From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID
	autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,998965b11075593f
X-Google-Attributes: gid103376,public
From: Matthew Heaney <matthew_heaney@acm.org>
Subject: Re: Beginner's question (array parameters to functions)
Date: 1998/11/29
Message-ID: <m3iufyykdy.fsf@mheaney.ni.net>#1/1
X-Deja-AN: 416712416
Sender: matt@mheaney.ni.net
References: <365F4360.F7E85C1E@columbia.edu> <m3iug0bbpf.fsf@mheaney.ni.net>
 <73qc0p$24e$1@mulga.cs.mu.OZ.AU>
NNTP-Posting-Date: Sun, 29 Nov 1998 00:26:18 PDT
Newsgroups: comp.lang.ada
Date: 1998-11-29T00:00:00+00:00
List-Id: <comp.lang.ada>

fjh@cs.mu.oz.au (Fergus Henderson) writes:

> Matthew Heaney <matthew_heaney@acm.org> writes:
> 
> >greg <ref15@columbia.edu> writes:
> >> One requirement of the assignment is that the
> >> sorting be done by a function which takes the array as a parameter and
> >> returns a sorted array.
> ...
> >> My question is the following:  As I understand it, in an Ada function, a
> >> parameter must be passed in "in" mode.  Therefore, I assume that in this
> >> assignment, the sorting function has to make a copy of the array
> >> parameter, sort the items in this second array, and finally return this
> >> second array to the caller subprogram, because, given that the array is
> >> passed as a parameter, there is no way to change the values of the array
> >> that is copied in for use in the function.
> >> 
> >> Is my understanding correct, or is there some way to avoid having to use
> >> two arrays?
> >
> >But there are two arrays no matter what: one is unsorted
> 
> When the sort function is called.
> 
> >and one is sorted.
> 
> When the sort function returns.
> 
> >Having two arrays is a condition of the problem statement,
> >and has nothing to do with the language.
> 
> This is not correct.
> 
> In some languages, e.g. C++, arrays can be passed and returned by
> reference, and can be modified, so (unless there was some additional
> constraint in the assignment, e.g. that the function's input argument
> not be modified) the two arrays could be the same array.

That is precisely that case.  The requirement is for a function, not a
procedure.  Therefore, there is a requirement that the original array
not be modified.

The problem is perhaps more clearly stated as like this.  "Given this
specification of this Ada function:

   function Sort (Items : Item_Array) return Item_Array;

Implement the body of this Ada function."

Implementing the body Sort requires implies the existence of another
array; specifically, the return value of the function.  When you call
the sort function:

  First_Array_Which_Is_Unsorted : Item_Array;
  ...
  Second_Array_Which_Is_Sorted : constant Item_Array :=
    Sort (First_Array_Which_Is_Unsorted);

There are two arrays.

Here are excepts from the original post:

(start of quote)
I have a homework assignment which involves writing an Ada program that
sorts an array of items.  One requirement of the assignment is that the
sorting be done by a function which takes the array as a parameter and
returns a sorted array.

My question is the following: As I understand it, in an Ada function, a
parameter must be passed in "in" mode.  Therefore, I assume that in this
assignment, the sorting function has to make a copy of the array
parameter, sort the items in this second array, and finally return this
second array to the caller subprogram, because, given that the array is
passed as a parameter, there is no way to change the values of the array
that is copied in for use in the function. [Matt - yes, everything in
the paragraph is true.]

(I realize that it would be possible to use a procedure instead of a
function, or to use a global variable rather than pass any parameter to
the sorting function, but let's assume that that would not be acceptable
in this assignment.)  [Matt - here is the explicit statement on the
requirement for a function, not a procedure.]

Is my understanding correct, or is there some way to avoid having to use
two arrays?  [Matt - yes, your understanding is correct.]
(end of quote)

> The fact that these two arrays have different contents is not a contraction,
> because the different contents occur at different times, and it's quite
> possible for a single array to have different contents at different times.

I don't understand this paragraph.