<markalanevans>
but for some reason its not downloading the entire gem or something?
vertis has joined #rubygems
_maes_ has quit [Ping timeout: 255 seconds]
adkron has joined #rubygems
<markalanevans>
How about this, how does gem install XXX work
sn0wb1rd has quit [Remote host closed the connection]
<markalanevans>
does it check to see if the gem file is downloaded and if it exists install it, else lookup some directory online for gems and then download them and install them?
sn0wb1rd has joined #rubygems
yerhot has joined #rubygems
the_mentat has quit [Quit: Computer has gone to sleep.]
vertis has quit [Quit: Leaving.]
ckrailo has quit [Quit: Computer has gone to sleep.]
yerhot has quit [Remote host closed the connection]
reset has quit [Quit: Leaving...]
the_mentat has joined #rubygems
yut148 has joined #rubygems
adkron has quit [Ping timeout: 245 seconds]
reset has joined #rubygems
adkron_ has joined #rubygems
newUser1234 has quit [Remote host closed the connection]
newUser1234 has joined #rubygems
newUser1234 has quit [Ping timeout: 276 seconds]
twoism has quit [Ping timeout: 252 seconds]
the_mentat has quit [Ping timeout: 252 seconds]
havenwood has quit [Remote host closed the connection]
the_mentat has joined #rubygems
newUser1234 has joined #rubygems
_maes_ has joined #rubygems
the_mentat has quit [Quit: Computer has gone to sleep.]
newUser1234 has quit [Remote host closed the connection]
newUser1234 has joined #rubygems
qmx|away is now known as qmx
sn0wb1rd has quit [Quit: sn0wb1rd]
Apes has joined #rubygems
newUser1234 has quit [Ping timeout: 245 seconds]
newUser1234 has joined #rubygems
ckelly has quit [Quit: Leaving...]
newUser1234 has quit [Remote host closed the connection]
newUser1234 has joined #rubygems
bjessbrown has joined #rubygems
newUser1234 has quit [Read error: Connection reset by peer]
newUser1234 has joined #rubygems
newUser1234 has quit [Remote host closed the connection]
newUser1234 has joined #rubygems
adkron__ has joined #rubygems
newUser1234 has quit [Read error: Connection reset by peer]
newUser1234 has joined #rubygems
adkron_ has quit [Ping timeout: 264 seconds]
newUser1234 has quit [Remote host closed the connection]
newUser1234 has joined #rubygems
sn0wb1rd has joined #rubygems
newUser1234 has quit [Read error: Connection reset by peer]
newUser1234 has joined #rubygems
sn0wb1rd has quit [Remote host closed the connection]
newUser1234 has quit [Remote host closed the connection]
newUser1234 has joined #rubygems
newUser1234 has quit [Read error: Connection reset by peer]
newUser1234 has joined #rubygems
mockra has quit [Remote host closed the connection]
newUser1234 has quit [Remote host closed the connection]
markalanevans has quit [Ping timeout: 240 seconds]
newUser1234 has joined #rubygems
ctracey` has quit [Read error: Connection reset by peer]
ctracey` has joined #rubygems
newUser1234 has quit [Read error: No route to host]
newUser1234 has joined #rubygems
benchMark has quit []
newUser1234 has quit [Ping timeout: 260 seconds]
newUser1234 has joined #rubygems
mootpointer has joined #rubygems
lsegal has quit []
qmx is now known as qmx|away
whit537 has joined #rubygems
dvu has joined #rubygems
tkramer has joined #rubygems
ckelly has joined #rubygems
<mootpointer>
Hey guys, I work for a hosting company in Australia, and we were looking at providing a public RubyGems mirror for Australians to use.
<drbrain>
evan: ↑
<drbrain>
mootpointer: evan has been handling our mirroring stuff in the past
<mootpointer>
I was wondering if you guys would be able to help me out in working out what would be required to implement such an endeavour.
<drbrain>
he might be eating dinner though
<mootpointer>
drbrain: Cool. I can hang around for a bit.
