In C# or php or other languages, there are 2 ways to pass a value to a function, pass it by value and pass it by referece.
Pass parameter by value make the value copied in the function, so this need a extra memory space although the memory space will be reclaimed after running outside the function.
But passing parameter by reference no need to copy a value, it's save the memory. From this perspective, can we say that using "pass by reference" is always better than "pass by value"?
Pass by reference and pass by value are semantically different and sometimes one is correct approach and sometimes the other one is. In many cases the task at hand already prescribes which approach is needed and in contexts where only one option is supported you often need to manually work around it (e.g., if you need a copy in Java you'll need to clone() the object).
In the context of generic functions the answer is rather the opposite way of your proposed preference: pass arguments of deduced type by value! The reason is that you can use something like std::ref() to obtain reference semantics but there is no way to get value semantics if the functions use reference semantics.
No.
There are tons of cases where you'd want to pass by value.
An example might be when you need both const Type& and Type&& overloads. Passing by value just handles both cases without having to duplicate any code:
void function(Object o) { do_something_with(std::move(o)); }
As opposed to:
void function(Object&& o) { do_something_with(std::move(o)); }
void function(const Object& o) { do_something_with(Object(o)); }
Of course there is much more to the subject, but since you're only asking for "is it always better?" I feel a single disproving example is enough. ;)
Edit: the question was originally tagged c++ hence my very specific answer.
Another, more language-agnostic example would be when you need to make a copy of your parameter because you don't want to modify the original object:
void function(int& val) { int v2 = val; modify(v2); use(v2); }
// vs
void function(int val) { modify(val); use(val); }
You get the idea...
Pass by reference requires copying a reference to the object. If that reference is comparable in cost to the object itself, then the benefit is illusory. Also, sometimes you need a copy of the object, and passing by value provides you one.
Also, there's a key error in the reasoning in the question. If passing by value, and there is no need to copy the value, nothing requires that the value actually be copied. Most languages have an "as-if" rule that states that the program only has to act as if the compiler did what you ask for. So if the copy can be avoided, the compiler is free to avoid it. If the copy can't be avoided, then you needed the copy.
Related
By default, when IPython displays an object, it seems to use __repr__.
__repr__ is supposed to produce a unique string which could be used to reconstruct an object, given the right environment.
This is distinct from __str__, which supposed to produce human-readable output.
Now suppose we've written a particular class and we'd like IPython to produce human readable output by default (i.e. without explicitly calling print or __str__).
We don't want to fudge it by making our class's __repr__ do __str__'s job.
That would be breaking the rules.
Is there a way to tell IPython to invoke __str__ by default for a particular class?
This is certainly possible; you just need implement the instance method _repr_pretty_(self). This is described in the documentation for IPython.lib.pretty. Its implementation could look something like this:
class MyObject:
def _repr_pretty_(self, p, cycle):
p.text(str(self) if not cycle else '...')
The p parameter is an instance of IPython.lib.pretty.PrettyPrinter, whose methods you should use to output the text representation of the object you're formatting. Usually you will use p.text(text) which just adds the given text verbatim to the formatted representation, but you can do things like starting and ending groups if your class represents a collection.
The cycle parameter is a boolean that indicates whether a reference cycle is detected - that is, whether you're trying to format the object twice in the same call stack (which leads to an infinite loop). It may or may not be necessary to consider it depending on what kind of object you're using, but it doesn't hurt.
As a bonus, if you want to do this for a class whose code you don't have access to (or, more accurately, don't want to) modify, or if you just want to make a temporary change for testing, you can use the IPython display formatter's for_type method, as shown in this example of customizing int display. In your case, you would use
get_ipython().display_formatter.formatters['text/plain'].for_type(
MyObject,
lambda obj, p, cycle: p.text(str(obj) if not cycle else '...')
)
with MyObject of course representing the type you want to customize the printing of. Note that the lambda function carries the same signature as _repr_pretty_, and works the same way.
All is in the subject, really.
I fail to see what the difference in behavior is between those two methods for x:
// first version
Method m(ByRef x As whatever)
{
// play with x
}
// second version
Method m(Output x As whatever)
{
// play with x
}
There must be some reason why both those modifiers exist, however my "mastery" (uhm) of the language is not enough to understand the difference. I have tried and read the documentation, search it etc, to no avail so far.
