|
you mean like javascript "rooms"  ?
#SupportHeForShe
Government can give you nothing but what it takes from somebody else. A government big enough to give you everything you want is big enough to take everything you've got, including your freedom.-Ezra Taft Benson
You must accept 1 of 2 basic premises: Either we are alone in the universe or we are not alone. Either way, the implications are staggering!-Wernher von Braun
|
|
|
|
|
TheGreatAndPowerfulOz wrote: you mean like javascript "rooms"
Yes, exactly like that. 
|
|
|
|
|
I'm learning Swift.
It's an interesting new(er) language that has some interesting new features.
Functions Have External Param Names
One of those features is the naming of external function params.
Background
Most of us are accustomed to named (internal) params. This is analogous to what we have in C# like the following:
int Add(int addend1, int addend2){
return addend1 + addend2;
}
Of course we call it like:
Add(2,3);
Those internal params are the names we use inside the function body. That all makes sense.
What About Swift?
But in Swift you can also name the external params. Actually, you have to name them in your function definition unless you use an underscore to tell the compiler you're not using an external name.
Here's the same function defined in Swift:
func Add(_ addend1: Int, _ addend2: Int) -> Int{
return addend1 + addend2;
}
You can call that method like :
Add(2,3)
Magical Underscore
However, if we do not supply the underscore, then we have to give the external param a name too, like the following:
func Add(a1 addend1: Int, a2 addend2: Int) -> Int{
return addend1 + addend2;
}
So now you have to call the Add function like the following (or it will fail to compile):
Add(a1:2, a2:3)
All That Culminates In This
The very good book I'm reading to learn Swift[^] has an example like the following:
func changeName(of d:Dog, to newName:String) {
d.name = newName
}
What changeName Does
The changeName function takes a Dog class and changes it's name property to the value that is sent in the 2nd param (String).
The external variable for the first param (of type Dog) is of and the internal name of that same param is d. The external name of the 2nd param (String) is to and its internal name is newName.
Now, check out what the final call to that method looks like. It's like readable English but hmm....it feels so odd to be code that it makes me stumble anyways and I'm not sure it really advances understanding.
changeName(of:d, to:"Rover")
Change name of d to rover.
 ? or  ?
modified 19-Nov-18 17:12pm.
|
|
|
|
|
I did kinda like that when I first learned it but then...
I think I'm too steeped in keeping code compact. Not Ruby-esque compact, but the need to explicitly name every input variable started grating.
I do prefer the C# way where naming input parameters is opt-in instead of opt-out.
cheers
Chris Maunder
|
|
|
|
|
Maunder said: the need to explicitly name every input variable started grating.
Yeah i wondered if it might be nifty the first dozen times then become too much.
And I think the example is nifty but so odd to name the vars to help the function call make a natural language sentence.
|
|
|
|
|
Yeah - like a fluid interface. But not really.
cheers
Chris Maunder
|
|
|
|
|
I use this in VB (6, Applications, .Net) sometimes, but I'd prefer to have the option. Most of the time it simply doesn't help the readability of the calling code.
|
|
|
|
|
Gross. So it is foisted onto the development team to making the decision as to which pattern to consistently use: _ so the syntax is changeName(d, "Rover") or to use named external parameters so the syntax is changeName(of:d, to:"Rover"). That is, if it's even a conscious decision.
But of course, the next team / outsource group / other library may use the opposite pattern, so your Swift code will be a smattering of the two, and it seems like the only way to tell which one you need to use is by inspecting the function declaration? Because I seriously doubt whatever editor you use has sophisticated intellisense to help you, or am I wrong in this?
So, what is the benefit of named external parameters? Merely the fact that you can also, if you're Hungarian or like RPN, write changeName(to:"Rover", of:d) ??? Is there any concept of optional parameters that have default values if not set?
And even worse, can you mix the two, so you could write changeName(d, to:"Rover") ?
Latest Article - A Concise Overview of Threads
Learning to code with python is like learning to swim with those little arm floaties. It gives you undeserved confidence and will eventually drown you. - DangerBunny
Artificial intelligence is the only remedy for natural stupidity. - CDP1802
|
|
|
|
|
Yes, you'd have to decide at the outset if you were really going to follow that type of pattern and it could get very mixed up.
Also, XCode does seem to help with a Intellisense type of functionality.
