06:24 <akerl> Hi! I'm running into issues where `bundle exec rake` loads my own gem automatically. It's messing up Coverage, since Coverage then doesn't get stats on my own lib/ files. Is there some change I need to make to get around that?

06:33 <akerl> Ahha, I think I figured it out. I was requiring my gem in the gemspec to get the version, and that was causing bundler to load it

18:51 <Who> indirect: hey

18:54 <indirect> Who: hey there

18:54 <indirect> have I mentioned I am super excited about your patch

18:54 <indirect> :D

18:54 <indirect> (I am teaching a RailsBridge workshop at this moment, so I haven't tried it out yet... sorry)

18:54 <Who> thanks

18:55 <Who> ah

18:55 <Who> I came to ask that

18:55 <Who> well I have been trying various combinations at my end, it seems to be working

18:56 <indirect> that is why I am excited :D

19:54 <Who> indirect: got some good news

19:54 <Who> http://pastebin.com/Duc0EQ2h

19:54 <Who> so I used this Gemfile

19:54 <indirect> yeah?

19:54 <Who> from one of my work projects

19:54 <Who> http://pastebin.com/LedPzh0L

19:54 <Who> http://pastebin.com/sX7hrd9h

19:54 <Who> and the diff

19:54 <Who> http://pastebin.com/yJ5Gearw

19:54 <indirect> (oh man, while you're being totally awesome, want to fix the child dep conflict message? :D :D :D)

19:54 <Who> nothing

19:54 <indirect> whoa

19:54 <indirect> you win

19:54 <indirect> :D

19:55 <indirect> I will run a prerelease with your resolver to solicit feedback, but that's super encouraging!

19:57 <Who> heh thanks

19:57 <Who> sure I would love to fix it

19:57 <Who> pretty late here

19:57 <Who> 1:30 AM

19:57 <Who> I would do it tomorrow (or something)

19:57 <Who> but I wanted to try this code out on a real gemfile

19:58 <Who> oh and before I forget, I also measure the time on a smaller gemfile

19:58 <Who> http://pastebin.com/YWGpws5X

19:58 <Who> real 0m38.036s (with patch)

19:58 <Who> http://pastebin.com/bhz2FNVL

20:04 <Who> Who: oh and before I forget, I also measure the time on a smaller gemfile

20:04 <Who> [01:27am] Who: http://pastebin.com/YWGpws5X

20:04 <Who> [01:27am] Who: http://pastebin.com/bhz2FNVL

20:04 <Who> [01:27am] Who: real 0m38.036s (with patch)

20:04 <Who> [01:28am] Who: real 0m56.524s

20:04 <Who> [01:28am] Who: without

20:04 <Who> [01:28am] Who: of course all gems were already fetched

20:04 <Who> [01:28am] Who: but the later could be slow because of delay in fetching the meta data

20:04 <Who> [01:28am] Who: but performance overall is comparable

20:04 <Who> [01:29am]

20:04 <indirect> yay

20:04 <indirect> that is pretty awesome

20:05 <indirect> we have some pathological cases left over from the last time someone optimized the resolver

20:05 <indirect> I will look up the ticket and mention it on your ticket

20:05 <indirect> if you want to try it out later

20:05 <indirect> thank you SO much for your help :)

20:06 <Who> sure

20:07 <Who> what I am really trying to do with this patch is to get simple algorithm in place which is much easier to understand

20:07 <Who> and test cases for the corner cases

20:07 <Who> with both in place refactoring or improving performance shouldn't it be too difficult

20:08 <indirect> yes

20:08 <indirect> that is amazing :D

20:10 <Who> I will look into the child dep conflict message tomorrow or on monday morning

20:13 <indirect> thanks a ton!

20:13 <indirect> you're great :D

20:14 <Who> oh, I just want to say this in the pull request but I was (and still am) waiting for the pull request to be approved (because you know it might be embarrassing otherwise)

20:14 <Who> can I say it here?

20:15 <indirect> oh, sure

20:15 <indirect> I mean, your pull request is approved... I am just going to look for things that might need to be tweaked

20:15 <indirect> we have wanted an iterative resolver for months

20:15 <indirect> and at least three people have taken stabs at it but not had time to make it work

20:16 <Who> ah, I see

20:19 <Who> well, so here is what I wanted to say. This patch I haven't done it alone, I had Headius giving me pointers (basically he told me to look at Rubygems code, when I told him I am thinking of working on bundler's resolver code) and then I had the amazing Rubygems code written by Evan as my reference (it helped a lot). I also have you guys reviewing my code (and not to forget you created bundler, which helps me at work everyday) and finally the users who actu

20:19 <Who> reported the issue because they use bundler.

20:20 <indirect> totally!

20:20 <Who> so what I mean is, behind a single patch there is lot of things going on and I am just happy to be a part of this

20:20 <indirect> the same is true for all of bundler

20:20 <indirect> but you were totally the first person to actually stick with it until things worked

20:20 <indirect> out of at least five people I know of, including me

20:20 <indirect> so yay you :D

20:22 <Who> thanks again, let's just write high quality software together.

20:23 <Who> and good night for now, today was an exciting day (I had fun testing all these gemfiles :) )

20:25 <indirect> good night!

20:25 <indirect> thanks again for all your testing

20:37 <ddd> hey indirect

20:38 <ddd> ltnt

20:39 <indirect> ddd: hi

20:39 <ddd> how's things?

21:18 <indirect> good I think