<enebo>
eregon: if you forget it will just be an error again on next sync so easily caught
<SynrG>
(note: i also tested without --1.8 in the command definition; same result)
<eregon>
ok :)
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] enebo pushed 1 new commit to master: http://git.io/vkHZA
<JRubyGithub>
jruby/master 95a9d88 Thomas E. Enebo: Module#class_variables should accept an inherit flag. Tests from MRI assume an order from class_variables (@eregon said he will patch upstream so we don't lose this change on next sync)
JRubyGithub has left #jruby [#jruby]
colinsurprenant has quit [Quit: colinsurprenant]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 3 new commits to master: http://git.io/vkHcv
<JRubyGithub>
jruby/master d32a698 Benoit Daloze: Split the Complex#hash spec into smaller examples.
<JRubyGithub>
jruby/master fe89353 Benoit Daloze: Add tags for some obvious failing specs.
JRubyGithub has left #jruby [#jruby]
<eregon>
enebo: I added a few tags for the new specs, but I left a few other failures undecided, please have a look and judge if worth fixing soon or just tag
<enebo>
eregon: ok
mcclurmc has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 1 new commit to master: http://git.io/vkHc1
<SynrG>
btw, it's not entirely hopeless about a recent java eventually being made for VMS. there seems to be strong community demand for java8, and it has been discussed at a recent OpenVMS Breakfast held jointly by HP and VSI (the company HP is handing off support for VMS to)
<SynrG>
so my purpose in asking about jruby 9000 is to be looking ahead a little bit to if-and-when we might later switch.
mrmargolis has quit [Remote host closed the connection]
<chrisseaton>
SynrG: it's definitely impossible to build the full version of JRuby 9k on Java 6 - you might be able to do it if you remove some functionality and patch some things yourself
PragTob has quit [Quit: Leaving]
vtunka has joined #jruby
<SynrG>
chrisseaton: ok. definitely not worth any effort by me, then, since at this stage i'm only evaluating possible future options, having no driving need to move forward.
<chrisseaton>
Can you not just build a recent OpenJDK on OpenVMS? It's just POSIX isn't it? Are there significant issues?
sdogruyol has quit [Remote host closed the connection]
<SynrG>
Considering the significant community interest, if it were that easy, I'd already know about it, as it would be very big news indeed.
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] bjfish pushed 1 new commit to master: http://git.io/vkHoM
<JRubyGithub>
jruby/master 3460e45 Brandon Fish: [Truffle] Add test_cgi_session to MRI test index
JRubyGithub has left #jruby [#jruby]
bzb has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] subbuss pushed 1 new commit to master: http://git.io/vkH6Z
<JRubyGithub>
jruby/master b665fa1 Subramanya Sastry: Run ensure blocks for redos as well
JRubyGithub has left #jruby [#jruby]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] enebo pushed 1 new commit to master: http://git.io/vkH62
<JRubyGithub>
jruby/master 325487f Thomas E. Enebo: Tag out failing spec
JRubyGithub has left #jruby [#jruby]
camlow325 has joined #jruby
<chrisseaton>
SynrG: what do you like about OpenVMS? Are there kernel features you can't get on for example Linux?
bzb has quit [Client Quit]
<SynrG>
chrisseaton: well, it pays the bills :)
<chrisseaton>
SynrG: but why do the companies want to run it? or is it legacy?
<SynrG>
chrisseaton: i personally consider it legacy. there are others who make a more compelling case for it.
djellemah has quit [Ping timeout: 265 seconds]
<SynrG>
chrisseaton: as for our company, our flagship product runs on it. i was hired in '85 to help with the port from RSTS/E
baroquebobcat has joined #jruby
<SynrG>
chrisseaton: that we are still alive today and still have clients on this is somewhat of a minor miracle
<SynrG>
i think this can be largely credited to the company being privately owned, diversified into other more money-making ventures, and it would likely be a bad move businesswise to suddenly drop support for our remaining clients and move on to something else
<SynrG>
meanwhile, having JRuby means at least we can build software for it that could run elsewhere.
<chrisseaton>
SynrG: but what I mean is what does OpenVMS actually give you - is it different system calls or something that wouldn't be available on Linux? I'm never sure what locks people into a platform like that?
