE * const outfile ) + { + enum { buffer_size = 16384 }; + uint8_t buffer[buffer_size]; + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_compress_write_size( encoder ) ); + if( size > 0 ) + { + len = fread( buffer, 1, size, infile ); + ret = LZ_compress_write( encoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) ) LZ_compress_finish( encoder ); + } + ret = LZ_compress_read( encoder, buffer, buffer_size ); + if( ret < 0 ) break; + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_compress_finished( encoder ) == 1 ) return 0; + } + return 1; + } +@end verbatim -@example -1) LZ_decompress_open -2) LZ_decompress_sync_to_member -3) go to step 6 if LZ_decompress_write_size returns 0 -4) LZ_decompress_write -5) if no more data to write, call LZ_decompress_finish -6) LZ_decompress_read -7) go back to step 3 until LZ_decompress_finished returns 1 -8) LZ_decompress_close -@end example + +@node File decompression +@section File decompression +@cindex file decompression + +File-to-file decompression using LZ_decompress_write_size. + +@verbatim +int ffdecompress( struct LZ_Decoder * const decoder, + FILE * const infile, FILE * const outfile ) + { + enum { buffer_size = 16384 }; + uint8_t buffer[buffer_size]; + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_decompress_write_size( decoder ) ); + if( size > 0 ) + { + len = fread( buffer, 1, size, infile ); + ret = LZ_decompress_write( decoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) ) LZ_decompress_finish( decoder ); + } + ret = LZ_decompress_read( decoder, buffer, buffer_size ); + if( ret < 0 ) break; + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_decompress_finished( decoder ) == 1 ) return 0; + } + return 1; + } +@end verbatim + + +@node File compression mm +@section File-to-file multimember compression +@cindex multimember compression + +Example 1: Multimember compression with members of fixed size +@w{(@var{member_size} < total output)}. + +@verbatim +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 ); + if( !encoder || LZ_compress_errno( encoder ) != LZ_ok ) + { fputs( "ffexample: Not enough memory.\n", stderr ); + LZ_compress_close( encoder ); return 1; } + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_compress_write_size( encoder ) ); + if( size > 0 ) + { + len = fread( buffer, 1, size, infile ); + ret = LZ_compress_write( encoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) ) LZ_compress_finish( encoder ); + } + ret = LZ_compress_read( encoder, buffer, buffer_size ); + if( ret < 0 ) break; + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_compress_member_finished( encoder ) == 1 ) + { + if( LZ_compress_finished( encoder ) == 1 ) { done = true; break; } + if( LZ_compress_restart_member( encoder, member_size ) < 0 ) break; + } + } + if( LZ_compress_close( encoder ) < 0 ) done = false; + return done; + } +@end verbatim @sp 1 @noindent -Example 8: Streamed decompression with automatic resynchronization to -next member in case of data error. +Example 2: Multimember compression (user-restarted members). +(Call LZ_compress_open with @var{member_size} > largest member). -@example -1) LZ_decompress_open -2) go to step 5 if LZ_decompress_write_size returns 0 -3) LZ_decompress_write -4) if no more data to write, call LZ_decompress_finish -5) if LZ_decompress_read produces LZ_header_error or LZ_data_error, - call LZ_decompress_sync_to_member -6) go back to step 2 until LZ_decompress_finished returns 1 -7) LZ_decompress_close -@end example +@verbatim +/* Compresses 'infile' to 'outfile' as a multimember stream with one member + for each line of text terminated by a newline character or by EOF. + Returns 0 if success, 1 if error. +*/ +int fflfcompress( struct LZ_Encoder * const encoder, + FILE * const infile, FILE * const outfile ) + { + enum { buffer_size = 16384 }; + uint8_t buffer[buffer_size]; + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_compress_write_size( encoder ) ); + if( size > 0 ) + { + for( len = 0; len < size; ) + { + int ch = getc( infile ); + if( ch == EOF || ( buffer[len++] = ch ) == '\n' ) break; + } + /* avoid writing an empty member to outfile */ + if( len == 0 && LZ_compress_data_position( encoder ) == 0 ) return 0; + ret = LZ_compress_write( encoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) || buffer[len-1] == '\n' ) + LZ_compress_finish( encoder ); + } + ret = LZ_compress_read( encoder, buffer, buffer_size ); + if( ret < 0 ) break; + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_compress_member_finished( encoder ) == 1 ) + { + if( feof( infile ) && LZ_compress_finished( encoder ) == 1 ) return 0; + if( LZ_compress_restart_member( encoder, INT64_MAX ) < 0 ) break; + } + } + return 1; + } +@end verbatim + + +@node Skipping data errors +@section Skipping data errors +@cindex skipping data errors + +@verbatim +/* Decompresses 'infile' to 'outfile' with automatic resynchronization to + next member in case of data error, including the automatic removal of + leading garbage. +*/ +int ffrsdecompress( struct LZ_Decoder * const decoder, + FILE * const infile, FILE * const outfile ) + { + enum { buffer_size = 16384 }; + uint8_t buffer[buffer_size]; + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_decompress_write_size( decoder ) ); + if( size > 0 ) + { + len = fread( buffer, 1, size, infile ); + ret = LZ_decompress_write( decoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) ) LZ_decompress_finish( decoder ); + } + ret = LZ_decompress_read( decoder, buffer, buffer_size ); + if( ret < 0 ) + { + if( LZ_decompress_errno( decoder ) == LZ_header_error || + LZ_decompress_errno( decoder ) == LZ_data_error ) + { LZ_decompress_sync_to_member( decoder ); continue; } + else break; + } + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_decompress_finished( decoder ) == 1 ) return 0; + } + return 1; + } +@end verbatim @node Problems @@ -1093,7 +1381,7 @@ for all eternity, if not longer. If you find a bug in lzlib, please send electronic mail to @email{lzip-bug@@nongnu.org}. Include the version number, which you can -find by running @w{@code{minilzip --version}} or in +find by running @w{@samp{minilzip --version}} or in @samp{LZ_version_string} from @samp{lzlib.h}. diff --git a/doc/minilzip.1 b/doc/minilzip.1 index adb6e33..13a2d6d 100644 --- a/doc/minilzip.1 +++ b/doc/minilzip.1 @@ -1,13 +1,23 @@ -.\" DO NOT MODIFY THIS FILE! It was generated by help2man 1.46.1. -.TH MINILZIP "1" "January 2019" "minilzip 1.11" "User Commands" +.\" DO NOT MODIFY THIS FILE! It was generated by help2man 1.47.16. +.TH MINILZIP "1" "January 2021" "minilzip 1.12" "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 lzlib compression library, fully +Minilzip is a test program for the compression library lzlib, fully compatible with lzip 1.4 or newer. +.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, chosen to maximize safety and +interoperability. 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 is better than gzip and bzip2 from +a data recovery perspective. 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. .SH OPTIONS .TP \fB\-h\fR, \fB\-\-help\fR @@ -41,7 +51,7 @@ keep (don't delete) input files set match length limit in bytes [36] .TP \fB\-o\fR, \fB\-\-output=\fR<file> -if reading standard input, write to <file> +write to <file>, keep input files .TP \fB\-q\fR, \fB\-\-quiet\fR suppress all messages @@ -69,6 +79,9 @@ alias for \fB\-9\fR .TP \fB\-\-loose\-trailing\fR allow trailing data seeming corrupt header +.TP +\fB\-\-check\-lib\fR +compare version of lzlib.h with liblz.{a,so} .PP If no file names are given, or if a file is '\-', minilzip compresses or decompresses from standard input to standard output. @@ -79,20 +92,29 @@ to 2^29 bytes. .PP The bidimensional parameter space of LZMA can't be mapped to a linear scale optimal for all files. If your files are large, very repetitive, -etc, you may need to use the \fB\-\-dictionary\-size\fR and \fB\-\-match\-length\fR -options directly to achieve optimal performance. +etc, you may need to use the options \fB\-\-dictionary\-size\fR and \fB\-\-match\-length\fR +directly to achieve optimal performance. +.PP +To extract all the files from archive 'foo.tar.lz', use the commands +\&'tar \fB\-xf\fR foo.tar.lz' or 'minilzip \fB\-cd\fR foo.tar.lz | tar \fB\-xf\fR \-'. .PP Exit status: 0 for a normal exit, 1 for environmental problems (file not found, invalid flags, I/O errors, etc), 2 to indicate a corrupt or invalid input file, 3 for an internal consistency error (eg, bug) which caused minilzip to panic. +.PP +The ideas embodied in lzlib are due to (at least) the following people: +Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for the +definition of Markov chains), G.N.N. Martin (for the definition of range +encoding), Igor Pavlov (for putting all the above together in LZMA), and +Julian Seward (for bzip2's CLI). .SH "REPORTING BUGS" Report bugs to lzip\-bug@nongnu.org .br Lzlib home page: http://www.nongnu.org/lzip/lzlib.html .SH COPYRIGHT -Copyright \(co 2019 Antonio Diaz Diaz. -Using lzlib 1.11 +Copyright \(co 2021 Antonio Diaz Diaz. +Using lzlib 1.12 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. @@ -1,20 +1,20 @@ -/* Lzlib - Compression library for the lzip format - Copyright (C) 2009-2019 Antonio Diaz Diaz. +/* Lzlib - Compression library for the lzip format + Copyright (C) 2009-2021 Antonio Diaz Diaz. - This library is free software. Redistribution and use in source and - binary forms, with or without modification, are permitted provided - that the following conditions are met: + This library is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ static int LZe_get_match_pairs( struct LZ_encoder * const e, struct Pair * pairs ) @@ -23,7 +23,7 @@ static int LZe_get_match_pairs( struct LZ_encoder * const e, struct Pair * pairs int32_t * ptr1 = ptr0 + 1; int32_t * newptr; int len = 0, len0 = 0, len1 = 0; - int maxlen = 0; + int maxlen = 3; /* only used if pairs != 0 */ int num_pairs = 0; const int pos1 = e->eb.mb.pos + 1; const int min_pos = ( e->eb.mb.pos > e->eb.mb.dictionary_size ) ? @@ -49,8 +49,8 @@ static int LZe_get_match_pairs( struct LZ_encoder * const e, struct Pair * pairs if( pairs ) { - int np2 = e->eb.mb.prev_positions[key2]; - int np3 = e->eb.mb.prev_positions[key3]; + const int np2 = e->eb.mb.prev_positions[key2]; + const int np3 = e->eb.mb.prev_positions[key3]; if( np2 > min_pos && e->eb.mb.buffer[np2-1] == data[0] ) { pairs[0].dis = e->eb.mb.pos - np2; @@ -60,19 +60,17 @@ static int LZe_get_match_pairs( struct LZ_encoder * const e, struct Pair * pairs if( np2 != np3 && np3 > min_pos && e->eb.mb.buffer[np3-1] == data[0] ) { maxlen = 3; - np2 = np3; - pairs[num_pairs].dis = e->eb.mb.pos - np2; - ++num_pairs; + pairs[num_pairs++].dis = e->eb.mb.pos - np3; } if( num_pairs > 0 ) { - const int delta = pos1 - np2; + const int delta = pairs[num_pairs-1].dis + 1; while( maxlen < len_limit && data[maxlen-delta] == data[maxlen] ) ++maxlen; pairs[num_pairs-1].len = maxlen; + if( maxlen < 3 ) maxlen = 3; if( maxlen >= len_limit ) pairs = 0; /* done. now just skip */ } - if( maxlen < 3 ) maxlen = 3; } e->eb.mb.prev_positions[key2] = pos1; @@ -1,20 +1,20 @@ -/* Lzlib - Compression library for the lzip format - Copyright (C) 2009-2019 Antonio Diaz Diaz. +/* Lzlib - Compression library for the lzip format + Copyright (C) 2009-2021 Antonio Diaz Diaz. - This library is free software. Redistribution and use in source and - binary forms, with or without modification, are permitted provided - that the following conditions are met: + This library is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ struct Len_prices diff --git a/encoder_base.c b/encoder_base.c index d385f2c..c1ef9ef 100644 --- a/encoder_base.c +++ b/encoder_base.c @@ -1,20 +1,20 @@ -/* Lzlib - Compression library for the lzip format - Copyright (C) 2009-2019 Antonio Diaz Diaz. +/* Lzlib - Compression library for the lzip format + Copyright (C) 2009-2021 Antonio Diaz Diaz. - This library is free software. Redistribution and use in source and - binary forms, with or without modification, are permitted provided - that the following conditions are met: + This library is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ static bool Mb_normalize_pos( struct Matchfinder_base * const mb ) @@ -58,7 +58,7 @@ static bool Mb_init( struct Matchfinder_base * const mb, const int before_size, mb->stream_pos = 