DomKM has quit [Quit: Connection closed for inactivity]
subbu has joined #jruby
subbu has quit [Ping timeout: 256 seconds]
pietr0 has quit [Quit: pietr0]
yfeldblum has joined #jruby
mcclurmc has joined #jruby
yfeldblum has quit [Ping timeout: 245 seconds]
mcclurmc_ has joined #jruby
mcclurmc has quit [Ping timeout: 258 seconds]
kotk_ has joined #jruby
kotk has quit [Ping timeout: 258 seconds]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] chrisseaton pushed 2 new commits to master: http://git.io/PLxF-w
<JRubyGithub>
jruby/master c1ed261 Chris Seaton: [Truffle] Always call block parameters 'block' so they aren't confused with final parameters that just happen to take a RubyProc.
<JRubyGithub>
jruby/master af77cd9 Chris Seaton: Merge branch 'master' of github.com:jruby/jruby
JRubyGithub has left #jruby [#jruby]
mcclurmc_ has quit [Remote host closed the connection]
ivan\ has quit [Ping timeout: 252 seconds]
ivan\ has joined #jruby
marr has quit [Ping timeout: 240 seconds]
ivan\ has quit [Ping timeout: 258 seconds]
ivan\ has joined #jruby
DomKM has joined #jruby
calavera has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<JRubyGithub>
jruby/master ba596b0 Thomas E. Enebo: Remove newline unwrapping because I cannot get any tests to fail with it gone
JRubyGithub has left #jruby [#jruby]
triple_b has joined #jruby
cprice404 has joined #jruby
benlovell has quit [Ping timeout: 245 seconds]
elia has joined #jruby
frobs has quit [Quit: Leaving]
<headius>
lopex: you implemented our symbol table, right?
nirvdrum_ has quit [Read error: Connection reset by peer]
<lopex>
headius: I dont think so, but I definitely changed some of that
<lopex>
headius: looking for bytelist based symbols ?
Felystirra has joined #jruby
yfeldblum has joined #jruby
<headius>
no, we're contemplating what would break if we eliminated the table altogether
<headius>
literal symbols are cached anyway, so the concern would mostly be about crafted symbols not being the same object and crafted symbols bloating up memory with multiple duplicates
<headius>
object_id is also an unsolved challenge :-\
<lopex>
headius: how mri deals with that now ? (the symbols)
<lopex>
there's must be some more special casing in their gc now I guess
<headius>
yeah I'm sure there is
<headius>
I guess it's a question about how idempotent they have to be
<headius>
we can guarantee literal symbols will remain the same object, but do we have to guarantee it for crafted symbols
<lopex>
maybe special version of equality ?
<lopex>
"non literal" aware
<lopex>
hm
yfeldblum has quit [Ping timeout: 255 seconds]
<headius>
we can fake out equality everywhere except identity at JVM level
<headius>
which is a concern
<headius>
and we'd have to clean up codebase to eliminate any other places we do == equality to compare symbols
<lopex>
and there's the hash and "eql?"
<lopex>
not sure if it complicates things
<headius>
not really
fniephaus has joined #jruby
elia has quit [Quit: Computer has gone to sleep.]
diegoviola has joined #jruby
elia has joined #jruby
subbu is now known as subbu|lunch
<lopex>
headius: looks like they treat fstrings quite specially now too
<lopex>
separate table for that
<lopex>
and there's the fstr in RSymbol
elia has quit [Client Quit]
<headius>
yeah I have an fstring table already
<headius>
ahh, fstring in symbol is a good idea...we do already cache the bytelist
<headius>
might as well do the fstring since symbols can GC
<headius>
I'm trying to think what the new table structure needs to be
<headius>
it's not pretty
<headius>
requirements:
<headius>
* weakly reference symbols
<headius>
* O(1) lookup from j.l.String
<headius>
* O(1) lookup from ByteList
<headius>
(these are current impl guarantees I'm not comfortable breaking)
<lopex>
encodings can complicate things
<lopex>
the lookups
<headius>
a bit, but MRI basically uses the raw bytes of the string's native encoding as the symbol
<headius>
mrmargolis: I know I can do alt + click to get old behavior as well
<mrmargolis>
ah, I didn't know that
yfeldblum has joined #jruby
benlovell has joined #jruby
<headius>
mrmargolis: I just started trying modifier keys and stumbled on it
<headius>
I think there's some others
<headius>
bleh, symbols are still increasing
subbu|lunch is now known as subbu
<headius>
guess I'll have to do some heap diving
<subbu>
brixen, started sometime july 2009 or so .. but mostly worked on it in spurts .. some in 2010, and again 2011/2012 and some 2014.
