Commit Graph

2242 Commits

Author SHA1 Message Date
Lasse Collin 107259e306 Fix alignment handling bugs in Subblock encoder.
This leaves one known alignment bug unfixed: If repeat count
doesn't fit into 28-bit integer, the encoder has to split
this to multiple Subblocks with Subblock Type `Repeating Data'.
The extra Subblocks may have wrong alignment. Correct alignment
is restored after the split Repeating Data has been completely
written out.

Since the encoder doesn't even try to fix the alignment unless
the size of Data is at least 4 bytes, to trigger this bug you
need at least 4 GiB of repeating data with sequence length of
4 or more bytes. Since the worst thing done by this bug is
misaligned data (no data corruption), this bug simply isn't
worth fixing, because a proper fix isn't simple.
2008-01-20 20:12:58 +02:00
Lasse Collin e141fe1895 Implemented LZMA_SYNC_FLUSH support to the Subblock encoder.
The API for handing Subfilters was changed to make it
consistent with LZMA_SYNC_FLUSH.

A few sanity checks were added for Subfilter handling. Some
small bugs were fixed. More comments were added.
2008-01-19 21:16:33 +02:00
Lasse Collin 23c227a864 Revised the Delta filter implementation. The initialization
function is still shared between encoder and decoder, but the
actual coding is in separate files for encoder and decoder.

There are now separate functions for the actual delta
calculation depending on if Delta is the last filter in the
chain or not. If it is the last, the new code copies the
data from input to output buffer and does the delta
calculation at the same time. The old code first copied the
data, then did the delta in the target buffer, which required
reading through the data twice.

Support for LZMA_SYNC_FLUSH was added to the Delta encoder.
This doesn't change anything in the file format.
2008-01-19 15:19:21 +02:00
Lasse Collin 61dc82f3e3 Added the debug directory and the first debug tool
(sync_flush). These tools are not built unless the
user runs "make" in the debug directory.
2008-01-18 20:18:08 +02:00
Lasse Collin 0ae3208db9 Added test files to test usage of flush marker in LZMA. 2008-01-18 20:13:00 +02:00
Lasse Collin ab5feaf1fc Fix LZMA_SYNC_FLUSH handling in LZ and LZMA encoders.
That code is now almost completely in LZ coder, where
it can be shared with other LZ77-based algorithms in
future.
2008-01-18 20:02:52 +02:00
Lasse Collin 079c4f7fc2 Don't add -g to CFLAGS when --enable-debug is specified.
It's the job of the user to put that in CFLAGS.
2008-01-18 17:21:24 +02:00
Lasse Collin 61d1784d8f Set stdin and stdout to binary mode on Windows. This patch is
a forward port of b7b22fcb979a16d3a47c8001f058c9f7d4416068
from lzma-utils-legacy.git. I don't know if the new code base
builds on Windows, but this is a start.
2008-01-18 14:17:37 +02:00
Lasse Collin c9cba97691 Added test_compress.sh and bunch of files needed by it.
This new set of tests compress and decompress several
test files with many different compression options.
This set of tests will be extended later.
2008-01-18 00:50:29 +02:00
Lasse Collin 33be3c0e24 Subblock decoder: Don't exit the main loop in decode_buffer()
too early if we hit End of Input while decoding a Subblock of
type Repeating Data. To keep the loop termination condition
elegant, the order of enumerations in coder->sequence were
changed.

To keep the case-labels in roughly the same order as the
enumerations in coder->sequence, large chunks of code was
moved around. This made the diff big and ugly compared to
the amount of the actual changes made.
2008-01-17 18:56:53 +02:00
Lasse Collin b254bd97b1 Fix wrong too small size of argument unfiltered_max
in ia64_coder_init(). It triggered assert() in
simple_coder.c, and could have caused a buffer overflow.

This error was probably a copypaste mistake, since most
of the simple filters use unfiltered_max = 4.
2008-01-17 17:39:42 +02:00
Lasse Collin 8f5794c8f1 Added --delta to the output of "lzma --help". 2008-01-17 17:27:45 +02:00
Lasse Collin f88590e001 Fix Subblock docoder: If Subblock filter was used with known
Uncompressed Size, and the last output byte was from RLE,
the code didn't stop decoding as it should have done.
2008-01-17 13:14:20 +02:00
Lasse Collin bc0b945ca3 Tiny non-technical edits to file-format.txt. 2008-01-16 16:33:37 +02:00
Lasse Collin 7599bb7064 Plugged a memory leak in stream_decoder.c. 2008-01-16 14:48:04 +02:00
Lasse Collin 0b58153931 Added memory leak detection to lzmadec.c. 2008-01-16 14:47:27 +02:00
Lasse Collin 5b5b13c7bb Added lzma_memlimit_count(). 2008-01-16 14:46:50 +02:00
Lasse Collin 19389f2b82 Added ARRAY_SIZE(array) macro. 2008-01-16 14:31:44 +02:00
Lasse Collin 9bc33a54cb Make Uncompresed Size validation more strict
in alone_decoder.c.
2008-01-16 13:27:03 +02:00
Lasse Collin 01d71d60b7 Free the allocated memory in lzmadec if debugging is
enabled. This should make it possible to detect possible
memory leaks with Valgrind.
2008-01-15 17:46:59 +02:00
Lasse Collin 8235e6e5b2 Fix memory leaks from test_block_header.c. 2008-01-15 16:25:38 +02:00
Lasse Collin f10fc6a69d Use fastpos.h when encoding LZMA dictionary size in
Filter Flags encoder.
2008-01-15 14:23:35 +02:00
Lasse Collin e5728142a2 Revised the fastpos code. It now uses the slightly faster
table-based version from LZMA SDK 4.57. This should be
fast on most systems.

