remove entry from text file [NOT A HOMEWORK]

[Roman Mashak]

Hello, All!

I implemented simple program to eliminate entry from the file having the
following structure (actually it's config file of 'named' DNS package for
those who care and know):

options {
directory "/var/named";
listen-on { 192.168.11.22; 127.0.0.1; };
forwarders { 168.126.63.1; };
pid-file "/var/run/named.pid";
};

controls {
unix "/var/run/ndc" perm 0600 owner 0 group 0;
};

zone "." {
type hint;
file "root.cache ";
};

zone "localhost" {
type master;
file "localhost" ;
};

zone "my.local" {
type master;
file "my.local";
allow-update { 127.0.0.1; 192.168.11.0/24; };
};
....

I bother only about "zone" entries and the goal of code to remove them only.
Here is the source code I'm done with. I'd like to hear reasonable criticism
and useful advices :-) on optimization, bugs etc.

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

#define BUFSIZE 100

int main(int argc, char **argv)
{
FILE *f; /* original */
FILE *fn; /* new */
char buf[BUFSIZE] = { 0 };
register char *idx;
char sidx[BUFSIZE];
int found = 0; /* set if zone found */
int i = 0;

if (argc != 3) {
fprintf(stderr, "Usage: %s <zonename> <configfile>\n" , *argv);
return (0);
}

printf("%s %s\n", argv[1], argv[2]);

/* open original */
if ( (f = fopen(argv[2], "r")) == NULL )
return -1;

if ( (fn = fopen("named.co nf.tmp", "w+")) == NULL )
return -1;

while ( !feof(f) ) {
fgets(buf, BUFSIZE, f);

/* skip leading spaces */
for (idx = buf; *idx != '\0' && (*idx == ' ' || *idx == '\t'); idx++)
;

/* look for 'zone' token */
if ( strncasecmp(idx , "zone ", sizeof("zone ") - 1) != 0 ) {
if (!found) {
fprintf(fn, buf);
fflush(fn);
}
continue;
}

idx += sizeof("zone ") - 1;

/* skip spaces inside of token */
for (; *idx != '\0' && (*idx == ' ' || *idx == '\t'); idx++)
;

/* extract token from commas */
if (*idx++ != '"')
continue;

/* put token into buffer to compare */
while (*idx != '\0' && *idx != '"')
sidx[i++] = *idx++;

sidx[i] = '\0'; i = 0;
if (strcasecmp(sid x, argv[1]) != 0) {
found = 0;
fprintf(fn, buf);
fflush(fn);
}
else
found = 1;
}

fclose(f);
fclose(fn);

return 0;
}

Thank you in advance.

With best regards, Roman Mashak. E-mail: mr*@tusur.ru

[tmp123]

Roman Mashak wrote:

Hello, All!

I implemented simple program to eliminate entry from the file having the
following structure (actually it's config file of 'named' DNS package for
those who care and know):

options {
directory "/var/named";
listen-on { 192.168.11.22; 127.0.0.1; };
forwarders { 168.126.63.1; };
pid-file "/var/run/named.pid";
};

controls {
unix "/var/run/ndc" perm 0600 owner 0 group 0;
};

zone "." {
type hint;
file "root.cache ";
};

zone "localhost" {
type master;
file "localhost" ;
};

zone "my.local" {
type master;
file "my.local";
allow-update { 127.0.0.1; 192.168.11.0/24; };
};
...

I bother only about "zone" entries and the goal of code to remove them only.
Here is the source code I'm done with. I'd like to hear reasonable criticism
and useful advices :-) on optimization, bugs etc.

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

#define BUFSIZE 100

int main(int argc, char **argv)
{
FILE *f; /* original */
FILE *fn; /* new */
char buf[BUFSIZE] = { 0 };
register char *idx;
char sidx[BUFSIZE];
int found = 0; /* set if zone found */
int i = 0;

if (argc != 3) {
fprintf(stderr, "Usage: %s <zonename> <configfile>\n" , *argv);
return (0);
}

printf("%s %s\n", argv[1], argv[2]);

/* open original */
if ( (f = fopen(argv[2], "r")) == NULL )
return -1;

if ( (fn = fopen("named.co nf.tmp", "w+")) == NULL )
return -1;

while ( !feof(f) ) {
fgets(buf, BUFSIZE, f);

/* skip leading spaces */
for (idx = buf; *idx != '\0' && (*idx == ' ' || *idx == '\t'); idx++)
;

