Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Dec 15 2019 14:37
    mpedrera removed as member
  • Jun 19 2019 12:30
    groyoh closed #98
  • Apr 05 2018 07:35
    plexus commented #20
  • Apr 03 2018 16:51
    DanielKuhn commented #20
  • Nov 13 2017 15:01

    plexus on master

    Travis: update ruby versions. M… (compare)

  • Nov 13 2017 14:58
    plexus closed #126
  • Nov 13 2017 14:58
    plexus commented #126
  • Nov 13 2017 14:56

    plexus on v0.13.0

    (compare)

  • Nov 13 2017 14:38

    plexus on master

    Bump version to 0.13.0 (compare)

  • Sep 23 2017 18:59
    jnylen opened #126
  • Jun 07 2017 07:15
    coveralls commented #110
  • Mar 16 2017 14:30
    fnordfish opened #125
  • Mar 07 2017 15:25
    plexus commented #124
  • Mar 07 2017 15:24
    plexus closed #123
  • Mar 07 2017 15:24

    plexus on master

    Update to anima 0.3.0 This dep… update sinatra dependency to al… remove old rubies from travis m… and 1 more (compare)

  • Mar 07 2017 15:24
    plexus closed #124
  • Mar 07 2017 15:24
    plexus commented #124
  • Mar 07 2017 15:22
    fnordfish commented #124
  • Mar 07 2017 15:20
    fnordfish edited #124
  • Mar 07 2017 15:18
    fnordfish commented #124
Arne Brasseur
@plexus
great for the motivation to see someone take interest :)
Piotr Solnica
@solnic
I might have a use case for yaks/rom hence my interest
Arne Brasseur
@plexus
well the good has already been done :)
Piotr Solnica
@solnic
my friend has an API project at work that he’d like to port to ROM, they are using AR/AMS right now and would love to run away from that combo
so yaks came to my mind instantly
Arne Brasseur
@plexus
ah yeah, Yaks was born in a fit of rage over AMS, so there's that
Piotr Solnica
@solnic
I can see that :)
Arne Brasseur
@plexus
are they interested in the hypermedia features, or just want general serialization?
Piotr Solnica
@solnic
they use Ember
so they’re interested in that
Arne Brasseur
@plexus
ah I see, we (Ticketsolve) also use Ember
Piotr Solnica
@solnic
right, I saw it in the README (I think)
Arne Brasseur
@plexus
we had to hack up our own "ember-hypermedia" layer which is actually pretty neat, but hasn't been open sourced (yet)
Piotr Solnica
@solnic
oh!
Arne Brasseur
@plexus
I've been thinking that I should really make a new example app to demonstrate what Yaks can do for you
Piotr Solnica
@solnic
marketing, it’s hard right ;)
Arne Brasseur
@plexus
in that case I could develop the API first, which with Yaks's HTML output is actually really easy and fun
and then build an ember frontend, which would be a good opportunity to extract/redo the hypermedia stuff we have
yeah man the marketing, it's killing me
but I did a lightning talk at RubyConf south africa, and I think people were really impressed with the HTML output, so that could be a good selling point
just added a bit about that to the README. it's cool because with a proper hypermedia API where all interactions are done following links and forms, you can completely consume the API from a browser
so API can be developed and tested separately from a client app
Piotr Solnica
@solnic
sounds fantastic
also, RubyConf South Africa, wow congrats
Arne Brasseur
@plexus
well I wasn't invited or anything :) just decided to go and signed up for a lightning talk
but I've just got my first conference invitation that isn't from someone I know personally, so that's cool
I proposed talking about "Hypermedia in Practice" (more marketing, yay). Still have to hear back if they like it
Piotr Solnica
@solnic
nice :) congrats
Carles Jove i Buxeda
@carlesjove
Hey! Cool to be here. Been busy for the last weeks, but I'm sure this room will help me being back to contributing ;-) So good that you had a chance to introduce Yaks at a conf!
Arne Brasseur
@plexus
I'd actually like to set up some mailing list or possibly a discourse for Yaks, would be a better channel for providing support and discussing features
Piotr Solnica
@solnic
I’m thinking about doing that for ROM. Lotus has a discourse forum already
google mailing lists are horrible
gitter has been doing wonders for the rom community but a forum is needed nevertheless
Arne Brasseur
@plexus
yay ROSSConf :)
Carles Jove i Buxeda
@carlesjove
Yeah! How did it go?
Arne Brasseur
@plexus
still working on stuff
some cool progress, yaks-rail, fix and improve use of attribute, test the README with Ataru
I'm starting a new topic to discuss how to handle errors
Arne Brasseur
@plexus
plexus/yaks#62
Yohan Robert
@groyoh
plexus/yaks#65 I'll fix it and create a pull request
Carles Jove i Buxeda
@carlesjove
Hey Arne, I'm just having a look on this #87 at the moment. Hope I can do something about it.
Yohan Robert
@groyoh
Guys I need your help on something. Basically, I have derive_mapper_from_item which calls .const_get("some_string", false). Mutant will try to mutate the second argument to true and nil. I managed to write a test that fails when set to true, but the behavior of const_get is identical for false and nil. Thus I could not find a proper way to make this code pass mutations except by creating a test that except this function to be called with the second argument set to false. Do you have any idea how to do it any other way?
Arne Brasseur
@plexus
that's an interesting case :)
Arne Brasseur
@plexus
I would say, replace it with nil. If it seems to work on all implementations we can roll with it. But I also created an issue over at Mutant to get some feedback mbj/mutant#326
Yohan Robert
@groyoh
Ok thanks, I will try with nil then ;)
Arne Brasseur
@plexus
Quoting the Mutant author

Mhh, you could always create a test like:

expect(that_object).to receive(:const_get).with(foo, false)

To "nail this specific API version".

But still yeah, we should have a mechanism to whitelist certain mutations. The infrastructure is there, the UI not.

so there you have it. Both are fine for me, nil or the explicit test.
Yohan Robert
@groyoh
I'd prefer to keep the explicit test so that it more clear that this argument should be set a false.
Arne Brasseur
@plexus
:+1: