claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
vtunka has quit [Quit: Leaving]
claudiuinberlin has joined #jruby
shellac has joined #jruby
drbobbeaty has joined #jruby
vtunka has joined #jruby
<GitHub32>
[jruby] iaddict opened pull request #4815: Fix spec of Thread#status for an externally killed thread that sleeps (master...fix_sleeping_killed_thread_status) https://git.io/vdKQr
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
drbobbeaty has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
claudiuinberlin has joined #jruby
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
claudiuinberlin has joined #jruby
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<kares>
enebo: AR definitely didn't have threading issues - at least no user reports/complains in a long time :)
<kares>
not reading through all the comm - you guys seem to be on the right track with Rob
<kares>
Postgre little gems (special handling) were all over the adapter .rb pieces - you can not just copy-pasta AR over
<kares>
execute_query_raw was introduced when execute_query started returning AR::Result (pbly way back to 3.0/3.1 or so)
<kares>
but the raw might still be a base for the "raw" connection.execute which if not used by API we still use internally
<kares>
regarding jdbc drivers - they're being updated/released from master (often tested on jdbc-xx_1-3 branches).
<kares>
its a bit complicated - waiting on latest jdbc-postgres release since for users to upgrade to latest 1.3.x (otherwise they will break)
<kares>
same (actually worse - a fix is only on 1-3-stable - not released yet) for jdbc-mariadb 2.x ;(
<kares>
the driver situation turned out unfortunate - somehow expected master (1.4 performance/cleanup work) to be released by now, but I never had proper time to finish it so it ended up what it is
<GitHub42>
[jruby-openssl] kares pushed 1 new commit to master: https://git.io/vd6vP
<GitHub42>
jruby-openssl/master 617ca56 roadrunner2: Fix loading of Subject/Issuer-Alt-Name extensions. (#144)...
clayton has joined #jruby
rdubya has joined #jruby
<rdubya>
enebo: this pr may fix the specs I mentioned yesterday about the transactions not rolling back when a thread is killed: https://github.com/jruby/jruby/pull/4815
drbobbeaty has joined #jruby
claudiuinberlin has joined #jruby
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
claudiuinberlin has joined #jruby
vtunka has quit [Quit: Leaving]
vtunka has joined #jruby
lance|afk is now known as lanceball
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
claudiuinberlin has joined #jruby
<enebo>
rdubya: coolio. I think it is ok but will asked Charlie to review...any changes to threading are touchy
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<enebo>
kares: re _raw I will be tracing through how invididual pieces are used by the main three adapters
<enebo>
kares: I am just looking towards possibility we can have one less code path. If we can't we can't and that is ok but I am looking around.
<enebo>
kares: This will be a new major number so we can kill things like generic adapter support. It was a good idea but I don't know if anyone even uses it now
claudiuinberlin has joined #jruby
claudiuinberlin has quit [Client Quit]
claudiuinberlin has joined #jruby
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
vtunka has quit [Quit: Leaving]
claudiuinberlin has joined #jruby
vtunka has joined #jruby
vtunka has quit [Quit: Leaving]
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<rdubya>
is there a way to pass a java object to a ruby method? or a way to wrap a java object as an IRubyObject? `callMethod` expects IRubyObjects to be passed for the arguments and I'm not sure the preferred way to work around that
<enebo>
rdubya: if you pass a Java object which cannot be coerced to a Ruby one it will just make a proxy object which wraps the Java one
<rdubya>
cool, thanks, I'll see if I can make headway with that
shellac has quit [Quit: Leaving]
claudiuinberlin has joined #jruby
claudiuinberlin has quit [Client Quit]
<enebo>
rdubya: @connection.execute_query_raw(sql, nil) do |*values|
<enebo>
I have been trying to understand the arguments to execute_query_{,raw}
<enebo>
abstract passes in [] as a second param for binds but not bound values and pg will pass in nil
<enebo>
I am thinking you should just remove that param
<rdubya>
enebo: where is that?
