If you to split long lines into several lines, as stated in the task description, then you must print '\n' instead of ',' when encountering the cut treshold.
You should not print the character on the output stream until you know whether it should be printed on the current line or the next line. Therefore, you must store all characters until you know which line it must be printed on.
You may want to use the function isblank to determine whether a character is a space or a tab character.
In accordance with the community guidelines on homework questions, I will not provide a full solution to the problem at this time. However, if required, I will provide further hints.
UPDATE 1:
The code in your most recend edit is not correct. According to the task description, you are supposed to split the line at the last blank before the treshold, not at the first blank after the threshold.
Most of the time when you encounter a new character, you cannot know whether it belongs in the current line or in a new line. Therefore, as already stated in my answer, you must remember all characters until you know which line they belong on. You should not print a character before you know whether it belongs on the current line or the next line.
You may find it easier to solve this problem if you use fgets instead of getchar. However, the problem can also be solved with getchar.
UPDATE 2:
Since you appear to be struggling with the problem for several days, I am now providing my solutions to the problem:
Here is my solution which uses getchar:
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <ctype.h>
#include <stdbool.h>
#include <assert.h>
#define CUT 10
int main( void )
{
//this memory buffer will hold the remembered
//line contents
char line[CUT+1];
//current index
int i = 0;
//whether a blank character has been encountered in
//the current line
bool found_blank = false;
//index of the last encountered blank character
int blank_index;
//this will store the current character
int c;
//this loop will read one character per loop iteration
while ( (c=getchar()) != EOF )
{
//if we encounter the end of the line, flush the
//buffer and start a new line
if ( c == '\n' )
{
line[i] = '\0';
puts( line );
i = 0;
found_blank = false;
continue;
}
//check if new character is blank and, if it is, then
//remember the index
if ( isblank(c) )
{
blank_index = i;
found_blank = true;
}
//write new character into array
line[i] = c;
if ( i == CUT - 1 )
{
if ( !found_blank )
{
//flush buffer
line[CUT] = '\0';
puts( line );
//reset line
i = 0;
found_blank = false;
continue;
}
else // found_blank == true
{
//remember character that is going to be overwritten
char temp = line[blank_index+1];
//output everything up to and including the last blank
line[blank_index+1] = '\0';
puts( line );
//move unwritten characters to start of next line
i = CUT - 1 - blank_index;
line[0] = temp;
memmove( line + 1, line+blank_index+2, i );
found_blank = false;
continue;
}
}
i++;
}
//check stream for error
if ( ferror(stdin) )
{
fprintf( stderr, "error reading input!\n" );
exit( EXIT_FAILURE );
}
//if we reach this, then we must have encountered
//end-of-file
assert( feof( stdin ) );
//flush buffer
line[i] = '\0';
puts( line );
}
Here is my solution which uses fgets:
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <ctype.h>
#include <stdbool.h>
#include <assert.h>
#define CUT 10
int main( void )
{
char line[CUT+2];
size_t leftovers = 0;
//in every iteration of the loop, we write one line of output
for (;;)
{
size_t blank_index;
bool found_blank = false;
if ( fgets( line + leftovers, sizeof line - leftovers, stdin ) == NULL )
{
//print all leftovers
line[leftovers] = '\0';
printf( "%s\n", line );
break;
}
size_t len = strlen( line );
//this check is probably not necessary
assert( len != 0 );
if ( line[len-1] == '\n' )
{
//we read a complete line, so print it
printf( "%s", line );
leftovers = 0;
continue;
}
//if we reach this, then input line was not read in
//completely, or we are dealing with the last line before
//end-of-file or input failure
//find last blank
for ( size_t i = 0; line[i] != '\0'; i++ )
{
if ( isblank( (unsigned char)line[i] ) )
{
found_blank = true;
blank_index = i;
}
}
if ( !found_blank )
{
if ( len <= CUT )
{
if ( ferror(stdin) )
{
fprintf( stderr, "error reading input!\n" );
exit( EXIT_FAILURE );
}
//we should only reach this if end-of-file was
//encountered in the middle of a line
assert( feof(stdin) );
printf( "%s\n", line );
break;
}
//remember last character
char temp = line[len-1];
line[len-1] = '\n';
printf( "%s", line );
line[0] = temp;
leftovers = 1;
}
else
{
//remember character after last blank
char temp = line[blank_index+1];
//replace character after last blank with terminator
line[blank_index+1] = '\0';
//print line up to and including last blank
printf( "%s\n", line );
if ( temp != '\0' )
{
//move unprinted characters to start of next line
line[0] = temp;
leftovers = len - blank_index - 1;
memmove( line + 1, line + blank_index + 2, leftovers - 1);
}
else
{
leftovers = 0;
}
}
}
}
It appears that my initial advice of using fgets instead of getchar was not very good, as the fgets solution turned out to be slightly more complicated than the getchar solution.
Here is some sample input and output of both programs (both programs behave the same way):
Sample input:
This is a line with several words.
Thisisalinewithonelongword.
Sample output:
This is a
line with
several
words.
Thisisalin
ewithonelo
ngword.
