summaryrefslogtreecommitdiff
path: root/xdelta3/xdelta3-main.h
AgeCommit message (Collapse)Author
2010-11-07Clean up and simplify main_set_source()josh.macdonald
2010-10-25Move block-cache code from -main.h to -blkcache.h. Previous merge wasjosh.macdonald
incorrect - contents were branched post-code-removal. To identify diffs in blkcache.h see previous version of -main.h
2010-08-24Fixes a bad optimization which caused raising -B to worsen compressionjosh.macdonald
in source-from-fifo mode (e.g., externally compressed inputs). Disables several minor optimizations to simplify the code.
2010-08-03release 3.0zjosh.macdonald@gmail.com
2010-08-02change external-compression warning textjosh.macdonald@gmail.com
2010-08-01address issue 109 adds a strong warning about external decompressionjosh.macdonald@gmail.com
2010-08-01patch for issue 111, compiles clean under cygwin gcc 4.3.4 toojosh.macdonald@gmail.com
2010-08-01Fixes for win7, especially named pipe handlingjosh.macdonald@gmail.com
2010-07-01Add -F command and properly handle trailing garbage. Manually tested.josh.macdonald
2010-02-23Add verbose messaging about externally compressed inputs/output.josh.macdonald
2010-02-21Fix the handling of external source decompression, which changed in 3.0wjosh.macdonald
and was inadequately tested. Mark version 3.0y.
2010-02-16Prepare release 3.0xjosh.macdonald
2010-02-16Whitespace changes.josh.macdonald
2010-02-16Fixes a memory corruption caused by recent changes (as early as 3.0w, josh.macdonald
it seems).
2010-02-07Implement srcwin_maxsz in xd3_source_match_setup(), which preventsjosh.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.
2010-01-06Fixes for gcc-4.2 on OS X 10.6, some comment cleanup.josh.macdonald
2009-11-08Rewrote main_set_source() to delay deciding lru_size (effectivelyjosh.macdonald
option_srcwinsz) until after decompression state is known. Working but definitely needs more testing.
2009-11-03Implement (and manually test) main_getblk_func() using FIFO instead ofjosh.macdonald
LRU for non-seekable sources (always, was not previously for the decoder, which caused problems).
2009-10-28Optimize decode for source copies.josh.macdonald
Add magic number for Xz.
2009-10-26Release 3.0wjosh.macdonald
2009-10-26Provide more information about the error.josh.macdonald
2009-10-26Fix the error message to suggest "try raising -B" when XD3_TOOFARBACK ↵josh.macdonald
happens in decode.
2009-10-26Fix xdelta3-main.h's lru[] implementation, new manual testing success.josh.macdonald
2009-10-26Debugging: add a big TODO!josh.macdonald
2009-10-26All tests pass.josh.macdonald
2009-10-26Fix the source_decompression test, which tests that appheader overrides ↵josh.macdonald
compression magic number.
2009-10-26Source from non-seekable input seems to work. Limited testing!josh.macdonald
2009-10-26This fixes main_getblk_func to supposely work with non-seekable inputs, ↵josh.macdonald
we'll see!
2009-10-25Fixes bugs in xd3_srcwin_move_point.josh.macdonald
2009-10-25Eliminates xd3_source->size, which is an _API Change_. Adds a new ↵josh.macdonald
requirement, that source->blocksize be a power-of-2. DOES NOT PASS ALL TESTS.
2009-10-24Move and re-invigorate the regtest, fix 1 harmless compiler warningjosh.macdonald
2009-02-12Fixes issue 79, adds more testing.josh.macdonald
2008-09-17Compile with josh.macdonald
-Wconversion -Wsign-compare -Wextra -Wno-unused-parameter in C and C++.
2008-09-13Fix examples build, all-targets build, run tests, set version=3.0u.josh.macdonald
2008-09-13Implement adler32 and appheader for merge command.josh.macdonald
2008-09-07Fixes two merge bugs:josh.macdonald
1. The whole_state struct now keeps an array of window sizes so that during reconstruction it can use the same window size as the target delta. The code was previously using dec_tgtlen, which was the window size of the last target delta window. 2. xd3_merge_copy_source(), which applys a source-copy instruction during merge was not properly translating target-copy instructions in the target. The solution here is SLOW and INEFFICIENT, but it at least allows the tests to pass. A big TODO here is to improve the algorithm: it has a potentially O(N) recursion for each target-copy that it sees, and the naive approach also can produce duplicate adds.
2008-09-06Improve "printdelta" to indicate source vs. target copies, widen size fieldsjosh.macdonald
to 6 chars.
2008-09-05Solves the issue reported in Issue 36. More merge tests now pass.josh.macdonald
2008-08-16Issue 77 -- VCDIFF tools #define issue.josh.macdonald
2008-08-16Fixes a error in the test. Merge test passes...???josh.macdonald
2008-08-09Fixes for merge of zero-byte files, allow merge w/o source.josh.macdonald
2008-08-08Fix error message in main.hjosh.macdonald
Port testing/random.h portable random number generator (corrected) back to xdelta3-test.h. The code copied from http://www.qbrundage.com was out of line with the original Mersenne Twister code and didn't work.
2008-07-07Make these files compile with C++.josh.macdonald
2008-07-05Fix stdin/stdout problem on Windows, issue 34.josh.macdonald
2008-06-27Adds new testing directory for a C++ test, makes xdelta3-test.h compile with ↵josh.macdonald
C++. Removes xdelta3-test2.h, moves contents into xdelta3-regtest.cc
2008-04-20It's plausible that this merge code actually works; untested.josh.macdonald
2008-04-19Cleanup of previous. New xd3_merge_input_output used in bothjosh.macdonald
main_merge_arguments and main_merge_output.
2008-04-19Straighten up the logic for calling xd3_merge_inputs in both josh.macdonald
main_merge_arguments and main_merge_output.
2008-04-15Fixes for issue 70. The test inputs tickled a case where thejosh.macdonald
non-blocking API would not make progress, searching for match on the page boundary and repeatedly asking the application to get another source block. This is only a non-blocking issue, glad it's fixed. I'll add encode_decode_test to my test suite before another release.
2008-04-09Add mechanism for processing merge arguments in order, memoryjosh.macdonald
management for xd3_whole_target structures being passed between arguments-processing and main merge_output function.