<GitHub68>
jruby/truffle-head ee52e2e Kevin Menard: [Truffle] Use the GetRubyEncodingNode anywhere we expect converting a jcodings Encoding -> Ruby encoding to be fast.
<GitHub68>
jruby/truffle-head aca2d0e Kevin Menard: [Truffle] Added a new GetRubyEncodingNode to have a fast path for converting jcodings encodings to Ruby encodings.
vtunka has quit [Quit: Leaving]
pawnbox has joined #jruby
pawnbox has quit [Remote host closed the connection]
<nirvdrum>
The code has been there for a while, so I tend to believe it's working. But it seems like this would be problematic for any bytes with the high bit set, since they're signed.
camlow325 has joined #jruby
swills has quit [Read error: Connection reset by peer]
swills has joined #jruby
thedarkone2 has quit [Quit: thedarkone2]
swills has quit [Read error: Connection reset by peer]
swills has joined #jruby
alex0ptr has joined #jruby
alex0ptr has quit [Ping timeout: 246 seconds]
<lopex>
nirvdrum: the equality should be ok, but not sure about the returned value
_whitelogger has joined #jruby
bga57 has joined #jruby
Guest25870 is now known as yipdw
yipdw has joined #jruby
yipdw has quit [Changing host]
swills has quit [Read error: Connection reset by peer]
swills has joined #jruby
atambo has joined #jruby
knowtheory has joined #jruby
bruceadams has joined #jruby
lopex has joined #jruby
deathy has joined #jruby
pawnbox has quit [Remote host closed the connection]
pawnbox has joined #jruby
thedarkone2 has joined #jruby
<nirvdrum>
lopex: It looks to me like if the first byte value is < 128 and the other is >= 128, you get back a bad value.
<lopex>
nirvdrum: but -12 - -12 is still zero
<lopex>
but yeah, you'll get bad diff returned
<nirvdrum>
Yes.
<lopex>
nirvdrum: probably newver used or always compared to zero :P
<GitHub116>
jruby/truffle-head 8197184 Chris Seaton: [Truffle] Core method nodes almost never need constructors.
<GitHub116>
jruby/truffle-head f215be1 Chris Seaton: [Truffle] Primitive nodes never need a source section constructor, and rarely need a constructor at all.
<GitHub116>
jruby/truffle-head 15ff866 Chris Seaton: [Truffle] YieldCoreMethodNode doesn't need a source constructor.
<GitHub131>
jruby/truffle-head 078586b Chris Seaton: [Truffle] Dispatch nodes are no longer a reason to need a constructor.
<GitHub131>
jruby/truffle-head 68ced32 Chris Seaton: [Truffle] Use RubyBaseNode and its system for finding and caching the context, for dispatch head nodes.
<GitHub138>
jruby/master b36a1c1 Thomas E. Enebo: Add comment and conditionalize how we treat expand_path on windows vs non-windows
enebo has quit [Quit: enebo]
lanceball is now known as lance|afk
brocktimus has joined #jruby
<brocktimus>
@headius @enebo I noticed this PR on rails which I figured would interest you https://github.com/rails/rails/pull/27363. Rails starting to use refinements.
<brocktimus>
Was going to @ you on GH but wasn't sure if that's poking my nose into another projects business