So, what is the difference between those two argument modifiers?
Well those are just "prettifiers", they don't do much in terms of actual language behaviour, and only used to provide documentation. Idea is that arguments documented as ByRef provide both input and output, for example you can pass an array to be sorted, and Output arguments only provide output, for example list of errors. Output modifier was introduced later, and a lot of system code still use ByRef for both use cases.
If argument is actually passed by reference is only determined by method caller, and keyword doesn't really matter. You will call your method as ..m(.parameter) to pass variable by reference, and ..m(parameter) to pass variable by value.
I have been reading about methods and functions in Scala. Jim's post and Daniel's complement to it do a good job of explaining what the differences between these are. Here is what I took with me:
functions are objects, methods are not;
as a consequence functions can be passed as argument, but methods can not;
methods can be type-parametrised, functions can not;
methods are faster.
I also understand the difference between def, val and var.
Now I have actually two questions:
Why can't we parametrise the apply method of a function to parametrise the function? And
Why can't the method be called by the function object to run faster? Or the caller of the function be made calling the original method directly?
Looking forward to your answers and many thanks in advance!
1 - Parameterizing functions.
It is theoretically possible for a compiler to parameterize the type of a function; one could add that as a feature. It isn't entirely trivial, though, because functions are contravariant in their argument and covariant in their return value:
trait Function1[+T,-R] { ... }
which means that another function that can take more arguments counts as a subclass (since it can process anything that the superclass can process), and if it produces a smaller set of results, that's okay (since it will also obey the superclass construct that way). But how do you encode
def fn[A](a: A) = a
in that framework? The whole point is that the return type is equal to the type passed in, whatever that type has to be. You'd need
Function1[ ThisCanBeAnything, ThisHasToMatch ]
as your function type. "This can be anything" is well-represented by Any if you want a single type, but then you could return anything as the original type is lost. This isn't to say that there is no way to implement it, but it doesn't fit nicely into the existing framework.
2 - Speed of functions.
This is really simple: a function is the apply method on another object. You have to have that object in order to call its method. This will always be slower (or at least no faster) than calling your own method, since you already have yourself.
As a practical matter, JVMs can do a very good job inlining functions these days; there is often no difference in performance as long as you're mostly using your method or function, not creating the function object over and over. If you're deeply nesting very short loops, you may find yourself creating way too many functions; moving them out into vals outside of the nested loops may save time. But don't bother until you've benchmarked and know that there's a bottleneck there; typically the JVM does the right thing.
Think about the type signature of a function. It explicitly says what types it takes. So then type-parameterizing apply() would be inconsistent.
A function is an object, which must be created, initialized, and then garbage-collected. When apply() is called, it has to grab the function object in addition to the parent.
I mean if there's some declarative way to prevent an object from changing any of it's members.
In the following example
class student(var name:String)
val s = new student("John")
"s" has been declared as a val, so it will always point to the same student.
But is there some way to prevent s.name from being changed by just declaring it like immutable???
Or the only solution is to declare everything as val, and manually force immutability?
No, it's not possible to declare something immutable. You have to enforce immutability yourself, by not allowing anyone to change it, that is remove all ways of modifying the class.
Someone can still modify it using reflection, but that's another story.
Scala doesn't enforce that, so there is no way to know. There is, however, an interesting compiler-plugin project named pusca (I guess it stands for Pure-Scala). Pure is defined there as not mutating a non-local variable and being side-effect free (e.g. not printing to the console)—so that calling a pure method repeatedly will always yield the same result (what is called referentially transparent).
I haven't tried out that plug-in myself, so I can't say if it's any stable or usable already.
There is no way that Scala could do this generally.
Consider the following hypothetical example:
class Student(var name : String, var course : Course)
def stuff(course : Course) {
magically_pure_val s = new Student("Fredzilla", course)
someFunctionOfStudent(s)
genericHigherOrderFunction(s, someFunctionOfStudent)
course.someMethod()
}
The pitfalls for any attempt to actually implement that magically_pure_val keyword are:
someFunctionOfStudent takes an arbitrary student, and isn't implemented in this compilation unit. It was written/compiled knowing that Student consists of two mutable fields. How do we know it doesn't actually mutate them?
genericHigherOrderFunction is even worse; it's going to take our Student and a function of Student, but it's written polymorphically. Whether or not it actually mutates s depends on what its other arguments are; determining that at compile time with full generality requires solving the Halting Problem.
