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=ham
	autolearn_force=no version=3.4.4
X-Google-Thread: 103376,8de7eedad50552f1
X-Google-Attributes: gid103376,public
X-Google-Language: ENGLISH,ASCII-7-bit
Path: 
 g2news1.google.com!news3.google.com!newsfeed2.dallas1.level3.net!news.level3.com!news.illinois.net!attcg1!ip.att.net!news.binc.net!kilgallen
From: Kilgallen@SpamCop.net (Larry Kilgallen)
Newsgroups: comp.lang.ada
Subject: Re: Ada bench : count words
Date: 22 Mar 2005 17:27:33 -0600
Organization: LJK Software
Message-ID: <qetH5JMsyrUV@eisner.encompasserve.org>
References: <ur7ib38sg.fsf@obry.net>
 <pan.2005.03.19.16.57.03.525542@linuxchip.demon.co.uk.uk.uk>
 <87vf7n5njs.fsf@code-hal.de> <umzsy3c8j.fsf@obry.net>
 <423f5813$0$9224$9b4e6d93@newsread4.arcor-online.net>
 <mailman.47.1111454204.23655.comp.lang.ada@ada-france.org>
 <18arnvu705ly4$.1wz6ybz1jt70y$.dlg@40tude.net>
 <mailman.48.1111492666.23655.comp.lang.ada@ada-france.org>
 <1q9cx4jt7802s.k45m6mcntl87$.dlg@40tude.net>
 <wccvf7jdboy.fsf@shell01.TheWorld.com>
NNTP-Posting-Host: eisner.encompasserve.org
X-Trace: grandcanyon.binc.net 1111534012 16226 192.135.80.34 (22 Mar 2005
 23:26:52 GMT)
X-Complaints-To: abuse@binc.net
NNTP-Posting-Date: Tue, 22 Mar 2005 23:26:52 +0000 (UTC)
Xref: g2news1.google.com comp.lang.ada:9767
Date: 2005-03-22T17:27:33-06:00
List-Id: <comp.lang.ada>

In article <wccvf7jdboy.fsf@shell01.TheWorld.com>, Robert A Duff <bobduff@shell01.TheWorld.com> writes:

> Text_IO is pretty broken, from both a usability perspective and an
> efficiency perspective.  I normally avoid it, and use pragma Import on
> the C routines (open, read, etc).  Suitably wrapped in a clean
> interface, of course.

Do those C routines handle line lengths, page lengths, and column
specifications ?

If not, perhaps you are looking for something different from the full
feature set of Text_IO.

But how Text_IO performs from an efficiency standpoint should depend
upon the particular implementation.  If you have a file system that
does not maintain record boundaries, then Text_IO must do more work.
But _something_ must handle the differences between various text file
formats (Stream-LF, Stream-CR, Stream-CRLF and out-of-band record
breaks).

The C routines with which I am familiar (VMS) seem to force the user
to convert everything into a stream of bytes and then the C routines
convert it back into the actual file format (like those listed above).
_That_ is my idea of inefficient.