ur5us has quit [Ping timeout: 260 seconds]
ur5us has joined #jruby
ur5us has quit [Ping timeout: 250 seconds]
_whitelogger has joined #jruby
_whitelogger has joined #jruby
travis-ci has joined #jruby
travis-ci has left #jruby [#jruby]
<travis-ci> jruby/jruby (master:e26ac02 by Karol Bucek): The build is still failing. https://travis-ci.org/jruby/jruby/builds/671952252 [171 min 23 sec]
travis-ci has joined #jruby
<travis-ci> jruby/jruby (master:e26ac02 by Karol Bucek): The build is still failing. https://travis-ci.org/jruby/jruby/builds/671952252 [171 min 23 sec]
travis-ci has left #jruby [#jruby]
drbobbeaty has quit [Ping timeout: 272 seconds]
<headius[m]> Hmm failing
claudiuinberlin has joined #jruby
ur5us has joined #jruby
shellac has joined #jruby
ur5us has quit [Ping timeout: 260 seconds]
claudiuinberlin has quit [Ping timeout: 265 seconds]
ur5us has joined #jruby
ur5us has quit [Ping timeout: 260 seconds]
travis-ci has joined #jruby
<travis-ci> jruby/jruby (master:e41e140 by Charles Oliver Nutter): The build is still failing. https://travis-ci.org/jruby/jruby/builds/672065817 [174 min 21 sec]
travis-ci has left #jruby [#jruby]
travis-ci has joined #jruby
<travis-ci> jruby/jruby (master:e7135b8 by Charles Oliver Nutter): The build is still failing. https://travis-ci.org/jruby/jruby/builds/672072634 [168 min 43 sec]
travis-ci has left #jruby [#jruby]
rwilliams[m] has left #jruby ["Kicked by @appservice-irc:matrix.org : Idle for 30+ days"]
rtyler1 has left #jruby ["Kicked by @appservice-irc:matrix.org : Idle for 30+ days"]
xardion has quit [Remote host closed the connection]
xardion has joined #jruby
victori has quit [Quit: ZNC 1.7.5 - https://znc.in]
victori has joined #jruby
shellac has quit [Ping timeout: 258 seconds]
subbu is now known as subbu|lunch
<headius[m]> fzakaria: around at all today?
<headius[m]> lopex: I will need to release joni to keep our maven dependencies aligned
<headius[m]> but you were still working on sjis
ur5us has joined #jruby
<headius[m]> ah well there's plenty of numbers, I'll just release joni
<lopex> headius[m]: not complete, but this should fix that failure
<lopex> unless there's others
<headius[m]> is it ready for a release?
<headius[m]> I'll do both if so
<headius[m]> I will test it in jruby
<lopex> yeah
<lopex> though only caseMap is used so it sohuld be ok
<lopex> joni has updated grapheme cluster bits for new enums
<headius[m]> ok
<headius[m]> oh lopex we are so close
<lopex> what encodings ?
<headius[m]> looks like euc-jp
<headius[m]> "\u{391}" and "\u{ff21}" with encode("euc-jp")
<headius[m]> enebo: I added that test about converging dependencies but this will be annoying
<enebo[m]> you mean maven deps
<enebo[m]> yeah it is annoying but I suppose having something tell us we are out of date is better
<enebo[m]> porting symbol type stuff and I will plug it into parser+
<headius[m]> oooo neat
<headius[m]> yeah we want this test but I wish there were a softer way for it to fail
<headius[m]> maybe we'll just get used to it
<enebo[m]> we need some build tool which knkows jffi needs to rev all up deps
<enebo[m]> It just versions and rebuilds
<headius[m]> I am tidying up some back PRs which will eventually include rebasing my load service work
<enebo[m]> half surprised someone has not made some weird maven plugin for this
<headius[m]> I think the main annoyance about this is that we don't check it locally
<headius[m]> so it will almost always fall out of CI some time after we actually update a library
<enebo[m]> can you forward reference break in Java...I don't think so
<headius[m]> eh
<headius[m]> what do you mean
<enebo[m]> in a loop/switch/whatever break foo where foo is further forward and outside the loop
<enebo[m]> I think it needs to be contained
<enebo[m]> break id;
<headius[m]> yeah can't break to a loop you're not in
<enebo[m]> yeah otherwise it would be a goto
<enebo[m]> I know this so I don't know why I said it
<headius[m]> wouldn't that make C porting so much easier
<enebo[m]> It just looks ugly
<enebo[m]> I think I will break and just use a boolean instead
<headius[m]> we should start using a java variant that supports goto
<enebo[m]> JRubySCala
subbu|lunch is now known as subbu
<headius[m]> lopex: find more code to be ported?
<headius[m]> lopex: releases are cheap and I want to move on to rebase some branches on a green master so I'll just spin a couple releases for now
<headius[m]> we can spin another one if you find euc-jp fixes
<lopex> headius[m]: yeah euc jp has the same impl
<headius[m]> lopex: should I hold off for a bit?
<lopex> headius[m]: yeah
<headius[m]> 👍
<lopex> headius[m]: pushed
<headius[m]> cool
<headius[m]> lopex: testing on master now and then I'll spin releases... thank you
<headius[m]> lopex: it works!
travis-ci has joined #jruby
travis-ci has left #jruby [#jruby]
<travis-ci> jruby/jruby (move_to_gh_actions:13e85a5 by Charles Oliver Nutter): The build failed. https://travis-ci.org/jruby/jruby/builds/672240018 [176 min 41 sec]
<lopex> headius[m]: cool!