Let's assume we could get around that (maybe we could set some secret flags that mean exceptions get raised if the s object is actually mutated, though personally I wouldn't find that good enough). What about that course field? Does course.someMethod() mutate it? That method call isn't invoked from s directly.
Worse than that, we only know that we'll have passed in an instance of Course or some subclass of Course. So even if we are able to analyze a particular implementation of Course and Course.someMethod and conclude that this is safe, someone can always add a new subclass of Course whose implementation of someMethod mutates the Course.
There's simply no way for the compiler to check that a given object cannot be mutated. The pusca plugin mentioned by 0__ appears to detect purity the same way Mercury does; by ensuring that every method is known from its signature to be either pure or impure, and by raising a compiler error if the implementation of anything declared to be pure does anything that could cause impurity (unless the programmer promises that the method is pure anyway).[1]
This is quite a different from simply declaring a value to be completely (and deeply) immutable and expecting the compiler to notice if any of the code that could touch it could mutate it. It's also not a perfect inference, just a conservative one
[1]The pusca README claims that it can infer impurity of methods whose last expression is a call to an impure method. I'm not quite sure how it can do this, as checking if that last expression is an impure call requires checking if it's calling a not-declared-impure method that should be declared impure by this rule, and the implementation might not be available to the compiler at that point (and indeed could be changed later even if it is). But all I've done is look at the README and think about it for a few minutes, so I might be missing something.
I had a doubt
I know that main difference between a function and procedure is
The function compulsory returns a value where as a procedure may or may not returns value.
But when we use a function of type void it returns nothing.
Can u people please clarify my doubt.
Traditionally, a procedure returning a value has been called a function (see below), however, many modern languages dispense with the term procedure altogether, preferring to use the term function for all named code blocks.
Read more at Suite101: Procedure, subroutine or function?: Programming terminology 101 - a look at the differences in approach and definition of procedures, subroutines and functions. http://www.suite101.com/content/procedure--subroutine-or-function--a8208#ixzz1GqkE7HjE
In C and its derivatives, the term "procedure" is rarely used. C has functions some of which return a value and some of which don't. I think this is an artefact of C's heritage where before the introduction of void in ANSI C, there was no way to not return a value. By default functions returned an int which you could ignore (can still) and might be some random number if no explicit return value was specified.
In the Pascal language family, the difference is explicit, functions return a value and procedures don't. A different keyword is used in each case for the definition. Visual Basic also differentiates with functions and subroutines(?).
Since we are talking about Objective-C, there are some further issues to confuse you. Functions associated with a class or object are known as "methods" (class methods and instance methods respectively).
Also, if we are being pedantic, you don't call Objective-C methods, you invoke them by sending a message to the object. The distinction is actually quite important because the message name (aka "selector") does not necessarily always refer to the same method, it can be changed at run time. This is fundamentally different to languages like Java and C++ where a particular method name for a particular class is really just a symbolic name for the address of the block of code constituting the body of the method.
Depending on the programming language, the distinction may be not so clear. Let's take a conservative language, Pascal:
procedure indeed has no return value. It is used for operations which do not have a return value, or have multiple return values. In the latter case, multiple arguments (the return-arguments or output-arguments) are passed by reference (using the var keyword) and their values are directly modified from inside the procedure. (Note that this latter case may not be considered good practice, depending on the circumstances).
function has a single return value, and usually we do not expect it to change the value of any of its arguments (which arguments may then be passed by value, or via the const keyword). Multiple return values may be returned by bundling them into a record.
C or Java does not distinguish syntactically, so a function of return type void can be thought of as a procedure. Scala distinguished between them by the presence of an equals sign between the method head and method body.
Generally, no matter how an actual language calls its construct, we would ideally expect that
A function takes arguments, doesn't modify any state (like mutating arguments, global variables, or printing info for the user to the console), and returns the result of computation.
A procedure takes arguments, performs operations which can have side-effects (writing to a database, printing to the console, maybe mutating variables), but hopefully doesn't mutate any arguments.
In practice however, depending on the situation, blends of these expectations can be observed. Sticking to these guidelines helps I think.