getting an error ???

For some weird reason I get an error when I try to compile my program . This is the bit of the code that it complains about :

Code:

---

.....

if (option==1) goto label_1;
else;
if (option==2) goto label_2;
// Help bit I still haven't added
label_1:

int ISO;
float aperture;

BOOL ReadSpeed = FALSE;
BOOL ReadAperture = FALSE;

.....

---

The compiler returns :

Code:

---

camera.c: In function ‘main’:
camera.c:33: error: expected expression before ‘int’
camera.c:48: error: ‘ISO’ undeclared (first use in this function)
camera.c:48: error: (Each undeclared identifier is reported only once
camera.c:48: error: for each function it appears in.)

---

What am I doing wrong ? I decelared the variables properly didn't I ?

Code:

---

if (option==1) goto label_1;

---

Nooooooooo! There is *never* a good reason to use goto! If you're using goto, you're doing something wrong.
Also, your else statement doesn't do anything and is superfluous. You should also use braces for your if statements! What is that code snippet trying to accomplish anyway?

QuantumPete

Code:

---

All fixed now :)

---

Alright, variables need to be declared at the beginning of a scope block in C (which I assume you're using). That means you need to declare *all* your variables first before you can call any functions or perform arithmetic operations:

Code:

---

int a = 0;
int b = 2;
int c = a  + b; /* This is ok */
printf ("%d\n", c); /* This too */
int d = 1; /* This isn't - needs to go before the printf */

---

Quote:

---

Dont worry about the goto

---

and don't be patronising! Goto is a bad choice for almost anything you're trying to accomplish and style and clarity are just as important in a program as getting it to run.

QuantumPete

Don't flush(stdin), see the FAQ.

Using goto isn't "bad" as such (In this case it is ;)) - but it does make your program hard to follow, but it does usually mean you have to rethink your design.
Consider indenting your code.

Quote:

---

Originally Posted by QuantumPete

Code:

---

if (option==1) goto label_1;

---

Nooooooooo! There is *never* a good reason to use goto! If you're using goto, you're doing something wrong.

---

Not true. Although there appears to be no good reason for the gotos in this particular example, on rare occasions they are useful, for example to break out of multiple nested loops.

Thanks, but I still don't see what I didn't declare properly . I declared ISO and aperture ,so why is it complaining about them.

Your code is fairly complex for such a simple problem. You can use arrays to define valid values for apertuers, ISO and shutter speed (when you get to the point of validating your shutter speed).

Is it just the formatting here on the site, or is your code actually not intended at all (that is, the left edge has everything to stuck to the edge)? If so, try to indent it so that it gets easier to follow whcih part belongs where, e.g

Code:

---

int main()
{
  ...
  if (something) {
      ...
  } else {
      if (someother) {
        while(condition) {
          ....
        }
      }
  }
}

---

This will obviously help you follow what's going on, but even more so, it will help OTHERS help you.

Finally, I think the reason your code doesn't compile is because you have "else;" fairly early in the code. That's not a valid syntax, there should be a statement after else - I think you meant for the next if-statement to be this statement, in which case just remove the ;, but since your code is poorly indented, it's nearly impossible to tell what you actually thought.

--
Mats

Quote:

---

Originally Posted by robatino

on rare occasions they are useful

---

Sure, there are exceptional circumstances, in which to use goto, but this is not one of them. Also, so far I have always managed to avoid it. All I was trying to get across to someone who's just started programming in C is not to get used to using goto! I'd also like to get across to use braces for if and for statements, but that never seems to get through to anyone...

QuantumPete

Quote:

---

Originally Posted by ki113r

Thanks, but I still don't see what I didn't declare properly . I declared ISO and aperture ,so why is it complaining about them.

---

In C90 variable declarations in each code block have to be at the beginning of the block. You can't place them in the middle, as in C99 or C++.

Quote:

---

Originally Posted by ki113r

Thanks, but I still don't see what I didn't declare properly . I declared ISO and aperture ,so why is it complaining about them.