0; mb->num_prev_positions23 = num_prev_positions23; mb->at_stream_end = false; - mb->flushing = false; + mb->sync_flush_pending = false; mb->buffer_size = max( 65536, buffer_size_limit ); mb->buffer = (uint8_t *)malloc( mb->buffer_size ); @@ -69,7 +69,7 @@ static bool Mb_init( struct Matchfinder_base * const mb, const int before_size, size = 1 << max( 16, real_bits( mb->dictionary_size - 1 ) - 2 ); if( mb->dictionary_size > 1 << 26 ) /* 64 MiB */ size >>= 1; - mb->key4_mask = size - 1; + mb->key4_mask = size - 1; /* increases with dictionary size */ size += num_prev_positions23; mb->num_prev_positions = size; @@ -118,7 +118,7 @@ static void Mb_reset( struct Matchfinder_base * const mb ) mb->pos = 0; mb->cyclic_pos = 0; mb->at_stream_end = false; - mb->flushing = false; + mb->sync_flush_pending = false; mb->dictionary_size = mb->saved_dictionary_size; Mb_adjust_array( mb ); mb->pos_limit = mb->buffer_size - mb->after_size; @@ -126,7 +126,7 @@ static void Mb_reset( struct Matchfinder_base * const mb ) } - /* End Of Stream mark => (dis == 0xFFFFFFFFU, len == min_match_len) */ +/* End Of Stream marker => (dis == 0xFFFFFFFFU, len == min_match_len) */ static void LZeb_try_full_flush( struct LZ_encoder_base * const eb ) { int i; @@ -149,33 +149,36 @@ static void LZeb_try_full_flush( struct LZ_encoder_base * const eb ) } - /* Sync Flush mark => (dis == 0xFFFFFFFFU, len == min_match_len + 1) */ -static bool LZeb_sync_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 ) { - int i; const int pos_state = Mb_data_position( &eb->mb ) & pos_state_mask; const State state = eb->state; + const unsigned min_size = eb->renc.ff_count + max_marker_size; if( eb->member_finished || - Cb_free_bytes( &eb->renc.cb ) < (2 * max_marker_size) + eb->renc.ff_count ) - return false; - for( i = 0; i < 2; ++i ) /* 2 consecutive markers guarantee decoding */ - { + Cb_free_bytes( &eb->renc.cb ) < min_size + max_marker_size ) return; + eb->mb.sync_flush_pending = false; + const unsigned long long old_mpos = Re_member_position( &eb->renc ); + do { /* size of markers must be >= rd_min_available_bytes + 5 */ Re_encode_bit( &eb->renc, &eb->bm_match[state][pos_state], 1 ); Re_encode_bit( &eb->renc, &eb->bm_rep[state], 0 ); LZeb_encode_pair( eb, 0xFFFFFFFFU, min_match_len + 1, pos_state ); Re_flush( &eb->renc ); } - return true; + while( Re_member_position( &eb->renc ) - old_mpos < min_size ); } static void LZeb_reset( struct LZ_encoder_base * const eb, const unsigned long long member_size ) { + const unsigned long long min_member_size = min_dictionary_size; + const unsigned long long max_member_size = 0x0008000000000000ULL; /* 2 PiB */ int i; Mb_reset( &eb->mb ); eb->member_size_limit = - min( member_size, 0x0008000000000000ULL ) - Lt_size - max_marker_size; + min( max( min_member_size, member_size ), max_member_size ) - + Lt_size - max_marker_size; eb->crc = 0xFFFFFFFFU; Bm_array_init( eb->bm_literal[0], (1 << literal_context_bits) * 0x300 ); Bm_array_init( eb->bm_match[0], states * pos_states ); diff --git a/encoder_base.h b/encoder_base.h index f0240cc..e727a7d 100644 --- a/encoder_base.h +++ b/encoder_base.h @@ -1,20 +1,20 @@ -/* Lzlib - Compression library for the lzip format - Copyright (C) 2009-2019 Antonio Diaz Diaz. +/* Lzlib - Compression library for the lzip format + Copyright (C) 2009-2021 Antonio Diaz Diaz. - This library is free software. Redistribution and use in source and - binary forms, with or without modification, are permitted provided - that the following conditions are met: + This library is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ enum { price_shift_bits = 6, @@ -197,7 +197,7 @@ static inline int price_symbol_reversed( const Bit_model bm[], int symbol, const bool bit = symbol & 1; symbol >>= 1; price += price_bit( bm[model], bit ); - model = ( model << 1 ) | bit; + model <<= 1; model |= bit; } return price; } @@ -240,7 +240,7 @@ struct Matchfinder_base int pos_array_size; int saved_dictionary_size; /* dictionary_size restored by Mb_reset */ bool at_stream_end; /* stream_pos shows real end of file */ - bool flushing; + bool sync_flush_pending; }; static bool Mb_normalize_pos( struct Matchfinder_base * const mb ); @@ -265,19 +265,20 @@ Mb_data_position( const struct Matchfinder_base * const mb ) { return mb->partial_data_pos + mb->pos; } static inline void Mb_finish( struct Matchfinder_base * const mb ) - { mb->at_stream_end = true; mb->flushing = false; } + { mb->at_stream_end = true; mb->sync_flush_pending = false; } static inline bool Mb_data_finished( const struct Matchfinder_base * const mb ) - { return mb->at_stream_end && !mb->flushing && mb->pos >= mb->stream_pos; } + { return mb->at_stream_end && mb->pos >= mb->stream_pos; } static inline bool Mb_flushing_or_end( const struct Matchfinder_base * const mb ) - { return mb->at_stream_end || mb->flushing; } + { return mb->at_stream_end || mb->sync_flush_pending; } static inline int Mb_free_bytes( const struct 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 ) +static inline bool +Mb_enough_available_bytes( const struct Matchfinder_base * const mb ) { return ( mb->pos + mb->after_size <= mb->stream_pos || ( Mb_flushing_or_end( mb ) && mb->pos < mb->stream_pos ) ); } @@ -426,28 +427,30 @@ static inline void Re_encode_bit( struct Range_encoder * const renc, static inline void Re_encode_tree3( struct Range_encoder * const renc, Bit_model bm[], const int symbol ) { - int model = 1; + int model; bool bit = ( symbol >> 2 ) & 1; - Re_encode_bit( renc, &bm[model], bit ); model = ( model << 1 ) | bit; + Re_encode_bit( renc, &bm[1], bit ); + model = 2 | bit; bit = ( symbol >> 1 ) & 1; - Re_encode_bit( renc, &bm[model], bit ); model = ( model << 1 ) | bit; + Re_encode_bit( renc, &bm[model], bit ); model <<= 1; model |= bit; Re_encode_bit( renc, &bm[model], symbol & 1 ); } static inline void Re_encode_tree6( struct Range_encoder * const renc, Bit_model bm[], const unsigned symbol ) { - int model = 1; + int model; bool bit = ( symbol >> 5 ) & 1; - Re_encode_bit( renc, &bm[model], bit ); model = ( model << 1 ) | bit; + Re_encode_bit( renc, &bm[1], bit ); + model = 2 | bit; bit = ( symbol >> 4 ) & 1; - Re_encode_bit( renc, &bm[model], bit ); model = ( model << 1 ) | bit; + Re_encode_bit( renc, &bm[model], bit ); model <<= 1; model |= bit; bit = ( symbol >> 3 ) & 1; - Re_encode_bit( renc, &bm[model], bit ); model = ( model << 1 ) | bit; + Re_encode_bit( renc, &bm[model], bit ); model <<= 1; model |= bit; bit = ( symbol >> 2 ) & 1; - Re_encode_bit( renc, &bm[model], bit ); model = ( model << 1 ) | bit; + Re_encode_bit( renc, &bm[model], bit ); model <<= 1; model |= bit; bit = ( symbol >> 1 ) & 1; - Re_encode_bit( renc, &bm[model], bit ); model = ( model << 1 ) | bit; + Re_encode_bit( renc, &bm[model], bit ); model <<= 1; model |= bit; Re_encode_bit( renc, &bm[model], symbol & 1 ); } @@ -460,7 +463,7 @@ static inline void Re_encode_tree8( struct Range_encoder * const renc, { const bool bit = ( symbol >> i ) & 1; Re_encode_bit( renc, &bm[model], bit ); - model = ( model << 1 ) | bit; + model <<= 1; model |= bit; } } @@ -474,7 +477,7 @@ static inline void Re_encode_tree_reversed( struct Range_encoder * const renc, const bool bit = symbol & 1; symbol >>= 1; Re_encode_bit( renc, &bm[model], bit ); - model = ( model << 1 ) | bit; + model <<= 1; model |= bit; } } @@ -561,7 +564,7 @@ static inline bool LZeb_init( struct LZ_encoder_base * const eb, } static inline bool LZeb_member_finished( const struct LZ_encoder_base * const eb ) - { return ( eb->member_finished && !Cb_used_bytes( &eb->renc.cb ) ); } + { return ( eb->member_finished && Cb_empty( &eb->renc.cb ) ); } static inline void LZeb_free( struct LZ_encoder_base * const eb ) { Re_free( &eb->renc ); Mb_free( &eb->mb ); } diff --git a/fast_encoder.c b/fast_encoder.c index 48f8368..bdcbb97 100644 --- a/fast_encoder.c +++ b/fast_encoder.c @@ -1,23 +1,23 @@ -/* Lzlib - Compression library for the lzip format - Copyright (C) 2009-2019 Antonio Diaz Diaz. +/* Lzlib - Compression library for the lzip format + Copyright (C) 2009-2021 Antonio Diaz Diaz. - This library is free software. Redistribution and use in source and - binary forms, with or without modification, are permitted provided - that the following conditions are met: + This library is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ -int FLZe_longest_match_len( struct FLZ_encoder * const fe, int * const distance ) +static int FLZe_longest_match_len( struct FLZ_encoder * const fe, int * const distance ) { enum { len_limit = 16 }; const uint8_t * const data = Mb_ptr_to_current_pos( &fe->eb.mb ); @@ -59,7 +59,7 @@ int FLZe_longest_match_len( struct FLZ_encoder * const fe, int * const distance } -bool FLZe_encode_member( struct FLZ_encoder * const fe ) +static bool FLZe_encode_member( struct FLZ_encoder * const fe ) { int rep = 0, i; State * const state = &fe->eb.state; @@ -86,7 +86,7 @@ bool FLZe_encode_member( struct FLZ_encoder * const fe ) while( !Mb_data_finished( &fe->eb.mb ) && Re_member_position( &fe->eb.renc ) < fe->eb.member_size_limit ) { - int match_distance; + int match_distance = 0; /* avoid warning from gcc 6.1.0 */ int main_len, pos_state; int len = 0; if( !Mb_enough_available_bytes( &fe->eb.mb ) || diff --git a/fast_encoder.h b/fast_encoder.h index 6066b85..1c3a6ff 100644 --- a/fast_encoder.h +++ b/fast_encoder.h @@ -1,20 +1,20 @@ -/* Lzlib - Compression library for the lzip format - Copyright (C) 2009-2019 Antonio Diaz Diaz. +/* Lzlib - Compression library for the lzip format + Copyright (C) 2009-2021 Antonio Diaz Diaz. - This library is free software. Redistribution and use in source and - binary forms, with or without modification, are permitted provided - that the following conditions are met: + This library is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ struct FLZ_encoder @@ -31,21 +31,19 @@ static inline void FLZe_reset_key4( struct FLZ_encoder * const fe ) fe->key4 = ( fe->key4 << 4 ) ^ fe->eb.mb.buffer[i]; } -int FLZe_longest_match_len( struct FLZ_encoder * const fe, int * const distance ); - static inline bool FLZe_update_and_move( struct FLZ_encoder * const fe, int n ) { + struct Matchfinder_base * const mb = &fe->eb.mb; while( --n >= 0 ) { - if( Mb_available_bytes( &fe->eb.mb ) >= 4 ) + if( Mb_available_bytes( mb ) >= 4 ) { - fe->key4 = ( ( fe->key4 << 4 ) ^ fe->eb.mb.buffer[fe->eb.mb.pos+3] ) & - fe->eb.mb.key4_mask; - fe->eb.mb.pos_array[fe->eb.mb.cyclic_pos] = fe->eb.mb.prev_positions[fe->key4]; - fe->eb.mb.prev_positions[fe->key4] = fe->eb.mb.pos + 1; + fe->key4 = ( ( fe->key4 << 4 ) ^ mb->buffer[mb->pos+3] ) & mb->key4_mask; + mb->pos_array[mb->cyclic_pos] = mb->prev_positions[fe->key4]; + mb->prev_positions[fe->key4] = mb->pos + 1; } - else fe->eb.mb.pos_array[fe->eb.mb.cyclic_pos] = 0; - if( !Mb_move_pos( &fe->eb.mb ) ) return false; + else mb->pos_array[mb->cyclic_pos] = 0; + if( !Mb_move_pos( mb ) ) return false; } return true; } diff --git a/ffexample.c b/ffexample.c new file mode 100644 index 0000000..2891f02 --- /dev/null +++ b/ffexample.c @@ -0,0 +1,300 @@ +/* File to file example - Test program for the library lzlib + Copyright (C) 2010-2021 Antonio Diaz Diaz. + + This program is free software: you have unlimited permission + to copy, distribute, and modify it. + + Try 'ffexample -h' for usage information. + + This program is an example of how file-to-file + compression/decompression can be implemented using lzlib. +*/ + +#define _FILE_OFFSET_BITS 64 + +#include <errno.h> +#include <limits.h> +#include <stdbool.h> +#include <stdint.h> +#include <stdio.h> +#include <stdlib.h> +#include <string.h> +#include <unistd.h> +#if defined(__MSVCRT__) || defined(__OS2__) || defined(__DJGPP__) +#include <fcntl.h> +#include <io.h> +#endif + +#include "lzlib.h" + +#ifndef min + #define min(x,y) ((x) <= (y) ? (x) : (y)) +#endif + + +static void show_help( void ) + { + printf( "ffexample is an example program showing how file-to-file (de)compression can\n" + "be implemented using lzlib. The content of infile is compressed,\n" + "decompressed, or both, and then written to outfile.\n" + "\nUsage: ffexample operation [infile [outfile]]\n" ); + printf( "\nOperation:\n" + " -h display this help and exit\n" + " -c compress infile to outfile\n" + " -d decompress infile to outfile\n" + " -b both (compress then decompress) infile to outfile\n" + " -m compress (multimember) infile to outfile\n" + " -l compress (1 member per line) infile to outfile\n" + " -r decompress with resync if data error or leading garbage\n" + "\nIf infile or outfile are omitted, or are specified as '-', standard input or\n" + "standard output are used in their place respectively.