is this correct

[Darklight]

If the user enters 100 or more characters after the first prompt, the
first `fgets' will read 99 characters, store them in the buffer, and
append a '\0' character. The remaining (again up to 99) characters will
be read by the second `fgets' call, so the user never has a chance to
enter a second string in this case. Part of what the user considers the
first string will be stored in `b'.

If this is not clear, just try it. Enter more than 99 characters and see
what happens.

Martin

Thanks for that but at this stage i just want the program to work.
But now i am aware of the problem thanks.

 

[Arthur J. O'Dwyer]

I.e. it's a bug in the software implementing it.

No, it's a feature. (How do you draw half a smiley?)

If it didn't make its way into an RFC, there must be a reason.

AFAIK, sliced bread doesn't have its own RFC, either. I
hardly think that is a valid argument.

Even the Followup-to header is a mistake, IMHO. The poster has no
business controlling the options of the people who decide to reply.

He may kindly suggest followups to a subset of the Newsgroups header or
even by private email, but not enforce it.

Duh. What do you think the "Followup-To" header is used for?
It's a standardized way of suggesting where "Followups" should go
"To." It would be impolite to force your readers to cut-and-paste
a list of followup groups from the message body, wouldn't it?
Usenet is a text medium, and text never "enforces" anything unless
the reader's news client wants it to.

-Arthur

 

[Dan Pop]

No, it's a feature. (How do you draw half a smiley?)


AFAIK, sliced bread doesn't have its own RFC, either.

If it were relevant in any way to Usenet, you may have had a point...

I hardly think that is a valid argument.

RFC 1036 is to the Usenet what the C standard is to C programming.
By your logic, HTML posts and MIME attachments should be fine, too.

Duh. What do you think the "Followup-To" header is used for?
It's a standardized way of suggesting where "Followups" should go
"To." It would be impolite to force your readers to cut-and-paste
a list of followup groups from the message body, wouldn't it?
Usenet is a text medium, and text never "enforces" anything unless
the reader's news client wants it to.

2.2.3. Followup-To

This line has the same format as "Newsgroups". If present, follow-
up messages are to be posted to the newsgroup or newsgroups listed
here. ^^^^^^^^^^^^^^^^

It's imperative to the client software, not a mere suggestion. And it's
not uncommon for clients to behave accordingly.

Dan

 

[Mark Henning]

RFC 1036 is to the Usenet what the C standard is to C programming.

By your logic, HTML posts and MIME attachments should be fine, too.

The C standard, like all 'standards' are mere guidelines not rules. Do you
also condemn all C compilers that offer extensions to the standard? Those
that make system calls outside the scope of standardised C?

 

[pete]

The C standard, like all 'standards' are mere guidelines not rules.

Please don't quote Pirates Of The Carribean
to substantiate your beliefs about C.

Do you also condemn all C compilers that offer extensions to the
standard?

There are rules about that.
If the behavior of any conforming program is not
altered by the extensions,
then extensions don't affect the conformance of the compiler.

Those
that make system calls outside the scope of standardised C?

Those programs are OK where they're OK,
but they're not OK on this newsgroup.

 

[Dan Pop]

The C standard, like all 'standards' are mere guidelines not rules. Do you
also condemn all C compilers that offer extensions to the standard?

If their extensions to the standard break correct C programs, yes, of
course. Otherwise, such extensions are *explicitly* allowed by the C
standard itself, so I have no reason to condemn them.

Those that make system calls outside the scope of standardised C?

I'm not sure what you mean here. It is C programs that may make
function calls outside the scope of standardised C, not C compilers.
For all we know, C compilers need not be even written in C.

I do condemn C implementations whose headers declare/define identifiers
not belonging to the implementation name space, when invoked in
conforming mode.

Dan

 

[those who know me have no need of my name]

[fu-t set]

in comp.lang.c i read:

In <[email protected]> those who know me have no need of my name

^^^^^^^^^^^^^^
Care to provide some concrete examples?

i believe you could find them yourself, and this is the wrong forum, but
i'll mention that rfc 1036 is the last rfc on the topic but is no longer
what is generally used, son-of-1036 is closer to what is actively used and
it is not an rfc (and is somewhat ignored because it doesn't provide enough
support and has flaws). if you want to learn how the usenet is operating
today and under what (non)standards the appropriate groups are
news.admin.technical (nearly dead) or news.software.nntp, or even the
venerable news.answers or news.newusers.questions are not inappropriate
even for an old-hand.

 

[Mark Henning]

Please don't quote Pirates Of The Carribean

to substantiate your beliefs about C.

I quote nothing.

 

[Ben Pfaff]

I quote nothing.

But you just did!