slyphon has joined #jruby
KeyJoo has joined #jruby
KeyJoo has quit [Ping timeout: 246 seconds]
KeyJoo has joined #jruby
KeyJoo has quit [Ping timeout: 246 seconds]
KeyJoo has joined #jruby
shellac has joined #jruby
shellac has quit [Quit: Computer has gone to sleep.]
KeyJoo has quit [Ping timeout: 250 seconds]
shellac has joined #jruby
KeyJoo has joined #jruby
Puffball has quit [Remote host closed the connection]
Puffball has joined #jruby
jmalves has joined #jruby
shellac_ has joined #jruby
shellac has quit [Ping timeout: 250 seconds]
drbobbeaty has quit [Ping timeout: 268 seconds]
KeyJoo has quit [Ping timeout: 272 seconds]
KeyJoo has joined #jruby
drbobbeaty has joined #jruby
shellac_ has quit [Quit: Computer has gone to sleep.]
KeyJoo has quit [Ping timeout: 250 seconds]
KeyJoo has joined #jruby
shellac has joined #jruby
shellac has quit [Quit: Computer has gone to sleep.]
KeyJoo has quit [Ping timeout: 268 seconds]
KeyJoo has joined #jruby
shellac has joined #jruby
travis-ci has joined #jruby
<travis-ci> jruby/jruby (ruby-2.6:da80382 by Thomas E Enebo): The build is still failing. (https://travis-ci.org/jruby/jruby/builds/466028972)
travis-ci has left #jruby [#jruby]
xardion has quit [Remote host closed the connection]
xardion has joined #jruby
Puffball has quit [Remote host closed the connection]
kitallis has joined #jruby
shellac has quit [Ping timeout: 252 seconds]
kitallis has quit [Quit: going, bye]
shellac has joined #jruby
shellac has quit [Quit: Computer has gone to sleep.]
<headius> hmmm
<headius> so I'm going to get stdlib updated for 9.2.6
<headius> trying to decide if I should pull HEAD from the MRI 2.5 branch or just align with current release 2.5.3
travis-ci has joined #jruby
<travis-ci> kares/jruby (method-missing-2:6503c28 by kares): The build has errored. (https://travis-ci.org/kares/jruby/builds/466108012)
travis-ci has left #jruby [#jruby]
<headius> oh good, no stdlib updates for 2.5.4 yet anyway
<headius> no worries
shellac has joined #jruby
travis-ci has joined #jruby
<travis-ci> kares/jruby (method-missing-2:6503c28 by kares): The build passed. (https://travis-ci.org/kares/jruby/builds/466108012)
travis-ci has left #jruby [#jruby]
KeyJoo has quit [Ping timeout: 250 seconds]
travis-ci has joined #jruby
<travis-ci> jruby/jruby (master:e2441d6 by Charles Oliver Nutter): The build was fixed. (https://travis-ci.org/jruby/jruby/builds/466134476)
travis-ci has left #jruby [#jruby]
subbu is now known as subbu|lunch
shellac has quit [Quit: Computer has gone to sleep.]
<headius> enebo: hey
<headius> so one of the failures is testing that ARGV values are tainted
<headius> apparently MRI wasn't doing it on windows so they fixed that and added this test
<headius> we don't do it at all, so we fail now
<headius> it's an easy fix either way, but I'm unsure whether to just fix ARGV or to fix all "external" strings
<headius> in MRI this method they use to turn a C string into a Ruby string always tains
<headius> taints
subbu|lunch is now known as subbu
<headius> that's roughly equivalent to the logic we use here for ARGV, newInternalFromJavaExternal
<headius> so I'm on the fence about whether to make all Java strings coming through that method get tainted
<enebo> hmm
<enebo> ARGV is less risky
<enebo> but you may fix more the other way
<enebo> I guess audit existing callers and see if we have obvious calls which do not expect taint
<enebo> ./lib/ruby/gems/shared/gems/activerecord-jdbc-adapter-51.1-java/src/java/arjdbc/jdbc/RubyJdbcConnection.java: return RubyString.newInternalFromJavaExternal(runtime, xml.getString());
<enebo> headius: so that may make you a little more nervous
<enebo> of course we can stop using that method and push a new point but any audit maybe should include a github check
<headius> well maybe it does
<headius> but if the native AR adapters are calling this function they're getting tainting too
<enebo> "we ... stop" == arjdbc
<enebo> yeah
<headius> you know we really don't do anything at all with taint so it probably doesn't matter either way
<enebo> I have not looked at what this string is for... "xml.getString()"
<enebo> we mark taint but I am not sure how complete we are
<headius> complete enough to pass tests and no more complete
<enebo> lol
<headius> yeah I'll just go with argv fix for now
travis-ci has joined #jruby
<travis-ci> nomadium/jruby (master:cfe635f by Charles Oliver Nutter): The build was fixed. (https://travis-ci.org/nomadium/jruby/builds/466156273)
travis-ci has left #jruby [#jruby]
<headius> if someone wants to raise an issue about a specific case of java to Ruby string that ought to be tainted, they can file it
<headius> I thought we had IRC notifiications set up to only come from our repo
<headius> I guess kares results have been showing up too
<enebo> headius: interesting discussion on taint if we actually cared about taint...we cannot really know from a Java call if we got the string from a trusted source
<headius> right
<headius> it's the C of JRuby so in theory we should be tainting
<enebo> We may be able to say for certain in limited ways we directly call some apis but generically they should all be tainted since we cannot know
<headius> but many of the reasons MRI taints don't apply to us (buffer overruns etc)
<enebo> a large part of me believes no one uses JRuby + tainting other than not wanting to see non-green runs on random gems they consume
<headius> what are the other reasons for tainting?
<headius> I mean I've never understood the concept, it seems comically stupid
<headius> so you're going to tell me that this string is evil...riiiiiight
<enebo> yeah it is such a fragile concept from end use and implementation
<headius> MRI basically doesn't do anything with it now either since they don't have safe levels
<headius> so I dunno
<headius> I assume there's some checks for using taint to launch subprocesses or whatever
<enebo> taint is so easy to get wrong I feel like it gives people an improper sense of security
<enebo> but my experience with taint was all with Perl so you know that was a long time ago
travis-ci has joined #jruby
<travis-ci> nomadium/jruby (ruby-2.6:da80382 by Thomas E Enebo): The build is still failing. (https://travis-ci.org/nomadium/jruby/builds/466156574)
travis-ci has left #jruby [#jruby]
<headius> enebo: I don't think anyone else has tried to use tainting as a security mechanism since then
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<headius> kares: lopex: if you guys want something small to work on there's new failures from updating MRI tests
<headius> enebo is also working on fixes
<lopex> doh
<lopex> new new failures or new specs ?
<lopex> er, I thought it's something new from stdlib update
shellac has joined #jruby
<headius> lopex: there were a couple stdlib things but I fixed those...the rest of these are just from updating the tests to 2.5 HEAD
<headius> so they're actual bugs that were fixed in core