---

Like I said, they need to be declared at the top, together with all other variables. C is demanding that way ;-)

edit:
Damn you robatino ;-) I have to learn to type faster

Quote:

---

Originally Posted by QuantumPete

Like I said, they need to be declared at the top, together with all other variables. C is demanding that way ;-)

edit:
Damn you robatino ;-) I have to learn to type faster

---

Well, you DID mention it first - I didn't see it in your earlier post.

Ah ok . That solved the problem , but now Im getting more errors :
camera.c:35: error: expected expression before ‘BOOL’
camera.c:38: error: ‘ReadSpeed’ undeclared (first use in this function)
camera.c:38: error: (Each undeclared identifier is reported only once
camera.c:38: error: for each function it appears in.)
camera.c:159: error: expected declaration or statement at end of input

... Is this because I coded this on a windows machibne and it ran fine but now im compiling on Unix ? I hope not

The BOOLs also need to be moved to the top.

I was going to post about the goto's too! Goto's are RARELY necessary, and if they are not necessary, they shouldn't be used.

The most common exception to "don't use goto's" is when there are several while/for/do-while loops that each would need an extra condition to detect that the loop needs to finish early because of some error condition deep inside the nested loops, or closely related where there are lots of conditions, where each of the conditions would need a "and not error" added to cope with some possible early error detection. In this case, it is OK to jump out from teh detection to a "return error" position.

Another possible case is where a function holds some lock or need to free some memory - in which case there's a good cause for having just one exit-point from the function, and thus if you need some

Code:

---

if (something) { cleanup; return; }

---

, it could be converted to

Code:

---

if (something) goto finish;

---

to make the code more maintainable (I've found more than one bug where a piece of code returns under some error condition, e.g. out of memory, and it fails to release a some resource (lock, handle, memory-allocation, or such) - having just one exit-point is preferrably for this reason, and often difficult to achieve in complex functions without using goto).

But as I said, this program shows no such code that motivates a single goto!

--
Mats

Hi everyone, thanks for your help.
If I understand correctly goto can be replaced by :

Code:

---

if (statement=true)
            {    some code to execute instead directly instead of goto                              }
else (statement=false)
            {      something else }
else if (statement=wrong input)
            {alert user                }
break;

---

so basically if i understand i should use { } and run the code directly instead ?

yes, that's correct :-)
As an aside, there is no statement in an else clause

Code:

---

if (statement == true) {
    /* Do something */
} else if (other_statement == true) {
  /* Do something else if another statement is true */
} else {
  /* goes here if all else fails */
}

---

QuantumPete

Thanks Pete , I appreciate it :)

I think that I got used to goto statements as I did a lot of basic programs , where I had to use it ( mostly programming graphics calculators :) )

Yeah, I know what you mean. I started out in BASIC as well, where you just can't do without them; but C is a more 'mature' language ;)

QuantumPete

Is there a way to check to see if the input number is a floating point number and ignore it if it is. I basically need a way to neglect input of a number unless it's a fraction or an integer . So I can accept 1/100 but not 0.01 ? Thanks :)

Quote:

---

Originally Posted by ki113r

Is there a way to check to see if the input number is a floating point number and ignore it if it is. I basically need a way to neglect input of a number unless it's a fraction or an integer . So I can accept 1/100 but not 0.01 ? Thanks :)

---

If you want input that is "1/125", then there's no direct support for inputting such things in C itself, you have to use a string as input and then parse the string to see that the shutter speed is correct (or at least forms "1/<some digits>" to begin with, then try to read the number after 1/ to see if it's one of the values you want, e.g. 100, 125, 250, 500, 1000, but not 65 or 550).

