is there a way to replace EOF in this loop..

we havent learned files
there is no End Of File in this program
can i change it into '\0'
or some like that?

Code:

---

for (i = 0; i < 39 && (ch = getchar()) != '\n' && ch != EOF; ++i)
            {
                input2[i] = ch;
            }

            input2[i] = '\0';

---

What?!

The current code will change the last valid character into NUL. Not what you want, infact you don't want anything.

i want to keep this loop doing what it does but
without EOF

It already does... perhaps brush up on your loops.

so i can delete this EOF condition?

No... because the condition stops you from putting EOF in the array.

getchar() returns EOF on *error* or when it reaches the EOF.

So getchar() will return EOF, even if there *is* no EOF char at the end of the file! :)

So I believe you will want to keep EOF just where it is, in your for statement.

Can you show the input that the for statement is not handling right?

this loop works fine

but i cant use the term EOF

i am looking for a way to replace it

Quote:

---

Originally Posted by transgalactic2

this loop works fine

but i cant use the term EOF

i am looking for a way to replace it

---

Why? Because your teacher says "You can't use that"? If so, remove it - but be aware that it IS correct to have EOF there.

--
Mats

is there a way to state EOF in getchar() terms

like

Code:

---

for (i = 0; i < 39 && (ch = getchar()) != '\n' && ch !=(getchar error); ++i)
            {
                input2[i] = ch;
            }

            input2[i] = '\0';

---

Quote:

---

Originally Posted by transgalactic2

is there a way to state EOF in getchar() terms

like

Code:

---

for (i = 0; i < 39 && (ch = getchar()) != '\n' && ch !=(getchar error); ++i)
            {
                input2[i] = ch;
            }

            input2[i] = '\0';

---

---

That is what EOF means - something went wrong when reading the next character [most often because you reached end of file or pressed CTRL-D/CTRL-Z on the keyboard - but anything that "fails" the read itself will return EOF from getchar()].

--
Mats

What will happen if i will remove this EOF condition regarding the string??

will it stop putting chars into the string and put '\0' like before

Quote:

---

Originally Posted by transgalactic2

What will happen if i will remove this EOF condition regarding the string??

will it stop putting chars into the string and put '\0' like before

---

In most situations NOTHING will change. However, if you happen to receive an EOF, it will continue to fill the string with EOF (converted to a character). And unless some part of your code detects and exits (or does something else meaningful) on EOF, it will loop forever.

--
Mats

how can i recieve EOF in the running of a program?
or in the string input part??

Can you actually read? I am having my doubts here...
You only seem to notice what you want and ignore everything else...

Quote:

---

Originally Posted by transgalactic2

this loop works fine

but i cant use the term EOF

i am looking for a way to replace it

---

Try replacing it with -1. iirc, that was the actual numerical value of EOF.

EOF is better to use, of course, but -1 may serve you well, if you can't use EOF.

Quote:

---

Originally Posted by Adak

Try replacing it with -1. iirc, that was the actual numerical value of EOF.

EOF is better to use, of course, but -1 may serve you well, if you can't use EOF.

---

EOF is only guaranteed to be negative, but that means that comparing for "less than 0" will do.

Quote:

---

Originally Posted by laserlight

EOF is only guaranteed to be negative, but that means that comparing for "less than 0" will do.

---

But that is REALLY not the right solution. If the original poster is not supposed to use EOF, then comparing with negative values is just cheating - you are assuming that EOF is a negative number [I'm not sure the spec says so - it probably does], and that anything else ever coming out of getchar that is negative is a failure.

I'd like to tell the teacher that set these rules that he/she is wrong in forbidding correctly written code (presumably "because we haven't talked about it yet") - it's one thing if the code is "cheating" by for example using more advanced functions when the student is supposed to learn how to do it at a lower level.

--
Mats

did i substituted correctly the EOF condition

Code:

---

for (i = 0; i < 39 && (ch = getchar()) != '\n' && ch >=0; ++i)
            {
                input2[i] = ch;
            }

            input2[i] = '\0';

---

Quote:

---

Originally Posted by transgalactic2

did i substituted correctly the EOF condition

Code:

---

for (i = 0; i < 39 && (ch = getchar()) != '\n' && ch >=0; ++i)
            {
                input2[i] = ch;
            }

            input2[i] = '\0';

---

---

Yes, that should work - but be aware that it's BAD code. But I guess it's a case of "anything to get past the teacher".

