razetime has quit [Remote host closed the connection]
jhawthorn has quit [Ping timeout: 248 seconds]
jhawthorn has joined #ruby
<ox1eef_>
openbsd-rails2 ? What's that ?
fruitf89 has joined #ruby
otisolsen70 has quit [Quit: Leaving]
<myappie>
just a set of updated instructions
<ox1eef_>
A guide to using Rails on OpenBSD?
fruitf89 has quit [Ping timeout: 246 seconds]
dionysus70 has joined #ruby
dionysus69 has quit [Ping timeout: 248 seconds]
dionysus70 is now known as dionysus69
Aminda has quit [Ping timeout: 255 seconds]
Aminda has joined #ruby
otisolsen70 has joined #ruby
infinityfye has quit [Read error: Connection reset by peer]
infinityfye has joined #ruby
John_Ivan has joined #ruby
shokohsc6 has quit [Ping timeout: 255 seconds]
<Bish>
hey, if i put a gem into my gemfile, with a git repository as a source, it ends up in another directory than all other gems
<Bish>
can somebody tell me why that is / how i can stop it from doing it / how to still use it?
_ht has quit [Ping timeout: 260 seconds]
<adam12>
Bish: A git repository isnt' a gem, so Bundler has to stash it somewhere then build/install it.
<adam12>
Bish: If it's not working for you, _make sure_ you're using `bundle exec` or a binstub generated by bundler, or `require "bundle/setup"` (depending on the nature of the library).
<rapha>
ah, ok, that's like, one distributor (?) of "a" (?) jdk
<rapha>
adam12: that site doesn't answer the question specifically with regard to jruby, though.
__ht is now known as _ht
<adam12>
rapha: I'm not sure it matters tho? Presumably they would be targetting the recommended version.
<rapha>
hmm perhaps this is an unfortunate preconception on my part, adam12
<adam12>
rapha: Ideally you get an answer. I'm not familiar with Java, but thought the whichjdk.com site might be the answer.
<rapha>
i've gotten so used to having to deal with java byterot where for some arcane reason it absolutely must be version 8
<rapha>
it's definitely a good site!
<adam12>
I ran into similar a long time ago with Amazon. I was doing an integration with their ordering API, and they were churning through tech leads while I was doing the integration. We were at the final integration stage and performing UA, but their webhooks were failing. Swore it was my fault ("We're Amazon, we make no mistakes. Your server must be
<adam12>
misconfigured"). After weeks of back and forth, it must of escalated high enough, and coincidentally found another Adam at Amazon who said "Oh yeah, we're on Java 6 and SNI support is broken. We had to disable it.". LOL.
<rapha>
what's UA?
<adam12>
rapha: User Acceptance.
<rapha>
ah ok
<rapha>
lol about the other Adam tho :)
<adam12>
We're ubiquitous, I guess.
<rapha>
same with jonathans
<rapha>
even my bff is called jonathan
<rapha>
which is often confusing to others, which we quite enjoy
<adam12>
I did chuckle, because this was ~ 2015 or whatever, and even though I knew I was right and everything was configured correct, I have Amazon telling me I'm wrong, so I went through every line in that nginx config multiple times.
<rapha>
surprised they didn't try to hire you
<adam12>
It basically came down to: I've tested this. I've seen it work. You test and show me it working for you.
<rapha>
so much "software" development is a lack or breakdown of human-to-human communication
<rapha>
basically Layer 8^2
<rapha>
or worse, Layer 8^n_people_involved
razetime has quit [Remote host closed the connection]
<adam12>
Indeed.
hightower2 has joined #ruby
<joto2>
hi guys, just noticed that rbenv install -l doesn't list Ruby's latest stable versions
<joto2>
latest one I see there is 3.1.2 while on the official site it's 3.2.1
<joto2>
Message telling me "Only latest stable releases for each Ruby implementation are shown."
<joto2>
does anybody know why that is?
<joto2>
okay so for anyone wondering
<joto2>
cd ~/.rbenv/plugins/ruby-build && git pull
<joto2>
that means an update to ruby-build was needed...
<adam12>
I believe it differs from ruby-install where it ships with definitions. ruby-install has definitions separate, and can just accept arbitrary version too.
<joto2>
aha
dviola has joined #ruby
teclator has quit [Ping timeout: 246 seconds]
<ox1eef_>
adam12: Adam is from the name of the first two humans ! A good name to have. For sure.
teclator has joined #ruby
<joto2>
yeah, I quite like the name too :)
FetidToot has quit [Ping timeout: 252 seconds]
FetidToot has joined #ruby
exclus39 has joined #ruby
FetidToot6 has joined #ruby
FetidToot has quit [Ping timeout: 255 seconds]
FetidToot6 is now known as FetidToot
<myappie>
Excuse me guys
<myappie>
That's not what his name is
<myappie>
Do you want the truth?
<myappie>
I said, DO YOU WANT THE TRUTH?
<rapha>
hmm
<myappie>
ox1eef_, joto2?
<myappie>
Just say yes already, damn
<ox1eef_>
I'm between channels. And no idea what joto2 is.