<enebo>
in query() in adapter.rb
<enebo>
in auditing all callers of this I only found two which have weird values
<enebo>
nil and []
<enebo>
I am guessing one difference is that by passing that in it will go down binds/prepared statement path
<enebo>
I am not sure if that is important but I want to define the API of this to not allow nil values to simplify things a tiny bit
<enebo>
So I guess passing in [] instead of nil will keep same functionality but allow me to document this with [] for binds or Integer for max_rows but nil for neither
<enebo>
or at least I would like to fully document this and would like to kill some possible values
<rdubya>
i'm all for simplifications
<GitHub194>
[jruby] headius pushed 1 new commit to jruby-9.1: https://git.io/vd6Fz
<GitHub194>
jruby/jruby-9.1 2781e22 Charles Oliver Nutter: Fixes to reduce access warnings on Java 9....
<enebo>
rdubya: I will change that in this changeset
<headius>
enebo: I'm working on getting a 9.1 that works mostly right on JDK9
<headius>
if you get a chance to review that we can discuss some of the changes
<headius>
our trouble spots mostly lie in digging out hidden fields like file descriptors...there may not be a way to recover the functionality we lose without making many remaining streams also native
<headius>
I think for sockets which are already fully selectable we can continue using what we have
<headius>
windows native process/IO stuff is going to be a bigger priority now
<enebo>
headius: can you review that PR on thread which just came in
<headius>
sure
<GitHub106>
[jruby] headius closed issue #4705: Thread.current.status is wrong for killed thread https://git.io/vQuI5
<GitHub95>
[jruby] headius closed pull request #4815: Fix spec of Thread#status for an externally killed thread that sleeps (master...fix_sleeping_killed_thread_status) https://git.io/vdKQr
<GitHub51>
jruby/master 6dd82b3 Charles Oliver Nutter: Merge pull request #4815 from iaddict/fix_sleeping_killed_thread_status...
<GitHub51>
jruby/master bacd973 Thomas Steinhausen: Fix failing spec on Thread#status for an externally killed thread that sleeps (Fixes #4705)
tartal has joined #jruby
<GitHub131>
[jruby] headius pushed 2 new commits to master: https://git.io/vd6AZ
<GitHub131>
jruby/master 404fdc9 Charles Oliver Nutter: Fixes to reduce access warnings on Java 9....
<GitHub131>
jruby/master 471b9c9 Charles Oliver Nutter: Update to latest jnr-ffi and jnr-posix.
<GitHub107>
[jruby] headius pushed 1 new commit to jruby-9.1: https://git.io/vd6AG
<GitHub107>
jruby/jruby-9.1 ffad568 Charles Oliver Nutter: Update to latest jnr-ffi and jnr-posix.
<headius>
enebo: seemed fine...those methods are a bit racy but *should* only be called from the thread being updated
<enebo>
headius: this fixes some racy test in AR
<enebo>
It may still have a race (I did not evaluate it) but this was someones conclusion from looking into it
<enebo>
Expected: ["EUC-JP"]
<enebo>
Actual: ["UTF-8"]
<enebo>
yay
tartal has quit [Ping timeout: 248 seconds]
<headius>
lovely
<GitHub181>
[jruby] headius created 9.1_2.3.5 (+2 new commits): https://git.io/vdiv1
<GitHub181>
jruby/9.1_2.3.5 d6e561d Charles Oliver Nutter: Update stdlib to match MRI 2.3.5 plus our patches.
<GitHub181>
jruby/9.1_2.3.5 a6fca15 Charles Oliver Nutter: Update tests from Ruby 2.3.5.
subbu is now known as subbu|lunch
<GitHub154>
[jruby] headius pushed 1 new commit to 9.1_2.3.5: https://git.io/vdikW
<GitHub154>
jruby/9.1_2.3.5 9f6f054 Charles Oliver Nutter: Update forwardable from MRI trunk....
<headius>
I look forward to this stuff being in gems finally
claudiuinberlin has joined #jruby
claudiuinberlin has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
subbu|lunch is now known as subbu
claudiuinberlin has joined #jruby
<GitHub13>
[jruby-openssl] kares pushed 1 new commit to master: https://git.io/vdiGx
<GitHub13>
jruby-openssl/master ec307fa Alex Coomans: normalize all constants in CipherStrings as public (#146)
<GitHub63>
[jruby-openssl] kares closed pull request #146: Normalize all constants in CipherStrings as public final static (master...alexc-public) https://git.io/vdoak