<drbrain>
mootpointer: it is 7PM here in Pacific Time, so if you don't see him tonight maybe try a few hours earlier
mockra has joined #rubygems
<mootpointer>
drbrain: Good call. Got to love that whole time zone thing.
newUser1234 has quit [Remote host closed the connection]
<drbrain>
mootpointer: I think he goes home a little after 5PM
<mootpointer>
Thanks.
<drbrain>
are you at RubyConf AU?
tkramer has left #rubygems [#rubygems]
newUser1234 has joined #rubygems
<evan>
mootpointer: it's pretty easy for us to do
<drbrain>
he appears!
<evan>
we just run nginx in a reverse proxy mode
<mootpointer>
drbrain: Yes, yes I am.
<evan>
and I can add records to redirect users to your mirror based on their geoip
<drbrain>
I wish I had gone, it sounds really cool
mockra has quit [Remote host closed the connection]
<mootpointer>
drbrain: It's been a really cool couple of days.
<mootpointer>
evan: Excellent. The trick (I assume) would be keeping up to date.
<evan>
actually no
<evan>
this nginx reverse proxy method does it automagically
<evan>
because the mirror is filled based on need
<evan>
rather than having to coordinate pushing everything
<evan>
so you can think of it being like a cache that lives in .au
<evan>
the first user pays the extra time
<evan>
but every user after that gets it fast.
<mootpointer>
Okay. That's pretty rocking.
blowmage has quit [Ping timeout: 252 seconds]
<raggi>
you can seed it with a cron if you wanted to
blowmage has joined #rubygems
<raggi>
realyl really easily
newUser1234 has quit [Ping timeout: 264 seconds]
<evan>
yep
<evan>
mootpointer: generally what I just need is access to a machine thats got 20G+ of disk space
<evan>
the machine will be doing pretty much 100% IO
<mootpointer>
evan: Easy.
<evan>
so it doesn't even need to be too beefy in terms of CPU
<evan>
if it's ubuntu, I can set it up in about 10 mins
<mootpointer>
Brilliant.
<evan>
and have users on it within minutes.
<raggi>
the only problems are, the cache format is not equal to that of the original, and it isn't resistant to source tampering
<mootpointer>
I'll need to run it past our overlords^w^wmy bosses.
<raggi>
but that's much less critical than making shit fast currently :)
<evan>
mootpointer: sure, just let me know.
<evan>
I like to have shell access to the mirrors
<mootpointer>
evan: I'll likely be in touch early next week.
<mootpointer>
evan: Makes sense.
<evan>
because if they have issues, I have to hunt things down to fix them
<evan>
mootpointer: feel free to email me @rubycentral.org
<evan>
when you hear
<mootpointer>
Will do. Drafting a convincing email now :)
<evan>
:)
<evan>
I'm happy to do a test run also
<evan>
so we can see what kind of traffic levels you all are comfortable with
<evan>
i've got pretty fine control over the percentage of traffic that the mirrors get.
<evan>
atm, I've got all of Japan and China using a mirror running in Linode Tokyo
<evan>
ok, off to dinner!
havenwood has joined #rubygems
newUser1234 has joined #rubygems
adkron__ has quit [Ping timeout: 256 seconds]
bcardarella has joined #rubygems
reset has quit [Ping timeout: 252 seconds]
mootpointer has quit [Read error: Connection reset by peer]
<pea53>
yeah, the analyzer is splitting on colons, so they are not searchable
teancom has quit [Remote host closed the connection]
<pea53>
qrush: is that the only term that the API key would be found under? `rubygems_api_key`
<qrush>
yes
<qrush>
i'd be amazed if there are only 3
<qrush>
pea53: if you find results (more than 3) please PM them to me
vanstee has joined #rubygems
whit537 has left #rubygems [#rubygems]
<qrush>
i'm going to reset the ones there
<pea53>
PM??
<pea53>
private message
<jm_>
Project Manage them to me plzzzz ;)
<pea53>
I'll throw them into a private gist and send a link via western union telegram
* drbrain
waves to pea53
<qrush>
please screenshot and fax
<nirvdrum>
qrush: Anything I can do to help? I'd hate to have just made your night shitty without compensating somehow.
