<JRubyGithub>
[jruby] chrisseaton pushed 2 new commits to truffle-head: http://git.io/FwrL
<JRubyGithub>
jruby/truffle-head 12ed071 Chris Seaton: [Truffle] Make coverage disabled by default.
<JRubyGithub>
jruby/truffle-head ec5b7a3 Chris Seaton: Merge branch 'master' into truffle-head...
JRubyGithub has left #jruby [#jruby]
iamjarvo has joined #jruby
dinfuehr has joined #jruby
mrmargolis has joined #jruby
vtunka has quit [Quit: Leaving]
ivan\ has quit [Read error: Connection reset by peer]
dinfuehr_ has quit [Ping timeout: 276 seconds]
dinfuehr has quit [Ping timeout: 264 seconds]
ivan\ has joined #jruby
marr has quit [Ping timeout: 245 seconds]
iamjarvo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
metadave has quit [Ping timeout: 255 seconds]
enebo has joined #jruby
skade has quit [Read error: Connection reset by peer]
slyphon has joined #jruby
mrmargolis has quit [Remote host closed the connection]
dinfuehr has joined #jruby
shellac has quit [Quit: Computer has gone to sleep.]
calavera has joined #jruby
elia has quit [Read error: Connection reset by peer]
elia has joined #jruby
slyphon has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
yfeldblum has joined #jruby
calavera has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
shellac has joined #jruby
calavera has joined #jruby
yfeldblum has quit [Ping timeout: 264 seconds]
e_dub has joined #jruby
e_dub has quit [Client Quit]
iamjarvo has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 3 new commits to master: http://git.io/FrTI
<JRubyGithub>
jruby/master 8a75264 Benoit Daloze: [Truffle] Add a missing form of eval.
<JRubyGithub>
jruby/master 457d3c7 Benoit Daloze: [Truffle] Get the right frame in module_eval.
<JRubyGithub>
jruby/master 0d15b09 Benoit Daloze: [Truffle] Implement Array#map! for empty Array, File.basename and File.open(..., no_block).
JRubyGithub has left #jruby [#jruby]
cremes has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] k77ch7 opened pull request #2537: Fix for issue 2524 on master: [BigDecimal] Loss of precision with different execution order (master...fix_GH-2524_on_master) http://git.io/FrLZ
JRubyGithub has left #jruby [#jruby]
mrmargolis has joined #jruby
slyphon has joined #jruby
e_dub has joined #jruby
noop has quit [Ping timeout: 246 seconds]
cprice404 has quit [Remote host closed the connection]
cprice has joined #jruby
cprice is now known as cprice404
paulswilliamsesq has joined #jruby
sferik has joined #jruby
JRubyGithub has joined #jruby
JRubyGithub has left #jruby [#jruby]
<JRubyGithub>
jruby/master 48dfab8 Benoit Daloze: [Truffle] No more a race condition since synchronizing on liveObjects.
<JRubyGithub>
[jruby] eregon pushed 1 new commit to master: http://git.io/Frs7
slyphon has quit [Ping timeout: 245 seconds]
cprice404 has quit [Read error: Connection reset by peer]
<enebo>
heh…I have not thought of that skit in a long time
Who has joined #jruby
<Who>
enebo: I am not done yet :P
<Who>
I still got some fight left
<enebo>
Go Who Go!
<enebo>
:)
<enebo>
Was eating lunch
<Who>
no problem :)
<Who>
do you have issues or features you want me to help out with? If not I was thinking of looking at LoadService and port MRI impl (at least the search_required method)
sferik has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
iamjarvo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<enebo>
Who: I don’t offhand. If you can solve that it would be truly great
<Who>
also there was the frozen kwargs issue
<Who>
and 2015 connected to it
calavera has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<enebo>
Who: I actually have it solved locally for interp but I don’t like it
<enebo>
Who: kwargs
<Who>
ah
<Who>
is there something better that can be done?
<enebo>
Who: That snippet I gave the other day is basically what I did and it works. I figured out a place to inject it for methods.