--
Mats

yep :)

so in a case of getchar error
this will work fine?

Quote:

---

Originally Posted by matsp

But that is REALLY not the right solution. If the original poster is not supposed to use EOF, then comparing with negative values is just cheating

---

Yes, it is cheating, but I am in Adak's good company :p

Quote:

---

Originally Posted by matsp

you are assuming that EOF is a negative number [I'm not sure the spec says so - it probably does], and that anything else ever coming out of getchar that is negative is a failure.

---

Ah yes, that is true: it is guaranteed by the standard that EOF is negative, but it is not guaranteed that anything that is negative is EOF. Consequently, the "cheat" is not guaranteed to work.

EDIT:
Oops, that is not true: the C standard does guarantee that anything that is negative is EOF, since "the fgetc function obtains that character as an unsigned char converted to an int" and "the getc function is equivalent to fgetc" and "the getchar function is equivalent to getc with the argument stdin".

what do you negative number is a failure
"
[I'm not sure the spec says so - it probably does], and that anything else ever coming out of getchar that is negative is a failure."

if i am going to enter a negative number into my string it will stop the input?

Quote:

---

Originally Posted by transgalactic2

if i am going to enter a negative number into my string it will stop the input?

---

If you enter a negative number as text, no. It would merely be interpreted as the characters of a string, within the ASCII range.

Quote:

---

Originally Posted by laserlight

Yes, it is cheating, but I am in Adak's good company :p


Ah yes, that is true: it is guaranteed by the standard that EOF is negative, but it is not guaranteed that anything that is negative is EOF. Consequently, the "cheat" is not guaranteed to work.

---

so i need to say that ch differs -1 like ADAK said
to be sure??

Actually, read my edit. The "cheat" is guaranteed to work (but that does not make it any less of a cheat).

thanks :)
the final nail in a coffin

Code:

---

for (i = 0; i < 39 && (ch = getchar()) != '\n' && ch >=0; ++i)
            {
                input2[i] = ch;
            }

            input2[i] = '\0';

---

comparing with -1 will simply EXPECT EOF to be -1 (which it probably is on 98% or more of all compiler&C Library implementations). Since getchar() is just a wrapper around fgetc() and fgetc() returns only positive numbers for valid characters, and EOF is the only defined negative value, comparing with >=0 should be fine.

However, you are still doing an EOF comparison - just avoiding the name EOF. Which is why we are calling it a cheat. Yes, it will work - but it's harder to understand, and you are still "using EOF" - just not by name. (And NOT using EOF means that you are actually hiding what that part of the code does - making it harder to understand. For example, the reader must know that EOF is a negative number).

--
Mats

You could do this, without assuming anything about the range of char

Code:

---

for (i = 0; i < 39 && (ch = getchar()) != '\n' ; ++i)
            {
                if ( feof(stdin) ) break;
                input2[i] = ch;
            }

---

Throw in an ferror() for good measure.

Quote:

---

Originally Posted by Salem

You could do this, without assuming anything about the range of char

Code:

---

for (i = 0; i < 39 && (ch = getchar()) != '\n' ; ++i)
            {
                if ( feof(stdin) ) break;
                input2[i] = ch;
            }

---

Throw in an ferror() for good measure.

---

But SURELY if you are not supposed to use EOF, then feof() is also bad, right? I mean the poor chap(ess) is not allowed to use fgets, hence the loop here [avoiding gets()].

--
Mats

Is this another one of those pointless "can you tie your shoelaces with one hand" type of assignments?

Quote:

---

Originally Posted by Salem

Is this another one of those pointless "can you tie your shoelaces with one hand" type of assignments?

---

Yes, that's my understanding.

As I said earlier, it's fair to say "don't use functions that do the job for you, when you are learning the basics", but certainly "correct code that solves the problem in the way that the teacher expects" should not be rejected just because it uses some "we haven't got there in the book yet". Fair enough that you can't do strcpy() when learning about strings and your task is to copy a string - but to check if the result of getchar() is not EOF in a loop is valid, and not a "trick" to solve the problem easier - it's just more correct.

--
Mats

I really don't think the teacher forbids the use of EOF, since it is the only way of finding the end-of-file and/or an error. There is no other way around. The "tricks" and "cheatings" use EOF, they just don't type it...