Re: += in ada

[Craig Carey <research@ijs.co.nz>]

On Wed, 01 Oct 2003 09:35:59 +0100, Peter Amey
<peter.amey-@-praxis-cs.co.uk> wrote to comp.lang.ada:
>Subject: Re: += in ada"
>>>
>>>    count = count + 1[;]
>>>
>>>The latter grates on my minimalist sensibilities like fingernails on a
>>>chalkboard

The arguments in favour of += that come from C programmers seem to
be arriving at comp.lang.ada.
They should be unpersuasive since ignoring the "renames" replacement for
the possible but missing "+=" operator.

Suppose this Ada code in the proposed C style, and += and -= are used:

01   A.Bcd += 1;
02   A := Z;
03   A.Bcd -= 1;

Now let A be pointer to a record containing a field named 'Bcd'.
This is the same code in Ada 95. Unlike real declarations, the renames
statement can associate a single variable name with any element of a
shiftinglist of possible memory locations. Ada has the much needed goto
which allows a solution that seems to be overly verbose:

01   No_Redo_Renaming : Boolean;
   ...
02   No_Redo_Renaming := False
03   <<REDO_RENAMING>>
04   declare
05      R : Integer renames A.all.Bcd;
06   begin
07      if No_Redo_Renaming then
08         R := R + 1;
09         A := Z;
10         No_Redo_Renaming := False;
11         goto REDO_RENAMING;
12      else
13         R := R - 1;
14      end if;
15   end;


If R is intended to denote all the different R's then the code need can
be briefer.

Arguments for new features that are inconsiderate could fail make it
all the way out to the end. Advocacy of += has so far failed to say
what is wrong with this syntax:

01   declare R : Integer renames A.all.Bcd; end;
02   R := R + 1;
03   A := Z;
04   declare R : Integer renames A.all.Bcd; end;
05   R := R - 1;

Variations on that could include:
 * "end" is replaced with "end declare;"
 * "declare is replaced with "begin declare".

What C++ programmers might now want, is to give up on saying that Ada
needs a += operator, since it seems to be a hopeless cause.

-----

This topic is also in the large thread that now has over 437 messages.
A message there that allocates failure to the "+=" proposal is this:


At 27 Sep 2003 17:43:42 -0500, "Randy Brukardt" <randy-@-rrsoftware.com>
 wrote to comp.lang.ada:
>Subject: Re: Is the Writing on the Wall for Ada? [although this thread
>      changed to something else a long time ago]
...
>In any case, I'm sympathic to the issue, but (a) I don't like this syntax,
>and there are other solutions, and (b) there are a hundred issues more
>important. I note that you didn't answer my note asking what you'd give up
>for this rather substantial change.
>
>>I've seen some weak arguments here, and this one is typical. What is
>>increasingly apparent to me is that Ada veterans are set in their
>>ways, and no amount of reason will budge them on the basic syntax of
>>Ada, even when if is deficient compared to the existing languages that
>>98% of programmers use.
...

It could be a joke: the set of unreasonable veterans is empty but the
description description parallels what suggesting big changes to C++
projects.

Mr Dewar here states that there is a need for some to get a "$" operator:

----------------------------
At 2002-11-09 22:10 -0500 Saturday, Robert Dewar wrote to
   ada-comment@ada-auth.org:
...
>Note incidentally that the grade school idea does not work for
>division, since the / operator has nothing to do with mathematics,
>where the horizontal line used for division uses a graphical technique
>to indicate precedence. In SNOBOL4 Griswold changed / to have lower
>precedence than multiplication
>
>so a*b / c*d means what it looks like, but this sure confused the
>Fortran folk.
>
><<P.S. This is fun.  I again apologize that it has little to do with the
>topic at hand.
>>>
>
>I am not sure that is the case. After all the actual topic at hand is
>useful additions to the language. Maybe allowing $ to be defined as an
>operator is just as interesting as bit string set packages :-)
>
----------------------------

Using my syntax

begin
   declare function "$" (L : lhs T; R : constant T) return T
      renames "~#="; end;

It might be nice to have constant to be like the "in" mode, except that
that objects underneath pointers are also constant, which better hide
when the objects are pointers or records.



>Ada has real technical issues that need fixing,


Some is online here:  http://www.ada-auth.org/ai-files/REST_AIS.ZIP


-----

On 24 Sep 2003 17:42:06 -0700, 18k11tm001-@-sneakemail.com (Russ) wrote
 to comp.lang.ada:

... which of the following is more readable [?]:
>
>lwienfowowoenfnowoqndfoowopqihjefhnowqoowldvno := 
>     lwienfowowoenfnowoqndfoowopqihjefhmowqoowldvno + 1
>
>or
>
>lwienfowowoenfnowoqndfoowopqihjefhnowqoowldvno += 1


Swapped an "n" with an "m".



Craig Carey