<SynrG>
what locks people into any legacy platform? a huge codebase with platform-specific code, i suspect
phlebas has quit [Ping timeout: 244 seconds]
<SynrG>
chrisseaton: it's not from a lack of trying that we're still on it. one of the more significant barriers is that having ported most of the old VMS BASIC codebase to C++ which is, in theory, portable, we still had no solution for the sizable chunk of DCL-based procedures, nor the batch queue oriented architecture of the jobs running within this system
rcvalle has joined #jruby
<SynrG>
chrisseaton: by the time we discovered that (which we had assumed we could later address) with a shrinking market for the kind of product we have, it was too late to invest the additional effort in that part of the port.
pjammer has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<SynrG>
anyway, i'm sure my tale of woe is uninteresting to most people here. but hope that answers your questions. :)
<SynrG>
there may be people who could wax eloquent about the benefits of VMS for them, and how it continues to be relevant in a modern age. I'm not one of them.
<JRubyGithub>
[jruby] enebo pushed 1 new commit to master: http://git.io/vkHAN
JRubyGithub has left #jruby [#jruby]
<JRubyGithub>
jruby/master dec4aca Thomas E. Enebo: Tag out lambda spec
<subbu>
chrisseaton, SynrG "System up-times of a decade or more[23] have been reported, and features such as Rolling Upgrades and clustering allow clustered applications and data to remain continuously accessible while operating system software and hardware maintenance and upgrades are performed, or when a whole data center is destroyed. Customers using OpenVMS include banks and financial services, hospitals and healthcare, network information services,
<subbu>
and large-scale industrial manufacturers of various products."
<SynrG>
headius: i don't know if "the community is quite interested in Java 8 on OpenVMS" will translate to "so VSI will put effort into making it a reality", but it's what we're hoping.
<nirvdrum>
Then the Encoding section starts off with "Regular expressions are assumed to use the source encoding." But that's not true either. If I pass in a UTF-8 String, the encoding is US-ASCII unless otherwise overridden.
<nirvdrum>
Maybe "source encoding" means something else, but the default encoding for source files is UTF-8. And my default* encodings are UTF-8. And my locale encoding is UTF-8.
<nirvdrum>
The ASCII part seems to be coming from kcode, which I thought was largely deprecated.
<nirvdrum>
Suffice to say, I don't understand any of this :-P
<headius>
nirvdrum: many cases will reduce UTF_8 to US-ASCII for simpler logic, if the characters all in 7-bit ASCII range
<nirvdrum>
Yeah. It's just strange to me that with String that's handled implicitly with code range and single byte optimizeable helpers, but with Regexp, the encoding is explicitly changed.
pitr-ch has joined #jruby
colinsurprenant has joined #jruby
<eregon>
enebo: done! Sorry for the typo path => patch
<lopex>
nirvdrum: isnt that the encoding is kind of lost since only buffer is being kept ?
sdogruyol has joined #jruby
<eregon>
enebo: wondering how you get cvars in reverse order of the module chain though :)
<nirvdrum>
lopex: The source is still kept, I believe.
sdogruyol has quit [Ping timeout: 265 seconds]
<lopex>
nirvdrum: as a bytelist, but yeah
<nirvdrum>
lopex: But the ByteList has the encoding still.
<nirvdrum>
lopex: So what are the rules for regexp encoding?
<lopex>
nirvdrum: afaik very complex :)
<lopex>
nirvdrum: afair there's some preprocessing done which can end up different encoding than the pattern
<lopex>
(string pattern)
<lopex>
nirvdrum: that's for dregexps
<lopex>
nirvdrum: also if the regexp literal is only acess it will be always ascii in spite of all encodings
<lopex>
so oyu have to force it via //u
<lopex>
nirvdrum: but I guess those are not new for you
yfeldblum has joined #jruby
<lopex>
s/acess/ascii/ doh
<nirvdrum>
I'm trying to follow RubyRegexp, which looks like it was a port of MRI. It's very hard to trace.
<lopex>
I guess preprocessDRegexp is one thing
pjammer has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
subbu is now known as subbu|lunch
<lopex>
and preprocess
<lopex>
nirvdrum: you want to reuse it ?
<headius>
hmm, my change fails one test
pjammer has joined #jruby
<nirvdrum>
lopex: I'd love to. The Truffle one is quite broken with regards to encodings.
<headius>
nirvdrum: where did the truffle one come from?
<nirvdrum>
I think chrisseaton just put something together lightly based on joni before the projects merged. It's been patched here and there since, but hasn't really been fully reexamined.
<nirvdrum>
I was hoping to share more like we did with String.
<nirvdrum>
But the division here looks a lot more complicated, even if the API surface area is smaller.
<headius>
ahh ok
<headius>
the logic for negotiating encodings in regexp has not had a re-port in a long time, so it's a bit hacked together
<nirvdrum>
And some of it seems to be in RegexpOptions, some in joni, some in RubyRegexp. I'm just trying to get a clear mental model at the moment.
