barubary
4ddd87206b
C#: the composite formats that were added in r48 should now be supported by the program. Not yet tested though.
...
Added support for the 'NULL' compression format; a built-in Nitro format that only contains plaintext (but includes the nitro compression header with magic byte 0)
2011-12-22 22:59:37 +00:00
barubary
6dc34d20a6
C#: Nitro-formats will now also try to decompress files with a plaintext size of more than 0x180000. However the Supports method will first try to partially decompress the file.
2011-11-23 10:35:29 +00:00
barubary
a30a7ac3e3
C#: implemented a compression algorithm for the LZE/Le compression scheme. (although it appears to perform worse than the original compression algorithm, it should still work)
2011-11-21 23:46:48 +00:00
barubary
4302517e58
C#: Added xml comments to methods that did not have any (in the main DSDecmp project).
...
The SupportsDecompression-flag is now also used in the main program, and not just in composite formats.
2011-11-14 17:33:30 +00:00
barubary
aff22d28da
C#: revised implementation of the tree-writing algorithm for the 8-bit huffman format. A test with a file containing all values from 0 to 0xFF evenly now does not result in a corrupt tree definition.
2011-11-13 22:21:18 +00:00
barubary
d64137221a
C#: A step into a new system, where game-specific formats can be defined as plugins, preventing users from having to download DSDecmp itself again and again. Such plugins must be stores in a 'Plugins' folder in the same location as the DSDecmp.exe.
...
To keep the base program mobile, all native NDS formats do not require plugins.
The 8-bit Huffman format cannot be used to compress at the moment; thanks to CUE a bug has been found where the offset-field of a node overflows, corrupting the compressed file.
2011-11-13 16:14:45 +00:00
barubary
c1994426f3
C#: Fix for the LZ-Ovl format; the compressed size may also be the file length, instead of the file length minus the header size.
2011-07-27 22:13:56 +00:00
barubary
0f9b368b04
C#: added the Golden Sun: Dark Dawn format in the new structure. This version hasn't been tested yet, and thus hasn't been included in the main program. I also need to think of how to include these game-specific formats in the main program, as they may conflict with built-in formats regarding their header (with GSDD I just try to decompress the file when checking if a file is supported).
2011-05-18 12:24:16 +00:00
barubary
42a6085222
C#: very minor tweak to the usage text to prevent an extra newline.
2011-05-15 15:53:00 +00:00
barubary
441848074d
C#: some fixes in the 'UI' part of the program.
2011-05-15 15:49:21 +00:00
barubary
3c0e60d574
C#: removed the old Program.cs from the project, and rebuilt the .exe to trim the filesize a bit.
2011-05-15 15:24:22 +00:00
barubary
3bf20290a3
C#: Disabled LZ-Overlay from 'decompressing' files that have four 0-bytes at the end. Also implemented the command-line input, and added a new release build to the bin/ folder.
2011-05-15 15:16:04 +00:00
barubary
591c44b309
C#: implemented Huffman compression for both 4-bit and 8-bit block sizes. It will not result in compressed files identical to those found in games, as I'm currently not in the mood to figure out exactly what heuristic they use to build the Huffman tree.
2011-05-15 11:40:14 +00:00
barubary
99366a7a9a
C#: made a start on the Huffman compression algorithm; the Huffman tree for 8-bit datablocks is now created.
2011-05-14 21:35:12 +00:00
barubary
a31e3b30a6
C#: the 'optimal' LZ-Overlay compression method will now stop with compressing when it's no longer useful to do so, allowing the start of the file to remain uncompressed.
2011-05-14 20:01:09 +00:00
barubary
48f189c991
C#: Added compression algorithms for the LZ-Overlay format. At the moment, the file will either be fully compressed or not compressed at all. The 'optimal' method also has a higher chance of being sub-optimal due to the nature of overlay files (usually double-compressed files).
2011-05-14 18:52:46 +00:00
barubary
b8d11b3e9d
C#: implemented the LZ-11 compression algorithms (the 'original' and the 'optimal' algorithms).
2011-05-14 16:00:50 +00:00
barubary
8892f527f5
C#: added a variation to the original LZ-10 compression algorithm that increases the compression rate, while still ensuring compatibility with the built-in decompression.
2011-05-14 14:19:11 +00:00
barubary
2c507a5765
C#: Tested the LZ-10 compression algorithm, and adjusted it to work the same as the original implementation. Decompress->Compress of an original file should thus result in the same original file.
2011-05-14 12:24:44 +00:00
barubary
1c2722315a
C#: added an initial compression algorithm for LZ-10. May not work yet, as it hasn't been tested.
2011-05-13 10:07:38 +00:00
barubary
69cd6ee199
C#: added built-in check for arm9.bin in LZ-Overlay decompression, as it has some extra bytes that should be ignored when decompressing. Only the file-based methods can handle that of course. Perhaps the extra bytes should also be copied to the output file however, which is currently not the case.
2011-04-21 14:16:59 +00:00
barubary
167300d5b8
C#: added usage description for the still-to-be-built console interface to the new compressor/decompressor structure.
2011-04-21 13:58:07 +00:00
barubary
80efded10b
C#: implemented the LZ-Overlay decompression in the new structure. (still untested again).