Marc Clifton wrote: And even worse, can you mix the two,
And, yes, you can definitely mix the two. The creator of the original function can use the _ to indicate the external name is not required and can use it on 0 or more params.
I can see that the language creators were trying to allow developers to convey more meaning but not sure external names really help.
It feels like what has recently been done with fluent interfaces[^]
mock.expects(once()).method("m").with( or(stringContains("hello"), stringContains("howdy")) );
The attempt to make it read like natural language. I think it works well with unit testing frameworks, but not sure about all code.
|
|
|
|
|
I remember that Algol 60 (used in my uni. days) had a different approach. When calling a function, you could separate parameters using either ',' or ') <letter string> :(' to give expressions like
CHANGENAME(D) THE DOG CHANGES NAME TO :( NEWNAME) (we only had uppercase characters  ). I didn't use that syntax very much - it was confusing because the ) looked like an end of a parameter list, not the start of an inline comment
|
|
|
|
|
|
I think partly this is a holdover from Objective-C. For example, this Cocoa method call:
UIAlertView* alert = [[UIAlertView alloc] initWithTitle:@"Hello!"
message:@"Hello, world!"
delegate:nil
cancelButtonTitle:@"Close"
otherButtonTitles:nil];
which calls initWithTitle to create a new pop-up with a message for the user. It uses the external notation and if I remember correctly (always problematic) the notation was required on all but the first argument, where it was optional.
This used to be more important when mixing Swift with Objective-C, but now most (if not all) of the various OS APIs are implemented in Swift, so it's less relevant. The exception is if you're using your own Objective-C code with Swift, or slowly converting a project over from one to the other.
|
|
|
|
|
Interesting, thanks for explaining by connecting it back to Objective-C.
|
|
|
|
|
What if it's a cat?
(Using C#, interfaces and "named parms")
Cat cat = new Cat(){ Name = "Whiskers" };
ChangeName( of: cat, to: "Fluffy" );
ChangeName( to: "Fluffy", of: cat );
public STATIC void ChangeName(IRenameable of, string to) {
of.Name = to;
}
public interface IRenameable {
string Name {get;set;}
}
public class Cat: IRenameable {
public string Name {get;set;}
}
"(I) am amazed to see myself here rather than there ... now rather than then".
― Blaise Pascal
|
|
|
|
|
Nice example.
Do you prefer code in that style?
|
|
|
|
|
Wow. I've recently learned Prolog (though I still have a much greater fondness for TLA+[^]) so I appreciate the _ "forget-about-it" convention but wow, requiring named parameters in absence of "anonymous" ones? FeelsBadMan
Just use F#, C#, or others depending on your needs. Apple doesn't do anything well including their phone which is based mostly on advertising.
|
|
|
|
|
Jon McKee wrote: requiring named parameters in absence of "anonymous" ones? FeelsBadMan
I agree.
|
|
|
|
|
|
Quote: IMHO it's requiring coders do what's basically IDE's job.
Ding ding, we have a winner. Did not expect this from Apple.
|
|
|
|
|
Why Apple? Idea is Jet Brains'.
|
|
|
|
|
|
My point was that language designers spent time to add a future which enforces some stupid rule on developers, instead spending that time implementing an IDE plugin which does the job (likely faster and better than developers anyway). Because everybody uses IDE nowadays, right? Uhm... wait.
And for the recored, no-single-liners and no-assigment-in-comparison policies suddenly started to make sense... idiots.
|
|
|
|
|
Wow, I hadn't seen that Apple bug you linked before.
I mean seriously, most C compilers will produce a warning for that, let alone Lint-style tools. For a company in there position not to be using at least those on crucial code like SSL shows that they really have no respect for their users.
"If you don't fail at least 90 percent of the time, you're not aiming high enough."
Alan Kay.
|
|
|
|
|
Smalltalk had similar readability, achieved differently.
In this case, an object kennel may define a method changeNameOf:To: would be invoked as follows:
kennel changeNameOf: d To: 'Fido'.
"If you don't fail at least 90 percent of the time, you're not aiming high enough."
Alan Kay.
|
|
|
|
|
Rob Grainger wrote: Smalltalk had similar readability, achieved differently
Interesting.
So what you're saying is that Swift is trying to copy syntax that is used by a 46 year old language.
Just kidding.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
