indirect changed the topic of #bundler to: Docs! http://bundler.io | Problems? http://bit.ly/bundler-issues | How to help: http://bit.ly/bundler-development | API Dashboard http://cl.ly/SBoH | #bundler logs: http://bit.ly/bundler-logs | Questions will be answered eventually, so hang out for a while
samphippen has joined #bundler
g0bl1n has joined #bundler
patcon has quit [Remote host closed the connection]
axl_ has quit [Quit: axl_]
havenwood has quit [Remote host closed the connection]
<indirect> kc8qvp_: bundler is pretty explicit about running sudo
rjhunter has joined #bundler
g0bl1n has quit [Quit: Ex-Chat]
samphippen has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
axl_ has joined #bundler
<kc8qvp_> indirect: I have mine as nopasswd and it never told me it was running commands as root
axl_ has quit [Quit: axl_]
axl_ has joined #bundler
<pipecloud> kc8qvp_: Who ever thought using nopasswd was a good idea? :\
<kc8qvp_> pipecloud: well the same thing would happen if you ran bundler within a few minutes of using sudo for something else
<kc8qvp_> since that's the default sudo configuration
<pipecloud> kc8qvp_: Gems can package anything they want. Talk to the author if you think that sudo isn't necessary for what they're trying to accomplich.
<pipecloud> accomplish*
<kc8qvp_> what author?
<kc8qvp_> bundler's author?
<kc8qvp_> and, you know what
<kc8qvp_> I've come to realize that it's not gem that's doing it
<kc8qvp_> it's bundler itself
<kc8qvp_> it installs packages system-wide by default, despite this being advised against
<kc8qvp_> and it will run sudo to do so w/o telling you.
<kc8qvp_> terrible design.
<pipecloud> Oh that's not bundler's fault. It's installing to your GEM_HOME, which requires root. So it asks for it.
<kc8qvp_> except it does not ask
<pipecloud> That's exected. You just use nopasswd, but your argument about the defaults makes some sense.
<pipecloud> kc8qvp_: It does ask, you just don't get asked because you use nopasswd.
<kc8qvp_> exactly, so it doesn't ask *me*
<pipecloud> It requests privilege escalation.
<kc8qvp_> or even *mention* it
<kc8qvp_> and I see you realize this affects more than just people who have nopasswd set
<pipecloud> My rubies and $GEM_HOME are owned by my user.
<pipecloud> So this never effects me.
<kc8qvp_> it really should not run sudo w/o saying so
<kc8qvp_> whether y'all choose to change it or not, I don't care
<pipecloud> So you'd rather bundler fail and say something about insufficient privileges?
<kc8qvp_> of course, yes!
<kc8qvp_> like every other program on my system does!
<kc8qvp_> haha, wtf
deepender has joined #bundler
<pipecloud> Not every one does, but most do.
<pipecloud> kc8qvp_: I'd open an issue with bundler with reproduction steps and as much info as you can provide. Not because this is particularly non-trivial, but it'll get more attention if it's better documented.
kc8qvp_ has left #bundler ["I'll take my trivial concerns elsewhere."]
axl_ has quit [Quit: axl_]
cwebber has joined #bundler
cwebber has quit []
<deepender> tmoore: there?
cwebber has joined #bundler
cwebber has quit []
rjhunter has quit [Remote host closed the connection]
JordanFaust_ has joined #bundler
JordanFaust_ has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
cha1tanya has joined #bundler
rjhunter has joined #bundler
rjhunter has quit [Ping timeout: 240 seconds]
rjhunter has joined #bundler
rjhunter has quit [Ping timeout: 246 seconds]
rjhunter has joined #bundler
rjhunter has quit [Ping timeout: 240 seconds]
r0bby_ has joined #bundler
robbyoconnor has quit [Read error: Connection reset by peer]
<tmoore> deepender: I am now
rjhunter has joined #bundler
rjhunter has quit [Ping timeout: 272 seconds]
<deepender> tmoore: now?
<tmoore> Yes hi
<pipecloud> tmoore: Would it be reasonable to open a feature request that has bundler bomb out with permission errors instead of using sudo to install gems?
<tmoore> I think there might already be one
<tmoore> I can't find it though
<tmoore> It's probably worth opening an issue at https://github.com/bundler/bundler-features to start a discussion
<tmoore> I guess that the drawback would be that if you need to install to a directory that's not writable by your user, then you need to run all of Bundler as root and not just the part that installs the gems
<tmoore> This comment explains some of the issues https://github.com/bundler/bundler/issues/890#issuecomment-588868
rjhunter has joined #bundler
<deepender> sorry tmoore: for late reply
<deepender> life here is very random because of classes
rjhunter has quit [Ping timeout: 245 seconds]
<deepender> just wanted to ask you did you check spec I mentioned there: https://github.com/bundler/bundler/pull/2787#issuecomment-31375959
<tmoore> When I said to write specs I meant rspec... the tests for the functionality
<tmoore> indirect is saying that the messages should be displayed in full if you pass the verbose option, but the code in that pull request doesn't do that
<deepender> it do that it shows message in full
<deepender> does it meant that
<deepender> ohh sorry about spec,I thought you were asking written I am really dumb.