<nirvdrum>
I've had a few false starts already.
<nirvdrum>
So I turned to the docs, but they seem to be out of date.
pitr-ch has quit [Ping timeout: 244 seconds]
x1337807x has joined #jruby
<enebo>
eregon: sorry was eating
<eregon>
enebo: sure, np :)
<eregon>
nirvdrum: are there any tests/specs which would describe the behavior?
<enebo>
eregon: is there some required ordering?
projectodd-ci has quit []
<enebo>
eregon: I was wondering but it seems this should really be a set and not a dumped array
<eregon>
enebo: I don't think so, but I would just expect to be in the order of ancestors somehow
<nirvdrum>
eregon: There are specs, but they're not very helpful I don't think.
projectodd-ci has joined #jruby
<eregon>
No ordering makes sense for cvars of a came class, but if one cvar is from one class and the other from the parent I suppose there could be some expectations
<nirvdrum>
it "ignores the third argument if it is 'e' or 'euc' (case-insensitive)"
projectodd-ci has quit [Client Quit]
<eregon>
nirvdrum: ahah, hopefully there are a few which are not about ignoring
<enebo>
eregon: also remember that two modules can have the same one defined and since this is set…
<eregon>
of a same* class
projectodd-ci has joined #jruby
<eregon>
enebo: right, I think the test fix is good in any case
<nirvdrum>
eregon: It just presupposes you both know how it's supposed to work and have something already working.
<eregon>
enebo: cvars in general seems very ill-defined and hopefully less and less used
<enebo>
eregon: yeah rspec allows =~ but I was not sure if there is some magic I don’t know in test/unit
<nirvdrum>
There's a whole set of specs like that.
<eregon>
nirvdrum: good time to contribute some good basic specs then :D
pitr-ch has joined #jruby
projectodd-ci has quit [Client Quit]
<eregon>
but yeah I could see the encoding as a side-effect of some optimization of the regexp engine, which is not exactly nice
<nirvdrum>
eregon: I'd agree, if I could figure the damn thing out!
projectodd-ci has joined #jruby
projectodd-ci has quit [Client Quit]
<eregon>
so no options seems to result in US-ASCII encoding if it's actually only ASCII
projectodd-ci has joined #jruby
projectodd-ci has quit [Client Quit]
simi has joined #jruby
projectodd-ci has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] enebo closed issue #3007: splats with non-array like values http://git.io/vkwil
JRubyGithub has left #jruby [#jruby]
projectodd-ci has quit [Client Quit]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius closed issue #3008: Module#included is private when called with super http://git.io/vkwPG
JRubyGithub has left #jruby [#jruby]
tcrawley is now known as tcrawley-away
projectodd-ci has joined #jruby
x1337807x has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<fuzzyhorns>
headius: when i tried before i couldnt get it to work :/
<headius>
hmm
<headius>
have you ever gotten the jruby beans to show up?
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] nirvdrum pushed 1 new commit to master: http://git.io/vk70b
<JRubyGithub>
jruby/master 2bab6f9 Kevin Menard: Removed dead code.
JRubyGithub has left #jruby [#jruby]
pjammer has quit [Quit: My Mac has gone to sleep. ZZZzzz…]
<fuzzyhorns>
headius: well, here's my current conundrum
<fuzzyhorns>
headius: i put in jndi datasource so i could easily get db conn data
yfeldblum has quit [Ping timeout: 265 seconds]
<fuzzyhorns>
and now when i run torquebox (on top of jboss, for the record) in the visual vm i dont see the jruby application processes eanymore
<fuzzyhorns>
maybe i missed some previous config i had
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] nirvdrum pushed 2 new commits to master: http://git.io/vk7Ez
<JRubyGithub>
jruby/master 27b4cc1 Kevin Menard: removed redundant modifier.
<JRubyGithub>
jruby/master 53342fb Kevin Menard: Inlined the getEncoding helper method.
JRubyGithub has left #jruby [#jruby]
<nirvdrum>
lopex, headius: I'm tidying up RubyRegexp a bit. Please let me know if any of those commits look bad.
<headius>
nirvdrum: ok
<headius>
fuzzyhorns: hmm
<headius>
you were able to see them under TB before this change?
<fuzzyhorns>
yeah headius
<fuzzyhorns>
like i said though, maybe i lost some jruby opt i needed or some other config change in the mix
<fuzzyhorns>
i dont _think_ so but i cant guarantee it
<nirvdrum>
headius: It looks like RubyRegexp calls RubyThread#executeBlockingTask, which is deprecated. I'm not confident enough to mess with that. But just a general FYI.
