<yopp>
For some reason I'm getting Bignum with value of like "700"
<headius>
yopp: that's weird
<headius>
yopp: try passing -Xjit.threshold=0 so there's ruby lines in that trace
<headius>
and just go ahead and open the bug please :-) it gives something for other people to find if they search
<headius>
brb
<yopp>
uh. I'm in the middle of writing 20+ page report :(
<headius>
ok whenever you can
<headius>
there's not really enough info in those links
<yopp>
And worst part: this is happens on production data set, and I'm not sure how to reproduce that on clean set
<yopp>
I have a feeling this bson gem fault at the end
vtunka has quit [Quit: Leaving]
<GitHub69>
[jruby] nirvdrum pushed 4 new commits to truffle-head: https://git.io/vPd0u
<GitHub69>
jruby/truffle-head 0f833b4 Kevin Menard: [Truffle] Register new mx command to deploy binary snapshots.
<GitHub69>
jruby/truffle-head e7ee5f9 Kevin Menard: [Truffle] Deploy all artifacts, not just the ones from the RUBY distribution.
<GitHub69>
jruby/truffle-head a10a78d Kevin Menard: [Truffle] Added maven snapshot deploys to mx config.
shellac has quit [Ping timeout: 244 seconds]
thedarkone2 has joined #jruby
<headius>
yopp: well, it might be
<headius>
however I suspect JRuby because this method does say it's for int32
<headius>
so something's not right
<headius>
the bson gem could be fixed to coerce that value properly though
<headius>
obj.convertToInteger().toIntValue
<headius>
actually just removing the cast might be good enough
<headius>
unsure
<headius>
brb
<GitHub151>
[jruby] headius opened issue #4240: Rescue optimization over-optimizes rescue body with $! https://git.io/vPd2J
pilhuhn is now known as pil-afk
sandelius has joined #jruby
claudiuinberlin has joined #jruby
sandelius has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
claudiuinberlin has quit [Ping timeout: 245 seconds]
<SynrG>
i asked some days ago if there's a pure ruby solution to using an alternate BouncyCastle provider. i think it went overlooked any clues? (yes, more OpenVMS weirdness, I think :)
<SynrG>
i.e. i've found a stackoverflow solution to using an alternate provider dynamically, but it involves some Java code. any pure ruby way to do the same? my Java is oooold, and apparently using the signed provider solves my DH keypair couldn't be generated error (according to stackoverflow)
<SynrG>
there's no chance i can use a newer Java than 6 :(
elia has quit [Read error: Connection reset by peer]
elia_ has joined #jruby
claudiuinberlin has joined #jruby
claudiuinberlin has quit [Ping timeout: 256 seconds]
<headius>
SynrG: hmm
<headius>
well there's no pure-ruby solution that covers all of what BC/openssl do
<headius>
krypt got a start on it but then has been mostly abandoned
<GitHub67>
[jruby] nirvdrum pushed 1 new commit to truffle-head: https://git.io/vPdod
<GitHub67>
jruby/truffle-head 4ff63a4 Kevin Menard: [Truffle] Allow LGPL v2.1 licensed projects to be pushed to Nexus repo.
<SynrG>
headius: no, that's not what i meant. i mean how, in ruby, do i get it to use the alternate BC provider? or do i need to write some (small amount, as per the article i found; i can find it again if you need it to see) Java to load the alternate provider?
<chrisseaton>
SynrG: Do you mean that you already have an alternative version of BC that you want to use?
<SynrG>
here, let me do a quick look. last week, when i asked, i was at the office. i just need to start my work browser and check the history :)
<chrisseaton>
I don't think headius or I really know what it is practically that you want to do
<chrisseaton>
So try explaining from scratch
<SynrG>
i have the barest of understanding of the crypto involved, so really, i need to refer to the article that tells me what to do
<SynrG>
but the user-facing problem is: a server offering a service for one of our clients stopped negotiating with software we wrote that uses https to contact it
<SynrG>
when i looked up the specific error, i found this article that matches our situation:
<SynrG>
I am told that if I can just use BouncyCastle's JCE implementation, it will work around the issue
<SynrG>
i have a bit of a problem with changing system files that belong to Java to workaround a problem in only one module for one client
<chrisseaton>
That doesn't make sense to me - JRuby already uses BouncyCastle
<SynrG>
yes, but BouncyCastle has multiple providers
<SynrG>
JRuby only bundles one.