<tmoore> no worries... I should have clarified
<deepender> do you mean this: Using uglifier (2.2.1) from /home/deepender/niveshi/bundle/ruby/2.0.0/specifications/uglifier-2.2.1.gemspec
<deepender> when bundle --verbose is done.
<deepender> its now give this:
<deepender> Using uglifier (2.2.1)
<deepender> tmoore:
<tmoore> just going back to the original discussion at https://github.com/bundler/bundler-features/issues/33
<tmoore> it should print the "Installing" message for each gem it installs if you pass --verbose
<tmoore> sorry I mean the "Using" message
<tmoore> and if --verbose isn't passed, it should print a summary of the number of gems that were already installed
<tmoore> and it should behave the same way in parallel or sequential mode
<tmoore> any other options shouldn't factor into it
cha1tanya has quit [Ping timeout: 245 seconds]
<deepender> cool
<deepender> then it is printing the message using in verbose mode
<tmoore> yeah that's right... it should act just like it does now in verbose mode
<deepender> so only change has to be made
<deepender> to make it work for paralled mode
<deepender> parallel*
<deepender> it works in sequential mode
<deepender> tmoore:
<tmoore> it needs to work in parallel mode, but it also has to work when flags other than verbose are passed
<deepender> tmoore: can I add you on gmail chat list?
<deepender> like
<tmoore> it should not be checking options.length at all
<tmoore> only options[:verbose]
<deepender> but suppose in quiet mode
<deepender> cool
<deepender> so you are saying any option other than verbose it should show the same behaviour
<tmoore> exactly
<deepender> for example there was an error
<deepender> while I tried this something like when bundle update was run then
<tmoore> quiet mode is already handled in cli.rb by setting the log level
<deepender> yeah
<deepender> I agree
<tmoore> actually for that matter, so is verbose mode
<deepender> no
<tmoore> so you probably shouldn't be looking at options at all, just using the appropriate log level
<deepender> are you sure
<deepender> as both bundle install
<deepender> and bundle --verbose give same output
<deepender> I need to look at options
<deepender> I tried it before there were 6 or 7 testcase which got failed
<tmoore> I haven't tested it but it looks that way to me
<tmoore> you can use Bundler.ui.debug to print verbose-only messages
<deepender> I will try this
<deepender> but just as a reason what is the problem with options.length =1
<deepender> all other cases will have the same behaviour
<deepender> as before
<tmoore> Maybe I'm missing something, but what does the number of options have to do with anything?
<deepender> see
<deepender> if if bundle --verbose is applied
<deepender> options.length = 2
<deepender> only in bundle install options length is 1
<deepender> which we need as there is no option passes
<deepender> passed
<tmoore> no that's not right... install is the command not an option
<deepender> yeah
<deepender> that's why onlu option present is no_cache => nil
<tmoore> and 'bundle --local' is one option but 'bundle --verbose --local' is 2
<deepender> in this also length of option will be more than
<deepender> 1
<deepender> {"local"=>true, "no-cache"=>true}
<deepender> I can do that also like bundle[:local]==true||bundle[:verbose] == true
<deepender> and oter
<deepender> other
<deepender> cases
<deepender> like update
<deepender> but with options.length it satisfy all
<tmoore> my point is that 'bundle --local' should show the summary but 'bundle --verbose' should not
<deepender> ohh
<deepender> cool
<deepender> you want bundle --local to also show
<tmoore> and the fact that no-cache is always present is incidental to this feature
<tmoore> and we're actually about to change that back
<deepender> ohh
<tmoore> yeah... so it should only matter whether --verbose is there or not
<deepender> In that case this hack is a bad hack
<deepender> cool
<deepender> tmoore just to summarize:
<deepender> we want bundle, bundle --local, bundle jobs to show
<deepender> summary
<tmoore> yeah we don't want hacks... we just want code that describes how it's supposed to work clearly :)
<deepender> and bundle verbose to show every
<deepender> thing as before
<tmoore> yeah but keep in mind that --verbose could be combined with other parameters
<deepender> then
<tmoore> so if you have parallel jobs without verbose, show the summary
<deepender> I will just write if(bindle[:verbose] == true)
<tmoore> if you have parallel jobs with verbose, show all of the Using methods
<deepender> cool
<tmoore> s/methods/messages/
<tmoore> but on second look, I think that rather than checking :verbose, it should just use Bundler.ui.debug to log the messages
<deepender> okey
<tmoore> for consistency with the rest of the code
<deepender> cool
<tmoore> thanks for taking the time on this one to get it right :)
<deepender> and what about this options clean and all
<deepender> no need for thanks
<deepender> this is a very good project of ruby
<deepender> and I should thank you for devoting your time with a novice like me
<deepender> like --shebang
<tmoore> you should be able to ignore all of the options... other than verbose they won't affect the output we want
<deepender> cool
<deepender> I am not sure then some test also need to be changed
<deepender> they failed initally when I did that
<deepender> may be when I do Bundler.u.debug they will pass
<deepender> rather than using puts
<deepender> cool
<deepender> tmoore: I will try and let you know
<tmoore> Thanks! I'll keep watching the pull request so you can comment there when you push more changes.
