lzip version 1.20 available (port) (Announce)
On 06 November 2018 I finished my 32-bit DOS port of Antonio Diaz Diaz' lzip version 1.20 using DJGPP version 2.05.
Binaries, user manuals in ASCII format, and fully configured sources plus some simple instructions on how to rebuild are available at http://www.bttr-software.de/ports/#lzip.
lzip changes since my last port (version 1.9):
2018-02-11 Antonio Diaz Diaz <antonio@gnu.org>
* Version 1.20 released.
* main.cc: Added new option '--loose-trailing'.
* Improved corrupt header detection to HD=3.
* main.cc: Show corrupt or truncated header in multimember file.
* main.cc (main): Option '-S, --volume-size' now keeps input files.
* encoder_base.*: Adjust dictionary size for each member.
* Replaced 'bits/byte' with inverse compression ratio in output.
* Show progress of decompression at verbosity level 2 (-vv).
* Show progress of (de)compression only if stderr is a terminal.
* main.cc: Show final diagnostic when testing multiple files.
* main.cc: Do not add a second .lz extension to the arg of -o.
* decoder.cc (verify_trailer): Show stored sizes also in hex.
Show dictionary size at verbosity level 4 (-vvvv).
* lzip.texi: Added chapter 'Output'.
2017-04-13 Antonio Diaz Diaz <antonio@gnu.org>
* Version 1.19 released.
* The option '-l, --list' has been ported from lziprecover.
* Don't allow mixing different operations (-d, -l or -t).
* Compression time of option '-0' has been slightly reduced.
* Decompression time has been reduced by 2%.
* main.cc: Continue testing if any input file is a terminal.
* main.cc: Show trailing data in both hexadecimal and ASCII.
* encoder.cc (Matchfinder_base): Verify size passed to new.
* file_index.cc: Improve detection of bad dict and trailing data.
* lzip.h: Unified messages for bad magic, trailing data, etc.
2016-05-14 Antonio Diaz Diaz <antonio@gnu.org>
* Version 1.18 released.
* main.cc: Added new option '-a, --trailing-error'.
* Decompression time has been reduced by 2%.
* decoder.cc (verify_trailer): Removed test of final code.
* main.cc (main): Delete '--output' file if infd is a terminal.
* main.cc (main): Don't use stdin more than once.
* Removed decompression support for version 0 files.
* lzip.texi: Added chapter 'Trailing data'.
* configure: Avoid warning on some shells when testing for g++.
* Makefile.in: Detect the existence of install-info.
* check.sh: A POSIX shell is required to run the tests.
* check.sh: Don't check error messages.
2015-07-12 Antonio Diaz Diaz <antonio@gnu.org>
* Version 1.17 released.
* Reorganization of the compression code.
* lzip.texi: Added chapter 'Quality assurance'.
* Makefile.in: Added new targets 'install*-compress'.
2014-08-26 Antonio Diaz Diaz <antonio@gnu.org>
* Version 1.16 released.
* Compression ratio of option '-9' has been slightly increased.
* Compression time has been reduced by 4%.
* Compression time of option '-0' has been reduced by 2%.
* main.cc (close_and_set_permissions): Behave like 'cp -p'.
* Minor improvements.
* lzip.texinfo: Renamed to lzip.texi.
* License changed to GPL version 2 or later.
2013-09-20 Antonio Diaz Diaz <antonio@gnu.org>
* Version 1.15 released.
* Show progress of compression at verbosity level 2 (-vv).
* main.cc (show_header): Don't show header version.
* Ignore option '-n, --threads' for compatibility with plzip.
* configure: Options now accept a separate argument.
* lzip.texinfo: Added chapter 'Stream format' and appendix
'Reference source code'.
2013-02-17 Antonio Diaz Diaz <ant_diaz@teleline.es>
* Version 1.14 released.
* Multi-step trials have been implemented.
* Compression ratio has been slightly increased.
* Compression time has been reduced by 5%.
* Decompression time has been reduced by 12%.
* Makefile.in: Added new target 'install-bin'.