\n" + "\nReport bugs to lzip-bug@nongnu.org\n" + "Lzlib home page: http://www.nongnu.org/lzip/lzlib.html\n" ); + } + + +int ffcompress( struct LZ_Encoder * const encoder, + FILE * const infile, FILE * const outfile ) + { + enum { buffer_size = 16384 }; + uint8_t buffer[buffer_size]; + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_compress_write_size( encoder ) ); + if( size > 0 ) + { + len = fread( buffer, 1, size, infile ); + ret = LZ_compress_write( encoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) ) LZ_compress_finish( encoder ); + } + ret = LZ_compress_read( encoder, buffer, buffer_size ); + if( ret < 0 ) break; + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_compress_finished( encoder ) == 1 ) return 0; + } + return 1; + } + + +int ffdecompress( struct LZ_Decoder * const decoder, + FILE * const infile, FILE * const outfile ) + { + enum { buffer_size = 16384 }; + uint8_t buffer[buffer_size]; + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_decompress_write_size( decoder ) ); + if( size > 0 ) + { + len = fread( buffer, 1, size, infile ); + ret = LZ_decompress_write( decoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) ) LZ_decompress_finish( decoder ); + } + ret = LZ_decompress_read( decoder, buffer, buffer_size ); + if( ret < 0 ) break; + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_decompress_finished( decoder ) == 1 ) return 0; + } + return 1; + } + + +int ffboth( struct LZ_Encoder * const encoder, + struct LZ_Decoder * const decoder, + FILE * const infile, FILE * const outfile ) + { + enum { buffer_size = 16384 }; + uint8_t buffer[buffer_size]; + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_compress_write_size( encoder ) ); + if( size > 0 ) + { + len = fread( buffer, 1, size, infile ); + ret = LZ_compress_write( encoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) ) LZ_compress_finish( encoder ); + } + size = min( buffer_size, LZ_decompress_write_size( decoder ) ); + if( size > 0 ) + { + ret = LZ_compress_read( encoder, buffer, size ); + if( ret < 0 ) break; + ret = LZ_decompress_write( decoder, buffer, ret ); + if( ret < 0 ) break; + if( LZ_compress_finished( encoder ) == 1 ) + LZ_decompress_finish( decoder ); + } + ret = LZ_decompress_read( decoder, buffer, buffer_size ); + if( ret < 0 ) break; + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_decompress_finished( decoder ) == 1 ) return 0; + } + return 1; + } + + +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 ); + if( !encoder || LZ_compress_errno( encoder ) != LZ_ok ) + { fputs( "ffexample: Not enough memory.\n", stderr ); + LZ_compress_close( encoder ); return 1; } + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_compress_write_size( encoder ) ); + if( size > 0 ) + { + len = fread( buffer, 1, size, infile ); + ret = LZ_compress_write( encoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) ) LZ_compress_finish( encoder ); + } + ret = LZ_compress_read( encoder, buffer, buffer_size ); + if( ret < 0 ) break; + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_compress_member_finished( encoder ) == 1 ) + { + if( LZ_compress_finished( encoder ) == 1 ) { done = true; break; } + if( LZ_compress_restart_member( encoder, member_size ) < 0 ) break; + } + } + if( LZ_compress_close( encoder ) < 0 ) done = false; + return done; + } + + +/* Compresses 'infile' to 'outfile' as a multimember stream with one member + for each line of text terminated by a newline character or by EOF. + Returns 0 if success, 1 if error. +*/ +int fflfcompress( struct LZ_Encoder * const encoder, + FILE * const infile, FILE * const outfile ) + { + enum { buffer_size = 16384 }; + uint8_t buffer[buffer_size]; + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_compress_write_size( encoder ) ); + if( size > 0 ) + { + for( len = 0; len < size; ) + { + int ch = getc( infile ); + if( ch == EOF || ( buffer[len++] = ch ) == '\n' ) break; + } + /* avoid writing an empty member to outfile */ + if( len == 0 && LZ_compress_data_position( encoder ) == 0 ) return 0; + ret = LZ_compress_write( encoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) || buffer[len-1] == '\n' ) + LZ_compress_finish( encoder ); + } + ret = LZ_compress_read( encoder, buffer, buffer_size ); + if( ret < 0 ) break; + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_compress_member_finished( encoder ) == 1 ) + { + if( feof( infile ) && LZ_compress_finished( encoder ) == 1 ) return 0; + if( LZ_compress_restart_member( encoder, INT64_MAX ) < 0 ) break; + } + } + return 1; + } + + +/* Decompresses 'infile' to 'outfile' with automatic resynchronization to + next member in case of data error, including the automatic removal of + leading garbage. +*/ +int ffrsdecompress( struct LZ_Decoder * const decoder, + FILE * const infile, FILE * const outfile ) + { + enum { buffer_size = 16384 }; + uint8_t buffer[buffer_size]; + while( true ) + { + int len, ret; + int size = min( buffer_size, LZ_decompress_write_size( decoder ) ); + if( size > 0 ) + { + len = fread( buffer, 1, size, infile ); + ret = LZ_decompress_write( decoder, buffer, len ); + if( ret < 0 || ferror( infile ) ) break; + if( feof( infile ) ) LZ_decompress_finish( decoder ); + } + ret = LZ_decompress_read( decoder, buffer, buffer_size ); + if( ret < 0 ) + { + if( LZ_decompress_errno( decoder ) == LZ_header_error || + LZ_decompress_errno( decoder ) == LZ_data_error ) + { LZ_decompress_sync_to_member( decoder ); continue; } + else break; + } + len = fwrite( buffer, 1, ret, outfile ); + if( len < ret ) break; + if( LZ_decompress_finished( decoder ) == 1 ) return 0; + } + return 1; + } + + +int main( const int argc, const char * const argv[] ) + { +#if defined(__MSVCRT__) || defined(__OS2__) || defined(__DJGPP__) + setmode( STDIN_FILENO, O_BINARY ); + 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 ) ? + fopen( argv[2], "rb" ) : stdin; + FILE * const outfile = ( argc >= 4 && strcmp( argv[3], "-" ) != 0 ) ? + fopen( argv[3], "wb" ) : stdout; + int retval; + + if( argc < 2 || argc > 4 || strlen( argv[1] ) != 2 || argv[1][0] != '-' ) + { show_help(); return 1; } + if( !encoder || LZ_compress_errno( encoder ) != LZ_ok || + !decoder || LZ_decompress_errno( decoder ) != LZ_ok ) + { fputs( "ffexample: Not enough memory.\n", stderr ); + LZ_compress_close( encoder ); LZ_decompress_close( decoder ); return 1; } + if( !infile ) + { fprintf( stderr, "ffexample: Can't open input file '%s': %s\n", + argv[2], strerror( errno ) ); return 1; } + if( !outfile ) + { fprintf( stderr, "ffexample: Can't open output file '%s': %s\n", + argv[3], strerror( errno ) ); return 1; } + + switch( argv[1][1] ) + { + case 'c': retval = ffcompress( encoder, infile, outfile ); break; + case 'd': retval = ffdecompress( decoder, infile, outfile ); break; + case 'b': retval = ffboth( encoder, decoder, infile, outfile ); break; + case 'm': retval = ffmmcompress( infile, outfile ); break; + case 'l': retval = fflfcompress( encoder, infile, outfile ); break; + case 'r': retval = ffrsdecompress( decoder, infile, outfile ); break; + default: show_help(); return ( argv[1][1] != 'h' ); + } + + if( LZ_decompress_close( decoder ) < 0 || LZ_compress_close( encoder ) < 0 || + fclose( outfile ) != 0 || fclose( infile ) != 0 ) retval = 1; + return retval; + } @@ -1,15 +1,14 @@ -/* Lzcheck - Test program for the lzlib library - Copyright (C) 2009-2019 Antonio Diaz Diaz. +/* Lzcheck - Test program for the library lzlib + Copyright (C) 2009-2021 Antonio Diaz Diaz. - This program is free software: you have unlimited permission - to copy, distribute and modify it. + This program is free software: you have unlimited permission + to copy, distribute, and modify it. - Usage is: - lzcheck filename.txt... + Usage: lzcheck [-m|-s] filename.txt... - This program reads each specified text file and then compresses it, - line by line, to test the flushing mechanism and the member - restart/reset/sync functions. + This program reads each text file specified and then compresses it, + line by line, to test the flushing mechanism and the member + restart/reset/sync functions. */ #define _FILE_OFFSET_BITS 64 @@ -21,21 +20,19 @@ #include <stdlib.h> #include <string.h> #include <unistd.h> +#include <sys/stat.h> #include "lzlib.h" -#ifndef min - #define min(x,y) ((x) <= (y) ? (x) : (y)) -#endif - +const unsigned long long member_size = INT64_MAX; enum { buffer_size = 32768 }; uint8_t in_buffer[buffer_size]; uint8_t mid_buffer[buffer_size]; uint8_t out_buffer[buffer_size]; -void show_line( const uint8_t * const buffer, const int size ) +static void show_line( const uint8_t * const buffer, const int size ) { int i; for( i = 0; i < size; ++i ) @@ -44,177 +41,281 @@ void show_line( const uint8_t * const buffer, const int size ) } -int lzcheck( FILE * const file, const int dictionary_size ) +static struct LZ_Encoder * xopen_encoder( const int dictionary_size ) { const int match_len_limit = 16; - const unsigned long long member_size = 0x7FFFFFFFFFFFFFFFULL; /* INT64_MAX */ - struct LZ_Encoder * encoder; - struct LZ_Decoder * decoder; - int retval = 0; - - encoder = LZ_compress_open( dictionary_size, match_len_limit, member_size ); + struct LZ_Encoder * const encoder = + LZ_compress_open( dictionary_size, match_len_limit, member_size ); if( !encoder || LZ_compress_errno( encoder ) != LZ_ok ) { - const bool mem_error = ( LZ_compress_errno( encoder ) == LZ_mem_error ); + const bool bad_arg = + encoder && ( LZ_compress_errno( encoder ) == LZ_bad_argument ); LZ_compress_close( encoder ); - if( mem_error ) + if( bad_arg ) { - fputs( "lzcheck: Not enough memory.\n", stderr ); - return 1; + fputs( "lzcheck: internal error: Invalid argument to encoder.\n", stderr ); + exit( 3 ); } - fputs( "lzcheck: internal error: Invalid argument to encoder.\n", stderr ); - return 3; + fputs( "lzcheck: Not enough memory.\n", stderr ); + exit( 1 ); } + return encoder; + } - decoder = LZ_decompress_open(); +static struct LZ_Decoder * xopen_decoder( void ) + { + struct LZ_Decoder * const decoder = LZ_decompress_open(); if( !decoder || LZ_decompress_errno( decoder ) != LZ_ok ) { LZ_decompress_close( decoder ); fputs( "lzcheck: Not enough memory.\n", stderr ); - return 1; + exit( 1 ); } + return decoder; + } - while( retval <= 1 ) - { - int l, r; - const int read_size = fread( in_buffer, 1, buffer_size, file ); - if( read_size <= 0 ) break; /* end of file */ - for( l = 0, r = 1; r <= read_size; l = r, ++r ) +static void xclose_encoder( struct LZ_Encoder * const encoder, + const bool finish ) + { + if( finish ) + { + unsigned long long size = 0; + LZ_compress_finish( encoder ); + while( true ) { - int in_size, mid_size, out_size; - while( r < read_size && in_buffer[r-1] != '\n' ) ++r; - in_size = LZ_compress_write( encoder, in_buffer + l, r - l ); - if( in_size < r - l ) r = l + in_size; - LZ_compress_sync_flush( encoder ); - mid_size = LZ_compress_read( encoder, mid_buffer, buffer_size ); - if( mid_size < 0 ) + const int rd = LZ_compress_read( encoder, mid_buffer, buffer_size ); + if( rd < 0 ) { - fprintf( stderr, "lzcheck: LZ_compress_read error: %s\n", + fprintf( stderr, "lzcheck: xclose: LZ_compress_read error: %s\n", LZ_strerror( LZ_compress_errno( encoder ) ) ); - retval = 3; break; + exit( 3 ); } - LZ_decompress_write( decoder, mid_buffer, mid_size ); - out_size = LZ_decompress_read( decoder, out_buffer, buffer_size ); - if( out_size < 0 ) + size += rd; + if( LZ_compress_finished( encoder ) == 1 ) break; + } + if( size > 0 ) + { + fprintf( stderr, "lzcheck: %lld bytes remain in encoder.\n", size ); + exit( 3 ); + } + } + if( LZ_compress_close( encoder ) < 0 ) exit( 1 ); + } + + +static void xclose_decoder( struct LZ_Decoder * const decoder, + const bool finish ) + { + if( finish ) + { + unsigned long long size = 0; + LZ_decompress_finish( decoder ); + while( true ) + { + const int rd = LZ_decompress_read( decoder, out_buffer, buffer_size ); + if( rd < 0 ) { - fprintf( stderr, "lzcheck: LZ_decompress_read error: %s\n", + fprintf( stderr, "lzcheck: xclose: LZ_decompress_read error: %s\n", LZ_strerror( LZ_decompress_errno( decoder ) ) ); - retval = 3; break; + exit( 3 ); } + size += rd; + if( LZ_decompress_finished( decoder ) == 1 ) break; + } + if( size > 0 ) + { + fprintf( stderr, "lzcheck: %lld bytes remain in decoder.