<enebo>
Who: well that is the question
<enebo>
Who: I thought about out-of-banding kwargs as it’s own parameter separate from args but that has a penalty in places
<Who>
ah, I was going to do the same
<enebo>
Who: I might land this as a stop-gap solution until we have no more kwargs issues
<enebo>
mpapis: 1.7.19 might be this afternoon or tomorrow morning
<Who>
sure, makes sense.
<Who>
so that crosses off two more issues
<enebo>
Who: I also solved the parser issues in test_keywords.rb
<enebo>
the def foo a:\nblah;end
<enebo>
That syntax looks really odd to me too
<mpapis>
enebo, oh then it's good I forgot to release :) I promise to push version with yours
<enebo>
mpapis: yeah worst case tomorrow if I run into something. Most of my testing is done
ahadding1 has quit [Remote host closed the connection]
<enebo>
weird
<enebo>
Who: I am really surprised that is broken
<Who>
yeah, it seems like the extra argument in front is messing it up
<enebo>
Who: ah
<enebo>
Who: I have a clue for you on this
<Who>
that would be helpful
<Who>
I can fix issue before I sleep
<Who>
couldn’t do that yesterday :D
<enebo>
Who: Signature/Arity probably thinks that it has a single required argument and is not thinking about whether it is a kwargs signature
iamjarvo has joined #jruby
<enebo>
Who: if you change to |foo, **b| it will work but you will see **b is {}
<enebo>
Who: proc logic is likely stripping off what it thinks is extra arguments
<Who>
I am debugging it and this is what signature is saying (pre=2,opt=0,post=0,rest=NONE,kwargs=false)
<enebo>
Who: yeah I am not sure Signature has much in the way of kwargs smarts atm
shellac has joined #jruby
e_dub has joined #jruby
<enebo>
Who Signature.from(IterNode) is asking argsnode if it has kwargs
<Who>
that’s the correct behaviour?
sferik has joined #jruby
<enebo>
Who: well I would have though pre=1, kwargs=true
<Who>
getPreCount from argsNode must be returning 2
<enebo>
Who: yeah like b: is not keyword arg
<enebo>
Who: I will show you some magic: jruby -S ast -e 'call_block { |foo, b:| puts foo, b }'
<Who>
that’s pretty useful!
<Who>
so AST parsing is spot on
<Who>
I see b: under kwargs
<enebo>
Who: well yeah but it might be an error in ArgsNode still
<enebo>
Who: it is definitiely in it but perhaps the processing of this in the constructor is missing it
<enebo>
Who: The puts of this is not saying what the second arraynode rpresents
<enebo>
Who: but we can also add —ir to this
<enebo>
HAHAH but I broke —ir apparently
<Who>
yeah asking for 3 args instead of 1
mrmargolis has joined #jruby
<enebo>
hmmm I fixed that error but we have this fairly complicated dry run stuff to initialize an extra map
<enebo>
Who: You can also run -Xir.compiler.debug=true to see this live but you need to redirect to a file since you see all of kernel loading
<enebo>
Who: For really simple scripts like this you can add -Xir.compiler.debug=true and -Xdebug.parser=true (which as a side-effect will not load any of our .rb kernel files)
ivan\ has quit [Remote host closed the connection]
<headius>
hello hello
<headius>
I'm on shitty bar wifi but I thought I'd stop in
<Who>
hey headius
<headius>
enebo: bottle shop right next to my hotel
ivan\ has joined #jruby
<Who>
enebo: I printout the args constructor and it’s giving preSize as 1
<Who>
and kwargs as true
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] undefinedvalue opened issue #2538: BigDecimal does not coerce properly during division http://git.io/FKnA
JRubyGithub has left #jruby [#jruby]
<enebo>
Who: ah interesting
<enebo>
Who: perhaps something is replacing it
<enebo>
headius: nice
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] undefinedvalue opened issue #2539: Incorrect error message when BigDecimal fails to coerce during comparison http://git.io/FKlk
JRubyGithub has left #jruby [#jruby]
paulswilliamsesq has joined #jruby
calavera has joined #jruby
<Who>
enebo: found the codebase
<Who>
er
<Who>
found the issue
<Who>
prepareArgs in RubyProc
<Who>
IRubyObject[] newArgs = Arrays.copyOf(args, required); is discard kwargs
<enebo>
bbrowning: so somehow dump_with_mocks and dump_without_mocks is the same method
<enebo>
bbrowning: what weirds me out is this has something to do with rubygems? Could this just be a coincidence or was this the same error as last time?
