This can be verified pretty easily in the shell, but there's an <a href="http://erlang

:erlang.phash2 does not uniformly distribute atoms across the hash range, should this be documented in README? about libring HOT 3 CLOSED

bitwalker commented on June 1, 2024

:erlang.phash2 does not uniformly distribute atoms across the hash range, should this be documented in README?

from libring.

Comments (3)

bitwalker commented on June 1, 2024

I'm fine with doing a pre-process step via :erlang.term_to_binary/1 if that actually solves the problem, otherwise I suppose we will need a new hash function

from libring.

jlerche commented on June 1, 2024

That should be ok as :erlang.term_to_binary is pretty fast right? Here's some testing I did

iex(1)> :erlang.phash2(:foo)
27999
iex(2)> :erlang.phash2(:erlang.term_to_binary(:foo))
86811783
iex(3)> :erlang.phash2(:erlang.term_to_binary(:fooaaaa))
43291300
iex(4)> :erlang.phash2(:erlang.term_to_binary(:ffgfgff))
30215940
iex(5)> :erlang.phash2(:erlang.term_to_binary(:a))      
64216263
iex(6)> :erlang.phash2(:erlang.term_to_binary(:aaaaaa))
86861957
iex(7)> :erlang.phash2(:erlang.term_to_binary(:b))     
14428281
iex(8)> :erlang.phash2(:erlang.term_to_binary(:b), trunc(:math.pow(2,32)-1))
3235653753
iex(9)> :erlang.phash2(:foo, trunc(:math.pow(2,32)-1))                      
27999
iex(10)> :erlang.phash2(:erlang.term_to_binary(:foo), trunc(:math.pow(2,32)-1))
355247239

which is a lot closer to expected behavior.

from libring.

jordanhubbard commented on June 1, 2024

Is this why elixir HEAD is currently generating one test failure?

test ensure function clauses are ordered (ModuleTest)
test/elixir/module_test.exs:289
Assertion with == failed
code: assert :erlang.phash2(atoms) == 61635213
left: 60755908
right: 61635213
stacktrace:
test/elixir/module_test.exs:297: (test)

from libring.

Recommend Projects