Digits after the decimal point

[Kuljit]

I am doing Engineering(B.T ech) in Computer Science.

I have a question for which i am struggling to write a C code(program).
It struck me when we were being taught about a program which counts the
number of digits in a given number.
I request to help me out in solving the below said question.

Ask the user to enter a decimal/float number(eg. 32.8952), then count
the number of digits in that number after the decimal point(4 in this
case).

eg: If user enters 45.99420 then we should get 5, which are the number
of digits after the decimal point.

- Kuljit

[Flash Gordon]

gamehack wrote:

Kuljit wrote:

I am doing Engineering(B.T ech) in Computer Science.

I have a question for which i am struggling to write a C code(program).
It struck me when we were being taught about a program which counts the
number of digits in a given number.
I request to help me out in solving the below said question.

Ask the user to enter a decimal/float number(eg. 32.8952), then count
the number of digits in that number after the decimal point(4 in this
case).

eg: If user enters 45.99420 then we should get 5, which are the number
of digits after the decimal point.

- Kuljit
Well...
I'll tell you the algorithm for doing it, you do the code :)
1. Allocate a buffer to hold the input as a string(eg char input[80])
2. Use scanf(hint: scanf("%s", input))

Bigger hint. Don't use scanf("%s", input) ever. It allows the user to
scribble all over your memory by typing in more characters than you have
allowed for.
3. Start walking the string character by character(assum es ASCII)
No need to assume ASCII. C provides perfectly good methods for
identifying digits, '+', '-' and '.'
4. When you encounter the char x = '.' start increment a counter
5. Finish incrementing when you've reached '\0'
You forgot to check for invalid input.
Hope you can write the code yourself.

Yes, the OP does have to attempt to write the code. Having done so
people here can assist with any C questions.
--
Flash Gordon
Living in interesting times.
Web site - http://home.flash-gordon.me.uk/
comp.lang.c posting guidlines and intro -
http://clc-wiki.net/wiki/Intro_to_clc

[Old Wolf]

gamehack wrote:

Kuljit wrote:

Ask the user to enter a decimal/float number(eg. 32.8952), then count
the number of digits in that number after the decimal point(4 in this
case).
I'll tell you the algorithm for doing it, you do the code :)
1. Allocate a buffer to hold the input as a string(eg char input[80])
2. Use scanf(hint: scanf("%s", input))

I must have missed the part where the OP wanted to cause
a buffer overflow. (Never use scanf with %s).
3. Start walking the string character by character(assum es ASCII)

You don't have to assume ASCII to walk a string character
by character.

[Barry Schwarz]

On 15 Feb 2006 10:03:18 -0800, "gamehack" <ga******@gmail .com> wrote:

Kuljit wrote:

I am doing Engineering(B.T ech) in Computer Science.

I have a question for which i am struggling to write a C code(program).
It struck me when we were being taught about a program which counts the
number of digits in a given number.
I request to help me out in solving the below said question.

Ask the user to enter a decimal/float number(eg. 32.8952), then count
the number of digits in that number after the decimal point(4 in this
case).

eg: If user enters 45.99420 then we should get 5, which are the number
of digits after the decimal point.

- Kuljit
Well...
I'll tell you the algorithm for doing it, you do the code :)
1. Allocate a buffer to hold the input as a string(eg char input[80])
2. Use scanf(hint: scanf("%s", input))

fgets() is a much better choice than using scanf() with an
unrestricted %s.
3. Start walking the string character by character(assum es ASCII)
4. When you encounter the char x = '.' start increment a counter
5. Finish incrementing when you've reached '\0'

Hope you can write the code yourself.

Remove del for email

[Kuljit]

Hello everyone,
Thankyou for your help. i have written the code as you have said to
accept it as a string and got the output, but as many of you have said
this is not a floating point problem.

So can we ask the user to enter the decimal number in floating point
value and then determine from that what the user typed.

Hope you wil be able to solve this problem also.

- Kuljit

[Vladimir S. Oka]

Kuljit wrote:

Hello everyone,
Thankyou for your help. i have written the code as you have said to
accept it as a string and got the output, but as many of you have said
this is not a floating point problem.
You should quote what you're replying to, even if you're replying to
yourself.
So can we ask the user to enter the decimal number in floating point
value and then determine from that what the user typed.

Hope you wil be able to solve this problem also.

Luckily for you, I'm using Google at the moment (grrrr) and can
actually see what the original problem was, so I may be able to help...

Are you now referring to numbers of the -1.23e-12 format? If yes:

The only difference is figuring out the rules that these numbers
follow, and checking the input against them (read in, and walk through
the string). The rules are (I may get this wrong, please do check):

<float> ::= <fixed><exponen t><integer>.
<fixed> ::= [<sign>]<unsigned>['.' [<unsigned>]].
<exponent> ::= 'e' | 'E' | 'd' | 'D'.
<integer> ::= [<sign>]<unsigned>.
<sign> ::= '+' | '-'.
<unsigned> ::= <digit>[<digit>]*.
<digit> ::= '0' ... '9'.

Now, using the grammar above, depending on where you encounter '\0',
you'll know exactly what sort of the number was entered (including
integers, both signed and unsigned).