<deepender> a bit late but happy new year
<deepender> :)
<tmoore> You too
r0bby_ is now known as robbyoconnor
deepender has quit [Ping timeout: 252 seconds]
<pipecloud> tmoore: I don't see why. I think you'd only need read permission to not have to run it as root all the time.
rjhunter has joined #bundler
<tmoore> I mean that the bundle command does a lot more than just write the gems
<tmoore> it writes the lock file in the current directory for one
axl_ has joined #bundler
axl_ has quit [Client Quit]
deepender has joined #bundler
axl_ has joined #bundler
samphippen has joined #bundler
axl_ has quit [Quit: axl_]
samphippen has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
r0bby_ has joined #bundler
robbyoconnor has quit [Read error: No route to host]
samphippen has joined #bundler
cwebber has joined #bundler
jwrigh13 has joined #bundler
jwrigh13 has left #bundler [#bundler]
chouhoulis has joined #bundler
r0bby_ has quit [Excess Flood]
r0bby_ has joined #bundler
samphippen has quit [Quit: Textual IRC Client: www.textualapp.com]
deivid-rodriguez has joined #bundler
deivid-rodriguez has left #bundler [#bundler]
axl_ has joined #bundler
deepender has quit [Ping timeout: 252 seconds]
patcon has joined #bundler
cha1tanya has joined #bundler
samphippen has joined #bundler
samphippen has quit [Client Quit]
cwebber has quit []
cha1tanya has quit [Ping timeout: 240 seconds]
retro|cz has joined #bundler
samphippen has joined #bundler
patcon_ has joined #bundler
patcon has quit [Ping timeout: 272 seconds]
patcon has joined #bundler
patcon_ has quit [Ping timeout: 245 seconds]
samphippen has quit [Ping timeout: 245 seconds]
samphippen has joined #bundler
deepender has joined #bundler
cwebber has joined #bundler
rjhunter has quit [Remote host closed the connection]
r0bby_ has quit [Excess Flood]
r0bby_ has joined #bundler
deepender has quit [Ping timeout: 260 seconds]
cha1tanya has joined #bundler
g0bl1n has joined #bundler
deepender has joined #bundler
rjhunter has joined #bundler
rjhunter has quit [Ping timeout: 246 seconds]
cha1tanya has quit [Ping timeout: 272 seconds]
cha1tanya has joined #bundler
g0bl1n has quit [Ping timeout: 245 seconds]
havenwood has joined #bundler
ckrailo has quit [*.net *.split]
ereslibre_laptop has quit [*.net *.split]
sneak_ has quit [*.net *.split]
tmoore has quit [*.net *.split]
pipecloud has quit [*.net *.split]
deepender has quit [Ping timeout: 246 seconds]
ckrailo has joined #bundler
sneak_ has joined #bundler
ereslibre_laptop has joined #bundler
tmoore has joined #bundler
pipecloud has joined #bundler
patcon has quit [Remote host closed the connection]
dwradcliffe_ is now known as dwradcliffe
patcon has joined #bundler
cwebber has quit []
patcon has quit [Remote host closed the connection]
patcon has joined #bundler
retro|cz has quit [Ping timeout: 240 seconds]
havenwood has quit [Remote host closed the connection]
havenwood has joined #bundler
g0bl1n has joined #bundler
r0bby_ is now known as robbyoconnor
axl_ has quit [Quit: axl_]
chouhoulis has quit [Ping timeout: 252 seconds]
axl_ has joined #bundler
ckrailo has quit [Ping timeout: 246 seconds]
henrikhodne has quit [Ping timeout: 272 seconds]
axl_ has quit [Quit: axl_]
JSharp has quit [Ping timeout: 272 seconds]
jds_ has quit [Ping timeout: 245 seconds]
ckrailo_ has joined #bundler
denysonique_ has quit [Ping timeout: 245 seconds]
pipecloud has quit [Ping timeout: 246 seconds]
JSharp has joined #bundler
jds_ has joined #bundler
henrikhodne has joined #bundler
cha1tanya has quit [Ping timeout: 260 seconds]
denysonique_ has joined #bundler
gazoombo has quit [Ping timeout: 240 seconds]
henrikhodne has quit [Ping timeout: 264 seconds]
ckrailo_ has quit [Ping timeout: 246 seconds]
axl_ has joined #bundler
ckrailo_ has joined #bundler
JSharp has quit [Ping timeout: 264 seconds]
tmoore has quit [Ping timeout: 246 seconds]
g0bl1n has quit [Read error: Operation timed out]
tmoore has joined #bundler
cwebber has joined #bundler
axl_ has quit [Quit: axl_]
ckrailo_ has quit [Ping timeout: 245 seconds]
ckrailo_ has joined #bundler
patcon has quit [Remote host closed the connection]
havenwood has quit [Remote host closed the connection]