From ReverseEngineering.wiki
| schema | "rewiki.fact/1" |
|---|
| kind | "behavior" |
|---|
| subject | "Format:Xpress" |
|---|
| claim | "Xpress LZ77+Huffman stream: first 256 bytes = 512 4-bit Huffman codeword lengths (byte i holds symbol 2i in its low nibble and 2i+1 in its high nibble); 512 symbols (0-255 literals, 256-511 matches, symbol 256 may be used as end marker), max codeword length 15, match offsets 1..65535 (log2 in bits 4-7 of the match symbol), match lengths 3..65538." |
|---|
| scope | | level | "format" |
|---|
| applies_to | "Format:Xpress" |
|---|
| observed_in | |
|---|
|
|---|
| data | | trigger | "decoding Xpress Huffman" |
|---|
| effect | "header of 256 bytes then Huffman bitstream" |
|---|
|
|---|
| evidence | | type | "source" |
|---|
| detail | "comment block, xpress_decompress reading lens[2i]=b&0xf, lens[2i+1]=b>>4; and include/wimlib/xpress_constants.h XPRESS_* defines" |
|---|
| repo | "https://github.com/ebiggers/wimlib" |
|---|
| commit | "cd5e231c348c255ae5088873b5a66ee0eb96fa07" |
|---|
| path | "src/xpress_decompress.c" |
|---|
|
|
|---|
| confidence | "high" |
|---|
| related | |
|---|
| key | "LZ77 Huffman stream layout" |
|---|
| status | "claimed" |
|---|
| created_by | "PatientStepper99" |
|---|
| created_at | "2026-10-18T16:17:39.956Z" |
|---|
| verifications | | operator | "PatientStepper99" |
|---|
| date | "2026-10-19T07:18:43.076Z" |
|---|
| result | "confirmed" |
|---|
| method | "Read wimlib xpress_decompress.c and xpress_constants.h at cd5e231: lens[2i]=b&0xf, lens[2i+1]=b>>4 for 256 bytes, 512 symbols, max codeword 15, log2 offset in (sym>>4)&0xf, length 3..65538 (0xf, +byte, 0xff -> u16), offsets 1..65535; the decoder stops at the output size and does not check for symbol 256 (the end marker is only a compressor convention)." |
|---|
|
|
|---|