<myappie>
Cool
<myappie>
I just love reading this sentence
<myappie>
"Disintegrin and metalloproteinase domain-containing protein 12 (previously Meltrin) is an enzyme that in humans is encoded by the ADAM12 gene."
<ox1eef_>
xD
<myappie>
:)
<rapha>
ox1eef_: weren't Adam's sons called Seth, Awan and Azura?
<rapha>
er, wait
<rapha>
1 son 2 daughters
<rapha>
ah, son was fathered by a demon i just saw ... so only 2 daughers, neither named Adam
<joto2>
yes!
<ox1eef_>
rapha: I don't know. But good topic to research.
<rapha>
ox1eef_: well ... wikipedia is your friend :)
<ox1eef_>
Good place to start, indeed.
dviola has quit [Ping timeout: 255 seconds]
eddof13 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
ur5us has joined #ruby
infinityfye has quit [Ping timeout: 255 seconds]
<joto2>
I have the following line in my Gemfile: gem "rails", "~> 7.0.4"
<joto2>
however when inspecting the accompanying lock file, it doesn't contain the latest point release which is 7.0.4.2. Would this be correct?
<joto2>
actually my question is, can I have 7.0.4. by using that line or do I need to explicitly ask for it like: gem "rails", "~> 7.0.4.2"
teclator has quit [Ping timeout: 255 seconds]
eddof13 has joined #ruby
caedmon has joined #ruby
caedmon has quit [Ping timeout: 255 seconds]
_ht has quit [Quit: _ht]
infinityfye has joined #ruby
otisolsen70 has quit [Quit: Leaving]
exclus39 has quit [Quit: Best CPUs can count to infinity twice without being bugged out by zero division errors or whatnot.]
<ox1eef_>
I think '~> 7.0.4' is the same as '~> 1.0'. Or more plainly, it will potentially match 7.0.5, 7.0.6, 7.0.4.2, 7.0.4.3, 7.0.4.5, etc.
<ox1eef_>
And I think it will match what's most recent. If 7.0.5 exists, it will pull that. If not, it will pull 7.0.4.X.
<ox1eef_>
And keep in mind that the lockfile is locked to what was resolved when 'bundle install' is run. If you want to update the lockfile, you have to run 'bundle update rails'.
<ox1eef_>
The update will be within the constraint of ~> 7.0.4, so it will follow the rules mentioned a bit above.
tomtmym has quit [Quit: Gone.]
hightower3 has joined #ruby
hightower2 has quit [Ping timeout: 255 seconds]
donofrio_ has joined #ruby
<ox1eef_>
Mileage may vary but so far I'm liking test-unit over rspec for a couple of projects. It's nice to use vanilla Ruby constructs to handle things like code re-use.
ruby[bot] has quit [Remote host closed the connection]
ruby[bot] has joined #ruby
<joto2>
ox1eef_ that was it. When running bundle install it wouldn't install the .2 point release when the line was "~> 7.0.4". Only after 'bundle update rails' it would install 7.0.4.2
<joto2>
come to think of it. Would prefer for it to be "~> 7.0". That would install anything below 7.1 right? So for example 7.0.6.2
dviola has joined #ruby
<ox1eef_>
Yeah, but when the lockfile is present the version specified there will always be installed. 'bundle update' would need to be run. '~> 7.0' is most likely what you want. It'd grabbed all those smaller point versions, and 7.1, 7.2, etc.
<ox1eef_>
And no, not anything below 7.1, it would install 7.1 and anything after it as well (but not 8 / 8.X).
<ox1eef_>
But I think ~> 7.0 is reasonable, because the lockfile is what protects you. You would need to opt into the update by running 'bundle update'.
<joto2>
oh okay
dionysus69 has joined #ruby
<ox1eef_>
At least for Rails projects it is common to have a Gemfile.lock checked into the source tree, and anyone running 'bundle install' will get the versions specidfied in Gemfile.lock. For libraries, it is more common to not check in Gemfile.lock but specify requirements in line with semantic versioning. In that case, ~> 1.0 should be safe, ~> 1.0.0 is safest, and ~> 1 is the most dangerous.
<ox1eef_>
Actually, = 1.0.0 is safest, and ~> 1.0.0 comes after that.
eddof13 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
<ox1eef_>
If you release a gem with a dependency on ~> 1.0.0, and some time passes, a person running 'gem install yourgem' might pick up 1.0.9 of that dependency. In theory it should be backwards compatible, but there's the potential for it not be. That's why = 1.0.0 is safest because regardless of time, the same version will always be pulled in.
<ox1eef_>
It can get complicated though. = 1.0.0 might be too strict and cause conflicts with other gems depending on ~> 1.0.1. As a general rule, just use ~> 1.0 or ~> 1.0.0.
eddof13 has joined #ruby
<joto2>
thanks ox1eef_! really appreciated your thorough explaining
<joto2>
I have it on => 1.0 now which will be safe enough for the rails gem itself
eddof13 has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
infinityfye has quit [Quit: Leaving]
<ox1eef_>
Yeah, ~> 1.0 should be safe in the case of Rails. They take backwards compatibility more seriously than most.