Note:
<> token
::= is
| one of
[] optional
* zero or more
. end of the rule

--
BR, Vladimir

[Vladimir S. Oka]

Vladimir S. Oka wrote:

Kuljit wrote:

Hello everyone,
Thankyou for your help. i have written the code as you have said to
accept it as a string and got the output, but as many of you have said
this is not a floating point problem.
You should quote what you're replying to, even if you're replying to
yourself.

So can we ask the user to enter the decimal number in floating point
value and then determine from that what the user typed.

Hope you wil be able to solve this problem also.

Luckily for you, I'm using Google at the moment (grrrr) and can
actually see what the original problem was, so I may be able to help...

Are you now referring to numbers of the -1.23e-12 format? If yes:

The only difference is figuring out the rules that these numbers
follow, and checking the input against them (read in, and walk through
the string). The rules are (I may get this wrong, please do check):

<float> ::= <fixed><exponen t><integer>.
<fixed> ::= [<sign>]<unsigned>['.' [<unsigned>]].
<exponent> ::= 'e' | 'E' | 'd' | 'D'.
<integer> ::= [<sign>]<unsigned>.
<sign> ::= '+' | '-'.
<unsigned> ::= <digit>[<digit>]*.
<digit> ::= '0' ... '9'.

Obviously:

<digit> ::= '0' | '1' | ... | '9'.

:-(

Now, using the grammar above, depending on where you encounter '\0',
you'll know exactly what sort of the number was entered (including
integers, both signed and unsigned).

Note:
<> token
::= is
| one of
[] optional
* zero or more
. end of the rule

--
BR, Vladimir

[Kuljit]

Dear Vladmir,
Thankyou for your reply.
I think you havent understood what i meant. My original text was

Kuljit wrote:

I have a question for which i am struggling to write a C code(program).
It struck me when we were being taught about a program which counts the
number of digits in a given number.
I request to help me out in solving the below said question. Ask the user to enter a decimal/float number(eg. 32.8952), then count
the number of digits in that number after the decimal point(4 in this
case).
eg: If user enters 45.99420 then we should get 5, which are the number
of digits after the decimal point.

so i want to count the number of digits after the decimal point as i
have said in my quoted text. But instead of accepting it in a string, i
want to accept it in float datatype.

So please try to help me out with this problem.

-Kuljit

[Richard G. Riley]

On 2006-02-24, Kuljit <ku**********@g mail.com> wrote:

Dear Vladmir,
Thankyou for your reply.
I think you havent understood what i meant. My original text was

Kuljit wrote:

I have a question for which i am struggling to write a C code(program).
It struck me when we were being taught about a program which counts the
number of digits in a given number.
I request to help me out in solving the below said question.

Ask the user to enter a decimal/float number(eg. 32.8952), then count
the number of digits in that number after the decimal point(4 in this
case).

eg: If user enters 45.99420 then we should get 5, which are the number
of digits after the decimal point.

so i want to count the number of digits after the decimal point as i
have said in my quoted text. But instead of accepting it in a string, i
want to accept it in float datatype.

So please try to help me out with this problem.

-Kuljit

You can read it into a string first, count your decimal places
and then convert the string to a float : google will help you convert a
string to a float. All the things to be careful about in parsing the
string to detect number of decimal places are already covered in the thread.

I can see no reason why you would want to calculate decimal places
from the float variable when you can have the string representation to
start with. You can store the input in a float afterwards maybe?

See here : http://tinyurl.com/heyck

This covers using scanf, strtod and others with the usual style and
robustness arguments prevalent in all threads to do with string to
number conversions :)
--
Remove evomer to reply

[Lew Pitcher]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Kuljit wrote:
[snip]

Ask the user to enter a decimal/float number(eg. 32.8952), then count
the number of digits in that number after the decimal point(4 in this
case).

eg: If user enters 45.99420 then we should get 5, which are the number
of digits after the decimal point.

so i want to count the number of digits after the decimal point as i
have said in my quoted text. But instead of accepting it in a string, i
want to accept it in float datatype.

Unfortunately, if (as you say) you want to count the number of digits
after the decimal point in a /floating point number/ (and not a string),
then your examples are a bit out of whack.

Consider that the user enters the string "32.8952", and you convert it
to a floatingpoint number (your first example). There are clearly 4
significant digits following the decimal point. *But*, 32.8952 /as a
rational number/ is the same as 32.89520, which (by your second example)
you wish to count as 5 digits after the decimal point. /And/, the number
is the same as 32.895200, which is 6 decimal places (by extension of
your two examples).

Do you see the problem? Which interpretation of the number is correct
for your count?

Also, when working with /floatingpoint/ numbers, floatingpoint is
inaccurate: a number /entered/ as 45.99420 might actually become
45.994197827734 , which will have a different count of significant
decimal places than you expect

So, to satisfy your count requirement, you cannot count the digits in
the floatingpoint datatype.

- --

Lew Pitcher, IT Specialist, Corporate Technology Solutions,
Enterprise Technology Solutions, TD Bank Financial Group

(Opinions expressed here are my own, not my employer's)
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.2 (MingW32)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFD/y9/agVFX4UWr64RAi9 eAJ4uhYQiboRiLl x6FjtRvS9sb8zGt wCfa81k
BMvnCHhESl8LOYm Rfeg0YeQ=
=Brvz
-----END PGP SIGNATURE-----