--
Mats

Quote:

---

Originally Posted by ki113r

Is there a way to check to see if the input number is a floating point number and ignore it if it is. I basically need a way to neglect input of a number unless it's a fraction or an integer . So I can accept 1/100 but not 0.01 ? Thanks :)

---

Would something like this work , if not can you please correct it :

Code:

---

{
char shutter[10]
scanf("%d", &shutter);
if (shutter[0] == '0'&&shutter[0] == '.')
{ printf("The entered number seems to be a decimal, Please input fractions or whole numbers ONLY");
break;
}
else continue;

if (shutter==1/4000||shutter==1/2000||shutter==1/1000||shutter==1/500||shutter==1/250||shutter==1/125||shutter==1/60||shutter==1/30||shutter==1/15||shutter==1/8||shutter==11/4||shutter==11/2||shutter==1||shutter==2||||shutter==4||shutter==8||shutter==15||shutter==30||shutter==60)
{printf("You have entered an incorrect shutter speed . Please try again")
}
else continue;

---

Sorry about the indentation I am just doing it quickly on notepad.

I've worked this bit out so far :

Code:

---

#include <stdio.h>
#include <string.h>

int main(int argc, char ** argv)
{

  FILE * output = stdout;
  FILE * input = stdin;


char shutter[10];
{
scanf("&#37;d", &shutter);
if (shutter[0] == '0'&&shutter[0] == '.') // If the string starts with 0 or . ignore
{ printf("The entered number seems to be a decimal, Please input fractions or whole numbers ONLY");
}

---

this works , but I have no idea how to compare ths string now , to determine if a valid shutter speed is entered .. I can see that I can't use

Code:

---

if (shutter!=correct_number)

---

Because its not an integer .

Or maybe Im going about this wrong . This advice I received from someone else :

" And no, that is not the best way of avoiding users entering decimals. The best way is to read the keyboard buffer and filter the input from there, parsing into a variable the contents you really want the user to enter, and of course, at the same time parsing it to the string, if desired. "

Ofcourse I have no idea how to implement that .

Quote:

---

Originally Posted by ki113r

" And no, that is not the best way of avoiding users entering decimals. The best way is to read the keyboard buffer and filter the input from there, parsing into a variable the contents you really want the user to enter, and of course, at the same time parsing it to the string, if desired. "

Ofcourse I have no idea how to implement that .

---

Hmm, well it's pretty much as he said. Have you considered learning a new function or two to help you with your problem? Lemme try:

Code:

---

#include <stdio.h>
#include <stdlib.h>

int verify_it (char *input)
{
  char *tail;
  long int res = strtol(input, &tail, 10);

  if (res && *tail == '\0') {
    return res;
  } else if (*tail == '/') {
    long int num, denom;
    res = sscanf(input, "%ld/%ld", &num, &denom);
    return res == 2;
  } else {
    return 0;
  }
}

int main(void)
{
  char *arr[] = {
    "1/2", "1./4", "1/60", "1/1000", "0.185", "405", "0xdeadbeef", "213",
  };
  unsigned long int i = 0;

  while (i < sizeof arr / sizeof *arr) {
    if (verify_it(arr[i]))
      printf("%s\t- passed validation!\n", arr[i]);
    i++;
  }

  return 0;
}

---

Here's the output:

1/2 - passed validation!
1/60 - passed validation!
1/1000 - passed validation!
405 - passed validation!
213 - passed validation!

Seems to filter out the bad stuff. If you want to accept only certain fractions or numbers, you'd have to implement something like a lookup table, maybe, to check that input matches, I think.

Quote:

---

Originally Posted by ki113r

Code:

---

if (shutter[0] == '0'&&shutter[0] == '.') // If the string starts with 0 or . ignore

---

this works...

---

No it doesn't. Something can never be equal to two different things simultaneously! As is obvious from the comment, it should be an OR (||), not an AND (&&).

Hmm so how would I do the Lookup table as you described above ?

Something similar to what citizen used to build the test-patterns, I would say.

