Race() accepts an empty array #57

Activity

1. jsor commented on May 4, 2016

   

   jsor

   on May 4, 2016

   Member

   More actions

   See domenic/promises-unwrapping#75
2. jsor

   added this to the v3.0This milestone has been deleted milestone on May 4, 2016
3. joshdifabio commented on May 4, 2016

   

   joshdifabio

   on May 4, 2016

   ContributorAuthor

   More actions

   I'm struggling to see the value in the never resolved promise approach, despite the arguments made in the linked discussion, which I think are fairly abstract. Can anyone think of a case that race([]) would happen other than in error? What do you think, @jsor, @clue?
4. joshdifabio commented on May 4, 2016

   

   joshdifabio

   on May 4, 2016

   ContributorAuthor

   More actions

   Note that both icicle and amp throw exceptions in such cases:

   https://github.com/amphp/amp/blob/v1.1.0/lib/functions.php#L388
   https://github.com/icicleio/icicle/blob/v0.9.6/src/Awaitable/functions.php#L301
5. jsor commented on May 4, 2016

   

   jsor

   on May 4, 2016

   Member

   More actions

   I'd vote for either

   1. Keep race(), make it ES6 compliant and add a warning to the docs (example).
   2. Remove race() in 3.0.
6. joshdifabio commented on May 4, 2016

   

   joshdifabio

   on May 4, 2016

   Last edited by joshdifabio

   ContributorAuthor

   More actions

   I'd vote for either

   1. Keep race(), make it ES6 compliant and add a warning to the docs (example).
   2. Remove race() in 3.0.

   What's the thinking behind removing it?

   Edit: The bluebird docs referenced in your post explain your reasoning.
7. jsor commented on May 4, 2016

   

   jsor

   on May 4, 2016

   Member

   More actions

   @joshdifabio yes 👍
8. clue commented on May 15, 2016

   

   clue

   on May 15, 2016

   Member

   More actions

   I'm struggling to see the value in the never resolved promise approach, despite the arguments made in the linked discussion, which I think are fairly abstract. Can anyone think of a case that race([]) would happen other than in error?

   Personally, I haven't had a use for the race() function so far, but this behavior is surprising to say the least, so it makes perfect sense raising this issue 👍

   For bluebird it certainly makes sense to follow ES6. However, this doesn't necessarily apply to this library, considering interoperability with javascript world is a rather moot point.

   I'm currently undecided, but I think we should address this somehow, docs could be a first step.

   if a necessary feature has a high astonishment factor, it may be necessary to redesign the feature
   https://en.wikipedia.org/wiki/Principle_of_least_astonishment
9. jsor commented on May 16, 2016

   

   jsor

   on May 16, 2016

   Member

   More actions

   For bluebird it certainly makes sense to follow ES6. However, this doesn't necessarily apply to this library, considering interoperability with javascript world is a rather moot point.

   Not sure what you are proposing by "not necessarily follow ES6": remove race or keep it in its current implementation?
10. clue commented on May 20, 2016

    

    clue

    on May 20, 2016

    Member

    More actions

    Giving I have no current use for race(), I'm undecided.

    Does removing it altogether really resolve the issue here? :)

    My main concern here would be lack of documentation, but I'm good with either solution 👍
11. jsor

    added a commit that references this issue on Jan 8, 2017

    Remove race()

    ...

    This issue will close once commit 7a797a5 is merged into the '3.x' branch.

    7a797a5
12. jsor

    closed this as completedin #83on Jan 14, 2017
13. clue

    added

    bug

    on Jun 8, 2022