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,
	MSGID_RANDY autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: 103376,c63aa81a67eceb8f
X-Google-Attributes: gid103376,public
From: Robert Dewar <robert_dewar@my-deja.com>
Subject: Re: Ragged Array Proposal
Date: 1999/09/25
Message-ID: <7sijl6$ta0$1@nnrp1.deja.com>#1/1
X-Deja-AN: 529392918
References: <37e7c08e@eeyore.callnetuk.com> <37EA9A72.594ED8F5@mitre.org>
 <37ebb120@eeyore.callnetuk.com>
X-Http-Proxy: 1.0 x40.deja.com:80 (Squid/1.1.22) for client 205.232.38.14
Organization: Deja.com - Before you buy.
X-Article-Creation-Date: Sat Sep 25 13:45:12 1999 GMT
X-MyDeja-Info: XMYDJUIDrobert_dewar
Newsgroups: comp.lang.ada
X-Http-User-Agent: Mozilla/4.04 [en] (OS/2; I)
Date: 1999-09-25T00:00:00+00:00
List-Id: <comp.lang.ada>

In article <37ebb120@eeyore.callnetuk.com>,
  "Nick Roberts" <nickroberts@callnetuk.com> wrote:

> Fruits: constant array (Positive range <>) of Unbounded_String
:=
>     (To_Unbounded_String("Apple"),
>      To_Unbounded_String("Orange"),
>      To_Unbounded_String("Pear"),
>      ...);
>
> This technique suffers from the same heavy syntax of the
allocator
> technique,

Well as I noted in an earlier message, the "light" syntax of
C hides a lot of sins, it is always puzzling to people who
are using C at first, that they cannot declare the Fruits
array as in our example, and then say


   if (Fruits [j] == "abc") ...

of course they *can* say it, it is perfectly legal and well
defined in C, it just does not do what the programmer expects
(generally the result will be false).

If you really get so upset by reading To_Unbounded_String, then
replace it with a unary "+" as in my previous example (I
actually wanted the unary "+" in the standard here, but
that proposal did not attract enough support). Indeed both
Jean Ichbiah and I have always liked the idea of using unary
plus as a quietish, but still explicit, conversion operator,
and we both pushed for adding one extra unary operator (obvious
choice the pillow or currency conversion symbol) which would
have no predefined meaning, but would be specifically intended
as a conversion operator of this kind, but that again did not
attract enough support.

> and while it saves having to explicitly declare an access
> type, it's almost certain to have an implementation that
> involves dynamic allocation (together with the further baggage
> of a controlled type).

Sure, but you would be surprised what a large percentage of
string processing stuff is quite happy with such overhead
(think of SPITBOL, and even Java).

> A very clever optimiser might be able to perform
> pre-allocation (as mentioned in
> other posts in this thread), but I suspect that most Ada
compilers, in
> reality, would not.

Waste of time I would think ...


Robert Dewar


Sent via Deja.com http://www.deja.com/
Before you buy.