As to filtering the input as it's typed, this will definitely belong in the non-portable section of coding, as there's no standard way to do this (unless you also involve some third-party cross-platform type libraries, ncurses for example). And even if you accept that it's non-portable, it will also require that you write a function to parse the input and deal with things like back-space to erase stuff and all sorts of other "user input fanciness" that you get "for free" when you use the standard input methods. It is certainly the way to go if you have a real product to produce (but then it should probably be using a graphical user interface and would most likely be reading EXIF information from real images (JPEG or Camera Raw formats) rathert thank asking the user for information.

--
Mats

Hmm that looks way too difficult for a beginner ... I don't understand my tutor said theres a much easier way to work out if the number entered is a decimal or not. He doesn't want to say what it is though :( And I certainly can't see any other way than entering it as a string and checking the first few characters . Myabe someone else has another idea how to do this ?

Quote:

---

Originally Posted by ki113r

Hmm that looks way too difficult for a beginner ... I don't understand my tutor said theres a much easier way to work out if the number entered is a decimal or not. He doesn't want to say what it is though :( And I certainly can't see any other way than entering it as a string and checking the first few characters . Myabe someone else has another idea how to do this ?

---

You can check if the number is decimal or not by trying to convert it to a decimal number using strtol(). One of the parameters to strtol is a char pointer, which points to "the last unconverted character" - if that character is NUL (0), then the whole string got converted.

See citizen's code.

--
Mats

Someone told me that I could also write my own function which checks the input byte for byte . Maybe someone else made a function that does this ? Keeping in mind that I have to make this as simple as possible ( and have an idea on what is happening ) which would you say is the best method to take ?

The "best" solution varies depending on what your goal is, but yes, you could write a function that steps through a string and checks that things are the way they should be.

--
Mats

Ok thanks matsp . For now I will leave that , as it is not required yet . This is what I came up with for my test code , I tried to make it as simple as possible , but it doesn't want to compile. It just says Error 42 Symbol Undefined _If. I presume I stuffed up something but compiler tells me nothing .

You have a "If" instead of "if" in your code, so the compiler says "Ok, there must be a function If that I need to call", and then the linker figures "There's no function If anywhere, I'll give an error".

If you add compiler warnings when you build your code, it would warn you about missing prototype for If, which isn't exactly the same as saying "You shouldn't captialize the word if", but it's going to show which line was broken.

--
Mats

Great matsp , youre a genius :) .... It compiles, but i realised I dont really need any of the Boolians , so I simplified it down to this , and made some changes to naming. Now Im having a problem with a while loop. I understand it goes something like this :

Code:

---

---

Yet , that just keeps looping like crazy even if I put in a right value. Am I doing this right --- (except for the mistake which is causing it to loop forever ) . Can you tell me where my mistake is ? Thanks a lot .

Code:

---

fflush(stdin);

---

is not correct code, see this FAQ:
http://faq.cprogramming.com/cgi-bin/...&id=1043284351

--
Mats

Quote:

---

Originally Posted by matsp

Code:

---

fflush(stdin);

---

is not correct code, see this FAQ:
http://faq.cprogramming.com/cgi-bin/...&id=1043284351

--
Mats

---

ok . i will replace that, but that still doesn't solve the problem of the loop. Can you see a mistake in the while loop thats causing it loop forever?

You're doing the scanf *outside* the while loop. Once the user enters a wrong value, he'll never get out of the loop, because the value of aperture never changes

QuantumPete

well i moved the scanf to inside the loop , but it still doesnt work 100% . If a wrong value is entered it works ok , but if a right value is entered it keeps looping this : aperture = 0 , incorrect aperture

Do you actually discard the input when you have a bad input?

--
Mats

Quote:

---

Originally Posted by matsp

Do you actually discard the input when you have a bad input?