<SynrG>
the bouncycastle website explains the whys and wherefores
<chrisseaton>
Ok, I'm out of my depth of BouncyCastle knowledge then
<chrisseaton>
It is just a switch of JAR?
<SynrG>
it's a switch of a jar *and* some Java code to make it only be loaded when needed. i wanted to replace that with "and some ruby code"
<SynrG>
https://www.bouncycastle.org/latest_releases.html links to all the various different jars, and you can see in their tables JCE is only one of the numerous options. it is not the one JRuby uses.
<SynrG>
because they are signed jars, i doubt if they could ship with JRuby (or rather, it would pose practical problems for those who wanted to rebuild)
<SynrG>
and in case you read through the whole article and thought "well, can't you just disable DHE, then?" you need to drill down into the comments and see that's only available in Java 7 and later :(
<chrisseaton>
It sounds like you'll have to rebuild JRuby to use this different jar - which isn't that bad but you'll need to get your hands dirty to do that
<SynrG>
though before committing to that, I'd like to know if it's likely to work if I just drop in the new jar files and edit java's security policy ...
<SynrG>
i could take a test machine and do that. but obviously, that has much broader impact than i'd like, so is not a solution i'd want to roll out to clients.
<SynrG>
"rebuild jruby" does sound a bit daunting. it means that with a less than complete java build environment on VMS, I'd have to retool using some different OS where currently upgrading JRuby releases on VMS is trivial. i think my boss will probably balk at that and opt for the less costly route of patching security files for the specific client(s) with the issue. unfortunately, that would leave me without a
<SynrG>
general solution across the whole user base :(
<enebo>
lopex: just had to exit low carb diet so I am having beer again
<lopex>
enebo: cool
<lopex>
enebo: all my life I've been heavily underweight
<enebo>
lopex: good for you
<lopex>
enebo: only lately I changed my diet to cheese and a lot of milk products, we'll se where it will go
<lopex>
enebo: oh, not that good
<enebo>
lopex: I did not really start to put on weight until my mid thirties to the point where I feel I have to fight it
<lopex>
enebo: I'm not anorectic, but I feel it's not healthy for me
<enebo>
lopex: In college I was really skinny but not unhealthy
<lopex>
enebo: in my case my definitely underweight in my late thirties
<lopex>
to the point I'm almost feeling weak
<lopex>
but not that bad
<lopex>
enebo: also, I eat a LOT of meat and it definitely doesn help me
<lopex>
like a double of a normal person
<lopex>
so I'm changing my diet to milk, cheese etc
<enebo>
lopex: fat will change insulin levels and make you feel full faster but then again most meat has fat in it
<lopex>
enebo: the only period I have put on weight was when I was training quite heavily
<lopex>
and it seems it was the only thing that made me put on weight
<lopex>
I can eat for three and I'm hungry after three hours
<lopex>
er, "put on weight" is an idiom in english right ?
<enebo>
lopex: I was reading an article about how genetics and gastro biome affects how much food is processed and it is a major factor in why some people are heavy or not
<enebo>
lopex: If I can find it I will forward it to you
<lopex>
sure
<enebo>
lopex: yeah that is an idiom
<lopex>
thx
<lopex>
enebo: but my father otoh has a real belly
<lopex>
:P
<lopex>
enebo: but only belly, otherwise he's thin
<enebo>
heh
<lopex>
but he eats a lot of bread
<lopex>
like LOTS
<enebo>
lopex: yeah I think carbs is the real issue for me and weight as well
<enebo>
lopex: US food is very carb heavy and I like to drink beer
<lopex>
enebo: also, I have problame maintaining muscle weight too
<lopex>
enebo: if I train I can go up a lot
<enebo>
lopex: you sound like you are practically in a natural atkin’s (ketogenic) diet
<lopex>
enebo: but otherwise it plunges down
<lopex>
no idea that that means
<lopex>
looking
<enebo>
lopex: reduce your carbohydrate intake enough and you are producing lots of ketones
<enebo>
lopex: one thing being in ketosis does is more effectively use your fat reserves for energy
<lopex>
enebo: in any ways pure protein diet is bad for me
<enebo>
lopex: yeah to be sure you need lots of stuff from non-protein sources
<lopex>
it make me not gain any weight
<lopex>
ok
<lopex>
well, good to know
<lopex>
enebo: thx
<lopex>
enebo: the other thing it's that I've not been aware of this problem for yeares
<enebo>
lopex: I guess early on in the atkin diet decades ago they did not eat as many veggies/fruits and counted fiber against their tally of grams of carbs per day
<enebo>
lopex: they realized over time that your body needs stuff and relaxed and changed their recommendations based on time and living in the diet
<enebo>
lopex: I actually love weight really fast on this diet and start to develop a rash (then I go off for a few weeks and repeat) so I can say I am not doing something right
<enebo>
s/love/lose
<enebo>
lopex: but it allows me to take off weight and if it is unhealthy it is only unhealth in two week doses?
