00:58 <postmodern> why doesn't bundler binstubs support the _VERSION_ argument?

00:58 <postmodern> at least check if the _VERSION_ argument matches what bundler's gem version

01:00 <postmodern> hmm is there a way to detect if the process is under bundler's control?

01:01 <postmodern> that way I can support system('foo',...) vs. system('foo','_VERSION_',...)

14:30 <Zero_Chaos> Is there a way to tell bundler where to put/read Gemfile.lock when it is generated? I need it to go in ~/ or something

15:41 <indirect> Zero_Chaos: BUNDLE_GEMFILE=/path/to/Gemfile bundle install

15:43 <Zero_Chaos> indirect: hmm, if I do that in env when I run install, how does it know to find it there later?

15:43 <t4nkd> Is there any reason why bundler would fail when compiling C extensions but rubygems would do just fine? The compile logs are saying that the newly created file has an older date than the distributed files

18:20 <indirect> Zero_Chaos: you have to always set that env (or pass --gemfile=/path) every time you run the command, iirc

18:20 <Zero_Chaos> indirect: also that path will be for the Gemfile and Gemfile.lock right? I cannot separate them?

18:22 <indirect> Zero_Chaos: iirc, yes, you can't separate them

18:22 <indirect> why do you want a lock separate from a gemfile?

18:24 <Zero_Chaos> indirect: because I'm packaging an app that uses bundler and I'm basically only using bundler to load the correctly versioned gems. So I don't want people editing the Gemfile, but it would be stupidly convenient to remove the Gemfile.lock and force it to regenerate every time the application is run because the gems on the system might change (but still meet the requirements in the Gemfile)

18:25 <Zero_Chaos> indirect: and yes, I realize that's not the design but it actually works really well thus far.

18:32 <indirect> Zero_Chaos: so write a script to boot your app that removes the gemfile?

18:34 <Zero_Chaos> indirect: that's the problem, the application runs as user and the Gemfile and Gemfile.lock are currently unwritable as the user.

18:35 <Zero_Chaos> indirect: moreover, if you run as one user, then another, that won't be permitted very easily.

18:35 <Zero_Chaos> indirect: I was hoping it might be possible to have the Gemfile.lock in some random spot (or /tmp) and avoid wierd scripts.

19:52 <crankharder> anyone else think it's a bit broken of bundler to attempt to update all dependent gems even if that update isn't required?

19:53 <crankharder> seems like it inadvertantly causes unnecessary changes that I'd expect