--
Mats

---

I got rid of flush(stdin) so that cant be causing a problem

Quote:

---

Originally Posted by ki113r

I got rid of flush(stdin) so that cant be causing a problem

---

Sorry, I thought the FAQ linked to the "solution" to: http://faq.cprogramming.com/cgi-bin/...&id=1043284392

--
Mats

Quote:

---

Originally Posted by matsp

Sorry, I thought the FAQ linked to the "solution" to: http://faq.cprogramming.com/cgi-bin/...&id=1043284392

--
Mats

---

Oh ok . I didn't think that flushing the buffer is important in my case. I will try this , and see what happens . Everything else looks ok to you ?

Yes, aside from the fact that I'd probably use an array of valid apertures, instead of a long list of or-ed together conditions. It's just harder to read, and that's quite important at times.

--
Mats

.....................

There are several things not quite right.

1. You use fgets(buf, ...) then scanf - do you mean to use sscanf(buf, ...0?

2. exposure is an integer, %f is reading a float. Then you print it as a float again.

3. Your "clear the buffer out" seems to be in the wrong place - you should do that when someone has given the wrong result. But if you use fgets consistently, that shouldn't be necessary.

4. There is a ", buf" at the end of a printf, that I don't think should be there.

5. For the aperture comparison to have a chance of working, use 1.2f, 1.4f, ... 32.0f. This makes sure that the compiler doesn't expand the floating point number to double before comparing - this expansion can lead to the precision of the number being different between your aperture variable and the constant in the code, which in turn means that it's not equal.

6. You have several sets of unnecessary braces.

--
Mats

Quote:

---

Originally Posted by matsp

There are several things not quite right.

1. You use fgets(buf, ...) then scanf - do you mean to use sscanf(buf, ...0?

2. exposure is an integer, %f is reading a float. Then you print it as a float again.

3. Your "clear the buffer out" seems to be in the wrong place - you should do that when someone has given the wrong result. But if you use fgets consistently, that shouldn't be necessary.

4. There is a ", buf" at the end of a printf, that I don't think should be there.

5. For the aperture comparison to have a chance of working, use 1.2f, 1.4f, ... 32.0f. This makes sure that the compiler doesn't expand the floating point number to double before comparing - this expansion can lead to the precision of the number being different between your aperture variable and the constant in the code, which in turn means that it's not equal.

6. You have several sets of unnecessary braces.

--
Mats

---

I'm just gonna rewrite this . probably the best option . If I do it in gvim it will indent it for me automatically , or I need to select some option ?

Quote:

---

Originally Posted by ki113r

I'm just gonna rewrite this . probably the best option . If I do it in gvim it will indent it for me automatically , or I need to select some option ?

---

Sorry, don't know that, I'm an emacs guy. [There is endless discussion on which is better, emacs or vi - and it's just as easily settled as the discussion of which form of christianity is better, catholisism or protestantism - the "right answer" depends on who you ask!].

I would expect it to be automagic. It is in emacs. [If you are using a Linux machine, it should have (x)emacs too

--
Mats

It seems like VIM is autoindenting always:
http://vim.sourceforge.net/tips/tip.php?tip_id=330

A second point, you don't really need to start over completely. Just fix up a few things. I took your code and made some adjustments, and it's working OK now.

--
Mats

Can you please show me the changes you did , in regards to clearing the buffer bit, because I basically just copied the FAQ example.

Quote:

---

Originally Posted by ki113r

Can you please show me the changes you did , in regards to clearing the buffer bit, because I basically just copied the FAQ example.

---

I outlined the changes in the "6 things that aren't quite right".

Let me explain a bit more:
1. I use only fgets() to read data from the user. This has several benefits: It's safe. It reads to newline (unless the user inputs more than BUFSIZ amount of data). Then use sscanf(buf, ..) to read the data out of the buffer.

2. I moved all variables to the top of the function - mostly because I wanted only one buf to use for both shutter and aperture.

3. I removed the stdin buffer clearing - it's not needed when using fgets().

4. I fixed your long list of "valid" apertures to have an f on the end, to make is float rather than double - this avoids conversion of the aperture value, so it will be less likely to mismatch simply because the number is converted.