<qrush>
wouldnt be the first time security issues have been reported via twitter
<qrush>
<_<
<raggi>
qrush: looks like you could kill those three accounts, they're unmaintained gems
<raggi>
or at least those api keys
<qrush>
yeah going to
<qrush>
resetting them now
<nirvdrum>
I'm new to this. Like I said, I didn't really think of it as RubyGems security issue per se. People are going to continue to do stupid things.
<qrush>
nirvdrum: scotch next time you see me will be fine :)
<nirvdrum>
Will do.
<pea53>
drbrain: hello!
teancom has joined #rubygems
<nirvdrum>
You have to come out this way though. I'm not heading up to Canada anytime soon.
freerobby has joined #rubygems
reset has joined #rubygems
reset has quit [Client Quit]
<nirvdrum>
qrush: I really have no idea how to guard against this going forward. Signing gems would help with the pushing problem. Yanking is still an issue if you have someone's API key though.
<nirvdrum>
And it seems there's an endless supply of people pushing their dotfiles.
mockra has quit [Remote host closed the connection]
<qrush>
evan was talking about using ssh auth for gem pushes
<qrush>
not sure where that ended p
<qrush>
and hey, i can see canada from my backyard :P
<nirvdrum>
Alas, those pushing dotfiles almost certainly are pushing their private SSH keys as well :-(
vanstee has quit [Quit: Computer has gone to sleep.]
<qrush>
if it helps pea53 it's going to be most likely in a file called "credentials"
<qrush>
~/.gem/credentials
<nirvdrum>
pea53, qrush: Although the 3 on the first search results page all have different paths.
<qrush>
only 2/3 of those API keys were valid, and i just reset both, emailed the users via help.rubygems.org
<pea53>
I'm doing some queries directly on the ElasticSearch cluster
<pea53>
finding 720 files, but there is a lot of noise in there
<qrush>
this is silly, but maybe download them all, see which YAML.parse ?
mockra has quit [Remote host closed the connection]
newUser1234 has quit [Ping timeout: 256 seconds]
<pea53>
my work here is done - time to get some sleep!
<pea53>
ping me on the twitters @pea53 if I can be of further assistance
<qrush>
cool, thanks man
pea53 has left #rubygems [#rubygems]
* jm_
waves
jm_ has quit [Quit: jm_]
adkron has quit [Ping timeout: 246 seconds]
newUser1_ has quit [Ping timeout: 245 seconds]
huoxito has quit [Quit: Leaving]
yerhot has joined #rubygems
nirvdrum has quit [Ping timeout: 248 seconds]
crandquist has quit [Quit: Leaving...]
skalnik has joined #rubygems
<skalnik>
qrush: yo bcardarella pinged me saying you were looking for someone from GitHub. You find someone?
<qrush>
yep, thanks
<skalnik>
Sweet
<skalnik>
Glad ya got it sorted :)
azool has quit [Ping timeout: 276 seconds]
azool has joined #rubygems
newUser1234 has joined #rubygems
yerhot has quit [Remote host closed the connection]
mockra has joined #rubygems
mockra has quit [Ping timeout: 252 seconds]
yerhot has joined #rubygems
yerhot has quit [Remote host closed the connection]
Apes has quit [Ping timeout: 244 seconds]
freerobby has quit [Quit: Leaving.]
nirvdrum has joined #rubygems
sn0wb1rd has joined #rubygems
nirvdrum has quit [Ping timeout: 252 seconds]
workmad3 has joined #rubygems
yerhot has joined #rubygems
yerhot has quit [Ping timeout: 252 seconds]
freerobby has joined #rubygems
freerobby has quit [Ping timeout: 252 seconds]
sferik has joined #rubygems
mlen has joined #rubygems
mlen has left #rubygems [#rubygems]
freerobby has joined #rubygems
freerobby has quit [Ping timeout: 255 seconds]
havenwood has quit [Remote host closed the connection]
workmad3 has quit [Ping timeout: 260 seconds]
ckelly has quit [Read error: Connection reset by peer]
bdrewery has quit [Read error: Connection reset by peer]
bdrewery has joined #rubygems
baburdick has quit [Quit: Leaving.]