\n", size ); + exit( 3 ); + } + } + if( LZ_decompress_close( decoder ) < 0 ) exit( 1 ); + } - if( out_size != in_size || memcmp( in_buffer + l, out_buffer, out_size ) ) - { - fprintf( stderr, "lzcheck: Sync error at pos %d in_size = %d, " - "out_size = %d\n", - l, in_size, out_size ); - show_line( in_buffer + l, in_size ); - show_line( out_buffer, out_size ); - retval = 1; - } + +/* Returns the next (usually newline-terminated) chunk of data from file. + The size returned in *sizep is always <= buffer_size. + If sizep is a null pointer, rewinds the file, resets state, and returns. + If file is at EOF, returns an empty line. */ +static const uint8_t * next_line( FILE * const file, int * const sizep ) + { + static int l = 0; + static int read_size = 0; + int r; + + if( !sizep ) { rewind( file ); l = read_size = 0; return in_buffer; } + if( l >= read_size ) + { + l = 0; read_size = fread( in_buffer, 1, buffer_size, file ); + if( l >= read_size ) { *sizep = 0; return in_buffer; } /* end of file */ + } + + for( r = l + 1; r < read_size && in_buffer[r-1] != '\n'; ++r ); + *sizep = r - l; l = r; + return in_buffer + l - *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(); + int retval = 0; + + while( retval <= 1 ) /* test LZ_compress_sync_flush */ + { + int in_size, mid_size, out_size; + int line_size; + const uint8_t * const line_buf = next_line( file, &line_size ); + if( line_size <= 0 ) break; /* end of file */ + + in_size = LZ_compress_write( encoder, line_buf, line_size ); + if( in_size < line_size ) + fprintf( stderr, "lzcheck: sync: LZ_compress_write only accepted %d of %d bytes\n", + in_size, line_size ); + LZ_compress_sync_flush( encoder ); + if( line_buf[0] & 1 ) /* read all data at once or byte by byte */ + mid_size = LZ_compress_read( encoder, mid_buffer, buffer_size ); + else for( mid_size = 0; mid_size < buffer_size; ) + { + const int rd = LZ_compress_read( encoder, mid_buffer + mid_size, 1 ); + if( rd > 0 ) mid_size += rd; + else { if( rd < 0 ) { mid_size = -1; } break; } + } + if( mid_size < 0 ) + { + fprintf( stderr, "lzcheck: LZ_compress_read error: %s\n", + LZ_strerror( LZ_compress_errno( encoder ) ) ); + retval = 3; break; + } + LZ_decompress_write( decoder, mid_buffer, mid_size ); + out_size = LZ_decompress_read( decoder, out_buffer, buffer_size ); + if( out_size < 0 ) + { + fprintf( stderr, "lzcheck: LZ_decompress_read error: %s\n", + LZ_strerror( LZ_decompress_errno( decoder ) ) ); + retval = 3; break; + } + + if( out_size != in_size || memcmp( line_buf, out_buffer, out_size ) ) + { + fprintf( stderr, "lzcheck: LZ_compress_sync_flush error: " + "in_size = %d, out_size = %d\n", in_size, out_size ); + show_line( line_buf, in_size ); + show_line( out_buffer, out_size ); + retval = 1; } } if( retval <= 1 ) { - rewind( file ); + int rd = 0; if( LZ_compress_finish( encoder ) < 0 || - LZ_compress_finish( encoder ) < 0 || - LZ_decompress_write( decoder, mid_buffer, - LZ_compress_read( encoder, mid_buffer, buffer_size ) ) < 0 || - LZ_decompress_read( decoder, out_buffer, buffer_size ) != 0 || - LZ_compress_finish( encoder ) < 0 || - LZ_compress_restart_member( encoder, member_size ) < 0 ) + ( rd = LZ_compress_read( encoder, mid_buffer, buffer_size ) ) < 0 ) { - fprintf( stderr, "lzcheck: Can't finish member: %s\n", - LZ_strerror( LZ_decompress_errno( decoder ) ) ); + fprintf( stderr, "lzcheck: Can't drain encoder: %s\n", + LZ_strerror( LZ_compress_errno( encoder ) ) ); retval = 3; } + LZ_decompress_write( decoder, mid_buffer, rd ); } - while( retval <= 1 ) + xclose_decoder( decoder, retval == 0 ); + xclose_encoder( encoder, retval == 0 ); + return retval; + } + + +/* Test member by member decompression without calling LZ_decompress_finish, + inserting leading garbage before some members, and resetting the + decompressor sometimes. Test that the increase in total_in_size when + syncing to member is equal to the size of the leading garbage skipped. +*/ +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(); + int retval = 0; + + while( retval <= 1 ) /* test LZ_compress_restart_member */ { - int l, r, size; - const int read_size = fread( in_buffer, 1, buffer_size / 2, file ); - if( read_size <= 0 ) break; /* end of file */ + unsigned long long garbage_begin = 0; /* avoid warning from gcc 3.3.6 */ + int leading_garbage, in_size, mid_size, out_size; + int line_size; + const uint8_t * const line_buf = next_line( file, &line_size ); + if( line_size <= 0 && /* end of file, write at least 1 member */ + LZ_decompress_total_in_size( decoder ) != 0 ) break; - for( l = 0, r = 1; r <= read_size; l = r, ++r ) + if( LZ_compress_finished( encoder ) == 1 ) { - int leading_garbage, in_size, mid_size, out_size; - while( r < read_size && in_buffer[r-1] != '\n' ) ++r; - leading_garbage = (l == 0) ? min( r, read_size ) / 2 : 0; - in_size = LZ_compress_write( encoder, in_buffer + l, r - l ); - if( in_size < r - l ) r = l + in_size; - LZ_compress_sync_flush( encoder ); - if( leading_garbage ) - memset( mid_buffer, in_buffer[0], leading_garbage ); - mid_size = LZ_compress_read( encoder, mid_buffer + leading_garbage, - buffer_size - leading_garbage ); - if( mid_size < 0 ) + if( LZ_compress_restart_member( encoder, member_size ) < 0 ) { - fprintf( stderr, "lzcheck: LZ_compress_read error: %s\n", + fprintf( stderr, "lzcheck: Can't restart member: %s\n", LZ_strerror( LZ_compress_errno( encoder ) ) ); retval = 3; break; } - LZ_decompress_write( decoder, mid_buffer, mid_size + leading_garbage ); - out_size = LZ_decompress_read( decoder, out_buffer, buffer_size ); - if( out_size < 0 ) + if( line_size >= 2 && line_buf[1] == 'h' ) + LZ_decompress_reset( decoder ); + } + in_size = LZ_compress_write( encoder, line_buf, line_size ); + if( in_size < line_size ) + fprintf( stderr, "lzcheck: member: LZ_compress_write only accepted %d of %d bytes\n", + in_size, line_size ); + LZ_compress_finish( encoder ); + if( line_size * 3 < buffer_size && line_buf[0] == 't' ) + { leading_garbage = line_size; + memset( mid_buffer, in_buffer[0], leading_garbage ); + garbage_begin = LZ_decompress_total_in_size( decoder ); } + else leading_garbage = 0; + mid_size = LZ_compress_read( encoder, mid_buffer + leading_garbage, + buffer_size - leading_garbage ); + if( mid_size < 0 ) + { + fprintf( stderr, "lzcheck: member: LZ_compress_read error: %s\n", + LZ_strerror( LZ_compress_errno( encoder ) ) ); + retval = 3; break; + } + LZ_decompress_write( decoder, mid_buffer, leading_garbage + mid_size ); + out_size = LZ_decompress_read( decoder, out_buffer, buffer_size ); + if( out_size < 0 ) + { + if( leading_garbage && + ( LZ_decompress_errno( decoder ) == LZ_header_error || + LZ_decompress_errno( decoder ) == LZ_data_error ) ) { - if( LZ_decompress_errno( decoder ) == LZ_header_error || - LZ_decompress_errno( decoder ) == LZ_data_error ) + LZ_decompress_sync_to_member( decoder ); /* skip leading garbage */ + const unsigned long long garbage_end = + LZ_decompress_total_in_size( decoder ); + if( garbage_end - garbage_begin != (unsigned)leading_garbage ) { - LZ_decompress_sync_to_member( decoder ); /* remove leading garbage */ - out_size = LZ_decompress_read( decoder, out_buffer, buffer_size ); - } - if( out_size < 0 ) - { - fprintf( stderr, "lzcheck: LZ_decompress_read error: %s\n", - LZ_strerror( LZ_decompress_errno( decoder ) ) ); + fprintf( stderr, "lzcheck: member: LZ_decompress_sync_to_member error:\n" + " garbage_begin = %llu garbage_end = %llu " + "difference = %llu expected = %d\n", garbage_begin, + garbage_end, garbage_end - garbage_begin, leading_garbage ); retval = 3; break; } + out_size = LZ_decompress_read( decoder, out_buffer, buffer_size ); } - - if( out_size != in_size || memcmp( in_buffer + l, out_buffer, out_size ) ) + if( out_size < 0 ) { - fprintf( stderr, "lzcheck: Sync error at pos %d in_size = %d, " - "out_size = %d, leading garbage = %d\n", - l, in_size, out_size, leading_garbage ); - show_line( in_buffer + l, in_size ); - show_line( out_buffer, out_size ); - retval = 1; + fprintf( stderr, "lzcheck: member: LZ_decompress_read error: %s\n", + LZ_strerror( LZ_decompress_errno( decoder ) ) ); + retval = 3; break; } } - if( retval >= 3 ) break; - if( LZ_compress_finish( encoder ) < 0 || - LZ_decompress_write( decoder, mid_buffer, - LZ_compress_read( encoder, mid_buffer, buffer_size ) ) < 0 || - LZ_decompress_read( decoder, out_buffer, buffer_size ) != 0 || - LZ_decompress_reset( decoder ) < 0 || - LZ_compress_restart_member( encoder, member_size ) < 0 ) + if( out_size != in_size || memcmp( line_buf, out_buffer, out_size ) ) { - fprintf( stderr, "lzcheck: Can't restart member: %s\n", - LZ_strerror( LZ_decompress_errno( decoder ) ) ); - retval = 3; break; - } - - size = min( 100, read_size ); - if( LZ_compress_write( encoder, in_buffer, size ) != size || - LZ_compress_finish( encoder ) < 0 || - LZ_decompress_write( decoder, mid_buffer, - LZ_compress_read( encoder, mid_buffer, buffer_size ) ) < 0 || - LZ_decompress_read( decoder, out_buffer, 0 ) != 0 || - LZ_decompress_sync_to_member( decoder ) < 0 || - LZ_compress_restart_member( encoder, member_size ) < 0 ) - { - fprintf( stderr, "lzcheck: Can't seek to next member: %s\n", - LZ_strerror( LZ_decompress_errno( decoder ) ) ); - retval = 3; break; + fprintf( stderr, "lzcheck: LZ_compress_restart_member error: " + "in_size = %d, out_size = %d\n", in_size, out_size ); + show_line( line_buf, in_size ); + show_line( out_buffer, out_size ); + retval = 1; } } - LZ_decompress_close( decoder ); - LZ_compress_close( encoder ); + xclose_decoder( decoder, retval == 0 ); + xclose_encoder( encoder, retval == 0 ); return retval; } @@ -223,7 +324,11 @@ int main( const int argc, const char * const argv[] ) { int retval = 0, i; int open_failures = 0; - const bool verbose = ( argc > 2 ); + const char opt = ( argc > 2 && + ( 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 ); if( argc < 2 ) { @@ -231,8 +336,10 @@ int main( const int argc, const char * const argv[] ) return 1; } - for( i = 1; i < argc && retval == 0; ++ i ) + for( i = first; i < argc && retval == 0; ++i ) { + struct stat st; + if( stat( argv[i], &st ) != 0 || !S_ISREG( st.st_mode ) ) continue; FILE * file = fopen( argv[i], "rb" ); if( !file ) { @@ -241,9 +348,14 @@ int main( const int argc, const char * const argv[] ) } if( verbose ) fprintf( stderr, " Testing file '%s'\n", argv[i] ); - retval = lzcheck( file, 65535 ); /* 65535,16 chooses fast encoder */ - if( retval == 0 ) - { rewind( file ); retval = lzcheck( file, 1 << 20 ); } + /* 65535,16 chooses fast encoder */ + if( opt != 'm' ) retval = check_sync_flush( file, 65535 ); + if( retval == 0 && opt != 'm' ) + { next_line( file, 0 ); retval = check_sync_flush( file, 1 << 20 ); } + if( retval == 0 && opt != 's' ) + { next_line( file, 0 ); retval = check_members( file, 65535 ); } + if( retval == 0 && opt != 's' ) + { next_line( file, 0 ); retval = check_members( file, 1 << 20 ); } fclose( file ); } if( open_failures > 0 && verbose ) @@ -1,20 +1,20 @@ -/* Lzlib - Compression library for the lzip format - Copyright (C) 2009-2019 Antonio Diaz Diaz. +/* Lzlib - Compression library for the lzip format + Copyright (C) 2009-2021 Antonio Diaz Diaz. - This library is free software. Redistribution and use in source and - binary forms, with or without modification, are permitted provided - that the following conditions are met: + This library is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ #ifndef max @@ -83,7 +83,7 @@ static inline int get_len_state( const int len ) { return min( len - min_match_len, len_states - 1 ); } static inline int get_lit_state( const uint8_t prev_byte ) - { return ( prev_byte >> ( 8 - literal_context_bits ) ); } + { return prev_byte >> ( 8 - literal_context_bits ); } enum { bit_model_move_bits = 5, @@ -197,7 +197,7 @@ static const uint8_t lzip_magic[4] = { 0x4C, 0x5A, 0x49, 0x50 }; /* "LZIP" */ typedef uint8_t Lzip_header[6]; /* 0-3 magic bytes */ /* 4 version */ - /* 5 coded_dict_size */ + /* 5 coded dictionary size */ enum { Lh_size = 6 }; static inline void Lh_set_magic( Lzip_header data ) @@ -255,9 +255,8 @@ static inline bool Lh_set_dictionary_size( Lzip_header data, const unsigned sz ) static inline bool Lh_verify( const Lzip_header data ) { - if( Lh_verify_magic( data ) && Lh_verify_version( data ) ) - return isvalid_ds( Lh_get_dictionary_size( data ) ); - return false; + return Lh_verify_magic( data ) && Lh_verify_version( data ) && + isvalid_ds( Lh_get_dictionary_size( data ) ); } @@ -1,20 +1,20 @@ -/* Lzlib - Compression library for the lzip format - Copyright (C) 2009-2019 Antonio Diaz Diaz. +/* Lzlib - Compression library for the lzip format + Copyright (C) 2009-2021 Antonio Diaz Diaz. - This library is free software. Redistribution and use in source and - binary forms, with or without modification, are permitted provided - that the following conditions are met: + This library is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ #include <stdbool.h> @@ -105,10 +105,11 @@ static bool verify_decoder( struct LZ_Decoder * const d ) } -/*------------------------- Misc Functions -------------------------*/ +/* ------------------------- Misc Functions ------------------------- */ -const char * LZ_version( void ) { return LZ_version_string; } +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 ) { @@ -135,7 +136,7 @@ int LZ_min_match_len_limit( void ) { return min_match_len_limit; } int LZ_max_match_len_limit( void ) { return max_match_len; } -/*---------------------- Compression Functions ----------------------*/ +/* --------------------- Compression Functions --------------------- */ struct LZ_Encoder * LZ_compress_open( const int dictionary_size, const int match_len_limit, @@ -226,8 +227,8 @@ int LZ_compress_restart_member( struct LZ_Encoder * const e, int LZ_compress_sync_flush( struct LZ_Encoder * const e ) { if( !verify_encoder( e ) || e->fatal ) return -1; - if( !Mb_flushing_or_end( &e->lz_encoder_base->mb ) ) - e->lz_encoder_base->mb.flushing = true; + if( !e->lz_encoder_base->mb.at_stream_end ) + e->lz_encoder_base->mb.sync_flush_pending = true; return 0; } @@ -235,24 +236,22 @@ int LZ_compress_sync_flush( struct LZ_Encoder * const e ) int LZ_compress_read( struct LZ_Encoder * const e, uint8_t * const buffer, const int size ) { - int out_size = 0; if( !verify_encoder( e ) || e->fatal ) return -1; if( size < 0 ) return 0; - do { + + { struct 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 ) + { if( ( e->flz_encoder && !FLZe_encode_member( e->flz_encoder ) ) || ( e->lz_encoder && !LZe_encode_member( e->lz_encoder ) ) ) { e->lz_errno = LZ_library_error; e->fatal = true; return -1; } - if( e->lz_encoder_base->mb.flushing && - Mb_available_bytes( &e->lz_encoder_base->mb ) <= 0 && - LZeb_sync_flush( e->lz_encoder_base ) ) - e->lz_encoder_base->mb.flushing = false; - out_size += Re_read_data( &e->lz_encoder_base->renc, - buffer + out_size, size - out_size ); + if( eb->mb.sync_flush_pending && Mb_available_bytes( &eb->mb ) <= 0 ) + LZeb_try_sync_flush( eb ); + out_size += Re_read_data( &eb->renc, buffer + out_size, size - out_size ); } - while( e->lz_encoder_base->mb.flushing && out_size < size && - Mb_enough_available_bytes( &e->lz_encoder_base->mb ) && - Re_enough_free_bytes( &e->lz_encoder_base->renc ) ); - return out_size; + return out_size; } } @@ -321,7 +320,7 @@ unsigned long long LZ_compress_total_out_size( struct LZ_Encoder * const e ) } -/*--------------------- Decompression Functions ---------------------*/ +/* -------------------- Decompression Functions -------------------- */ struct LZ_Decoder * LZ_decompress_open( void ) { @@ -401,10 +400,11 @@ int LZ_decompress_read( struct LZ_Decoder * const d, { int result; if( !verify_decoder( d ) ) return -1; + if( size < 0 ) return 0; if( d->fatal ) /* don't return error until pending bytes are read */ { if( d->lz_decoder && !Cb_empty( &d->lz_decoder->cb ) ) goto get_data; return -1; } - if( d->seeking || size < 0 ) return 0; + if( d->seeking ) return 0; if( d->lz_decoder && LZd_member_finished( d->lz_decoder ) ) { @@ -1,35 +1,39 @@ -/* Lzlib - Compression library for the lzip format - Copyright (C) 2009-2019 Antonio Diaz Diaz. +/* Lzlib - Compression library for the lzip format + Copyright (C) 2009-2021 Antonio Diaz Diaz. - This library is free software. Redistribution and use in source and - binary forms, with or without modification, are permitted provided - that the following conditions are met: + This library is free software. Redistribution and use in source and + binary forms, with or without modification, are permitted provided + that the following conditions are met: - 1. Redistributions of source code must retain the above copyright - notice, this list of conditions and the following disclaimer. + 1. Redistributions of source code must retain the above copyright + notice, this list of conditions, and the following disclaimer. - 2. Redistributions in binary form must reproduce the above copyright - notice, this list of conditions and the following disclaimer in the - documentation and/or other materials provided with the distribution. + 2. Redistributions in binary form must reproduce the above copyright + notice, this list of conditions, and the following disclaimer in the + documentation and/or other materials provided with the distribution. - This library is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. + This library is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. */ #ifdef __cplusplus extern "C" { #endif -#define LZ_API_VERSION 1 +/* LZ_API_VERSION was first defined in lzlib 1.8 to 1. + Since lzlib 1.12, LZ_API_VERSION is defined as (major * 1000 + minor). */ -static const char * const LZ_version_string = "1.11"; +#define LZ_API_VERSION 1012 + +static const char * const LZ_version_string = "1.12"; 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 }; +int LZ_api_version( void ); /* new in 1.12 */ const char * LZ_version( void ); const char * LZ_strerror( const enum LZ_Errno lz_errno ); @@ -41,7 +45,7 @@ int LZ_min_match_len_limit( void ); int LZ_max_match_len_limit( void ); -/*---------------------- Compression Functions ----------------------*/ +/* --------------------- Compression Functions --------------------- */ struct LZ_Encoder; @@ -71,7 +75,7 @@ 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 ); -/*--------------------- Decompression Functions ---------------------*/ +/* -------------------- Decompression Functions -------------------- */ struct LZ_Decoder; @@ -1,24 +1,24 @@ -/* Minilzip - Test program for the lzlib library - Copyright (C) 2009-2019 Antonio Diaz Diaz. +/* Minilzip - Test program for the library lzlib + Copyright (C) 2009-2021 Antonio Diaz Diaz. - This program is free software: you can redistribute it and/or modify - it under the terms of the GNU General Public License as published by - the Free Software Foundation, either version 2 of the License, or - (at your option) any later version. + This program is free software: you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation, either version 2 of the License, or + (at your option) any later version. - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU General Public License for more details. + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. - You should have received a copy of the GNU General Public License - along with this program. If not, see <http://www.gnu.org/licenses/>. + You should have received a copy of the GNU General Public License + along with this program. If not, see <http://www.gnu.org/licenses/>. */ /* - Exit status: 0 for a normal exit, 1 for environmental problems - (file not found, invalid flags, I/O errors, etc), 2 to indicate a - corrupt or invalid input file, 3 for an internal consistency error - (eg, bug) which caused minilzip to panic. + Exit status: 0 for a normal exit, 1 for environmental problems + (file not found, invalid flags, I/O errors, etc), 2 to indicate a + corrupt or invalid input file, 3 for an internal consistency error + (eg, bug) which caused minilzip to panic. */ #define _FILE_OFFSET_BITS 64 @@ -74,19 +74,21 @@ #define min(x,y) ((x) <= (y) ? (x) : (y)) #endif -void cleanup_and_fail( const int retval ); -void show_error( const char * const msg, const int errcode, const bool help ); -void show_file_error( const char * const filename, const char * const msg, - const int errcode ); -void internal_error( const char * const msg ); +static void cleanup_and_fail( const int retval ); +static void show_error( const char * const msg, const int errcode, + const bool help ); +static void show_file_error( const char * const filename, + const char * const msg, const int errcode ); +static void internal_error( const char * const msg ); +static const char * const mem_msg = "Not enough memory."; int verbosity = 0; -const char * const program_name = "minilzip"; -const char * const program_year = "2019"; -const char * invocation_name = 0; +static const char * const program_name = "minilzip"; +static const char * const program_year = "2021"; +static const char * invocation_name = "minilzip"; /* default value */ -const struct { const char * from; const char * to; } known_extensions[] = { +static const struct { const char * from; const char * to; } known_extensions[] = { { ".lz", "" }, { ".tlz", ".tar" }, { 0, 0 } }; @@ -101,15 +103,24 @@ enum Mode { m_compress, m_decompress, m_test }; /* Variables used in signal handler context. They are not declared volatile because the handler never returns. */ -char * output_filename = 0; -int outfd = -1; -bool delete_output_on_interrupt = false; +static char * output_filename = 0; +static int outfd = -1; +static bool delete_output_on_interrupt = false; static void show_help( void ) { - printf( "Minilzip is a test program for the lzlib compression library, fully\n" + printf( "Minilzip is a test program for the compression library lzlib, fully\n" "compatible with lzip 1.4 or newer.\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, chosen to maximize safety and\n" + "interoperability. Lzip can compress about as fast as gzip (lzip -0) or\n" + "compress most files more than bzip2 (lzip -9). Decompression speed is\n" + "intermediate between gzip and bzip2. Lzip is better than gzip and bzip2 from\n" + "a data recovery perspective. Lzip has been designed, written, and tested\n" + "with great care to replace gzip and bzip2 as the standard 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" @@ -122,7 +133,7 @@ static void show_help( void ) " -F, --recompress force re-compression of compressed files\n" " -k, --keep keep (don't delete) input files\n" " -m, --match-length=<bytes> set match length limit in bytes [36]\n" - " -o, --output=<file> if reading standard input, write to <file>\n" + " -o, --output=<file> write to <file>, keep input files\n" " -q, --quiet suppress all messages\n" " -s, --dictionary-size=<bytes> set dictionary size limit in bytes [8 MiB]\n" " -S, --volume-size=<bytes> set volume size limit in bytes\n" @@ -132,7 +143,8 @@ static void show_help( void ) " --fast alias for -0\n" " --best alias for -9\n" " --loose-trailing allow trailing data seeming corrupt header\n" - "If no file names are given, or if a file is '-', minilzip compresses or\n" + " --check-lib compare version of lzlib.h with liblz.{a,so}\n" + "\nIf no file names are given, or if a file is '-', minilzip compresses or\n" "decompresses from standard input to standard output.\n" "Numbers may be followed by a multiplier: k = kB = 10^3 = 1000,\n" "Ki = KiB = 2^10 = 1024, M = 10^6, Mi = 2^20, G = 10^9, Gi = 2^30, etc...\n" @@ -140,12 +152,19 @@ static void show_help( void ) "to 2^29 bytes.\n" "\nThe bidimensional parameter space of LZMA can't be mapped to a linear\n" "scale optimal for all files. If your files are large, very repetitive,\n" - "etc, you may need to use the --dictionary-size and --match-length\n" - "options directly to achieve optimal performance.\n" + "etc, you may need to use the options --dictionary-size and --match-length\n" + "directly to achieve optimal performance.\n" + "\nTo extract all the files from archive 'foo.tar.lz', use the commands\n" + "'tar -xf foo.tar.lz' or 'minilzip -cd foo.tar.lz | tar -xf -'.\n" "\nExit status: 0 for a normal exit, 1 for environmental problems (file\n" "not found, invalid flags, I/O errors, etc), 2 to indicate a corrupt or\n" "invalid input file, 3 for an internal consistency error (eg, bug) which\n" "caused minilzip to panic.\n" + "\nThe ideas embodied in lzlib are due to (at least) the following people:\n" + "Abraham Lempel and Jacob Ziv (for the LZ algorithm), Andrey Markov (for the\n" + "definition of Markov chains), G.N.N. Martin (for the definition of range\n" + "encoding), Igor Pavlov (for putting all the above together in LZMA), and\n" + "Julian Seward (for bzip2's CLI).