A simpler and smaller alternative version is also provided.
On some CPUs this can be even a little faster than the
default table-based version (see comments in fastpos.h),
but on most systems the table-based code is faster.
2008-01-15 14:02:22 +02:00
Lasse Collin 10437b5b56 Added bsr.h. 2008-01-15 13:32:13 +02:00
Lasse Collin f3c88e8b8d Fixed assembler detection in configure.ac, and added
detection for x86_64.
2008-01-15 13:29:14 +02:00
Lasse Collin 54ec204f58 Omit invalid space from printf() format string
in price_table_gen.c.
2008-01-15 12:20:41 +02:00
Lasse Collin 01b4b19f49 Removed a few unused macros from lzma_common.h. 2008-01-15 09:54:34 +02:00
Lasse Collin 19bd7f3cf2 Fix a typo in lzma_encoder.c. 2008-01-15 08:37:42 +02:00
Lasse Collin 9f9b198301 Convert bittree_get_price() and bittree_reverse_get_price()
from macros to inline functions.
2008-01-15 08:36:25 +02:00
Lasse Collin 78e85cb1a7 Fix CRC code in case --enable-small is used. 2008-01-15 07:44:59 +02:00
Lasse Collin 949d4346e2 Fix typo in test_index.c. 2008-01-15 07:41:39 +02:00
Lasse Collin d13d693155 Added precomputed range coder probability price table. 2008-01-15 07:40:21 +02:00
Lasse Collin 362dc3843b Remove RC_BUFFER_SIZE from lzma_encoder_private.h
and replace it with a sanity check.
2008-01-14 13:42:43 +02:00
Lasse Collin e22b37968d Major changes to LZ encoder, LZMA encoder, and range encoder.
These changes implement support for LZMA_SYNC_FLUSH in LZMA
encoder, and move the temporary buffer needed by range encoder
from lzma_range_encoder structure to lzma_lz_encoder.
2008-01-14 13:39:54 +02:00
Lasse Collin b59ef39737 Added one assert() to process.c of the command line tool. 2008-01-14 13:34:29 +02:00
Lasse Collin 9547e734a0 Don't use coder->lz.stream_end_was_reached in assertions
in match_c.h.
2008-01-14 12:09:52 +02:00
Lasse Collin 3e09e1c058 In lzma_read_match_distances(), don't use
coder->lz.stream_end_was_reached. That variable
will be removed, and the check isn't required anyway.
Rearrange the check so that it doesn't make one to
think that there could be an integer overflow.
2008-01-14 12:08:02 +02:00
Lasse Collin a670fec802 Small LZMA_SYNC_FLUSH fixes to Block and Single-Stream encoders. 2008-01-14 11:56:41 +02:00
Lasse Collin 3599dba957 More fixes to LZMA decoder's flush marker handling. 2008-01-14 11:54:56 +02:00
Lasse Collin f73c2ab607 Eliminate lzma_lz_encoder.must_move_pos. It's needed
only in one place which isn't performance criticial.
2008-01-10 17:13:42 +02:00
Lasse Collin 382808514a Define HAVE_ASM_X86 when x86 assembler optimizations are
used. This #define will be useful for inline assembly.
2008-01-09 20:05:57 +02:00
Lasse Collin 0e70fbe403 Added good-single-none-empty_3.lzma and
bad-single-none-empty.lzma.
2008-01-09 12:06:46 +02:00
Lasse Collin 379fbbe84d Take advantage of return_if_error() in block_decoder.c. 2008-01-08 23:11:59 +02:00
Lasse Collin 97d5fa8207 Updated tests/files/README. 2008-01-08 23:10:57 +02:00
Lasse Collin 3bb9bb3109 Added test files with empty Compressed Data. 2008-01-08 23:05:40 +02:00
Lasse Collin 7054c5f588 Fix decoding of Blocks that have only Block Header. 2008-01-08 22:58:42 +02:00
Lasse Collin 753e4d95cd Added good-single-subblock_implicit.lzma. 2008-01-08 22:27:46 +02:00
Lasse Collin faeac7b7ac Disable CRC32 from Block Headers when --check=none
has been specified.
2008-01-08 18:50:30 +02:00
Lasse Collin a751126dbb Fixed encoding of empty files. Arguments to is_size_valid()
were in wrong order in block_encoder.c.
2008-01-08 13:36:29 +02:00
Lasse Collin 9080267603 Added a few test files. 2008-01-08 13:35:36 +02:00