sferik has quit [Quit: Computer has gone to sleep.]
lsegal has quit [Quit: Quit: Quit: Quit: Stack Overflow.]
freerobby has joined #rubygems
freerobby has quit [Ping timeout: 244 seconds]
sferik has joined #rubygems
Elhu has joined #rubygems
workmad3 has joined #rubygems
whit537 has joined #rubygems
Elhu has quit [Quit: Computer has gone to sleep.]
havenwood has joined #rubygems
Elhu has joined #rubygems
freerobby has joined #rubygems
freerobby has quit [Ping timeout: 245 seconds]
yut148 has quit [Quit: Leaving...]
tbuehlmann has joined #rubygems
freerobby has joined #rubygems
aspiers has quit [Ping timeout: 276 seconds]
freerobby has quit [Ping timeout: 256 seconds]
aspiers has joined #rubygems
terceiro has joined #rubygems
sferik has quit [Quit: Computer has gone to sleep.]
newUser1234 has quit [Remote host closed the connection]
huoxito has joined #rubygems
markstarkman has quit [Remote host closed the connection]
markstarkman has joined #rubygems
markstarkman has quit [Remote host closed the connection]
markstarkman has joined #rubygems
rcvalle has joined #rubygems
rmartin has joined #rubygems
tcopeland has quit [Quit: Leaving.]
markstarkman has quit [Remote host closed the connection]
Elhu has quit [Quit: Computer has gone to sleep.]
rcvalle has quit [Quit: Leaving]
qmx|away is now known as qmx
whit537 has quit [Ping timeout: 255 seconds]
whit537_ has joined #rubygems
_maes_ has quit [Ping timeout: 264 seconds]
rcvalle has joined #rubygems
markstarkman has joined #rubygems
qmx has quit [Changing host]
qmx has joined #rubygems
Elhu has joined #rubygems
benchMark has joined #rubygems
vanstee has joined #rubygems
markstarkman has quit [Remote host closed the connection]
markstarkman has joined #rubygems
indirect_ has joined #rubygems
drbrain has quit [Ping timeout: 248 seconds]
indirect has quit [Ping timeout: 248 seconds]
indirect_ is now known as indirect
drbrain has joined #rubygems
therealadam has joined #rubygems
markstarkman has quit [Remote host closed the connection]
newUser1234 has joined #rubygems
jcaudle has joined #rubygems
kgrz has joined #rubygems
freerobby has joined #rubygems
tkramer has joined #rubygems
yerhot has joined #rubygems
tcopeland has joined #rubygems
freerobby has quit [Ping timeout: 276 seconds]
teancom has quit [Remote host closed the connection]
teancom has joined #rubygems
teancom has quit [Ping timeout: 248 seconds]
whit537_ has quit [Quit: BLAM!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!]
havenwood has quit [Remote host closed the connection]
therealadam has quit [Remote host closed the connection]
therealadam has joined #rubygems
therealadam has quit [Read error: Connection reset by peer]
stevenharman has quit [Quit: Leaving...]
yasho99 has quit [Remote host closed the connection]
yerhot has quit [Remote host closed the connection]
freerobby has joined #rubygems
teancom has joined #rubygems
kgrz has quit [Read error: Connection reset by peer]
newUser1234 has quit [Remote host closed the connection]
qmx is now known as qmx|away
cowboyd has joined #rubygems
nirvdrum has joined #rubygems
_maes_ has joined #rubygems
freerobby has quit [Quit: Leaving.]