And I tidied up your extra braces that aren't doing anything. This of course has no real effect on the execution, just makes it easier to follow and fewer lines.

[I also formatted it in emacs, so I could at all read it - I'm so unused to reading code that is flat, that I can't do it.]

--
Mats

Thanks for explaining that matsp. I'll try my best to get this thing to work . I wish I could use goto lol -- it seems a lot easier than doing the loop in this case .

/me looks for a newspaper to smack ki113r with ;)

Quote:

---

Originally Posted by QuantumPete

/me looks for a newspaper to smack ki113r with ;)

---

lol Pete , is it just me or can you smell out a post that mentions gotos .

Quote:

---

Originally Posted by ki113r

lol Pete , is it just me or can you smell out a post that mentions gotos .

---

It's a sixth sense (and instant post notification via gmail) ;)

QuantumPete

Ah nvm I think I worked it out ... fingers crossed

Code:

---

int exposure=0;
float aperture=0;
...
scanf("%f",&exposure);
...
scanf("%d",&aperture);

---

You've got them the wrong way round, exposure is declared as an int, but you scanf for a float. aperture is declared as a float, but you're scanning for an int.
Hence, when scanf reads in '1.2' it will do a conversion to int and give you '1'. That's probably why it doesn't work. Change the f to d and the d to f and you should be fine. ;)

QuantumPete

Ok so what I figured out is that I was using &#37;f for my aperture - which is incorrect as I need %d as its an integer not a float. The second mistake is that when I enter 1.4 the program reads it as 1.400000001 ( maybe ) so that is not a perfect match . I guess thats what you were talking about matsp when you said

Quote:

---

4. I fixed your long list of "valid" apertures to have an f on the end, to make is float rather than double - this avoids conversion of the aperture value, so it will be less likely to mismatch simply because the number is converted.

---

Can you show me how to do that exactly please, because that should fix everything . I just want to get this thing done :)

Code:

---

#include <stdio.h>
int main(void)
{
int exposure=0;
float aperture=0;
int choice = 0;

printf("\nEnter Exposure Time: ");
scanf("&#37;d",&exposure);

printf("The selected Exposure time is %d\n", exposure);

{
while (choice<1)
{
printf ("Please enter the aperture size ");
scanf("%d",&aperture);

if ((aperture==1.2) ||(aperture==1.4) ||(aperture==1.8) ||(aperture==2.0) ||(aperture==2.8) ||(aperture==2.8) ||(aperture==4.0) ||

(aperture==5.6) ||(aperture==8.0) ||(aperture==11)||(aperture==16)||(aperture==22)||(aperture==32))
{
printf("You have entered a correct value ");
choice=2;
}
else
{
printf("Invalid aperture=%d\nPlease Try Again\n", aperture);
choice=0;
}
}
}
}

---

Code:

---

scanf("%d",&aperture);

---

This should be %f, not %d.

And you should relly add an f on the end of all your numbers comparing the aperture (and you will need to add a .0 to the ones that don't have a . in them, as for example 4f isn't a valid float, 4.0f is.

--
Mats

Quote:

---

Originally Posted by matsp

Code:

---

scanf("%d",&aperture);

---

This should be %f, not %d.

And you should relly add an f on the end of all your numbers comparing the aperture (and you will need to add a .0 to the ones that don't have a . in them, as for example 4f isn't a valid float, 4.0f is.

--
Mats

---

But if ((aperture==1.2f. . . ) doesn't work

Quote:

---

Originally Posted by ki113r

But if ((aperture==1.2f. . . ) doesn't work

---

In what sense does it not work. It works on my machine.

--
Mats

Quote:

---

Originally Posted by matsp

In what sense does it not work. It works on my machine.

--
Mats

---

It won't compile for me , it gives me an error.

Quote:

---

Originally Posted by ki113r

It won't compile for me , it gives me an error.