benlovell has quit [Ping timeout: 244 seconds]
yfeldblum has quit [Ping timeout: 272 seconds]
e_dub has quit [Quit: e_dub]
diegoviola has quit [Quit: WeeChat 1.0.1]
<headius>
heh
<headius>
fixed the references, started my test, and my kid spilled my coffee...came back and heap is level, so I guess this is working
<headius>
symbol gc: DONE
<headius>
asarih, mpapis: I was pulled away yesterday...what's the status?
diegoviola has joined #jruby
mcclurmc has quit [Remote host closed the connection]
lanceball is now known as lance|afk
marr has quit [Ping timeout: 245 seconds]
<mpapis>
headius, bug in rvm is one thing, I will be working on it now ... hopefully
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius pushed 1 new commit to master: http://git.io/bbtgsA
<JRubyGithub>
jruby/master 9a73dec Charles Oliver Nutter: enumerator, rational, and complex are considered loaded at boot.
JRubyGithub has left #jruby [#jruby]
pietr0 has joined #jruby
<headius>
mpapis: please do :-)
mattwildig has quit []
elia has quit [Quit: Computer has gone to sleep.]
yfeldblum has joined #jruby
tlarevo has quit [Remote host closed the connection]
tlarevo has joined #jruby
calavera has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
calavera has joined #jruby
e_dub has joined #jruby
yfeldblum has quit [Ping timeout: 244 seconds]
johnmuhl has quit [Quit: Connection closed for inactivity]
noop1 has quit [Ping timeout: 264 seconds]
JohnBat26 has joined #jruby
tlarevo_ has joined #jruby
tlarevo has quit [Ping timeout: 240 seconds]
rcvalle has joined #jruby
subbu is now known as subbu|afk
<mpapis>
headius, asarih bug in RVM fixed, proper jruby will be installed now, need to fix https://github.com/wayneeseguin/rvm/issues/3244 and will release version, might be in ~12h if I do not manage to do it shortly ... but it's not so big ... anyway please test "rvm get head" to be sure there is no other problem
<headius>
ok
<headius>
thank you!
<asarih>
I'm off until Monday, so the release would be probably Tuesday.
Hobogrammer has joined #jruby
<mpapis>
so no hurry ok
calavera has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
calavera has joined #jruby
yfeldblum has joined #jruby
calavera has quit [Client Quit]
tlarevo_ has quit [Ping timeout: 256 seconds]
<headius>
mpapis: if it's out, though, people can update rvm locally or on travis to fix the problem
<headius>
still requires intervention by users though
diegoviola has quit [Quit: WeeChat 1.0.1]
subbu|afk is now known as subbu
yfeldblum has quit [Ping timeout: 255 seconds]
<brixen>
subbu: ah, thanks, I thought it was about then
marr has joined #jruby
<mpapis>
headius, the problem is 1) that travis does not give you option to update rvm before selecting ruby 2) it would not matter as the version used while building was faulty and the "bad" jruby is shipped in images --- basically images rebuild is needed
erikhatcher has quit [Quit: erikhatcher]
<chrisseaton>
has anyone seen mkristian around here recently? I had a Maven question he might have been able to help me with
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] rdp opened issue #2387: absolute paths without drive letter don't work in windows http://git.io/LpEfrQ
JRubyGithub has left #jruby [#jruby]
mrmargolis has quit [Read error: Connection reset by peer]
mrmargolis has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] enebo pushed 3 new commits to master: http://git.io/S98ESA
<JRubyGithub>
jruby/master 6cfa5bb Thomas E. Enebo: Next parser generation will pick up these comments but they are only for my notes
<JRubyGithub>
jruby/master 8d89919 Thomas E. Enebo: All ArrayNodes for args list are lightweight. Simplify buildCallArgs a little more
<JRubyGithub>
jruby/master 1e3d305 Thomas E. Enebo: Remove some arity-split AST nodes which we no longer use
JRubyGithub has left #jruby [#jruby]
Felystirra has quit []
yfeldblum has joined #jruby
<headius>
mpapis: if anyone's playing along at home it would help them, but I doubt most folks are using --1.8
<headius>
chrisseaton: not since before xmas
<mpapis>
headius, I think ou should discuss this with travis, not sure if it makes sense to build new binaries with the new versions when there is 1.9 compatibility by default already ...
