roadroller VS wordle-trie-packing

Crinkler [1] is a popular compressor-linker for 1--8 KB demos and its decompressor (partially embedded in a PE header) is probably around 1--200 bytes. Later efforts like oneKpaq [2] also have a comparable decompressor size.

If you don't mind a shameless plug and a slightly larger decompressor (about 500 bytes in JS) for better compression, my Roadroller [3] might fit the bill as well.

[1] https://github.com/runestubbe/Crinkler

[2] https://github.com/temisu/oneKpaq

[3] https://lifthrasiir.github.io/roadroller/

Roadroller [1] is probably a borderline general purpose compression algorithm, and with some automatic parameter tuning it results in 12,170 bytes estimated [2], at the expense of a lot of memory. "Estimated" because the algorithm was originally meant to be recompresssed in a ZIP file, so it doesn't bother to generate the smallest JS file in terms of uncompressed size (yet).

[1] https://lifthrasiir.github.io/roadroller/ (the exact parameters: golf.horse dataset; input mode text; action write to document; # contexts 12 with 12,15,49,50,70,79,96,97,131,154,292,353; pollute the global scope; max memory usage 150 MB; precision 16; learning rate 1333; model max count 11; model base divisor 14; dynamic model flags -1; # abbreviations 64)

And compressed final code even more with Roadroller (reduced 1.3 KB in the zip). And used ECT to recompress the zip (reduced 300 KB).

I'm honored that Q1K3, in addition to many other entries to js13kGames 2021, is compresed with Roadroller [1], which is the reason that its source code is basically a random-looking string plus some bit of JS code.

[1] https://github.com/lifthrasiir/roadroller/

```

Doing this, and then doing the variable length encoding I got the file down to 13,181 bytes (it was 13,180 bytes, but I needed to add a 7 bit termination string so that you can properly decode the file after you write it to disk, otherwise when the file rounds to the nearest byte you have random 0s that get decoded).

I'm sure with some twiddling of the alphabets some more you could save a few more bytes, but this does better than both Brotli on a ASCII trie and the Huffman Trie by almost 1KB (https://github.com/adamcw/wordle-trie-packing#all-words), so I'm very happy.