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=-1.9 required=5.0 tests=BAYES_00
	autolearn=unavailable autolearn_force=no version=3.4.4
Path: 
 border1.nntp.dca3.giganews.com!border2.nntp.dca3.giganews.com!border4.nntp.dca.giganews.com!border2.nntp.dca.giganews.com!nntp.giganews.com!usenet.blueworldhosting.com!feeder01.blueworldhosting.com!eternal-september.org!feeder.eternal-september.org!news.eternal-september.org!.POSTED!not-for-mail
From: Simon Wright <simon@pushface.org>
Newsgroups: comp.lang.ada
Subject: Re: Optimizing Ada
Date: Wed, 02 Oct 2013 08:16:02 +0100
Organization: A noiseless patient Spider
Message-ID: <lyvc1gta6l.fsf@pushface.org>
References: <8aca502c-bba8-4af5-9192-459c15fe048b@googlegroups.com>
Mime-Version: 1.0
Content-Type: text/plain
Injection-Info: mx05.eternal-september.org;
 posting-host="efa74e8377ee8d65bf4cce5dc399bc92";
 logging-data="28059"; mail-complaints-to="abuse@eternal-september.org";
 posting-account="U2FsdGVkX19OYaEuKJ7Z6zcEMjiuKzL0RRL8QVdO0Zo="
User-Agent: Gnus/5.13 (Gnus v5.13) Emacs/24.3 (darwin)
Cancel-Lock: sha1:t7a6sn3qidHrEe9XHOf9SAyfFCE=
 sha1:fBdQz13at5yht40OxQ+V1EspIXw=
X-Original-Bytes: 2289
Xref: number.nntp.dca.giganews.com comp.lang.ada:183567
Date: 2013-10-02T08:16:02+01:00
List-Id: <comp.lang.ada>

kennethesills@gmail.com writes:

> So I was just wondering if anyone could help me optimize my Ada code a
> bit, as well as tell me how I'm doing in terms of general code style
> and idiomatic code writing.

Looks pretty good to me (we disagree about having a space between the
subprogram name and the opening paren, but then the ALRM is wrong about
that too :-)

I'm not sure that pragma Inline is meant to work when applied to a
subprogram body, so I'd apply it to the spec (I always write specs,
because I've set the GNAT style options to standard (-gnaty), and that
warns me about missing specs - and about missing spaces, see above).

GNAT may not have implemented your Inline anyway. There's a GNAT pragma
Inline_Always, and there are switches (-gnatn, -gnatN I think) that
affect inlining. The last time I did a check, inlining made my program
slower - cache effects, presumably. That was on powerpc.

AdaCore's house rules don't allow I as a variable name (confusion
potential), they start at J.

I wonder whether your performance problem is caused by using a function?
The internal Word_Map is, I think, going to be copied to the destination
and then finalized. Could you use an out parameter? (remembering to
clear it before adding ne new content).