<postmodern>
what do people use instead of HashWithIndifferentAccess?
bf4 has joined #rom-rb
bf4 has quit [Ping timeout: 272 seconds]
<dkubb>
postmodern: I use whatever the web framework provivdes me
<dkubb>
in Rails 4 I use ActionController::Parameters
<dkubb>
if it's just an object interface, I wouldn't use indifferent access. I would just make it a part of the documented interface that I expect the keys to be Strings or Symbols, not both
<lgierth>
dkubb: since mbj just left: i haven't gotten anything useful out of the heisen mutant debugging
<lgierth>
and i'll be at couchhack for the week, so nothing more before next week :)
<lgierth>
s/week/weekend/
<dkubb>
lgierth: was that with one of those mutation that disappeared when you started to look for it?
CraigBuchek has quit [Quit: Leaving.]
<dkubb>
I've only been watching on my perhiphery
<dkubb>
lgierth: btw, how do you like mutant? any ideas for missing mutations?
<lgierth>
the vanilla spec in the beginning, without mutations, was "randomly" failing
<lgierth>
and i was looking into getting more output from rspec
<lgierth>
oh i love it! :)
<lgierth>
i know i noticed a couple of things
<lgierth>
will write those down next week to get some comments. some will probably be cases of non-obvious wrong-doing on my side :)
<dkubb>
yeah, but that's also a good way to learn
<lgierth>
exactly
<dkubb>
I would love to see more discussion about new mutators. every new one we add makes all the software that uses mutant stronger
<dkubb>
once we hit critical mass, in the same way rubocop has, we should be able to add a handful of mutators every month which will continuously break out specs in in ways
lgierth has quit [Read error: Operation timed out]
<dkubb>
I'd like to see an mspec strategy too and see if we can get mutant running against rubyspec and rbx
<dkubb>
dbussink: ^^ do you think there's any interest in doing mutation testing on rubyspec and rbx?
lgierth has joined #rom-rb
<dkubb>
lgierth: you dropped out when I was talking about mutant and rubyspec/rbx :)
<lgierth>
covering the stdlib would be great!
<lgierth>
that'd get you the ruby heroes badge for sure :P
<dkubb>
lgierth: I already won a Ruby Hero award :P
<dkubb>
I can't win again because I judge it now
<lgierth>
hehe i see
<dkubb>
I still think it's worth doing. I would love to see rubysl gems become mutation tested
<dkubb>
even parts of rbx itself could be tested with mutant
<lgierth>
lots of work to do then! :)
<lgierth>
and a way to introduce new contributors
<lgierth>
to the respective projects
<dkubb>
yeah, I likely won't be able to actually *do* the mutation testing, but I think we could get the ball rolling
<lgierth>
there'll be enough mutations for everybody
<lgierth>
yeah
<dkubb>
it'll show people what parts of the gems are still in need of mutation testing
<dkubb>
in addition to coverage testing, which I'm not sure if that's being done or not
<dkubb>
I see coverage testing as a really weak indicator of spec quality; but there's not a lot of point in doing mutation testing on a piece of code until it's covered
<lgierth>
i'm going to bed, heading out to vienna early tomorrow
<lgierth>
that's exactly what i use simplecov for: show me which class to take on next