ivan\ has quit [Read error: Connection reset by peer]
<bbrowning>
this was the same error I got last time RG was upgraded
subbu has joined #jruby
<bbrowning>
enebo: hey - if this file were getting loaded twice, would it cause the infinite loop?
<bbrowning>
headius said the issue with require loading files twice is a problem in 1.7 as well
<bbrowning>
well, the underlying issue is there but it hasn't really been exposed in 1.7
<enebo>
bbrowning: yeah it would
<enebo>
bbrowning: you could add a guard to prove it
<enebo>
bbrowning: set a value and put the aliases in a conditional
<enebo>
bbrowning: hell even a puts would work
ivan\ has joined #jruby
ivan\ has joined #jruby
ivan\ has quit [Changing host]
<bbrowning>
enebo: yep, I see it being called twice
<bbrowning>
and that also explains a lot of new uninitialized constant errors I'm seeing that I've ignored
<bbrowning>
stuff like [WARNING] /home/bbrowning/src/torquebox3/integration-tests/target/rubygems-test/gems/rspec-expectations-2.7.0/lib/rspec/expectations/version.rb:5 warning: already initialized constant STRING
<enebo>
bbrowning: ok well I guess we downgrade for another version
subbu has quit [Ping timeout: 246 seconds]
<enebo>
bbrowning: I am going to revert RG commit and it would be good to make sure we are cool after that
<bbrowning>
has anyone else tried 1_7 with real apps and everything's working fine, by chance?
rsim has quit [Quit: Leaving.]
<enebo>
bbrowning: I ran rails with it
<enebo>
bbrowning: made an app went through that whole process
<bbrowning>
huh
<enebo>
bbrowning: but perhaps the right mix of loading hits this and I am doing nothing which exercises that
<bbrowning>
only gems that do stuff w/ alias_method, undef_method, etc on load would cause the issue
<enebo>
bbrowning: but rake, rails, bundler, gems are all part of it
<enebo>
bbrowning: I run it on MacOS and Windows
<bbrowning>
activesupport 4.1.8 has the same issue on 9k at least, but only if you load it a certain way
<enebo>
bbrowning: but you would think I would see double constant warns
<enebo>
bbrowning: which is a good point
<enebo>
I am using 4.2
<enebo>
bbrowning: I am going to revert rg patch
<enebo>
bbrowning: if you are close to knocking off then just run some tests tomorrow
<bbrowning>
I can try it again today
noop has quit [Ping timeout: 276 seconds]
<Who>
enebo: I am running specs now
<Who>
I am adding +1 if kwargs are present
<Who>
while creating arity in signature
<enebo>
Who: but that will require one more or only one more if there happens to be one more?
yfeldblum has joined #jruby
pgokeeffe has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] enebo pushed 1 new commit to jruby-1_7: http://git.io/F6vZ
<JRubyGithub>
jruby/jruby-1_7 e835fd1 Thomas E. Enebo: Revert "Update to RubyGems 2.4.5."...
JRubyGithub has left #jruby [#jruby]
<enebo>
bbrowning: bamm
<Who>
enebo: I should add the size of kwargs right?
balo_ is now known as balo
<enebo>
Who: For required?
<Who>
yeah or it should be 1?
<enebo>
Who: required is +1 if there are required kwarg args but not if there is optional
<Who>
okay I will need to capture that one detail (if there are keywords basically)
<enebo>
not if there is only optional kwargs
<enebo>
because |a, b:1| would be required=1 but also be optional
<enebo>
because it was optional I think we pass it all in
<enebo>
I don’t remember if we differentiate on rest or restkwargs
<enebo>
if we do then optional may be +1 if no reqd kwarg but if there are opt kwargs
<enebo>
Who: did that make sense?
