If anything this seems more suited to be a warning rather than something deprecated and slated for removal. It would be similar to how the compiler suggests pattern matching on an empty list instead of length(list). The difference being that there are performance differences for pattern matching a list and using unless/else is purely confusing to humans.
Regardless of how it would be done, I don’t think this change *should* be done. > On Jan 3, 2019, at 10:26, Chris Keathley <[email protected]> wrote: > > Given that it's a breaking change this wouldn't happen till at least some > future version of Elixir so its not really a real discussion. That said, > generally speaking the amount of times you need to break an existing api are > close to 0 and definitely shouldn't be driven by what the community believes > is a good or bad pattern at the moment. If people want to lock down subsets > of the language then there are tools to do that. I'm not really defending > `unless` `else` or its usage. But for better or worse its a part of the > language and breaking the language should effectively never be an option. > >> On Thu, Jan 3, 2019 at 1:02 PM Josh Adams <[email protected]> wrote: >> I really appreciate this suggestion. It would be a breaking change to the >> API, so that's a negative and implies that it couldn't actually happen until >> Elixir 2.0 I would assume. However, I can confirm that every single time >> I've seen an `unless/else` it has resulted in confusing logic and removing >> the option entirely will just make codebases clearer imho. >> >> Josh >> >>> On Thu, Jan 3, 2019 at 11:26 AM Mikael Muszynski <[email protected]> wrote: >>> I would like to propose the removal of the ability to use `Kernel.unless/2` >>> with an `else`-clause. >>> >>> This proposal is inspired by the seeming consensus between the most >>> prominent style guides for Elixir, along with similar consensus in the most >>> often cited style guides for Ruby (where this ability exists as well). >>> >>> Some examples for Elixir: >>> https://github.com/lexmag/elixir-style-guide#no-else-with-unless >>> https://github.com/christopheradams/elixir_style_guide#unless-with-else >>> https://github.com/rrrene/elixir-style-guide#no-unless-with-else >>> Similar examples for Ruby: >>> https://github.com/rubocop-hq/ruby-style-guide#no-else-with-unless >>> https://github.com/airbnb/ruby#no-unless-with-else >>> https://shopify.github.io/ruby-style-guide/ >>> I have not researched the topic to the degree that I can point to some >>> scientific study regarding how comprehensible negated booleans are. >>> However, if the style guides are an indicator of community preferences, >>> then anywhere they are enforced, `unless-else` will likely be banned. >>> >>> I have nothing but intuition to go by when it comes to how many Elixir >>> users are actually using style guides like the ones above (with or without >>> automatic checking), but the people and companies I've worked with >>> generally use these guides. >>> >>> Therefore, this proposal is mostly a question whether this one rule should >>> be enforced everywhere, at the potential expense of those who don't care >>> much about style guides. I guess that also opens discussion about whether >>> other rules could receive similar treatment, but that's a discussion for a >>> different thread. >>> -- >>> 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/6fd77557-3080-496b-af5f-fe51cf9de79b%40googlegroups.com. >>> For more options, visit https://groups.google.com/d/optout. >> >> >> -- >> Josh Adams >> -- >> 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/CAA1-O0wBG_KDg140BCgBmRKN5ggpJoYjfsK%3D7Lt6qw97NBXtOg%40mail.gmail.com. >> For more options, visit https://groups.google.com/d/optout. > > > -- > Chris > -- > 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/CAGjNwxosn5L_EMuuBrpnCYSV91U4_QGBDm2URC1%2BCLj1qzM9gA%40mail.gmail.com. > 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/855E9B37-B56F-4821-93EC-D320FB2EFC0A%40sorentwo.com. For more options, visit https://groups.google.com/d/optout.