2011-04-21 11:52:09 +00:00
barubary
426f89f37c
C#: added a check in LZOvl.Support to make sure the 'compressed size' is correct (fits within the given file).
2011-04-13 12:24:20 +00:00
barubary
8d1a6abd9d
C#: started on the new implementation for the LZ-overlay format. Only the Supports-method is implemented at the moment.
2011-04-08 08:14:44 +00:00
barubary
5f77ee332e
C#: updated the project file to include the IO-utils class.
2011-04-07 13:55:10 +00:00
barubary
468e2e645e
C#: removed the BitConverter reference in the Huffman decompressor, and replaced the call with one to a local utility method (to be sure that the byte*->u32 conversion always uses the correct endianness.
2011-04-07 13:51:09 +00:00
barubary
8a5583c308
C#: fixed The new LZ-0x10 and LZ-0x11 decompressors. Only LZ-0x10 hasn't been tested yet.
2011-04-06 16:42:27 +00:00
barubary
5ecc296c1a
C#: fixed the new Huffman decompressor.
2011-04-06 16:00:04 +00:00
barubary
fa10d8a515
C#: Extended the generic Supports(Stream) method with the length of the input stream, since Overlay-LZ compresses from end to start.
2011-04-05 16:50:17 +00:00
barubary
3a6221a0c6
C#: Fixed the writing of the header of the RLE compressed files; it should state the decompressed size, and not the compressed size. Also added an exception that indicates there is still some input data to read after the compressed stream.
...
The current RLE implementation passed the first test: compression->decompression resulted in the same file for a manually created test file. (and the compressed file also has the expected format)
2011-04-05 16:44:41 +00:00
barubary
2c6ef5570a
C#: updated the generic compression/decompression functions to make sure the folder of the given output file exists before creating the file.
2011-04-05 16:41:43 +00:00
barubary
bd9c45a89b
C#: improved the RLE compressor, even though it's still untested. (should no longer read beyond the end of the input stream)
2011-04-05 16:18:03 +00:00
barubary
edc3d2da24
C#: added an untested compressed algorithm for RLE. (all new decompression algorithms are also untested)
2011-04-04 13:07:07 +00:00
barubary
da75d115e0
C#: completed the implementation for the decompression of the Huffman format.
2011-04-01 14:12:23 +00:00
barubary
1c194134ac
C#: started on re-implementing the huffman decompression in the mew format.
2011-03-23 12:04:59 +00:00
barubary
ba18c44249
C#: the project file is now for VS2010. Removed the unused using-statement for Linq to get .NET 2.0 compatibility again.
2011-03-23 09:53:25 +00:00
barubary
0e5fe616c3
C#: Re-implemented the RLE decompression in the new format. The new format file has not been added to the project file, as I can only work in VS2010 where I am at the moment (and manually editing the file seems to break the autorun every time I try it).
2011-03-22 14:24:22 +00:00
barubary
1bd750c76d
C# (previous one as well): re-implemented the LZ-0x11 algorithm in the new structure. Also updated the GBATEK-like format description of the format, hopefully correct this time.
2011-03-22 11:31:48 +00:00
barubary
8c777e2c02
The start of an attempt to make this program structured better, as well as for easier addition of more formats AND compression for supported formats.
2011-03-21 19:18:07 +00:00
barubary
052b9c0677
Augh. idiot-bug fixed. (the last 0xC bytes of arm9.bin should be ignored, no the first ones)
2011-01-16 00:38:02 +00:00
barubary
a19592264e
Last adjustment to the LZ-Overlay decompression for now, until I can compare a properly decompressed file with one that doesn't come out right with this program.
2011-01-16 00:25:46 +00:00
barubary
e8b3913596
Minor adjustment in LZ-Overlay decompressing, catching more Disp-values that are too large.
2011-01-16 00:11:04 +00:00
barubary
11564be73d
Fixed some bugs in the LZ-Overlay code.
2011-01-15 23:51:06 +00:00
barubary
2334261982
Added support for compression used in overlay files of the DS, as well as the arm9.bin file. The format is mostly LZ-0x10 in reverse, but has a small difference in the calculation of the Disp value for compressed blocks. The complete file also need not be compressed.
2011-01-15 21:38:13 +00:00
barubary
0e13ad064e
Added a new compression format in the C# source: type 0x40 LZ77/LZSS. It differs not very much from type 0x11 in compression size, but has an odd way of storing which blocks are compressed.
...
A Java-version of the scheme may come at a later time.
I _think_ this is a built-in compression, but it could also be a format Camelot made and used only for the latest Golden Sun installment.
Oh, I've also added the MIT licence to the C# source code file.
2010-10-28 23:24:56 +00:00
barubary
4500dfc1e1
Upgraded sln to VS2008.
...
Also updated the source to match my local copy. No changes in decompression, only usability,
2010-09-21 10:05:10 +00:00
barubary
a597cd23a9
added source code for both C# and Java.
...
Note that the Java algorithms are a port of the C# ones, and only the LZ10 and LZ11 have been tested of the Java implementation.
2009-06-07 13:40:49 +00:00