<Who>
ah
<Who>
we do
<Who>
restkwargs are keyrest
<Who>
so if |a, b:1| is there I should not do +1 right?
<Who>
otherwise it seems to be straightforward
<Who>
but not sure if there optional kwargs
<enebo>
Who for that required is not +1 but optional may be +1
<enebo>
ah so I think req + 1 is reqdkwargs; opt + 1 is no reqdkwargs but optkwargs
<enebo>
or perhaps we don’t care on optional
<enebo>
although there is some logic in checkArity just asking for kwargs
<enebo>
wow this method also sort of sucks because we basically convert to hash just to check arity?
<enebo>
perhaps MRI does that too
<Who>
probably need to check that
mitchellhenke has joined #jruby
<enebo>
In my mind this code feels wrong
<enebo>
like does an arity check really need to know if last argument is a hash to print this out
<enebo>
Sort of seems if we have required kwargs we know we need +1
<enebo>
I guess if we have optional kwargs and last arg is not a hash it should error but I think we will already error in prologue of the IR instrs binding values
pgokeeffe has joined #jruby
e_dub has quit [Quit: e_dub]
paulswilliamsesq has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
elia has joined #jruby
<bbrowning>
enebo: integs are running - already past the previous spot that gives the StackOverflowError
subbu has joined #jruby
drbobbeaty_ has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<enebo>
Actually I am confused. I only see checkArity called if it is a lambda
<enebo>
This code I don’t think is correct but I also don’t think it is making the proc provide wrong thing
<Who>
enebo: you mean to say required arity?
<Who>
I am almost done
<Who>
but it’s almost 3 AM, so I will take a look tomorrow before rasiing the PR
<enebo>
Who: yeah I can understand that. It is 3 pm here and my brain is already fried :)
<Who>
hehe
pglombardo has joined #jruby
<Who>
well I did finish it
<Who>
and it seems to be working
<Who>
running tests
mrmargolis has quit [Remote host closed the connection]
<bbrowning>
enebo: TB3 integs all passed w/ the RG version revert
<enebo>
bbrowning: thanks a ton. We will go with that for one more point release and I guess hopefully we can narrow down the double require
<Who>
enebo: I am just raising the PR, comment if you find anything a miss (you might find lot of it :P), I will check back tomorrow
<bbrowning>
yeah that requires a lot of loading logic change, apprently?
<enebo>
Who: ok will look at it
skade has joined #jruby
<Who>
good night
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] Who828 opened pull request #2540: Fixed arity require in case of required keyword arguments (master...2518_bugfix) http://git.io/F6RX
JRubyGithub has left #jruby [#jruby]
Who has quit [Quit: Who]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] enebo pushed 1 new commit to jruby-1_7: http://git.io/F6gs
<JRubyGithub>
jruby/jruby-1_7 8dfa73a Thomas E. Enebo: Stop using deprecated lines method
JRubyGithub has left #jruby [#jruby]
dinfuehr_ has quit [Remote host closed the connection]
Aethenelle has quit [Quit: Aethenelle]
slyphon has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
spectre256 has quit [Quit: leaving]
rsim has quit [Quit: Leaving.]
ivan\ has quit [Read error: Connection reset by peer]
ivan\ has joined #jruby
baroquebobcat has quit [Quit: baroquebobcat]
baroquebobcat has joined #jruby
sferik has quit [Read error: Connection reset by peer]
sferik has joined #jruby
rsim has joined #jruby
tenderlove has quit [Quit: Leaving...]
_gautam_ has joined #jruby
tenderlove has joined #jruby
tcrawley is now known as tcrawley-away
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] nirvdrum pushed 1 new commit to master: http://git.io/F66O
<JRubyGithub>
jruby/master 995aa83 Kevin Menard: [Truffle] Tagged all language specs failing on Windows.