bartblast
Keyword inside tuple shorthand syntax {a: 1, b: 2}
This could resolve to {[a: 1, b: 2]}. Was it ever considered to allow such syntax? Notice this: {:abc, a: 1, b: 2} and this: my_fun(:abc, a: 1, b: 2) work as I described. As I understand, the compiler is able to figure out that the second part is a keyword list.
I can see that it is similar to %{a: 1, b: 2}, and it may be misleading for people coming from Ruby (hash syntax).
Context: I’m working on Hologram’s template engine. If the shorthand syntax worked for tuples as well, it would simplify a few things since Hologram uses tuples for interpolation and prop syntax, e.g.
<MyComponent my_prop={:my_value} />
or
<button $click={:do_something, a: 1, b: 2}>Click me</button>
or shorthand class attribute syntax:
<div class={button: true, "button-active": false}></div>
This issue can be handled by the template parser, but I feel that it may be beneficial to support it in the Elixir compiler.
Most Liked
al2o3cr
My supported-only-by-vibes take is that when a user writes {a: 1, b: 2} they are far more likely to be typoing either a map literal (missing %) or a kwlist literal (wrong kind of bracket). 
LostKobrakai
Looking at Syntax reference — Elixir v1.16.3 I’m wondering if the syntax sugar in tuples should rather be removed. The atom shorthand within […] and %{…} works the same, but is considered a feature available for both those types and is not available for tuples.
The syntax sugar for skipping square brackets is documented to apply for the last parameter of calls. Creating a tuple is not a call. The closest comparison to the creation of a tuple would be the creation of a binary <<…>>.
Eiji
@gregvaughn Yes, it’s indeed confusing - the syntax, see:
iex> {:do_something, a: 1, b: 2}
{:do_something, [a: 1, b: 2]}
iex> {a: 1, b: 2}
** (SyntaxError) invalid syntax found on iex:2:1:
error: unexpected keyword list inside tuple. Did you mean to write a map (using %{...}) or a list (using [...]) instead? Syntax error after: '{'
│
2 │ {a: 1, b: 2}
│ ^
│
└─ iex:2:1
(iex 1.17.0-rc.1) lib/iex/evaluator.ex:295: IEx.Evaluator.parse_eval_inspect/4
(iex 1.17.0-rc.1) lib/iex/evaluator.ex:187: IEx.Evaluator.loop/1
(iex 1.17.0-rc.1) lib/iex/evaluator.ex:32: IEx.Evaluator.init/5
(stdlib 6.0) proc_lib.erl:329: :proc_lib.init_p_do_apply/3
How about something like this?
<div {:class, button: true, "button-active": false}></div>
Eiji
It’s more rather about consistency. As mentioned this is supported in other cases, so I think same i.e. it should be supported in tuples.
sodapopcan
I can get on board with the consistency argument but I think the desire for this would be realllllly niche. I’ve had the exact thought as @bartblast were I was thinking it would be nice shorthand for this to work in HEEx, but that is literally the only time. I think it’s more beneficial not introduce the ambiguity of %{a: :b} vs {a: :b} for a niche purpose.
Popular in Proposals: Ideas
Other popular topics
Latest on Elixir Forum
Sponsor Spotlight
Practical resources that improve the lives of professional developers. By developers, for developers.
Categories:
Sub Categories:
Forums
Our Sponsors
Build Elixir applications with speed and confidence.
Supporting innovation across the BEAM ecosystem.
We build reliable cloud platforms for business-critical systems.
Catch errors, track performance, monitor hosts and more.
Practical resources that improve the lives of professional developers.
Producing high quality Elixir screencasts since 2017.
The deployment platform built for Elixir: PaaS ease, VPS control.
Courses that'll move you from confusion to "Aha, now I get it!"