<lopex>
enebo: in theire terms what does it mean to be unhealthy ?
<enebo>
lopex: well I develop a rash every time I am on the diet in about 2 weeks
<enebo>
lopex: I am guessing it must be unhealthy by that alone
<lopex>
enebo: what's a rash ?
<enebo>
lopex: my theory is that I lose weight fast enough my body has a reaction
<enebo>
lopex: some other guy has this problem like me and made a web site (note: he is not a doctor :) )
<lopex>
ah
<lopex>
enebo: is it backed up by scientific research ?
<enebo>
lopex: no in fact he was unsatisfied with the doctors from what I remember. They kept giving an obviously wrong diagnosis
akp_ has joined #jruby
<lopex>
enebo: jeeez I'm so ashamed alwas almost ssold to the global worming thingy
<lopex>
enebo: remember ?
<enebo>
lopex: I am not quite parsing that. I know you used to think it was not a real problem
akp_ has quit [Read error: Connection reset by peer]
<enebo>
lopex: then you changed your mind
<lopex>
enebo: immediately
<lopex>
enebo: yeah, it's was the "controversy" thingy
<enebo>
lopex: poland is mostly coal/lignite still right?
<lopex>
enebo: germany/ western europe accuse poland often for amittin coal etc
akp has quit [Ping timeout: 260 seconds]
<enebo>
lopex: yeah while importing it at the same time
<enebo>
lopex: although the germans are trying to eventually divest from it
<lopex>
enebo: in general poland is not clean it's a lot of low emission towns
<lopex>
enebo: it's politically complex actually
<lopex>
enebo: the proble is the winds
<lopex>
enebo: which steadly blow from west to east
<lopex>
enebo: whatever you do, the winds will affect you political neighbours
<enebo>
lopex: sounds like a good time for wind power!
<lopex>
sure
<lopex>
enebo: but we're no rich enough country
<lopex>
someone wanting to invest ?
<enebo>
wind should be cost competitive with coal now
<lopex>
enebo: in western europe definitely yes
<enebo>
in US wind is quite a bit cheaper where there is reasonable wind. In places like NL I think it is considerably cheaper
<lopex>
and they do
<enebo>
solar also is at utility scale
elia has joined #jruby
<lopex>
enebo: the governemnt is also very depandant on coal mining incomes
<lopex>
enebo: it will take time
<lopex>
the coal minig in poland IS declining but it;s the majority of income for local populations
<lopex>
enebo: ^^^
<lopex>
enebo: and in most cases it's political dependency
<enebo>
lopex: yeah. in US coal has been declining for decades due to price pressures. The industry is basically collapsing and Obama paid some lip service to us divesting from coal.
<lopex>
enebo: the mines alre also steadily subsidized
<lopex>
enebo: like antrack ar whateve it was calles
<lopex>
enebo: btw Wendover Productions is a really cool channel
<enebo>
lopex: I am listening to this but I need to stop a bit as I am hacking on something and it is pulling my attention but so far it is pretty good
<lopex>
enebo cool
<lopex>
enebo: the author is definitely an American
<enebo>
lopex: yeah I can tell from accent
<enebo>
lopex: or guess anyways
<flavorjones>
headius: just saw your AR question. Open a Nokogiri issue? I can take a look. It's possible that whitespace is meaningfully inconsistent.
akp has joined #jruby
akp has quit [Ping timeout: 256 seconds]
<GitHub49>
[jruby] bjfish pushed 1 new commit to truffle-head: https://git.io/vPF3J