\n" "\nReport bugs to lzip-bug@nongnu.org\n" "Lzlib home page: http://www.nongnu.org/lzip/lzlib.html\n" ); } @@ -162,16 +181,43 @@ static void show_version( void ) } +int check_lib() + { + bool warning = false; + if( strcmp( LZ_version_string, LZ_version() ) != 0 ) + { warning = true; + if( verbosity >= 0 ) + printf( "warning: LZ_version_string != LZ_version() (%s vs %s)\n", + LZ_version_string, LZ_version() ); } +#if defined LZ_API_VERSION && LZ_API_VERSION >= 1012 + if( LZ_API_VERSION != LZ_api_version() ) + { warning = true; + if( verbosity >= 0 ) + printf( "warning: LZ_API_VERSION != LZ_api_version() (%u vs %u)\n", + LZ_API_VERSION, LZ_api_version() ); } +#endif + if( verbosity >= 1 ) + { + printf( "Using lzlib %s\n", LZ_version() ); +#if !defined LZ_API_VERSION + fputs( "LZ_API_VERSION is not defined.\n", stdout ); +#elif LZ_API_VERSION >= 1012 + printf( "Using LZ_API_VERSION = %u\n", LZ_api_version() ); +#else + printf( "Compiled with LZ_API_VERSION = %u. " + "Using an unknown LZ_API_VERSION\n", LZ_API_VERSION ); +#endif + } + return warning; + } + + /* assure at least a minimum size for buffer 'buf' */ static void * resize_buffer( void * buf, const unsigned min_size ) { if( buf ) buf = realloc( buf, min_size ); else buf = malloc( min_size ); - if( !buf ) - { - show_error( "Not enough memory.", 0, false ); - cleanup_and_fail( 1 ); - } + if( !buf ) { show_error( mem_msg, 0, false ); cleanup_and_fail( 1 ); } return buf; } @@ -186,8 +232,7 @@ struct Pretty_print }; static void Pp_init( struct Pretty_print * const pp, - const char * const filenames[], - const int num_filenames ) + const char * const filenames[], const int num_filenames ) { unsigned stdin_name_len; int i; @@ -208,8 +253,8 @@ static void Pp_init( struct Pretty_print * const pp, if( pp->longest_name == 0 ) pp->longest_name = stdin_name_len; } -static inline void Pp_set_name( struct Pretty_print * const pp, - const char * const filename ) +static void Pp_set_name( struct Pretty_print * const pp, + const char * const filename ) { unsigned name_len, padded_name_len, i = 0; @@ -227,7 +272,7 @@ static inline void Pp_set_name( struct Pretty_print * const pp, pp->first_post = true; } -static inline void Pp_reset( struct Pretty_print * const pp ) +static void Pp_reset( struct 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 ) @@ -258,7 +303,7 @@ static void show_header( const unsigned dictionary_size ) int i; for( i = 0; i < 8 && ( num > 9999 || ( exact && num >= factor ) ); ++i ) { num /= factor; if( num % factor != 0 ) exact = false; p = prefix[i]; np = ""; } - fprintf( stderr, "dictionary %s%4u %sB, ", np, num, p ); + fprintf( stderr, "dict %s%4u %sB, ", np, num, p ); } @@ -321,7 +366,7 @@ static int get_dict_size( const char * const arg ) const long bits = strtol( arg, &tail, 0 ); if( bits >= LZ_min_dictionary_bits() && bits <= LZ_max_dictionary_bits() && *tail == 0 ) - return ( 1 << bits ); + return 1 << bits; dictionary_size = getnum( arg, LZ_min_dictionary_size(), LZ_max_dictionary_size() ); if( dictionary_size == 65535 ) ++dictionary_size; /* no fast encoder */ @@ -329,7 +374,7 @@ static int get_dict_size( const char * const arg ) } -void set_mode( enum Mode * const program_modep, const enum Mode new_mode ) +static void set_mode( enum Mode * const program_modep, const enum Mode new_mode ) { if( *program_modep != m_compress && *program_modep != new_mode ) { @@ -363,7 +408,7 @@ static void set_c_outname( const char * const name, const bool force_ext, strlen( known_extensions[0].from ) + 1 ); strcpy( output_filename, name ); if( multifile ) strcat( output_filename, "00001" ); - if( force_ext || multifile || extension_index( output_filename ) < 0 ) + if( force_ext || multifile ) strcat( output_filename, known_extensions[0].from ); } @@ -395,7 +440,7 @@ 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 bool recompress, const bool to_stdout ) + const bool one_to_one, const bool recompress ) { int infd = -1; if( program_mode == m_compress && !recompress && eindex >= 0 ) @@ -416,14 +461,12 @@ static int open_instream( const char * const name, struct stat * const in_statsp const bool can_read = ( i == 0 && ( S_ISBLK( mode ) || S_ISCHR( mode ) || S_ISFIFO( mode ) || S_ISSOCK( mode ) ) ); - const bool no_ofile = ( to_stdout || program_mode == m_test ); - if( i != 0 || ( !S_ISREG( mode ) && ( !can_read || !no_ofile ) ) ) + if( i != 0 || ( !S_ISREG( mode ) && ( !can_read || one_to_one ) ) ) { if( verbosity >= 0 ) fprintf( stderr, "%s: Input file '%s' is not a regular file%s.\n", - program_name, name, - ( can_read && !no_ofile ) ? - ",\n and '--stdout' was not specified" : "" ); + program_name, name, ( can_read && one_to_one ) ? + ",\n and neither '-c' nor '-o' were specified" : "" ); close( infd ); infd = -1; } @@ -433,11 +476,11 @@ static int open_instream( const char * const name, struct stat * const in_statsp } -static bool open_outstream( const bool force, const bool from_stdin ) +static bool open_outstream( const bool force, const bool protect ) { const mode_t usr_rw = S_IRUSR | S_IWUSR; const mode_t all_rw = usr_rw | S_IRGRP | S_IWGRP | S_IROTH | S_IWOTH; - const mode_t outfd_mode = from_stdin ? all_rw : usr_rw; + const mode_t outfd_mode = protect ? usr_rw : all_rw; int flags = O_CREAT | O_WRONLY | O_BINARY; if( force ) flags |= O_TRUNC; else flags |= O_EXCL; @@ -456,25 +499,6 @@ static bool open_outstream( const bool force, const bool from_stdin ) } -static bool check_tty( const char * const input_filename, const int infd, - const enum Mode program_mode ) - { - if( program_mode == m_compress && isatty( outfd ) ) - { - show_error( "I won't write compressed data to a terminal.", 0, true ); - return false; - } - if( ( program_mode == m_decompress || program_mode == m_test ) && - isatty( infd ) ) - { - show_file_error( input_filename, - "I won't read compressed data from a terminal.", 0 ); - return false; - } - return true; - } - - static void set_signals( void (*action)(int) ) { signal( SIGHUP, action ); @@ -483,7 +507,7 @@ static void set_signals( void (*action)(int) ) } -void cleanup_and_fail( const int retval ) +static void cleanup_and_fail( const int retval ) { set_signals( SIG_IGN ); /* ignore signals */ if( delete_output_on_interrupt ) @@ -500,7 +524,7 @@ void cleanup_and_fail( const int retval ) } -void signal_handler( int sig ) +static void signal_handler( int sig ) { if( sig ) {} /* keep compiler happy */ show_error( "Control-C or similar caught, quitting.", 0, false ); @@ -508,7 +532,35 @@ void signal_handler( int sig ) } - /* Set permissions, owner and times. */ +static inline void set_retval( int * retval, const int new_val ) + { if( *retval < new_val ) *retval = new_val; } + + +static bool check_tty_in( const char * const input_filename, const int infd, + const enum Mode program_mode, int * const retval ) + { + if( ( program_mode == m_decompress || program_mode == m_test ) && + isatty( infd ) ) /* for example /dev/tty */ + { show_file_error( input_filename, + "I won't read compressed data from a terminal.", 0 ); + close( infd ); set_retval( retval, 1 ); + if( program_mode != m_test ) cleanup_and_fail( *retval ); + return false; } + return true; + } + +static bool check_tty_out( const enum Mode program_mode ) + { + if( program_mode == m_compress && isatty( outfd ) ) + { show_file_error( output_filename[0] ? + output_filename : "(stdout)", + "I won't write compressed data to a terminal.", 0 ); + return false; } + return true; + } + + +/* Set permissions, owner, and times. */ static void close_and_set_permissions( const struct stat * const in_statsp ) { bool warning = false; @@ -602,7 +654,7 @@ static int do_compress( struct LZ_Encoder * const encoder, { unsigned long long partial_volume_size = 0; enum { buffer_size = 65536 }; - uint8_t buffer[buffer_size]; + uint8_t buffer[buffer_size]; /* read/write buffer */ if( verbosity >= 1 ) Pp_show_msg( pp, 0 ); while( true ) @@ -658,7 +710,7 @@ static int do_compress( struct LZ_Encoder * const encoder, close_and_set_permissions( in_statsp ); if( !next_filename() ) { Pp_show_msg( pp, "Too many volume files." ); return 1; } - if( !open_outstream( true, !in_statsp ) ) return 1; + if( !open_outstream( true, in_statsp ) ) return 1; } } size = min( member_size, volume_size - partial_volume_size ); @@ -726,7 +778,8 @@ static int do_decompress( struct LZ_Decoder * const decoder, const int infd, const bool loose_trailing, const bool testing ) { enum { buffer_size = 65536 }; - uint8_t buffer[buffer_size]; + uint8_t buffer[buffer_size]; /* read/write buffer */ + unsigned long long total_in = 0; /* to detect library stall */ bool first_member; for( first_member = true; ; ) @@ -777,17 +830,16 @@ static int do_decompress( struct LZ_Decoder * const decoder, const int infd, if( verbosity >= 4 ) show_header( LZ_decompress_dictionary_size( decoder ) ); if( data_size == 0 || member_size == 0 ) - fputs( "no data compressed. ", stderr ); + fputs( "no data compressed. ", stderr ); else - fprintf( stderr, "%6.3f:1, %5.2f%% ratio, %5.2f%% saved. ", + fprintf( stderr, "%6.3f:1, %5.2f%% ratio, %5.2f%% saved. ", (double)data_size / member_size, ( 100.0 * member_size ) / data_size, 100.0 - ( ( 100.0 * member_size ) / data_size ) ); if( verbosity >= 4 ) fprintf( stderr, "CRC %08X, ", LZ_decompress_data_crc( decoder ) ); if( verbosity >= 3 ) - fprintf( stderr, "decompressed %9llu, compressed %8llu. ", - data_size, member_size ); + fprintf( stderr, "%9llu out, %8llu in. ", data_size, member_size ); fputs( testing ? "ok\n" : "done\n", stderr ); Pp_reset( pp ); } } @@ -840,8 +892,7 @@ static int do_decompress( struct LZ_Decoder * const decoder, const int infd, else break; /* trailing data */ return 2; } - if( lz_errno == LZ_mem_error ) - { Pp_show_msg( pp, "Not enough memory." ); return 1; } + if( lz_errno == LZ_mem_error ) { Pp_show_msg( pp, mem_msg ); return 1; } if( verbosity >= 0 ) { Pp_show_msg( pp, 0 ); @@ -853,7 +904,11 @@ static int do_decompress( struct LZ_Decoder * const decoder, const int infd, } if( LZ_decompress_finished( decoder ) == 1 ) break; if( in_size == 0 && out_size == 0 ) - internal_error( "library error (stalled)." ); + { + const unsigned long long size = LZ_decompress_total_in_size( decoder ); + if( total_in == size ) internal_error( "library error (stalled)." ); + total_in = size; + } } if( verbosity == 1 ) fputs( testing ? "ok\n" : "done\n", stderr ); return 0; @@ -868,7 +923,7 @@ static int decompress( const int infd, struct Pretty_print * const pp, int retval; if( !decoder || LZ_decompress_errno( decoder ) != LZ_ok ) - { Pp_show_msg( pp, "Not enough memory." ); retval = 1; } + { Pp_show_msg( pp, mem_msg ); retval = 1; } else retval = do_decompress( decoder, infd, pp, ignore_trailing, loose_trailing, testing ); LZ_decompress_close( decoder ); @@ -876,7 +931,8 @@ static int decompress( const int infd, struct Pretty_print * const pp, } -void show_error( const char * const msg, const int errcode, const bool help ) +static void show_error( const char * const msg, const int errcode, + const bool help ) { if( verbosity < 0 ) return; if( msg && msg[0] ) @@ -889,8 +945,8 @@ void show_error( const char * const msg, const int errcode, const bool help ) } -void show_file_error( const char * const filename, const char * const msg, - const int errcode ) +static void show_file_error( const char * const filename, + const char * const msg, const int errcode ) { if( verbosity >= 0 ) fprintf( stderr, "%s: %s: %s%s%s\n", program_name, filename, msg, @@ -899,7 +955,7 @@ void show_file_error( const char * const filename, const char * const msg, } -void internal_error( const char * const msg ) +static void internal_error( const char * const msg ) { if( verbosity >= 0 ) fprintf( stderr, "%s: internal error: %s\n", program_name, msg ); @@ -924,12 +980,14 @@ int main( const int argc, const char * const argv[] ) { 3 << 23, 132 }, /* -8 */ { 1 << 25, 273 } }; /* -9 */ struct Lzma_options encoder_options = option_mapping[6]; /* default = "-6" */ - const unsigned long long max_member_size = 0x0008000000000000ULL; - const unsigned long long max_volume_size = 0x4000000000000000ULL; + 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 = ""; - const char ** filenames = 0; + static struct Arg_parser