therealadam has joined #rubygems
adkron has joined #rubygems
dwradcliffe has joined #rubygems
dwradcliffe has quit [Remote host closed the connection]
dwradcliffe has joined #rubygems
martinisoft has quit [Quit: martinisoft]
newUser1234 has joined #rubygems
baburdick has joined #rubygems
therealadam has quit [Ping timeout: 244 seconds]
baburdick has quit [Remote host closed the connection]
baburdick has joined #rubygems
adkron has quit [Ping timeout: 252 seconds]
adkron has joined #rubygems
dwradcliffe has quit [Remote host closed the connection]
dwradcliffe has joined #rubygems
stevenharman has joined #rubygems
the_mentat has joined #rubygems
newUser1234 has quit [Remote host closed the connection]
dwradcliffe has quit [Remote host closed the connection]
dwradcliffe has joined #rubygems
dwradcliffe has quit [Remote host closed the connection]
newUser1234 has joined #rubygems
jcaudle_ has joined #rubygems
freerobby has joined #rubygems
jcaudle has quit [Ping timeout: 252 seconds]
jcaudle_ is now known as jcaudle
dwradcliffe has joined #rubygems
dwradcliffe has quit [Remote host closed the connection]
adkron has quit [Ping timeout: 252 seconds]
adkron has joined #rubygems
freerobby has left #rubygems [#rubygems]
postmodern has quit [Quit: Leaving]
tbuehlmann has quit [Remote host closed the connection]
fromonesrc has joined #rubygems
nmeum has quit [Ping timeout: 252 seconds]
havenwood has joined #rubygems
x0F_ has joined #rubygems
x0F has quit [Disconnected by services]
x0F_ is now known as x0F
dwradcliffe has joined #rubygems
dwradcliffe has quit [Remote host closed the connection]
dwradcliffe has joined #rubygems
benchMar_ has joined #rubygems
dwradcliffe has quit [Remote host closed the connection]
dwradcliffe has joined #rubygems
benchMark has quit [Ping timeout: 264 seconds]
benchMar_ is now known as benchMark
_maes_ has quit [Ping timeout: 256 seconds]
_maes_ has joined #rubygems
Elhu has quit [Quit: Computer has gone to sleep.]
Elhu has joined #rubygems
tubbo has joined #rubygems
<tubbo>
hi i'm trying to use `gem install ..` when i once had rbenv on my machine (but not anymore), and for some reason RubyGems keeps generating binstubs that look like this: https://gist.github.com/tubbo/5014855
<tubbo>
anyone know how i can make it go back to just using #!/usr/bin/env ruby?
the_mentat has quit [Quit: Computer has gone to sleep.]
ckrailo has joined #rubygems
adkron has quit [Quit: leaving]
adkron has joined #rubygems
dvu has quit [Remote host closed the connection]
tcopeland has quit [Ping timeout: 244 seconds]
whit537 has joined #rubygems
graphex has joined #rubygems
hahuang65 has joined #rubygems
<hahuang65>
is there a best practice around writing optional features for a gem and requiring dependencies? For example I'd like to include an async module that utilizes Resque/Sidekiq in my gem, but don't want to make it a dependency of the gem.
newUser1234 has quit [Read error: Connection reset by peer]
newUser1234 has joined #rubygems
newUser1_ has joined #rubygems
therealadam has quit [Ping timeout: 260 seconds]
<raggi>
hahuang65: best practice is to have the user declaratively load your stuff, and just have those entries expect the deps to be present
<raggi>
hahuang65: e.g. require 'mything/resque', and inside mything/resque you can just do a plain `require "resque"`
newUser1234 has quit [Ping timeout: 244 seconds]
<raggi>
hahuang65: but people do much more complicated things like catching loaderror and retrying, and all kinds of other dynamic stuff
<raggi>
i'd imagine rails users will cry if you ask them to call require
<raggi>
rathern than provide some kind of ""DSL""
<hahuang65>
raggi: hmm I was thinking about doing an unless defined? Resque and throw some exception saying you need resque to run this module of the gem
<raggi>
hahuang65: why?
<raggi>
hahuang65: do developers not understand LoadError?
<hahuang65>
@raggi: because I want to load the file with the gem, but not have it error out
<raggi>
hahuang65: why do you want to load it?
