Those are both valid syntaxes. I don't think Elixir nor Credo should warn.
If another serialization becomes wide-spread in the future in a way it also
causes confusion with Elixir syntax, should we also deprecate that other
part of the Elixir syntax? The rationale of why it works in Elixir is
because we allow :"such an atom" and keywords keep the same syntax except
":" is used at the end.
If you are writing a lot of JSON, maybe one idea is to use a sigil so you
can literally write JSON?
~j"""
{
json: "this is literally json"
}
"""
The json can be converted to a data structure at compile-time if you don't
have an interpolation. And if you have interpolation, you can probably
change each interpolated argument to encode to json. So you can even do
things such as:
sub = ~j"""
{foo: "bar"}
"""
~j"""
{total: #{sub}}
"""
*José Valim*
www.plataformatec.com.br
Skype: jv.ptec
Founder and Director of R&D
On Thu, Sep 14, 2017 at 2:02 AM, Allen Madsen <[email protected]> wrote:
> I've thought about this and rather than warning when using trailing colons
> on quoted atoms. It could make sense to warn when using quoted atoms when
> the characters are valid for a non-quoted atom.
>
> %{"this": "warns"}
> %{"this one": "doesn't"}
>
> I'm also exploring adding this to credo, in case it doesn't make sense for
> Elixir core.
>
> Allen Madsen
> http://www.allenmadsen.com
>
> On Wed, Sep 13, 2017 at 6:16 PM, Allen Madsen <[email protected]> wrote:
>
>> At work we deal with a lot of JSON and so it's common to have the
>> equivalent of JSON in Elixir. A common problem people on my team have when
>> they write out the Elixir structure is that they miss converting trailing
>> commas into hash rockets.
>>
>> They have this:
>>
>> %{"x": 1}
>>
>> Instead of:
>>
>> %{"x" => 1}
>>
>> Because both look like a string, it's very easy to miss. I'm suggesting
>> warning about the latter syntax as it can be a source of errors.
>>
>> My goal is that it become more obvious that you've made this type of
>> error. If people don't like this suggestion, I'd love to hear other ideas
>> about ways this problem could be addressed.
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "elixir-lang-core" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to [email protected].
>> To view this discussion on the web visit https://groups.google.com/d/ms
>> gid/elixir-lang-core/14344c22-06a8-42c2-a827-29359a826ad0%
>> 40googlegroups.com
>> <https://groups.google.com/d/msgid/elixir-lang-core/14344c22-06a8-42c2-a827-29359a826ad0%40googlegroups.com?utm_medium=email&utm_source=footer>
>> .
>> For more options, visit https://groups.google.com/d/optout.
>>
>
> --
> You received this message because you are subscribed to the Google Groups
> "elixir-lang-core" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> To view this discussion on the web visit https://groups.google.com/d/
> msgid/elixir-lang-core/CAK-y3Cv%2B9yb0XwRh9tBNiKBE%
> 3D09RL9m7F2OXK6pUJxpNRA4%3DFQ%40mail.gmail.com
> <https://groups.google.com/d/msgid/elixir-lang-core/CAK-y3Cv%2B9yb0XwRh9tBNiKBE%3D09RL9m7F2OXK6pUJxpNRA4%3DFQ%40mail.gmail.com?utm_medium=email&utm_source=footer>
> .
>
> For more options, visit https://groups.google.com/d/optout.
>
--
You received this message because you are subscribed to the Google Groups
"elixir-lang-core" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion on the web visit
https://groups.google.com/d/msgid/elixir-lang-core/CAGnRm4JAqPxwr7z6W2tAWgUk%3DEU5hj2LM00tQii5ymC64AJ1XA%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.
