Line Counting

00Sven · 03-31-2006, 08:02 PM

I have been looking and determined that there is no C function that counts the number of lines in a file. I would like to make a function to do this but do not really know how. So far I have this

Code:

int linecount (FILE *file){
	char ar[1024];
	int count=0;
	while(feof(file)==0){
		fgets(ar,1024,file);
		count++;
	}
    return(count);
}

This for some reason always seems to return 0 though. Can someone please help with this. Thanks.
~Sven

cdalten · 03-31-2006, 09:11 PM

Maybe something like this

input file

line1
line2
line3
line4

Code:

/*Count number of lines*/

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

#define BORED "/home/cdalten/cstuff/cstuff.txt"
#define MAXLINE 400

int linecount(FILE *fp) {
    char buff[MAXLINE];
    int count = 0;

    while(fgets(buff,MAXLINE,fp) != NULL) {
	count++;
    }
    return count;
}


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

    FILE *fp;

    int value = 0;

    if ((fp=fopen(BORED,"r")) != NULL) {
	  value = linecount(fp);
    }

    if(feof(fp)){
        printf("The value is: %d \n", value);
	return 0;
    }

    if(ferror(fp)){
	perror(argv[0]);
	return -1;
    }

    fclose(fp);
    return 0;
}

$gcc -Wall lc.c -o lc
$./lc
The value is: 4
$

enggabhinandan · 03-31-2006, 11:27 PM

Hey, I think u can go for one more way of doing it by searching for "\n" and increase the count by one whenever u find "\n". This may work fine

WaltP · 04-01-2006, 12:24 AM

Quote:

Originally Posted by 00Sven

I have been looking and determined that there is no C function that counts the number of lines in a file. I would like to make a function to do this but do not really know how. So far I have this

Code:

int linecount (FILE *file){
	char ar[1024];
	int count=0;
	while(feof(file)==0){
		fgets(ar,1024,file);
		count++;
	}
    return(count);
}

This for some reason always seems to return 0 though. Can someone please help with this. Thanks.
~Sven

Your concept is fine, with a couple small problems.

Problem #1 is your use of feof(). See this FAQ for an explanation.

Next, what if a line is 1123 characters? fgets() will read the first 1023 and the next fgets() will read the last 100 characters. Test the last character in your buffer (using strlen()) for the \n and increase the count only if it's there.

cdalten · 04-01-2006, 02:12 PM

Quote:

Originally Posted by WaltP

Your concept is fine, with a couple small problems.

Problem #1 is your use of feof(). See this FAQ for an explanation.

Next, what if a line is 1123 characters? fgets() will read the first 1023 and the next fgets() will read the last 100 characters. Test the last character in your buffer (using strlen()) for the \n and increase the count only if it's there.

I haven't test this, but wouldn't something like this also work

Code:

int linecount = 0; 
int lastchar = '\n'; 

while (fgets(buffer, MAXLINE, fp)) 
         if (strchr(buffer, lastchar)) 
             ++count; 
 
    if (ferror(fp)) { 
         fprintf(stderr, "Error reading the input file\n"); 
         exit(EXIT_FAILURE); 
     } 
 
   if (!strchr(buffer, lastchar)) 
         ++count;

**Salem** · 04-01-2006, 02:19 PM

Why do you have two strchr() calls?
What does the last one do for you?

cdalten · 04-01-2006, 02:30 PM

It was the first thing that came to my snoodle when I thought about treating any trailing characters without a new-line as a line.

**Salem** · 04-01-2006, 02:41 PM

True, but you also have to contend with a file which may be empty as well.

cdalten · 04-01-2006, 02:49 PM

Code:

while (fgets(buffer, MAXLINE, fp))

is the same as

Code:

while (fgets(buffer, MAXLINE, fp) != NULL)

If the file exists and there is nothing in the file, the code should fail. Right? Again, I didn't test this.

dwks · 04-01-2006, 03:08 PM

Yes, fgets() returns NULL if no characters were stored in the string before EOF was encountered.

This code:

Code:

while(fgets(line, sizeof(line), fp)) count ++;

is better than just counting newlines, because you can have a line in a file that isn't terminated by a newline. That is, if you have a file with the contents "foo", the fgets() method will claim the file has one line (which is probably what you want), whereas just counting the newlines would give you zero lines.

Sysop_fb · 04-01-2006, 05:01 PM

Quote:

Originally Posted by dwks

Yes, fgets() returns NULL if no characters were stored in the string before EOF was encountered.

This code:

Code:

while(fgets(line, sizeof(line), fp)) count ++;

is better than just counting newlines, because you can have a line in a file that isn't terminated by a newline. That is, if you have a file with the contents "foo", the fgets() method will claim the file has one line (which is probably what you want), whereas just counting the newlines would give you zero lines.

True but what happens if the line exceeds the array limits? It would continue to pull in data until it reached the new-line character each time doing a call to fgets and incrementing the counter each time as well?

cdalten · 04-01-2006, 07:13 PM

Hmm....I never thought about using sizeof(). Hahaha. Inexperience rears its ugly head huh?

Tonto · 04-01-2006, 08:08 PM

Well this is tough because we can't just check for like;

Code:

if(line[strlen(line) - 1] == '\n')
{
	count++;
}

Because of the aforementioned single line then EOF thing. Is there any sort of straight-forward solution? Is fgets a bad way to do this maybe?

whoie · 04-02-2006, 12:18 AM

Quote:

...Because of the aforementioned single line then EOF thing. Is there any sort of straight-forward solution? Is fgets a bad way to do this maybe?