mrmargolis has quit [Read error: Connection reset by peer]
mrmargolis has joined #jruby
yfeldblum has quit [Ping timeout: 240 seconds]
<mpapis>
headius, 1.8 compatibility is one of the new selling points of rbx ;)
<mpapis>
headius, maybe they have some stats on the ruby usage ... that would make sense to collect this data
<headius>
mpapis: agreed about travis
<mpapis>
:)
<headius>
I don't really care because we've already dropped 1.8
<headius>
numbers won't change anyhting
<chrisseaton>
well we can still support 1.8 on the 1.7 branch - and that's all MRI does - maintain an old branch
<headius>
they don't maintain it
<headius>
it's EOL
Hobogrammer has quit [Ping timeout: 265 seconds]
<chrisseaton>
i thought it got essential security fixes? does it not even get that?
<chrisseaton>
it must do from people like canonical
<headius>
I suppose we could patch it forever toward eliminating all feature deltas, but 1.8 is a dead duck
<chrisseaton>
and red hat?
<headius>
third parties maintain security fixes
<headius>
if third parties want to maintain jruby --1.8 they're welcome to do so
<JRubyGithub>
[jruby] rjnienaber opened issue #2389: Kernel.system does not utilize tailing arguments in 1.9 mode http://git.io/Fy9fIw
JRubyGithub has left #jruby [#jruby]
yfeldblum has quit [Ping timeout: 244 seconds]
havenwood has quit []
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] rjnienaber opened issue #2390: BigDecimal can be provoked into a big infinite loop http://git.io/ea1q9g
JRubyGithub has left #jruby [#jruby]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] rjnienaber opened issue #2391: File.dirname normalization differs from MRI http://git.io/zL1tMQ
JRubyGithub has left #jruby [#jruby]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] rjnienaber opened issue #2392: Closure conversion doesn't play nice with some interface method names http://git.io/mK6EVw
JRubyGithub has left #jruby [#jruby]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] rjnienaber opened issue #2393: Unable to hide a singleton method created by an extend http://git.io/qjVZag
JRubyGithub has left #jruby [#jruby]
pietr0 has quit [Quit: pietr0]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] rjnienaber opened issue #2394: Compiler trips up on a combination of send and eval (concocted example) http://git.io/DIQM3Q
JRubyGithub has left #jruby [#jruby]
tcrawley-away is now known as tcrawley
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] rjnienaber opened issue #2395: Setup a SecurityManager from JRuby http://git.io/WFtkLA
JRubyGithub has left #jruby [#jruby]
nateberkopec has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] rjnienaber opened issue #2396: String does not use <=> to compare two string instances http://git.io/DpHyhw
JRubyGithub has left #jruby [#jruby]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] rjnienaber opened issue #2397: Make String/ByteList creation facilities part of RedBridge APIs http://git.io/uJSsCQ
JRubyGithub has left #jruby [#jruby]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] rjnienaber opened issue #2398: Ruby.newInstance(RubyInstanceConfig) should not unconditionally setGlobalRuntime http://git.io/mmG1EQ
JRubyGithub has left #jruby [#jruby]
fniephaus has quit [Quit: fniephaus]
<rjnienaber>
headius: I've recreated 11 of the 12 Open issues in the Codehaus Jira on Github. I've left http://jira.codehaus.org/browse/JRUBY-7131 as there's no been no response from the original reporter