susbtring function

[HSeganfredo@gmail.com]

Hello again,

I´ve made a small substring() function that shoudl give back the a
susbtring from a string, given its start and end positions.

Is there any chance of the code below give an output containing data
from another memory area that is completely unrelated to the input
string? This is happening and the start/end parameters are acceptable
(within bounds) for the input string...any idea?


char *substring(char *string, int start, int end){
printf("Input %s:", string);

char *result = (char *)malloc((end - start + 1)*sizeof(char) );
if(result == NULL){
return(NULL);
}

string = string + start;
while(start < end){
*result = *string;
result++;
string++;
start++;
}
printf("Output %s:", result);
return(result);
}

[Keith Thompson]

Re: susbtring function

Ian Collins <ian-news@hotmail.co mwrites:

HSeganfredo@gma il.com wrote:

[...]

> printf("Input %s:", string);
>>

>
You forgot the '\n':
>
printf("Input %s:", string);

So did you. Just as a matter of esthetics, putting the ':' at the end
seems odd. I'd write this as:

printf("Input: \"%s\"\n", string);

[...]

> return(NULL);

>
return isn't a function
>
return NULL;

To expand on that a bit: the syntax of a return statement is
return ;
or
return <expression;

The parentheses are not necessary, but they are permitted. If you use

return(NULL);

then the parentheses are part of the expression, not part of the
syntax of the return statement itself.

I prefer *not* to use extraneous parentheses; a return statement isn't
a function call, so it shouldn't look like one.

But using parentheses isn't wrong. Even the examples in K&R1 use
parentheses on return statements (though K&R2 changed this).

[...]

--
Keith Thompson (The_Other_Keit h) kst-u@mib.org <http://www.ghoti.net/~kst>
San Diego Supercomputer Center <* <http://users.sdsc.edu/~kst>
"We must do something. This is something. Therefore, we must do this."
-- Antony Jay and Jonathan Lynn, "Yes Minister"

[CBFalconer]

Re: susbtring function

HSeganfredo@gma il.com wrote:

>
I´ve made a small substring() function that shoudl give back the a
susbtring from a string, given its start and end positions.
>
Is there any chance of the code below give an output containing data
from another memory area that is completely unrelated to the input
string? This is happening and the start/end parameters are acceptable
(within bounds) for the input string...any idea?
>
char *substring(char *string, int start, int end){

start and end are indices into *string, so should be type size_t

printf("Input %s:", string);
>
char *result = (char *)malloc((end - start + 1)*sizeof(char) );

You can't put this after code. Declare result before the printf
statement. Delete the cast. Eliminate "*sizeof(char)" , because
that is 1 by definition.

if(result == NULL){
return(NULL);
}
>
string = string + start;

Where have you checked that start is a valid index for this
string? Maybe "if (start < strlen(string)) ... " would do.

while(start < end){

Where have you checked that end is a valid index for this string?
Where have you checked that end is larger than start? See above.

*result = *string;

Maybe this expression should include start?

result++;

Maybe this action should disappear?

string++;
start++;

Indenting code controlled by a conditional is considered cool.

}
printf("Output %s:", result);
return(result);

This isn't the value malloc returned. You have a problem when you
free, or you have a memory leak.

}

Enjoy.

--
Chuck F (cbfalconer at maineline dot net)
Available for consulting/temporary embedded and systems.
<http://cbfalconer.home .att.net>


--
Posted via a free Usenet account from http://www.teranews.com

[Ian Collins]

Re: susbtring function

Charlie Gordon wrote:

>
All these corrections are fine and dandy...

I hope the were more than that.

to complete the answer for the
OP, there are 2 reasons you can get data in the output unrelated to in
input:
- you forgot the final '\0' after the copy loop. Insert *result = '\0';
after the }.
- you don't check that start and end fall within the boundaries of the
string. If not, you are attempting to read from beyond the end of the
string or before its beginning or indeed from no particular location... This
invokes undefined behaviour and does not necessarily cause your program to
crash.
>

There were at least three, as I pointed out, he was returning one past
the end of the result.

--
Ian Collins.

[Michal Nazarewicz]

Re: susbtring function

HSeganfredo@gma il.com writes:

I´ve made a small substring() function that shoudl give back the a
susbtring from a string, given its start and end positions.
>
Is there any chance of the code below give an output containing data
from another memory area that is completely unrelated to the input
string?

As others described it is. :)

This is happening and the start/end parameters are acceptable
(within bounds) for the input string...any idea?
>
char *substring(char *string, int start, int end){
printf("Input %s:", string);
>
char *result = (char *)malloc((end - start + 1)*sizeof(char) );
if(result == NULL){
return(NULL);
}
>
string = string + start;
while(start < end){
*result = *string;
result++;
string++;
start++;
}
printf("Output %s:", result);
return(result);
}

You may use strncpy() but remember to put 0 byte at the end of the
string:

#v+
char *substring(cons t char *string, size_t start, size_t end) {
return strcnpy(calloc( end-start+1), string+start, end-start);
}
#v-

Exercise: make the code readable, check if memory allocation succeed,
check if start is valid string's index and [start, end] is
non-empty set.

--
Best regards, _ _
.o. | Liege of Serenly Enlightened Majesty of o' \,=./ `o
..o | Computer Science, Michal "mina86" Nazarewicz (o o)
ooo +--<mina86*tlen.pl >---<jid:mina86*chr ome.pl>--ooO--(_)--Ooo--

[santosh]

Re: susbtring function

Charlie Gordon wrote:

Michal Nazarewicz wrote:

<snip>

>#v+
>char *substring(cons t char *string, size_t start, size_t end) {
> return strcnpy(calloc( end-start+1), string+start, end-start);
>}
>#v-

>
Assuming strcncpy was a typo for the infamous strncpy,

There is a typo in the correction to the previous typo :)

<snip>