I don't see a compelling reason to use fgets here. After all, all versions posted so far simply discard the input. Using [f]scanf to discard the characters might yield a nicer solution.

The call scanf("%*[^\n]"); means "read all characters up to but *not* including the newline character and discard them". The '*' in this case means assignment suppression. A following call to getchar() removes the '\n' if it was there, otherwise it simply does nothing and evaluates to EOF.

Any solution should handle:
1) Empty files
2) Files with blank lines
3) Files with ill-formed last lines (no newline character before EOF)
4) Well formed last lines (newline character appears before EOF)

Perhaps even more, but I can't think of any off the top of my head.

HTH,
Will

WaltP · 04-02-2006, 12:57 AM

fgets() is the way to do this. After you read a line:
1) if the end of the text read ends with \n, count it. You read a line.
2) if the end of the text read does not end with \n, don't count it. You read the start but not the end of the line. The input buffer will be full in this case.
3) if you get EOF, you have reached the end of the file, exit the read loop. The input buffer still has the previous line read. Now test the end character again. If it does not contain a \n, count it. It's the last line.

03-31-2006, 11:27 PM	#3
enggabhinandan Registered User Join Date: Mar 2006 Location: Bangalore, INDIA Posts: 43	Hey, I think u can go for one more way of doing it by searching for "\n" and increase the count by one whenever u find "\n". This may work fine
enggabhinandan is offline

04-01-2006, 02:19 PM	#6
Salem and the hat of Jobseeking Join Date: Aug 2001 Location: The edge of the known universe Posts: 21,639	Why do you have two strchr() calls? What does the last one do for you? __________________ If you dance barefoot on the broken glass of undefined behaviour, you've got to expect the occasional cut.
Salem is offline

04-01-2006, 02:30 PM	#7
cdalten Registered User Join Date: Jan 2006 Location: Berkeley, Ca Posts: 196	It was the first thing that came to my snoodle when I thought about treating any trailing characters without a new-line as a line.
cdalten is offline

04-01-2006, 02:41 PM	#8
Salem and the hat of Jobseeking Join Date: Aug 2001 Location: The edge of the known universe Posts: 21,639	True, but you also have to contend with a file which may be empty as well. __________________ If you dance barefoot on the broken glass of undefined behaviour, you've got to expect the occasional cut.
Salem is offline

04-01-2006, 02:49 PM	#9
cdalten Registered User Join Date: Jan 2006 Location: Berkeley, Ca Posts: 196	Code: while (fgets(buffer, MAXLINE, fp)) is the same as Code: while (fgets(buffer, MAXLINE, fp) != NULL) If the file exists and there is nothing in the file, the code should fail. Right? Again, I didn't test this.
cdalten is offline

04-01-2006, 03:08 PM	#10
dwks Frequently Quite Prolix Join Date: Apr 2005 Location: Canada Posts: 7,698	Yes, fgets() returns NULL if no characters were stored in the string before EOF was encountered. This code: Code: while(fgets(line, sizeof(line), fp)) count ++; is better than just counting newlines, because you can have a line in a file that isn't terminated by a newline. That is, if you have a file with the contents "foo", the fgets() method will claim the file has one line (which is probably what you want), whereas just counting the newlines would give you zero lines. __________________ dwk Seek and ye shall find. quaere et invenies. "Simplicity does not precede complexity, but follows it." -- Alan Perlis "Testing can only prove the presence of bugs, not their absence." -- Edsger Dijkstra "The only real mistake is the one from which we learn nothing." -- John Powell Other boards: DaniWeb, TPS Unofficial Wiki FAQ: cpwiki.sf.net My website: http://dwks.theprogrammingsite.com/ Projects: codeform, xuni, atlantis, nort, etc.
dwks is offline

04-01-2006, 07:13 PM	#12
cdalten Registered User Join Date: Jan 2006 Location: Berkeley, Ca Posts: 196	Hmm....I never thought about using sizeof(). Hahaha. Inexperience rears its ugly head huh?
cdalten is offline

04-01-2006, 08:08 PM	#13
Tonto Registered User Join Date: Jun 2005 Location: New York Posts: 1,465	Well this is tough because we can't just check for like; Code: if(line[strlen(line) - 1] == '\n') { count++; } Because of the aforementioned single line then EOF thing. Is there any sort of straight-forward solution? Is fgets a bad way to do this maybe?
Tonto is offline

04-02-2006, 12:57 AM	#15
WaltP Been here, done that. Join Date: May 2003 Posts: 1,036	fgets() is the way to do this. After you read a line: 1) if the end of the text read ends with \n, count it. You read a line. 2) if the end of the text read does not end with \n, don't count it. You read the start but not the end of the line. The input buffer will be full in this case. 3) if you get EOF, you have reached the end of the file, exit the read loop. The input buffer still has the previous line read. Now test the end character again. If it does not contain a \n, count it. It's the last line. __________________ There are only 10 types of people in the world -- those that use binary, and those that don't
WaltP is offline

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Imposing Line Numbers automatically in the C code	cavestine	C Programming	14	10-15-2007 12:41 AM
Adding Line numbers in Word	Mister C	A Brief History of Cprogramming.com	24	06-24-2004 08:45 PM
Trouble replacing line of file	Rpog	C Programming	4	04-19-2004 10:22 AM
Contest Results - May 27, 2002	ygfperson	A Brief History of Cprogramming.com	18	06-18-2002 01:27 PM
follow up with char I/O and line, word, and char counting	Led Zeppelin	C Programming	1	03-23-2002 09:26 PM