Age | Commit message (Collapse) | Author | |
---|---|---|---|
2015-10-01 | LARGEWINDOW->LARGESIZET, now passing regtests and selftests | Josh MacDonald | |
2015-09-29 | Merge from HEAD | Josh MacDonald | |
2015-05-04 | More followup on issue 188, correction to frontier_blkno calculation (was ↵ | Joshua MacDonald | |
wrong when last block is full) | |||
2015-05-02 | Off-by-one error leading to XD3_TOOFARBACK when a potential match is found ↵ | Joshua MacDonald | |
in the 32nd block prior to EOF | |||
2015-03-30 | Update Copyright dates | Joshua MacDonald | |
2015-03-10 | Functions out of macros for xd3_decode_offset, to address warnings created ↵ | Joshua MacDonald | |
by new definition of xoff_t | |||
2015-03-06 | Builds clean Apple LLVM version 6.0 (clang-600.0.56) (based on LLVM 3.5svn) | Joshua MacDonald | |
2014-11-01 | Fix 'xdelta3: -B: maximum value: 9223372036854775808' | Josh MacDonald | |
2014-10-28 | Fix an overflow in setting xd3_source.maskby | Josh MacDonald | |
2014-10-28 | Added 64-bit checksum for largewindow support; not ready | Josh MacDonald | |
2014-10-25 | Checksum test cleanups & dev | Josh MacDonald | |
2014-10-23 | checksum_test is working (apparently) | Josh MacDonald | |
2014-10-19 | Support for 64bit usize_t w/ XD3_USE_LARGEWINDOW64 | Josh MacDonald | |
2014-10-18 | Some -Wconversion cleanups | Josh MacDonald | |
2014-10-18 | Remove generic encoding table support | josh.macdonald | |
2014-04-03 | Revert recent (unreleased) change in hash_value formula; make several tests ↵ | josh.macdonald | |
use the average over 16 runs | |||
2014-03-09 | Fix issue in xd3_encode_memory causing error when IOPT buffer capacity ↵ | josh.macdonald | |
reached and a copy reaches the end-of-source, causing unexpected request for second source block. New tests to cover this case. Removed the IOPT limit in xd3_encode_memory (improves compression). Doubles sizing of the large-checksum hash table (fewer collisions, better compression). | |||
2013-05-14 | 3.0.7 RC0 | josh.macdonald | |
2013-05-13 | Repair and improve the regtest | josh.macdonald | |
2013-01-21 | Revert to r386; prepare 3.0.6 release | josh.macdonald | |
2013-01-21 | Park these changes; tremendous slowdown, will revert to 386 | josh.macdonald | |
2013-01-21 | fix main_getblk_lru for decode | josh.macdonald | |
2013-01-21 | Further conversion cleanups. | josh.macdonald | |
2013-01-21 | Audit -Wconversion results | josh.macdonald | |
2013-01-20 | Partial Win64 build fixes. | josh.macdonald | |
2013-01-20 | Fix some TODOs (option_srcwinsz up to 4GB? | josh.macdonald | |
2013-01-20 | Fix regression in SVN 376 (3.0.5) | josh.macdonald | |
2012-11-17 | Make -B a size_t, build and test w/ -m64 | josh.macdonald | |
2012-11-13 | Fix an off-by-one error in preventing > max_srcwinsize backward seeks | josh.macdonald | |
2012-07-19 | Fix for build w/o liblzma present | josh.macdonald@gmail.com | |
2012-07-19 | Add support for lzma secondary compression (-S lzma) | josh.macdonald | |
2012-07-15 | Compiles on gcc-4.6 | josh.macdonald@gmail.com | |
2012-06-18 | Fixes for CC=clang | josh.macdonald | |
2012-06-17 | Issue 120: compile with -pedantic | josh.macdonald | |
2012-06-16 | Updates for iOS build, Automake setup | josh.macdonald | |
2010-11-07 | Whitespace cleanups, comments | josh.macdonald | |
2010-10-25 | Conditional compile fixes. | josh.macdonald | |
2010-08-01 | Fixes for win7, especially named pipe handling | josh.macdonald@gmail.com | |
2010-02-16 | Undo recent "off-by-one" fix, it was incorrect (the bug was actually | josh.macdonald | |
the issue about updating the value of xd3_source_eof() in the same change). | |||
2010-02-16 | Whitespace changes. | josh.macdonald | |
2010-02-16 | Fix an off-by-one error in srcwin_move_point, and prevent updating | josh.macdonald | |
xd3_source_eof () when eof is known in xd3_getblk. Now performance is the same for FIFO and no. | |||
2010-02-16 | Fixes a memory corruption caused by recent changes (as early as 3.0w, | josh.macdonald | |
it seems). | |||
2010-02-15 | Misc whitespace changes. | josh.macdonald | |
2010-02-11 | Running xdelta3-regtest.py. Observed a bug in recently-added | josh.macdonald | |
match-setup code, using source_eof instead of frontier position. | |||
2010-02-07 | Implement srcwin_maxsz in xd3_source_match_setup(), which prevents | josh.macdonald | |
the encoder from seeking backwards further than this parameter. This was supposed to be implemented long ago, but was especially problematic in the recent release, which implements source-from-FIFO. | |||
2010-01-06 | Fixes for gcc-4.2 on OS X 10.6, some comment cleanup. | josh.macdonald | |
2009-11-08 | Rewrote main_set_source() to delay deciding lru_size (effectively | josh.macdonald | |
option_srcwinsz) until after decompression state is known. Working but definitely needs more testing. | |||
2009-11-05 | Cleanups in xd3_decode_output_halfinst() and xd3_decode_emit(). | josh.macdonald | |
2009-11-03 | Implement (and manually test) main_getblk_func() using FIFO instead of | josh.macdonald | |
LRU for non-seekable sources (always, was not previously for the decoder, which caused problems). | |||
2009-11-03 | Fixes for issue 94. | josh.macdonald | |