<headius>
yeah I need to fix it
<headius>
fuzzyhorns: well the --manage flag just sets the OpenJDK property for RMI and the jruby property jruby.management.enabled=true
<headius>
if those are both set and you still don't see them I'm not sure what's wrong
<bbrowning>
fuzzyhorns: when you run jruby apps in torquebox on top of jboss you won't see separate jruby processes in the visual vm process list
<bbrowning>
you'll have to connect to the jboss process and inspect the jruby mbeans from there
<bbrowning>
jboss has its own JMX remoting protocol as well, which complicates things
<fuzzyhorns>
bbrowning: hmm, what was i seeing before then?
<fuzzyhorns>
headius: hmm i dont think i have --manage on anymore
<headius>
brb
<bbrowning>
fuzzyhorns: I have no idea - if you want to inspect anything in your torquebox 3 app via jmx you have to connect to that org.jboss.modules.Main process
projectodd-ci has quit [Remote host closed the connection]
projectodd-ci has joined #jruby
_JKL_ has quit [Read error: Connection reset by peer]
iamjarvo has joined #jruby
marr has joined #jruby
projectodd-ci has quit [Remote host closed the connection]
subbu|lunch is now known as subbu
projectodd-ci has joined #jruby
_JKL_ has joined #jruby
projectodd-ci has quit []
projectodd-ci has joined #jruby
projectodd-ci has quit [Remote host closed the connection]
projectodd-ci has joined #jruby
projectodd-ci has quit [Client Quit]
projectodd-ci has joined #jruby
vikaton has quit []
elia has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] headius closed issue #3010: 9k has different scoping behavior than MRI http://git.io/vkMAz
JRubyGithub has left #jruby [#jruby]
vikaton has joined #jruby
skade has joined #jruby
mrmargol_ has joined #jruby
_JKL_ has quit [Ping timeout: 244 seconds]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 1 new commit to master: http://git.io/vk71X
<nirvdrum>
headius, lopex: I'm thinking of creating a RegexpSupport holder for some of these static & private methods, much like StringSupport. Any objections or concerns?
<lopex>
nirvdrum: I guess that's the way to go
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 1 new commit to master: http://git.io/vk7FR
<JRubyGithub>
jruby/master 195761e Benoit Daloze: [Truffle] Full Module#append_features.
JRubyGithub has left #jruby [#jruby]
djbkd has quit [Remote host closed the connection]
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon pushed 1 new commit to master: http://git.io/vk7bc
<JRubyGithub>
jruby/master 0430387 Benoit Daloze: [Truffle] JT: remove print in favor of puts.
JRubyGithub has left #jruby [#jruby]
iamjarvo has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
dinfuehr has joined #jruby
<enebo>
some day out entire codebase will just be static imports
<nirvdrum>
enebo: A side effect free implementation!
elia has quit [Quit: Computer has gone to sleep.]
rsim has quit [Quit: Leaving.]
kstuart has quit [Quit: WeeChat 1.2]
kstuart has joined #jruby
djbkd has joined #jruby
JRubyGithub has joined #jruby
<JRubyGithub>
[jruby] eregon opened issue #3013: [Truffle] Kernel#send should be more transparent http://git.io/vk7pF
JRubyGithub has left #jruby [#jruby]
mistergibson has quit [Ping timeout: 245 seconds]
camlow325 has quit [Ping timeout: 244 seconds]
x1337807x has quit [Read error: Connection reset by peer]
x1337807x has joined #jruby
<chrisseaton>
enebo: funny you should say that - we're removing a lot of our instance classes at the moment like RubyArray etc
mistergibson has joined #jruby
<enebo>
chrisseaton: yeah we thought of doing the same thing a long time ago and passing the data holder into methods as paramter
<Antiarc>
`NoVarsDynamicScope.java:70:in `getValueOneDepthZeroOrNil': java.lang.RuntimeException: OneVarDynamicScope does not support scopes with two or more variables` - what on earth?
sdogruyol has joined #jruby
<Antiarc>
(1.7.19)
mrmargol_ has joined #jruby
<nirvdrum>
Cool. I've been meaning to look into Packer.
x1337807x has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<nirvdrum>
Antiarc: As we've gotten older, "PC LOAD LETTER" has just gotten more advanced.
<Antiarc>
heh
<nirvdrum>
I had a few "Office Space" moments earlier today.
<Antiarc>
I'm just not even sure what that assert *means*, let alone where it was thrown from