This is an implementation of LZW compression in Parrot assembly. The fact that pack() can't handle null bytes makes it a bit more complicated (and limited) than it has to be, but for just text files, it seems to work just fine. It's probably a decent stress test for the hash PMC, but especially for strings, since I probably go through more strings than I have to.
It's also good for showing interesting performance effects: Compressing 20875 bytes: 5.3 sec Compressing 19646 bytes: 1.4 sec I haven't looked into exactly why, but it looks like compact_string_pool() is going insane (see attachment). /s
ziptest.tgz
Description: Binary data
Smaller file: % cumulative self self total time seconds seconds calls us/call us/call name 31.03 0.18 0.18 130 1384.62 1384.62 compact_string_pool 10.34 0.24 0.06 239357 0.25 0.25 singlebyte_decode 10.34 0.30 0.06 221558 0.27 0.27 buffer_lives 5.17 0.33 0.03 299167 0.10 0.10 singlebyte_skip_forward 5.17 0.36 0.03 168 178.57 535.37 mark_hash 5.17 0.39 0.03 1 30000.00 251150.69 cg_core Larger file: % cumulative self self total time seconds seconds calls us/call us/call name 70.30 0.71 0.71 217 3271.89 3271.89 compact_string_pool 5.94 0.77 0.06 168 357.14 475.98 mark_hash 2.97 0.80 0.03 85754 0.35 8.74 Parrot_allocate_string 2.97 0.83 0.03 1 30000.00 530991.70 cg_core
