NetNews Offline 2/NetNews Offline Volume 2.iso/news/comp/lang/c-part2/11604

Internet Message Format | 1996-08-05 | 2.6 KB

Path: mail2news.demon.co.uk!genesis.demon.co.uk
From: Lawrence Kirby <[e-mail]>
Newsgroups: comp.lang.c
Subject: Re: Determining the length of an int in string form
Date: Mon, 25 Mar 96 12:48:09 GMT
Organization: none
Message-ID: <[e-mail]>
References: <[e-mail]> <4i7uth$[e-mail]> <[e-mail]> <4in1ga$[e-mail]> <[e-mail]> <[e-mail]> <[e-mail]> <[e-mail]>
Reply-To: [e-mail]
X-NNTP-Posting-Host: genesis.demon.co.uk
X-Newsreader: Demon Internet Simple News v1.27
X-Mail2News-Path: genesis.demon.co.uk

In article <[e-mail]> [e-mail] "Larry Weiss" writes:

>We all discussed this, maybe a year ago, and what I was remembering was
>the issue with the representation of end-of-line on the file system, and
>whether this affected the return value of fprintf().

7.9.3:

"All input takes place as if characters were read by successive calls to
the fgetc function; all output takes place as if characters were written
by successive calls to the fputc function."

The only figure with any meaning is the number of effective calls to fputc.
Once you start getting into the lower level text file line formatting C
ceases to ascribe any meaningful values to 'characters transferred'. So
IMHO the return value of fprintf can only refer to the number of effective
fputc calls which means '\n' always corresponds to precisely 1 character.

>Seems safe enough
>that if you never indicate the end-of-line condition (usually via the
>\n in in format string) then that is not an issue (even on a mainframe,
>with those block-mode fixed-length records).

The problem there is that in block-mode fixed-length records you may well
get *no* characters written until the whole block is assembled (i.e. you
wrote a '\n'). Perhaps a more obvious format to show this is a variable
length block with the block length written at the start: nothing can be
written until the block length is known.

>Now what I'd really like to see in the next revision of the standard library
>would be an snprintf() with the extra argument to declare only so many
> writeable
>chars in the buffer, and a return code to indicate logical attempts to write
>beyond that limit.

Indeed, a return value indicating how many characters would have been
written to the buffer if it were large enough.

--
-----------------------------------------
Lawrence Kirby | [e-mail]
Wilts, England | [e-mail]
-----------------------------------------