---

Which error?

--
Mats

..................

Quote:

---

Originally Posted by matsp

Which error?

--
Mats

---

Lexican Error

Quote:

---

Originally Posted by ki113r

Lexican Error

---

I've never heard of a compiler giving a "lexican" error. When we ask you to post the error, it's not to make you look stupid, but because it tells us exactly where the error is and more importantly *what* the error is.
So, copy and paste please!

QuantumPete

If you don't immediately remove those goto's, I'm not going to help you further. goto's are NOT the way to solve the problem.

Also, post the if-statement that gives your error (with the f's).

--
Mats

Quote:

---

Originally Posted by QuantumPete

I've never heard of a compiler giving a "lexican" error. When we ask you to post the error, it's not to make you look stupid, but because it tells us exactly where the error is and more importantly *what* the error is.
So, copy and paste please!

QuantumPete

---

Yeh no problem Pete . Thats what it gave me , lemme try it again ...
wow it worked now . This is crazy :-|

Unfortunately it still does the infinite loop Btw this is how it looks when I run it :
http://img119.imageshack.us/img119/3227/coders4.jpg

Quote:

---

Originally Posted by ki113r

wow it worked now . This is crazy :-|

---

Forgot to save before compiling? ;)

QuantumPete

Unless you have changed it just this moment, the scanf() still needs to be %f, not %d. If you scanf with %d and enter 1.2 (or anything else with a . in it), scanf will get stuck on the . in the input, and since you have no code to consume the remaining input, you get an infinite loop.

I did suggest to use fgets() and sscanf(), rather than scanf(), because it avoids many of the problems of "garbage in the input causes strange problems".

--
Mats

Quote:

---

Originally Posted by matsp

Unless you have changed it just this moment, the scanf() still needs to be %f, not %d. If you scanf with %d and enter 1.2 (or anything else with a . in it), scanf will get stuck on the . in the input, and since you have no code to consume the remaining input, you get an infinite loop.

I did suggest to use fgets() and sscanf(), rather than scanf(), because it avoids many of the problems of "garbage in the input causes strange problems".

--
Mats

---

I have to use scanf , its one of the requirements from my lecturer

Wohoo , it sort of worked

Code:

---

C:\Users\Ben\dm\bin>test3

Enter Exposure Time: 12
The selected Exposure time is 12
Please enter the aperture size 1.4
Invalid aperture=1610612736
Please Try Again
Please enter the aperture size 12
You have entered a correct value

---

Quote:

---

Originally Posted by matsp

If you don't immediately remove those goto's, I'm not going to help you further. goto's are NOT the way to solve the problem.

Also, post the if-statement that gives your error (with the f's).

--
Mats

---

I was just trying it with gotos , in case its the while loop thats causing the problem . :) , but it seems to be fine

I take it you didn't change the printf to use %f - that looks like a floating point number printed with %d.

The reason the language has while and do-while and a few other constructs is exactly to avoid having goto's. They do the same thing in a more easy to read fashion.

--
Mats

...................

Yes, if you enter f (or any other non-float-number stuff) as part of your input, then it will get stuck - like I said, scanf() can be very temperamental, and will quite easily get confused by "bad input".

If it is a requirement to actually be able to tolerate user errors, then fgets() is the right way to go.

--
Mats

Ok , well ill get some sleep now, and ask my tutor about using fgets , thanks a lot for your help :)

Turns out the user must enter the f value , and the program must process the whole thing . I think the way I will handle this is to use string compare. I think my tutor would prefer that to fgets. When I askd him this is the reply I rceeived :
> Am I allowed to use a more advanced way if it simplifies the
> procedure ?
No. Look at style ( that he teaches in lectures and practicals ) NOT the code you use.

What do you think is easier and more logical way for a beginner to use for my problem , to convert the input to a string use string compare to compare the apertures with valid values , or to use fgets ? We haven't been taught neither of the ways . Any suggestions ?