[swift-users] if-case syntax ambiguity

[Nicholas Outram]

Thanks for that, it was really helpful. I see it now.

I’m happy to admit I had not fully appreciated the notion of pattern matching, and I very much doubt I'm alone either.

The Swift Programming book goes some way to explain it (in the language reference section) - maybe this is an area that needs some attention (from an educational perspective).

Many thanks for replying.

Nick

p.s. I looked up some Erlang examples - it did help 








> On 9 Nov 2016, at 12:50, Michael Nisi <michael.nisi at gmail.com> wrote:
> 
> Yes, equals isn't assignment but pattern matching here. Reminds of Erlang, which I love.
> 
> Michael
> 
> On 9 Nov 2016, at 12:57, Zhao Xin via swift-users <swift-users at swift.org <mailto:swift-users at swift.org>> wrote:
> 
>> The `if case` is the same meaning as `switch-case`, so I don't think there is anything ambitious. For `switch-case`, it is not equal, it is matching.
>> 
>> Zhaoxin
>> 
>> On Wed, Nov 9, 2016 at 7:17 PM, Nicholas Outram via swift-users <swift-users at swift.org <mailto:swift-users at swift.org>> wrote:
>> Hi
>> 
>> I’ve been drilling down on the syntax of enumerated types with associated data in the current release version of Swift 3.
>> I’ve pasted below a section of a Playground that captures an issue I’d like to raise.
>> 
>> In summary:
>> 
>> Consider the following 
>> enum Vehicle {
>>    case car(petrol: Bool, sizeCC: Int)
>>    case plane(engines : Int)
>>    case other(String)
>>    case none
>> }
>> let myJourney : Vehicle = .other("pogo stick")
>> 
>> Whereas the following is clear
>> if case .other(_) = myJourney
>> 
>> the following shorthand equivalent is potentially confusing for the sake of 3 characters
>> if case .other = myJourney
>> 
>> - In the first case, the presence of the underscore does communicate that something is being assigned, but dropped.
>> - In the second case, the reader could easily be mislead into thinking that = was supposed to be == as there no apparent place to assign anything.
>> 
>> My suggestion would simply be to drop the shorthand as it’s ambiguous?
>> 
>> 
>> Nick Outram
>> 
>> 
>> 
>> 
>> import Foundation
>> 
>> //: Consider the following enumerated type with associated data
>> enum Vehicle {
>>    case car(petrol: Bool, sizeCC: Int)
>>    case plane(engines : Int)
>>    case other(String)
>>    case none
>> }
>> 
>> //: Let's pick an example
>> let myJourney : Vehicle = .other("pogo stick")
>> 
>> //: I now want to test what case `myJourney` is.
>> //:
>> //: We cannot use the `==` operator because `Vehicle` has associated data. Instead we use `if case` and *simply drop the associated value* with `_` as shown above
>> if case .other(_) = myJourney {
>>    print("Somewhere nice?")
>> } else {
>>    print("Ok, it's a secret?")
>> }
>> //:The above is clear enough once you get used to the syntax. The `_` communicates that a value has been dropped.
>> //:
>> //: **However**, Swift 3 allows us to drop the parenthesis altogether and use the following shorthand:
>> if case .other = myJourney {
>>    print("Somewhere nice?")
>> } else {
>>    print("Ok, it's a secret?")
>> }
>> //: *Unlike the previous example, I do wonder if this is a language feature that needs review?*
>> //:
>> //: - On face value, reading this code as is there is an assignment operator `=` with nothing apparently being assigned.
>> //: - It also reads as if `=` should be `==`
>> 
>> 
>> 
>> 
>> 
>> 
>> _______________________________________________
>> swift-users mailing list
>> swift-users at swift.org <mailto:swift-users at swift.org>
>> https://lists.swift.org/mailman/listinfo/swift-users <https://lists.swift.org/mailman/listinfo/swift-users>
>> 
>> 
>> _______________________________________________
>> swift-users mailing list
>> swift-users at swift.org <mailto:swift-users at swift.org>
>> https://lists.swift.org/mailman/listinfo/swift-users <https://lists.swift.org/mailman/listinfo/swift-users>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.swift.org/pipermail/swift-users/attachments/20161109/3620acbb/attachment.html>