* main.cc: Use 'setmode' instead of '_setmode' on Windows and OS/2.
* main.cc: Define 'strtoull' to 'std::strtoul' on Windows.
2012-02-24 Antonio Diaz Diaz <ant_diaz@teleline.es>
* Version 1.13 released.
* Lziprecover has been moved to its own package.
* main.cc (close_and_set_permissions): Inability to change output
file attributes has been downgraded from error to warning.
* Compression time of option '-0' has been reduced by 2%.
* Reorganization of the compression code.
* Small change in '--help' output and man page.
* Changed quote characters in messages as advised by GNU Standards.
* configure: 'datadir' renamed to 'datarootdir'.
* 'unzcrash.cc' has been moved to package 'lziprecover'.
2011-04-30 Antonio Diaz Diaz <ant_diaz@teleline.es>
* Version 1.12 released.
* main.cc: Added new option '-F, --recompress'.
* encoder.h (update_prices): Update high length symbol prices
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%.
* Compression time of option '-0' has been reduced by 2%.
* main.cc (decompress): Print only one status line for each
multimember file when only one '-v' is specified.
* main.cc (decompress): Print up to 6 bytes of trailing data
when '-vvvv' is specified.
* main.cc (open_instream): Don't show the message
" and '--stdout' was not specified" for directories, etc.
* lziprecover.cc: If '-v' is not specified show errors only.
* unzcrash.cc: Use Arg_parser.
* unzcrash.cc: Added new option '-b, --bits'.
* unzcrash.cc: Added new option '-p, --position'.
* unzcrash.cc: Added new option '-s, --size'.
2010-09-16 Antonio Diaz Diaz <ant_diaz@teleline.es>
* Version 1.11 released.
* Added new option '-0' which produces a compression speed and
ratio comparable to those of 'gzip -9'.
* fast_encoder.h fast_encoder.cc: New files.
* main.cc: Match length limit set by options -1 to -8 has been
reduced to extend range of use towards gzip. Lower numbers now
compress less but faster. (-1 now takes 43% less time for only
20% larger compressed size).
* Compression ratio of option '-9' has been slightly increased.
* lziprecover.cc: Added new option '-m, --merge' which tries to
produce a correct file merging the good parts of two or more
damaged copies.
* lziprecover.cc: Added new option '-R, --repair' for repairing
a 1-byte error in single-member files.
* decoder.cc (decode_member): Detect file errors earlier to
improve efficiency of lziprecover's new repair capability.
This change also prevents (harmless) access to uninitialized
memory when decompressing a corrupt file.
* lziprecover.cc: Added new option '-f, --force'.
* lziprecover.cc: Added new option '-o, --output'.
* lziprecover.cc: Added new option '-s, --split' to select the
until now only operation of splitting multimember files.
* lziprecover.cc: If no operation is specified, warn the user
and do nothing.
* main.cc: Fixed warning about fchown's return value being ignored.
* decoder.cc: '-tvvvv' now also shows compression ratio.
* main.cc: Set stdin/stdout in binary mode on MSVC and OS2.
* lzip.texinfo: Added new examples.
* testsuite: 'test1' renamed to 'test.txt'. Added new tests.
* Matchfinder types HC4 (4 bytes hash-chain) and HT4 (4 bytes
hash-table) have been tested and found no better than the
current BT4.
2010-04-05 Antonio Diaz Diaz <ant_diaz@teleline.es>
* Version 1.10 released.
* decoder.h: Input_buffer integrated in Range_decoder.
* main.cc: File specified with option '-o' is now created with
mode 0666 if umask allows it, deleted if interrupted by user.
* main.cc: New constant 'o_binary'.
* main.cc: Dictionary size for options -2, -3, -4 and -8 has
been changed to improve linearity of compressed sizes.
* lzip.h: Fixed warnings produced by over-optimization (-O3).
* Makefile.in: Added quotes to directory names.
---
Forum admin
Complete thread:
- lzip version 1.20 available (port) - rr, 06.11.2018, 22:37 (Announce)