<raggi>
require is a perfectly reasonable interface for loading features
<hahuang65>
because it'd be a module that they can include in their models... so unless I load the file, that module will be undefined
<hahuang65>
normally, the sync feature would be
bcardarella has left #rubygems [#rubygems]
<hahuang65>
include Gem::SyncFeature, but I want them to optionally be able to do include Gem::AsyncFeature, which means I have to load the async_feature.rb with my gem. If I do a require there for Resque, they're shit outta luck if they don't have/want Resque.
<raggi>
the only other option i agree with is creating a new gem wiht the different dependencies
<raggi>
e.g. mygem-resque
<hahuang65>
why is this approach frowned upon?
the_mentat has joined #rubygems
<raggi>
but to magically resolve this elsehow is goign to be a combination of bug ridden, and overengineered
<raggi>
every time i have to fix someones magical other solution to this, it makes me angry
k2rebina has left #rubygems [#rubygems]
<raggi>
and given that most users cannot solve the bugs that arise from those magic solutions, that happens way more than you'd think
<hahuang65>
Why is this a magical solution?
<raggi>
hahuang65: because you're going ot be altering your object schema at load time based on facts that may or may not be true
<hahuang65>
my object schema?
<raggi>
yes
<hahuang65>
there's no object schema
<raggi>
look, lets not get into a semantic argument
<raggi>
if you feel you must implement magic
<raggi>
just go do it, but i won't help you
<hahuang65>
Nono, I want to fully understand
<raggi>
whether or not Resque is defined can be caused by a LOT of different aspects
<hahuang65>
I don't want to do something ridiculous, but there must be a good reason... Maybe I should describe what I am thinking further
<raggi>
far far beyond whether the gem is listed in the gemfile
<raggi>
things are loaded in orders
<raggi>
that order in rails is *very poorly defined*
<raggi>
i can't understate how poorly defined it is
<raggi>
you can simplify this problem drastically, by making your load order 1) base don ruby-core apis alone (i.e. require), and 2) well defined (i.e. require), and your interface is also then *very* simple and takes no effort to implement
<raggi>
what happens to users when these magic solutions occur
<raggi>
is that they'll do something odd (probably for what seems like a good reason), like have a resque extension in lib
<raggi>
that defines module Resque; module SomeOtherThing; ...; end; end
<raggi>
your code will get added to the gemfile *above* gem 'resque'
<raggi>
and then your code will see defined?(Resque) => trueish value
<raggi>
and try to use the interface it really cares about in resque
<raggi>
before resque is loaded by Bundler.require
<raggi>
and the world ends in a way that the users have *nochance* of understanding, more well over 95% of gem/rails users
<hahuang65>
Well the only thing you said that applies here is if they manually declare some Resque class... Since my code runs at runtime, the defined won't get checked until the method is actually run, which guarantees that all the gems are loaded already no?
<raggi>
hahuang65: that's just one example of how this happens
<raggi>
hahuang65: there are many other ways you can get out of order execution
fromonesrc has quit [Ping timeout: 248 seconds]
<raggi>
hahuang65: you have no control how apps are booted unless you take control of loading your dependencies
<hahuang65>
but my proposed solution won't run into this order problem
<raggi>
choose: bugs or correctness
<hahuang65>
because it's only checks at runtime, not boot tiem
<raggi>
hahuang65: i've seen apps that reflect on models at boot time
* raggi
tries to remember an open source example
<raggi>
the site_admin rails engine
<raggi>
there you go.
<hahuang65>
Rails loads all the gems when they boot up the server... this means that by runtime, Resque is guaranteed to be there if it was part of the gems
<raggi>
hahuang65: not if it's there by side-effect, where someones keeping a gemfile lightweight
marc_online_ has joined #rubygems
<raggi>
hahuang65: e.g. if a user includes `gem 'capistrano-resque'`
<hahuang65>
by side-effect? what does that mean?