<lopex> headius[m]: mri has noop there https://github.com/ruby/ruby/blob/master/enc/euc_jp.c#L384
travis-ci has joined #jruby
travis-ci has left #jruby [#jruby]
<travis-ci> jruby/jruby (move_to_gh_actions:e69489c by Charles Oliver Nutter): The build failed. https://travis-ci.org/jruby/jruby/builds/672242075 [149 min 24 sec]
<headius[m]> lopex: releases out and master updated
travis-ci has joined #jruby
travis-ci has left #jruby [#jruby]
<travis-ci> jruby/jruby (master:546a0f6 by Charles Oliver Nutter): The build has errored. https://travis-ci.org/jruby/jruby/builds/672255726 [178 min 38 sec]
travis-ci has joined #jruby
<travis-ci> jruby/jruby (move_to_gh_actions:69ea545 by Charles Oliver Nutter): The build is still failing. https://travis-ci.org/jruby/jruby/builds/672260329 [154 min 23 sec]
travis-ci has left #jruby [#jruby]
lucasb has joined #jruby
<byteit101[m]> I think I'm done with the major work of concrete java extension annotations. Only major thing left is to decide which api(s) to implement for field annotations as per my PR
travis-ci has joined #jruby
<travis-ci> jruby/jruby (move_to_gh_actions:73377d6 by Charles Oliver Nutter): The build failed. https://travis-ci.org/jruby/jruby/builds/672265876 [145 min 1 sec]
travis-ci has left #jruby [#jruby]
<headius[m]> lopex: I might be seeing a hang in an euc-jp test
<lopex> headius[m]: I'm suspicious about this code https://github.com/ruby/ruby/blob/master/enc/euc_jp.c#L387
<headius[m]> jruby test/mri/runner.rb --excludes=test/mri/excludes/ test/mri/ruby/enc/test_euc_jp.rb
<lopex> set len to 1 and then inscrease ?
<headius[m]> I can't even get that to start running
<headius[m]> it sits at 100% and never runs the test
<lopex> where does it hang ?
<headius[m]> kr works ok so it's not the suite
<headius[m]> I'll get a jit trace
<headius[m]> it's not in any Ruby code because this is hanging at parse time
<lopex> oh, so it's a regexp thing ?
<lopex> yeah, that trace is not very helpful
<lopex> we miss some thing from euc-jp and sjis
<lopex> co that bits might cause that
<lopex> any chance to reduce that ?
travis-ci has joined #jruby
<travis-ci> jruby/jruby (move_to_gh_actions:4bfc7a3 by Charles Oliver Nutter): The build failed. https://travis-ci.org/jruby/jruby/builds/672281121 [145 min 31 sec]
travis-ci has left #jruby [#jruby]
<headius[m]> probably
<headius[m]> that file is small
<lopex> or it might be that petty broken char thing
<headius[m]> oh hmm I'm on branch... hold please this may be fixed with jcodings update
<headius[m]> ok it is not
<lopex> the new release has if (len == ErrorCodes.ERR_INVALID_CODE_POINT_VALUE) len = 1;
<headius[m]> ```
<headius[m]> # vim: set fileencoding=euc-jp
<headius[m]> p /(ａ)(a)\1\2/i
<headius[m]> that file is sufficient to hang
<headius[m]> in case formatting screws it up: https://gist.github.com/headius/646ded7437808b1c05624b0b88a23882
<headius[m]> note one of those "a" is a special "a"
<lopex> yeah, I know
<lopex> well, I can even see that
<headius[m]> ok
<lopex> what's java equivalent of euc-jp ?
<lopex> for getBytes
<lopex> aah
<lopex> I just copied taht via utf-8 from here
<headius[m]> $ jruby -e 'p org.jcodings.specific.EUCJPEncoding::INSTANCE.charset'
<headius[m]> #<Java::SunNioCsExt::EUC_JP:0x58a90037>
<headius[m]> I guess "EUC_JP" for charset will do it
<lopex> I guess i need those bytes ?
<lopex> headius[m]: whoat I mean are they transcoded properly to utf-8 here ?
<lopex> doh I can lok at the tests themselves
<lopex> er, what test is that ?
<headius[m]> test/mri/ruby/enc/test_euc_jp.rb
<lopex> ah, I missed that somehow
<lopex> and it's not a regression ?
<headius[m]> I think it is
<headius[m]> no excludes changed for this file
<headius[m]> I will merge this GH actions PR shortly and confirm it's still failing on master
<headius[m]> FYI @kare
<headius[m]> oops
<lopex> headius[m]: doesnt hang on regexp parse for me
<lopex> byte[]pattern = new byte[] {40, (byte)194, (byte)163, (byte)195, (byte)161, 41, 40, 97, 41, 92, 49, 92, 50};
<lopex> Regex regex = new Regex(pattern, 0, pattern.length, Option.NONE, EUCJPEncoding.INSTANCE, Syntax.RUBY);
<headius[m]> hmm
<lopex> the bytes are from String#bytes
<lopex> # coding: euc-jp
<lopex> from "(£á)(a)\\1\\2"
<travis-ci> jruby/jruby (move_to_gh_actions:585183d by Charles Oliver Nutter): The build has errored. https://travis-ci.org/jruby/jruby/builds/672293358 [161 min 6 sec]
travis-ci has joined #jruby
travis-ci has left #jruby [#jruby]