summaryrefslogtreecommitdiff
path: root/xdelta3/xdelta3-main.h
AgeCommit message (Collapse)Author
2007-03-24Rename #define P to #define DPjosh.macdonald
2007-02-18pre 3.0p, all tests passjosh.macdonald
2007-02-18remove PROMOTE, SSMATCH, TRYLAZY, raise default IOPT_SIZEjosh.macdonald
2007-02-18raise default sprevsz, add command-line flag -P to setjosh.macdonald
2007-02-17generic encode tables compiled in setup.py; comments on HARDMAXWINSZ; more ↵josh.macdonald
winsize tests
2007-02-13Adds xd3_flags for setting compression level in xd3_encode_memory().josh.macdonald
2007-02-12Fix and recomment xd3_srcwin_move_point() after changes in SVN 127 to use ↵josh.macdonald
src->blksize as the advance size.
2007-02-11Fixes bugs in XDELTA environment variable.josh.macdonald
Removes stream.srcwin_size variable, uses stream.src->blksize instead, to address performance regression from xdelta1. Changes the order of checksum computation to be like xdelta1, insert checksums in reverse to give preference to earlier matches. Removes svn:executable from files.
2007-02-08Bugfix XDELTA environment var.josh.macdonald
2007-02-08XDELTA environment variable lets you do stuff like this:josh.macdonald
XDELTA="-s source-dir.tar" tar --use-compress-program=xdelta3 -cf - target-dir
2007-02-08Make "option ignored" less verbose.josh.macdonald
2007-02-08Python swig module working with xd3_encode_memory(), xd3_decode_memory(),josh.macdonald
and xd3_main_cmdline(). module added to xdelta3-regtest.py run-length experiments. XD3_INVALID_INPUT code added to distinguish decoder errors from invalid stream-states. xdelta3-test.py unittest added for python module. xd3_encode_memory() now limits config.winsize to <= 1MB: xd3_process_memory() now properly splits the input buffer into windows.
2007-02-07Ready to release 3.0o. Still working on swig.josh.macdonald
2007-02-07Update FASTEST, DEFAULT compression level. Comments.josh.macdonald
2007-02-07Fixes print commands on Windows, the needed GetStartupInfo() and ↵josh.macdonald
stdin/stdout handles set after the change in 112 (missing its comment) which converts the print commands to use snprintf() and the default I/O routines. Addresses issue 14.
2007-02-06(no commit message)josh.macdonald
2007-02-04Cleanup www (it's on the code.google.com/p/xdelta/wiki/list now). josh.macdonald
Rekey.
2007-02-04Adds a new compression level (-6), makes -1 faster and -9 better.josh.macdonald
2007-02-04Decoder memory error, related to secondary compression, fixed. Valgrind josh.macdonald
made this easy to spot.
2007-02-01Fixes build on Cygwin, some work on -regtest.pyjosh.macdonald
2007-01-29Updates on G:josh.macdonald
remove missing file from vcproj compiles under cygwin
2007-01-29Change _completely() to _completely_stream(), to introduce a new, josh.macdonald
simpler in-memory API.
2007-01-22Release 3.0mdotdotisdead
2007-01-21Update configs, comments, etc.dotdotisdead
2007-01-20Changes to support -I IOPT_SIZE on the commandline. New support for ↵dotdotisdead
unlimited instruction buffer size set by -I 0. The default remains 4096.
2007-01-16Increase default IOPT buffer size to 4096 entries: there is still no dotdotisdead
command-line flag for this setting! Bump version to 3.0l.
2007-01-14Cleanup the calculation of -B and -W, get the minimum source buffer size dotdotisdead
(-B) to work, enforce a maximum input window size (-W). Add some debugging output. Correct a bug in setting the large hashtable size. I've noticed some odd behavior with the -W setting, more tests to do.
2007-01-13Edit xdelta3-cfgs, not exactly scientific tuning but much better for dotdotisdead
now.
2007-01-10Work on tuning tests.dotdotisdead
2007-01-08Experimenting with -C soft-config, trying to tune the -1..-9 values.dotdotisdead
Right now -9 is definitely slower but not usually better.
2007-01-08Set source buffer size to min(source->size) and if the command-line isdotdotisdead
not set, scale to source->size up to 64MB.
2007-01-07Removes the -M memsize flag, the #define, and stream_config field of the dotdotisdead
same name. Now computes small/large hash tables based on 1 slot per input-size/hash-width.
2007-01-07Fixes spurious warning from printhdr.dotdotisdead
2007-01-07Fixes the python module, there were some bugs in main.dotdotisdead
2007-01-07Fixes source_decompression_test. I thought I fixed this once before:dotdotisdead
the main external-decompression waitpid() was called before the output file is closed. The fix was lost. Makes all the external-command tests conditional on !WIN32.
2006-12-173.0idotdotisdead
2006-12-16Make badcopy use xoff_tjosh.macdonald
2006-12-16Adds /DXD3_WIN32 using native open/read/write/seek, etc, also fixing ↵josh.macdonald
get_errno and xd3_strerror. Still not 64bit tested.
2006-12-16(no commit message)josh.macdonald
2006-12-10Release 3.0hdotdotisdead
2006-12-10Adds some comments about LZMA external compression.dotdotisdead
2006-12-10Builds on WIN32.dotdotisdead
2006-12-10Preliminary WIN32 support.dotdotisdead
2006-09-28Update command-line summary.dotdotisdead
2006-09-28Add XDELTA1PATH external compression option.dotdotisdead
2006-09-25Oooops.dotdotisdead
2006-09-25Cleanups, working on the python module, which has assertion failures.dotdotisdead
2006-09-24Cleanups for release 3.0g.dotdotisdead
2006-09-24Cleanups involving xd3_smatcher and soft-config variables. Adds -M memsizedotdotisdead
to allocate larger/smaller source-checksum hash table.
2006-09-24Fix a bug in xd3_source_extend_match, causing stream->input_position todotdotisdead
move backward when a large-cksum collision causes a short match in lazy mode, where there was a small match_back. Reproduce w/ testcases/7.
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-07 15:56:21 +0000