parser; /* static because valgrind complains */ + static struct Pretty_print pp; /* and memory management in C sucks */ + static const char ** filenames = 0; int num_filenames = 0; enum Mode program_mode = m_compress; int argind = 0; @@ -944,20 +1002,19 @@ int main( const int argc, const char * const argv[] ) bool recompress = false; bool stdin_used = false; bool to_stdout = false; - struct Pretty_print pp; - enum { opt_lt = 256 }; + enum { opt_chk = 256, opt_lt }; const struct ap_Option options[] = { { '0', "fast", ap_no }, - { '1', 0, ap_no }, - { '2', 0, ap_no }, - { '3', 0, ap_no }, - { '4', 0, ap_no }, - { '5', 0, ap_no }, - { '6', 0, ap_no }, - { '7', 0, ap_no }, - { '8', 0, ap_no }, + { '1', 0, ap_no }, + { '2', 0, ap_no }, + { '3', 0, ap_no }, + { '4', 0, ap_no }, + { '5', 0, ap_no }, + { '6', 0, ap_no }, + { '7', 0, ap_no }, + { '8', 0, ap_no }, { '9', "best", ap_no }, { 'a', "trailing-error", ap_no }, { 'b', "member-size", ap_yes }, @@ -976,20 +1033,14 @@ int main( const int argc, const char * const argv[] ) { 't', "test", ap_no }, { 'v', "verbose", ap_no }, { 'V', "version", ap_no }, + { opt_chk, "check-lib", ap_no }, { opt_lt, "loose-trailing", ap_no }, - { 0 , 0, ap_no } }; + { 0, 0, ap_no } }; - struct Arg_parser parser; - - invocation_name = argv[0]; - - if( LZ_version()[0] != LZ_version_string[0] ) - internal_error( "bad library version." ); - if( strcmp( PROGVERSION, LZ_version_string ) != 0 ) - internal_error( "bad library version_string." ); + if( argc > 0 ) invocation_name = argv[0]; if( !ap_init( &parser, argc, argv, options, 0 ) ) - { show_error( "Not enough memory.", 0, false ); return 1; } + { show_error( mem_msg, 0, false ); return 1; } if( ap_error( &parser ) ) /* bad option */ { show_error( ap_error( &parser ), 0, true ); return 1; } @@ -1015,7 +1066,8 @@ int main( const int argc, const char * const argv[] ) getnum( arg, LZ_min_match_len_limit(), LZ_max_match_len_limit() ); break; case 'n': break; - case 'o': default_output_filename = arg; break; + case 'o': if( strcmp( arg, "-" ) == 0 ) to_stdout = true; + else { default_output_filename = arg; } break; case 'q': verbosity = -1; break; case 's': encoder_options.dictionary_size = get_dict_size( arg ); break; @@ -1023,11 +1075,27 @@ int main( const int argc, const char * const argv[] ) case 't': set_mode( &program_mode, m_test ); break; case 'v': if( verbosity < 4 ) ++verbosity; break; case 'V': show_version(); return 0; + case opt_chk: return check_lib(); case opt_lt: loose_trailing = true; break; default : internal_error( "uncaught option." ); } } /* end process options */ + if( strcmp( PROGVERSION, LZ_version_string ) != 0 ) + internal_error( "wrong PROGVERSION." ); +#if !defined LZ_API_VERSION || LZ_API_VERSION < 1012 +#error "lzlib 1.12 or newer needed." +#else + if( LZ_api_version() < 1012 ) /* minilzip passes null to LZ_decompress_read */ + { show_error( "lzlib 1.12 or newer needed. Try --check-lib.", 0, false ); + return 1; } + if( LZ_api_version() != LZ_API_VERSION ) show_error( + "warning: wrong library API version. Try --check-lib.", 0, false ); + else +#endif + if( strcmp( LZ_version_string, LZ_version() ) != 0 ) show_error( + "warning: wrong library version_string. Try --check-lib.", 0, false ); + #if defined(__MSVCRT__) || defined(__OS2__) || defined(__DJGPP__) setmode( STDIN_FILENO, O_BINARY ); setmode( STDOUT_FILENO, O_BINARY ); @@ -1043,16 +1111,31 @@ int main( const int argc, const char * const argv[] ) if( strcmp( filenames[i], "-" ) != 0 ) filenames_given = true; } - if( program_mode == m_test ) - outfd = -1; + if( program_mode == m_compress ) + { + if( volume_size > 0 && !to_stdout && default_output_filename[0] && + num_filenames > 1 ) + { show_error( "Only can compress one file when using '-o' and '-S'.", + 0, true ); return 1; } + } + else volume_size = 0; + if( program_mode == m_test ) to_stdout = false; /* apply overrides */ + if( program_mode == m_test || to_stdout ) default_output_filename = ""; - if( !to_stdout && program_mode != m_test && - ( filenames_given || default_output_filename[0] ) ) + output_filename = resize_buffer( output_filename, 1 ); + output_filename[0] = 0; + if( to_stdout && program_mode != m_test ) /* check tty only once */ + { outfd = STDOUT_FILENO; if( !check_tty_out( program_mode ) ) return 1; } + else outfd = -1; + + const bool to_file = !to_stdout && program_mode != m_test && + default_output_filename[0]; + if( !to_stdout && program_mode != m_test && ( filenames_given || to_file ) ) set_signals( signal_handler ); Pp_init( &pp, filenames, num_filenames ); - output_filename = resize_buffer( output_filename, 1 ); + const bool one_to_one = !to_stdout && program_mode != m_test && !to_file; for( i = 0; i < num_filenames; ++i ) { const char * input_filename = ""; @@ -1060,68 +1143,48 @@ int main( const int argc, const char * const argv[] ) int tmp; struct stat in_stats; const struct stat * in_statsp; - output_filename[0] = 0; - if( !filenames[i][0] || strcmp( filenames[i], "-" ) == 0 ) + Pp_set_name( &pp, filenames[i] ); + if( strcmp( filenames[i], "-" ) == 0 ) { if( stdin_used ) continue; else stdin_used = true; infd = STDIN_FILENO; - if( program_mode != m_test ) - { - if( to_stdout || !default_output_filename[0] ) - outfd = STDOUT_FILENO; - else - { - if( program_mode == m_compress ) - set_c_outname( default_output_filename, false, volume_size > 0 ); - else - { - output_filename = resize_buffer( output_filename, - strlen( default_output_filename ) + 1 ); - strcpy( output_filename, default_output_filename ); - } - if( !open_outstream( force, true ) ) - { - if( retval < 1 ) retval = 1; - close( infd ); - continue; - } - } - } + if( !check_tty_in( pp.name, infd, program_mode, &retval ) ) continue; + if( one_to_one ) { outfd = STDOUT_FILENO; output_filename[0] = 0; } } else { const int eindex = extension_index( input_filename = filenames[i] ); infd = open_instream( input_filename, &in_stats, program_mode, - eindex, recompress, to_stdout ); - if( infd < 0 ) { if( retval < 1 ) retval = 1; continue; } - if( program_mode != m_test ) + eindex, one_to_one, recompress ); + if( infd < 0 ) { set_retval( &retval, 1 ); continue; } + if( !check_tty_in( pp.name, infd, program_mode, &retval ) ) continue; + if( one_to_one ) /* open outfd after verifying infd */ { - if( to_stdout ) outfd = STDOUT_FILENO; - else - { - if( program_mode == m_compress ) - set_c_outname( input_filename, true, volume_size > 0 ); - else set_d_outname( input_filename, eindex ); - if( !open_outstream( force, false ) ) - { - if( retval < 1 ) retval = 1; - close( infd ); - continue; - } - } + if( program_mode == m_compress ) + set_c_outname( input_filename, true, volume_size > 0 ); + else set_d_outname( input_filename, eindex ); + if( !open_outstream( force, true ) ) + { close( infd ); set_retval( &retval, 1 ); continue; } } } - Pp_set_name( &pp, input_filename ); - if( !check_tty( pp.name, infd, program_mode ) ) + if( one_to_one && !check_tty_out( program_mode ) ) + { set_retval( &retval, 1 ); return retval; } /* don't delete a tty */ + + if( to_file && outfd < 0 ) /* open outfd after verifying infd */ { - if( retval < 1 ) retval = 1; - if( program_mode == m_test ) { close( infd ); continue; } - cleanup_and_fail( retval ); + if( program_mode == m_compress ) set_c_outname( default_output_filename, + false, volume_size > 0 ); + else + { output_filename = resize_buffer( output_filename, + strlen( default_output_filename ) + 1 ); + strcpy( output_filename, default_output_filename ); } + if( !open_outstream( force, false ) || !check_tty_out( program_mode ) ) + return 1; /* check tty only once and don't try to delete a tty */ } - in_statsp = input_filename[0] ? &in_stats : 0; + in_statsp = ( input_filename[0] && one_to_one ) ? &in_stats : 0; if( program_mode == m_compress ) tmp = compress( member_size, volume_size, infd, &encoder_options, &pp, in_statsp ); @@ -1129,29 +1192,24 @@ int main( const int argc, const char * const argv[] ) tmp = decompress( infd, &pp, ignore_trailing, loose_trailing, program_mode == m_test ); if( close( infd ) != 0 ) - { - show_error( input_filename[0] ? "Error closing input file" : - "Error closing stdin", errno, false ); - if( tmp < 1 ) tmp = 1; - } - if( tmp > retval ) retval = tmp; + { show_file_error( pp.name, "Error closing input file", errno ); + set_retval( &tmp, 1 ); } + set_retval( &retval, tmp ); if( tmp ) { if( program_mode != m_test ) cleanup_and_fail( retval ); else ++failed_tests; } - if( delete_output_on_interrupt ) + if( delete_output_on_interrupt && one_to_one ) close_and_set_permissions( in_statsp ); - if( input_filename[0] ) - { - if( !keep_input_files && !to_stdout && program_mode != m_test && - ( program_mode != m_compress || volume_size == 0 ) ) - remove( input_filename ); - } + if( input_filename[0] && !keep_input_files && one_to_one && + ( program_mode != m_compress || volume_size == 0 ) ) + remove( input_filename ); } - if( outfd >= 0 && close( outfd ) != 0 ) + if( delete_output_on_interrupt ) close_and_set_permissions( 0 ); /* -o */ + else if( outfd >= 0 && close( outfd ) != 0 ) /* -c */ { show_error( "Error closing stdout", errno, false ); - if( retval < 1 ) retval = 1; + set_retval( &retval, 1 ); } if( failed_tests > 0 && verbosity >= 1 && num_filenames > 1 ) fprintf( stderr, "%s: warning: %d %s failed the test.\n", diff --git a/testsuite/check.sh b/testsuite/check.sh index 026ea2b..44e2428 100755 --- a/testsuite/check.sh +++ b/testsuite/check.sh @@ -1,9 +1,9 @@ #! /bin/sh # check script for Lzlib - Compression library for the lzip format -# Copyright (C) 2009-2019 Antonio Diaz Diaz. +# Copyright (C) 2009-2021 Antonio Diaz Diaz. # # This script is free software: you have unlimited permission -# to copy, distribute and modify it. +# to copy, distribute, and modify it. LC_ALL=C export LC_ALL @@ -11,6 +11,7 @@ objdir=`pwd` testdir=`cd "$1" ; pwd` LZIP="${objdir}"/minilzip BBEXAMPLE="${objdir}"/bbexample +FFEXAMPLE="${objdir}"/ffexample LZCHECK="${objdir}"/lzcheck framework_failure() { echo "failure in testing framework" ; exit 1 ; } @@ -32,10 +33,13 @@ cd "${objdir}"/tmp || framework_failure cat "${testdir}"/test.txt > in || framework_failure in_lz="${testdir}"/test.txt.lz -test2="${testdir}"/test2.txt +in_em="${testdir}"/test_em.txt.lz +fox_lf="${testdir}"/fox_lf +fox_lz="${testdir}"/fox.lz fail=0 test_failed() { fail=1 ; printf " $1" ; [ -z "$2" ] || printf "($2)" ; } +"${LZIP}" --check-lib # just print warning printf "testing lzlib-%s..." "$2" "${LZIP}" -fkqm4 in @@ -59,10 +63,22 @@ done [ $? = 2 ] || test_failed $LINENO "${LZIP}" -dq -o in < "${in_lz}" [ $? = 1 ] || test_failed $LINENO +"${LZIP}" -dq -o in "${in_lz}" +[ $? = 1 ] || test_failed $LINENO +"${LZIP}" -dq -o out nx_file.lz +[ $? = 1 ] || test_failed $LINENO +[ ! -e out ] || test_failed $LINENO +"${LZIP}" -q -o out.lz nx_file +[ $? = 1 ] || test_failed $LINENO +[ ! -e out.lz ] || test_failed $LINENO +"${LZIP}" -qf -S100k -o out in in +[ $? = 1 ] || test_failed $LINENO # these are for code coverage "${LZIP}" -cdt "${in_lz}" > out 2> /dev/null [ $? = 1 ] || test_failed $LINENO -"${LZIP}" -t -- nx_file 2> /dev/null +"${LZIP}" -t -- nx_file.lz 2> /dev/null +[ $? = 1 ] || test_failed $LINENO +"${LZIP}" -t "" < /dev/null 2> /dev/null [ $? = 1 ] || test_failed $LINENO "${LZIP}" --help > /dev/null || test_failed $LINENO "${LZIP}" -n1 -V > /dev/null || test_failed $LINENO @@ -86,15 +102,22 @@ printf "LZIP\001+.............................." | "${LZIP}" -t 2> /dev/null printf "\ntesting decompression..." -"${LZIP}" -t "${in_lz}" || test_failed $LINENO -"${LZIP}" -cd "${in_lz}" > copy || test_failed $LINENO -cmp in copy || test_failed $LINENO +for i in "${in_lz}" "${in_em}" "${testdir}"/test_sync.lz ; do + "${LZIP}" -t "$i" || test_failed $LINENO "$i" + "${LZIP}" -d "$i" -o copy || test_failed $LINENO "$i" + cmp in copy || test_failed $LINENO "$i" + "${LZIP}" -cd "$i" > copy || test_failed $LINENO "$i" + cmp in copy || test_failed $LINENO "$i" + "${LZIP}" -d "$i" -o - > copy || test_failed $LINENO "$i" + cmp in copy || test_failed $LINENO "$i" + "${LZIP}" -d < "$i" > copy || test_failed $LINENO "$i" + cmp in copy || test_failed $LINENO "$i" + rm -f copy || framework_failure +done -"${LZIP}" -t "${testdir}"/test_sync.lz || test_failed $LINENO -"${LZIP}" -cd "${testdir}"/test_sync.lz > copy || test_failed $LINENO -cmp in copy || test_failed $LINENO +lines=$("${LZIP}" -tvv "${in_em}" 