<raggi>
which is perfectly valid
<raggi>
Bundler.require is a disease
<hahuang65>
people will do includes at runtime?
workmad3 has joined #rubygems
fromonesrc has joined #rubygems
<hahuang65>
raggi: i feel like this is a simple and elegant solution because 1) it doesn't require people to install another gem just to use an async feature, 2) it doesn't require making another gem that's just 10 lines of code, 3) it'\]\
<hahuang65>
3) it doesn't force them to install resque if they don't want to, especially since I'd like for them to be able to choose between Resque and Sidekiq
<hahuang65>
The only thing that concerns me in your statements is if they actually declare some class named Resque or Sidekiq
<hahuang65>
nothing else should apply because it's checked at runtime when the method is invoked, so all gems (included by Bundler or manually) should be loaded by then
<hahuang65>
Am I incorrect?
<hahuang65>
I feel like I'm not understanding your explanation and I'm a complete idiot for missing some key point that you are making.
<hltbra>
I want comments about it, and see what should I change
<drbrain>
looking
torarne1 has joined #rubygems
<drbrain>
I would move line 12 of the first chunk to execute
<drbrain>
line 12 in the first chunk
<drbrain>
all the other commands extract entries from options in execute
<hltbra>
so no @host= at initialize?
<drbrain>
correct
<drbrain>
to keep the style consistent
<hltbra>
anything else?
<drbrain>
also, can you use rubygems.example instead of fake host?
<drbrain>
fakehost
<hltbra>
sure :)
<hltbra>
drbrain: what about test methods names?
<drbrain>
I think we will collide on test/rubygems/test_gem_commands_push_command.rb, I believe I deleted one of those and renamed the other
<drbrain>
checking names
<hltbra>
what will we collide on?
<drbrain>
you'll get a merge conflict on that file since that test no longer exists
<drbrain>
it looks like you don't pass @host into sign_in for the owner command
<hltbra>
drbrain: no, I did not commit yet
<drbrain>
or to rubygems_api_request
<hltbra>
drbrain: both use @host. should I really change to pass that arg?
<drbrain>
yes, otherwise it uses the default host
<drbrain>
you can check it against your private gemcutter host with ruby -Ilib bin/gem owner --host http://your.host.example some_gem
<hltbra>
sorry drbrain, but I didnt understand if you are saying that if I don't pass @host to sign_in or rubygems_api_request they wouldnt work or if I should pass just as a style matter
<hltbra>
(all tests pass here, ok? that patch I showed you works)
<drbrain>
due to the suspicious tests you found in push_command I think it may only work in the tests
vertis has joined #rubygems
<drbrain>
using ENV['RUBYGEMS_HOST'] it should work regardless of --host
<drbrain>
or even passing in @host
<drbrain>
anyhow, if it works for you I can double check it
<hltbra>
drbrain: my patch works as is, not only at tests.
<hltbra>
just checked
<drbrain>
ok
<hltbra>
but I think it is a good idea to pass @host anyway
<drbrain>
maybe there's extra magic in gemcutter_utilities that I am not spotting
<hltbra>
def sign_in sign_in_host = self.host
<hltbra>
and at rubygems_api_request it uses self.host
<drbrain>
yes, but I'm not sure how --host can set Gem::GemcutterUtilities#host without passing in @host
<drbrain>
does --host set Gem.host?
<hltbra>
no
<drbrain>
anyhow, I believe you
<hltbra>
drbrain: does ruby evaluates default args everytime it is called? python does that only when the method is defined.
<drbrain>
every time
<hltbra>
maybe that's your confusion?
<drbrain>
def x y = (raise "missing x"); end; x
tkramer has quit [Quit: Leaving]
<drbrain>
should be "missing y"
<hltbra>
missing x
<drbrain>
I mean, I put a typo in the exception message
<drbrain>
if you x(5) you get no exception
<hltbra>
drbrain: should I use self.host instead?
<drbrain>
I don't think so
<drbrain>
if your patch works, please make a pull request
<hltbra>
great then
<hltbra>
because I got a warning with what you requested
<hltbra>
warning: instance variable @host not initialized