Thread: Two questions. (Sorry for lots of them =( )

The end of a string is not NULL - NULL, when it is used, is for comparing with pointers that point to nothing, such as:

Code:

   char *p;
   p = strchr(somestring, somechar);
   if (p == NULL)
      ... didn't find somechar in somestring. ...

The zero at the end of a string is SOMETIMES called 'NUL'. There is a FAQ for "What's the difference between NUL and NULL?" or some such.

Essentially, both are zero, but one is a char, the other is a pointer (or something that can be compared to a pointer).

--
Mats

As to your fscanf() question - fscanf will skip whitespaces until the first non-whitespace. When it hits that, it will TRY to read that as a number. Since the first non-whitespace in your text is not a number, your number never gets set - it happens to be whatever was in the variable before, and by the looks of the value, it was a local variable that wasn't given a value, so it contains some "large sort of random number".

scanf() and related functions are quite "picky" on the format of the input, and can easily be confused by bad input.

Check the return value from scanf, and I bet you it will be zero in your example, because it failed to read a number.

--
Mats

Originally Posted by ozumsafa

Other questions ? =)
I will also ask why do we need to use pointers for structs when it is adequate to define standart struct variables?

Eh? You need pointers to structs when you want to pass or keep track of a structure without making a copy of the structure itself. Pointers to structures are also useful when you don't know at programming time how many you need, and you want to dynamically create space for the structures using malloc.

--
Mats

Yes, every line as a newline on the end. Some input functions (such as fgets()) will include the newline in the input [it is useful to see if you got the whole line or not].

Technically, the "newline" in a system is dependant on which system the file is on, in Windows (and DOS, OS/2, CP/M and several others) the newline is actually CR/LF, whilst in Unix, it's only LF, and in Apple's older OS's it was only CR.

CR stands for "Carriage Return", and comes from the old "typewriter" style teletype terminals that existed in the 1950-1970 timeframe - it means "return the typing carriage to the left-hand side of the paper".

LF stands for "LineFeed", which is easier to follow: push the paper forward one line.

Fortunately for us, as long as you use text-mode, all these differences are hidden behind the scenes, so the string you read will have a newline (LF) at the end, whether it is a Windows, Linux or some other OS you're running on.

--
Mats

Originally Posted by ozumsafa

Is the samething true about fscanf for dat files? Because in a book , I see people reading integers with fscanf , but it doesnt work for me..

Ehm ".dat" files aren't a very defined format - .dat just means that it's some for of data - it could be plain text, quoted text, XML or any form of binary data for all I know, and of course, each of those file-formats require a different type of read function.

--
Mats