2>&1 | wc -l) || test_failed $LINENO +[ "${lines}" -eq 8 ] || test_failed $LINENO "${lines}" -rm -f copy || framework_failure cat "${in_lz}" > copy.lz || framework_failure "${LZIP}" -dk copy.lz || test_failed $LINENO cmp in copy || test_failed $LINENO @@ -112,12 +135,20 @@ cat "${in_lz}" > copy.lz || framework_failure cmp in copy || test_failed $LINENO printf "to be overwritten" > copy || framework_failure +"${LZIP}" -d -o copy < "${in_lz}" 2> /dev/null +[ $? = 1 ] || test_failed $LINENO "${LZIP}" -df -o copy < "${in_lz}" || test_failed $LINENO cmp in copy || test_failed $LINENO - -rm -f copy || framework_failure -"${LZIP}" -s16 < in > anyothername || test_failed $LINENO -"${LZIP}" -dv --output copy - anyothername - < "${in_lz}" 2> /dev/null || +rm -f out copy || framework_failure +"${LZIP}" -d -o ./- "${in_lz}" || test_failed $LINENO +cmp in ./- || test_failed $LINENO +rm -f ./- || framework_failure +"${LZIP}" -d -o ./- < "${in_lz}" || test_failed $LINENO +cmp in ./- || test_failed $LINENO +rm -f ./- || framework_failure + +cat "${in_lz}" > anyothername || framework_failure +"${LZIP}" -dv - anyothername - < "${in_lz}" > copy 2> /dev/null || test_failed $LINENO cmp in copy || test_failed $LINENO cmp in anyothername.out || test_failed $LINENO @@ -152,19 +183,18 @@ done cmp in copy || test_failed $LINENO cat in in > in2 || framework_failure -cat "${in_lz}" "${in_lz}" > in2.lz || framework_failure -"${LZIP}" -t in2.lz || test_failed $LINENO -"${LZIP}" -cd in2.lz > copy2 || test_failed $LINENO +"${LZIP}" -t "${in_lz}" "${in_lz}" || test_failed $LINENO +"${LZIP}" -cd "${in_lz}" "${in_lz}" -o out > copy2 || test_failed $LINENO +[ ! -e out ] || test_failed $LINENO # override -o cmp in2 copy2 || test_failed $LINENO - -"${LZIP}" -s16 --output=copy2.lz < in2 || test_failed $LINENO -"${LZIP}" -t copy2.lz || test_failed $LINENO -"${LZIP}" -cd copy2.lz > copy2 || test_failed $LINENO +rm -f copy2 || framework_failure +"${LZIP}" -d "${in_lz}" "${in_lz}" -o copy2 || test_failed $LINENO cmp in2 copy2 || test_failed $LINENO +rm -f copy2 || framework_failure +cat "${in_lz}" "${in_lz}" > copy2.lz || framework_failure printf "\ngarbage" >> copy2.lz || framework_failure "${LZIP}" -tvvvv copy2.lz 2> /dev/null || test_failed $LINENO -rm -f copy2 || framework_failure "${LZIP}" -atq copy2.lz [ $? = 2 ] || test_failed $LINENO "${LZIP}" -atq < copy2.lz @@ -178,68 +208,91 @@ rm -f copy2 || framework_failure printf "to be overwritten" > copy2 || framework_failure "${LZIP}" -df copy2.lz || test_failed $LINENO cmp in2 copy2 || test_failed $LINENO -rm -f in2 copy2 || framework_failure +rm -f copy2 || framework_failure printf "\ntesting compression..." +"${LZIP}" -c -0 in in in -S100k -o out3.lz > copy2.lz || test_failed $LINENO +[ ! -e out3.lz ] || test_failed $LINENO # override -o and -S +"${LZIP}" -0f in in --output=copy2.lz || test_failed $LINENO +"${LZIP}" -d copy2.lz -o out2 || test_failed $LINENO +cmp in2 out2 || test_failed $LINENO +rm -f out2 copy2.lz || framework_failure + "${LZIP}" -cf "${in_lz}" > out 2> /dev/null # /dev/null is a tty on OS/2 [ $? = 1 ] || test_failed $LINENO -"${LZIP}" -cFvvm36 -s16 "${in_lz}" > out 2> /dev/null || test_failed $LINENO +"${LZIP}" -Fvvm36 -o - -s16 "${in_lz}" > out 2> /dev/null || test_failed $LINENO "${LZIP}" -cd out | "${LZIP}" -d > copy || test_failed $LINENO cmp in copy || test_failed $LINENO +"${LZIP}" -0 -o ./- in || test_failed $LINENO +"${LZIP}" -cd ./- | cmp in - || test_failed $LINENO +rm -f ./- || framework_failure +"${LZIP}" -0 -o ./- < in || test_failed $LINENO # don't add .lz +[ ! -e ./-.lz ] || test_failed $LINENO +"${LZIP}" -cd ./- | cmp in - || test_failed $LINENO +rm -f ./- || framework_failure + for i in s4Ki 0 1 2 3 4 5 6 7 8 9 ; do "${LZIP}" -k -$i -s16 in || test_failed $LINENO $i mv -f in.lz copy.lz || test_failed $LINENO $i printf "garbage" >> copy.lz || framework_failure "${LZIP}" -df copy.lz || test_failed $LINENO $i cmp in copy || test_failed $LINENO $i -done -for i in s4Ki 0 1 2 3 4 5 6 7 8 9 ; do - "${LZIP}" -c -$i -s16 in > out || test_failed $LINENO $i + "${LZIP}" -$i -s16 in -c > out || test_failed $LINENO $i + "${LZIP}" -$i -s16 in -o o_out || test_failed $LINENO $i # don't add .lz + [ ! -e o_out.lz ] || test_failed $LINENO + cmp out o_out || test_failed $LINENO $i + rm -f o_out || framework_failure printf "g" >> out || framework_failure "${LZIP}" -cd out > copy || test_failed $LINENO $i cmp in copy || test_failed $LINENO $i -done -for i in s4Ki 0 1 2 3 4 5 6 7 8 9 ; do "${LZIP}" -$i -s16 < in > out || test_failed $LINENO $i "${LZIP}" -d < out > copy || test_failed $LINENO $i cmp in copy || test_failed $LINENO $i -done -for i in s4Ki 0 1 2 3 4 5 6 7 8 9 ; do + rm -f out.lz || framework_failure + printf "to be overwritten" > out || framework_failure # don't add .lz "${LZIP}" -f -$i -s16 -o out < in || test_failed $LINENO $i - "${LZIP}" -df -o copy < out.lz || test_failed $LINENO $i + [ ! -e out.lz ] || test_failed $LINENO + "${LZIP}" -df -o copy < out || test_failed $LINENO $i cmp in copy || test_failed $LINENO $i done -rm -f out.lz || framework_failure +rm -f out out.lz || framework_failure cat in in in in in in in in > in8 || framework_failure "${LZIP}" -1s12 -S100k in8 || test_failed $LINENO "${LZIP}" -t in800001.lz in800002.lz || test_failed $LINENO "${LZIP}" -cd in800001.lz in800002.lz | cmp in8 - || test_failed $LINENO +[ ! -e in800003.lz ] || test_failed $LINENO rm -f in800001.lz in800002.lz || framework_failure -"${LZIP}" -1s12 -S100k -o out.lz < in8 || test_failed $LINENO +"${LZIP}" -1s12 -S100k -o out.lz in8 || test_failed $LINENO +# ignore -S +"${LZIP}" -d out.lz00001.lz out.lz00002.lz -S100k -o out || test_failed $LINENO +cmp in8 out || test_failed $LINENO "${LZIP}" -t out.lz00001.lz out.lz00002.lz || test_failed $LINENO -"${LZIP}" -cd out.lz00001.lz out.lz00002.lz | cmp in8 - || test_failed $LINENO -rm -f out.lz00001.lz out.lz00002.lz || framework_failure +[ ! -e out.lz00003.lz ] || test_failed $LINENO +rm -f out out.lz00001.lz out.lz00002.lz || framework_failure "${LZIP}" -1ks4Ki -b100000 in8 || test_failed $LINENO "${LZIP}" -t in8.lz || test_failed $LINENO -"${LZIP}" -cd in8.lz | cmp in8 - || test_failed $LINENO -rm -f in8 || framework_failure +"${LZIP}" -cd in8.lz -o out | cmp in8 - || test_failed $LINENO # override -o +[ ! -e out ] || test_failed $LINENO "${LZIP}" -0 -S100k -o out < in8.lz || test_failed $LINENO "${LZIP}" -t out00001.lz out00002.lz || test_failed $LINENO "${LZIP}" -cd out00001.lz out00002.lz | cmp in8.lz - || test_failed $LINENO +[ ! -e out00003.lz ] || test_failed $LINENO rm -f out00001.lz || framework_failure "${LZIP}" -1 -S100k -o out < in8.lz || test_failed $LINENO "${LZIP}" -t out00001.lz out00002.lz || test_failed $LINENO "${LZIP}" -cd out00001.lz out00002.lz | cmp in8.lz - || test_failed $LINENO +[ ! -e out00003.lz ] || test_failed $LINENO rm -f out00001.lz out00002.lz || framework_failure "${LZIP}" -0 -F -S100k in8.lz || test_failed $LINENO "${LZIP}" -t in8.lz00001.lz in8.lz00002.lz || test_failed $LINENO "${LZIP}" -cd in8.lz00001.lz in8.lz00002.lz | cmp in8.lz - || test_failed $LINENO +[ ! -e in8.lz00003.lz ] || test_failed $LINENO rm -f in8.lz00001.lz in8.lz00002.lz || framework_failure "${LZIP}" -0kF -b100k in8.lz || test_failed $LINENO "${LZIP}" -t in8.lz.lz || test_failed $LINENO @@ -247,12 +300,33 @@ rm -f in8.lz00001.lz in8.lz00002.lz || framework_failure rm -f in8.lz in8.lz.lz || framework_failure "${BBEXAMPLE}" in || test_failed $LINENO -"${BBEXAMPLE}" out || test_failed $LINENO -"${BBEXAMPLE}" ${test2} || test_failed $LINENO +"${BBEXAMPLE}" "${in_lz}" || test_failed $LINENO +"${BBEXAMPLE}" "${fox_lf}" || test_failed $LINENO + +"${FFEXAMPLE}" -h > /dev/null || test_failed $LINENO +"${FFEXAMPLE}" > /dev/null && test_failed $LINENO +rm -f out || framework_failure +"${FFEXAMPLE}" -b in out || test_failed $LINENO +cmp in out || test_failed $LINENO +"${FFEXAMPLE}" -b in | cmp in - || test_failed $LINENO +"${FFEXAMPLE}" -b in8 | cmp in8 - || test_failed $LINENO +"${FFEXAMPLE}" -b "${fox_lf}" | cmp "${fox_lf}" - || test_failed $LINENO +"${FFEXAMPLE}" -d "${in_lz}" - | cmp in - || test_failed $LINENO +"${FFEXAMPLE}" -d "${in_em}" - | cmp in - || test_failed $LINENO +"${FFEXAMPLE}" -c in | "${FFEXAMPLE}" -d | cmp in - || test_failed $LINENO +"${FFEXAMPLE}" -m in | "${FFEXAMPLE}" -d | cmp in - || test_failed $LINENO +"${FFEXAMPLE}" -l in | "${FFEXAMPLE}" -d | cmp in - || test_failed $LINENO +cat "${fox_lf}" "${in_lz}" | "${FFEXAMPLE}" -r | cmp in - || test_failed $LINENO +cat in8 "${in_lz}" | "${FFEXAMPLE}" -r | cmp in - || test_failed $LINENO +cat "${in_lz}" "${fox_lf}" "${in_lz}" | "${FFEXAMPLE}" -r - | cmp in2 - || + test_failed $LINENO +cat "${in_lz}" in8 "${in_lz}" | "${FFEXAMPLE}" -r - - | cmp in2 - || + test_failed $LINENO "${LZCHECK}" in || test_failed $LINENO -"${LZCHECK}" out || test_failed $LINENO -"${LZCHECK}" ${test2} || test_failed $LINENO +"${LZCHECK}" "${in_lz}" || test_failed $LINENO +"${LZCHECK}" "${fox_lf}" || test_failed $LINENO +rm -f in8 || framework_failure printf "\ntesting bad input..." @@ -301,6 +375,21 @@ else fi rm -f int.lz || framework_failure +for i in fox_v2.lz fox_s11.lz fox_de20.lz \ + fox_bcrc.lz fox_crc0.lz fox_das46.lz fox_mes81.lz ; do + "${LZIP}" -tq "${testdir}"/$i + [ $? = 2 ] || test_failed $LINENO $i +done + +"${LZIP}" -cd "${fox_lz}" > fox || test_failed $LINENO +for i in fox_bcrc.lz fox_crc0.lz fox_das46.lz fox_mes81.lz ; do + "${LZIP}" -cdq "${testdir}"/$i > out + [ $? = 2 ] || test_failed $LINENO $i + cmp fox out || test_failed $LINENO $i +done +rm -f fox out || framework_failure + +cat "${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=14752 count=1 2> /dev/null && [ -e trunc.lz ] && cmp in2.lz trunc.lz > /dev/null 2>&1 ; then @@ -323,13 +412,23 @@ rm -f in2.lz in3.lz trunc.lz out || framework_failure cat "${in_lz}" > ingin.lz || framework_failure printf "g" >> ingin.lz || framework_failure cat "${in_lz}" >> ingin.lz || framework_failure +"${LZIP}" -atq ingin.lz +[ $? = 2 ] || test_failed $LINENO +"${LZIP}" -atq < ingin.lz +[ $? = 2 ] || test_failed $LINENO +"${LZIP}" -acdq ingin.lz > out +[ $? = 2 ] || test_failed $LINENO +"${LZIP}" -adq < ingin.lz > out +[ $? = 2 ] || test_failed $LINENO "${LZIP}" -t ingin.lz || test_failed $LINENO +"${LZIP}" -t < ingin.lz || test_failed $LINENO "${LZIP}" -cd ingin.lz > copy || test_failed $LINENO cmp in copy || test_failed $LINENO -"${LZIP}" -t < ingin.lz || test_failed $LINENO "${LZIP}" -d < ingin.lz > copy || test_failed $LINENO cmp in copy || test_failed $LINENO -rm -f copy ingin.lz || framework_failure +"${FFEXAMPLE}" -d ingin.lz | cmp in - || test_failed $LINENO +"${FFEXAMPLE}" -r ingin.lz | cmp in2 - || test_failed $LINENO +rm -f copy ingin.lz in2 out || framework_failure echo if [ ${fail} = 0 ] ; then diff --git a/testsuite/fox.lz b/testsuite/fox.lz Binary files differnew file mode 100644 index 0000000..509da82 --- /dev/null +++ b/testsuite/fox.lz diff --git a/testsuite/fox_bcrc.lz b/testsuite/fox_bcrc.lz Binary files differnew file mode 100644 index 0000000..8f6a7c4 --- /dev/null +++ b/testsuite/fox_bcrc.lz diff --git a/testsuite/fox_crc0.lz b/testsuite/fox_crc0.lz Binary files differnew file mode 100644 index 0000000..1abe926 --- /dev/null +++ b/testsuite/fox_crc0.lz diff --git a/testsuite/fox_das46.lz b/testsuite/fox_das46.lz Binary files differnew file mode 100644 index 0000000..43ed9f9 --- /dev/null +++ b/testsuite/fox_das46.lz diff --git a/testsuite/fox_de20.lz b/testsuite/fox_de20.lz Binary files differnew file mode 100644 index 0000000..10949d8 --- /dev/null +++ b/testsuite/fox_de20.lz diff --git a/testsuite/test2.txt b/testsuite/fox_lf index a0b11b5..a0b11b5 100644 --- a/testsuite/test2.txt +++ b/testsuite/fox_lf diff --git a/testsuite/fox_mes81.lz b/testsuite/fox_mes81.lz Binary files differnew file mode 100644 index 0000000..d50ef2e --- /dev/null +++ b/testsuite/fox_mes81.lz diff --git a/testsuite/fox_s11.lz b/testsuite/fox_s11.lz Binary files differnew file mode 100644 index 0000000..dca909c --- /dev/null +++ b/testsuite/fox_s11.lz diff --git a/testsuite/fox_v2.lz b/testsuite/fox_v2.lz Binary files differnew file mode 100644 index 0000000..8620981 --- /dev/null +++ b/testsuite/fox_v2.lz diff --git a/testsuite/test_em.txt.lz b/testsuite/test_em.txt.lz Binary files differnew file mode 100644 index 0000000..7e96250 --- /dev/null +++ b/testsuite/test_em.txt.lz |