/* look for 'zone' token */
if ( strncasecmp(idx , "zone ", sizeof("zone ") - 1) != 0 ) {
if (!found) {
fprintf(fn, buf);
fflush(fn);
}
continue;
}

idx += sizeof("zone ") - 1;

/* skip spaces inside of token */
for (; *idx != '\0' && (*idx == ' ' || *idx == '\t'); idx++)
;

/* extract token from commas */
if (*idx++ != '"')
continue;

/* put token into buffer to compare */
while (*idx != '\0' && *idx != '"')
sidx[i++] = *idx++;

sidx[i] = '\0'; i = 0;
if (strcasecmp(sid x, argv[1]) != 0) {
found = 0;
fprintf(fn, buf);
fflush(fn);
}
else
found = 1;
}

fclose(f);
fclose(fn);

return 0;
}

Thank you in advance.

With best regards, Roman Mashak. E-mail: mr*@tusur.ru

Look for references about "lex" C source generator (or even "yacc" if
you have more complex requirements like this one)

[Sirius Black]

Roman Mashak wrote:

Hello, All!

I implemented simple program to eliminate entry from the file having the
following structure (actually it's config file of 'named' DNS package for
those who care and know):
<snip>
if ( (fn = fopen("named.co nf.tmp", "w+")) == NULL )
return -1;

Here, because of opening the file for writing in present directory, it
will restrict the directory from which a user can run your program. If a
user does n't have write permission for the present directory, then he
will have to change the directory and run your program.

Probably you should have explained algorithm used in your program in
short paragraph.

[boa]

Roman Mashak wrote:

Hello, All!

I implemented simple program to eliminate entry from the file having the
following structure (actually it's config file of 'named' DNS package for
those who care and know):

options {
directory "/var/named";
listen-on { 192.168.11.22; 127.0.0.1; };
forwarders { 168.126.63.1; };
pid-file "/var/run/named.pid";
};

controls {
unix "/var/run/ndc" perm 0600 owner 0 group 0;
};

zone "." {
type hint;
file "root.cache ";
};

zone "localhost" {
type master;
file "localhost" ;
};

zone "my.local" {
type master;
file "my.local";
allow-update { 127.0.0.1; 192.168.11.0/24; };
};
...

I bother only about "zone" entries and the goal of code to remove them only.
Here is the source code I'm done with. I'd like to hear reasonable criticism
and useful advices :-) on optimization, bugs etc.

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

#define BUFSIZE 100

int main(int argc, char **argv)
{
FILE *f; /* original */
FILE *fn; /* new */
char buf[BUFSIZE] = { 0 };
register char *idx;
char sidx[BUFSIZE];
int found = 0; /* set if zone found */
int i = 0;

if (argc != 3) {
fprintf(stderr, "Usage: %s <zonename> <configfile>\n" , *argv);
return (0);
Maybe return EXIT_FAILURE to indicate an error?
}

printf("%s %s\n", argv[1], argv[2]);

/* open original */
if ( (f = fopen(argv[2], "r")) == NULL )
return -1;
Again, EXIT_FAILURE.

if ( (fn = fopen("named.co nf.tmp", "w+")) == NULL )
How about using tmpnam() instead?
return -1;

while ( !feof(f) ) {
fgets(buf, BUFSIZE, f);
See http://c-faq.com/stdio/feof.html

/* skip leading spaces */
for (idx = buf; *idx != '\0' && (*idx == ' ' || *idx == '\t'); idx++)
;

/* look for 'zone' token */
if ( strncasecmp(idx , "zone ", sizeof("zone ") - 1) != 0 ) {
strlen("zone"), not sizeof().
if (!found) {
fprintf(fn, buf);
fflush(fn);
Why do you need to call fflush()?

}
continue;
}

idx += sizeof("zone ") - 1;

/* skip spaces inside of token */
for (; *idx != '\0' && (*idx == ' ' || *idx == '\t'); idx++)
;
How about a function that strips the white space?


/* extract token from commas */
if (*idx++ != '"')
continue;

/* put token into buffer to compare */
while (*idx != '\0' && *idx != '"')
sidx[i++] = *idx++;

sidx[i] = '\0'; i = 0;
if (strcasecmp(sid x, argv[1]) != 0) {
found = 0;
fprintf(fn, buf);
fflush(fn);
}
else
found = 1;
}

fclose(f);
fclose(fn);

return 0;
}

That's a start, HTH.

Boa

[Micah Cowan]

"Roman Mashak" <mr*@tusur.ru > writes:

Hello, All!

I implemented simple program to eliminate entry from the file having the
following structure (actually it's config file of 'named' DNS package for
those who care and know):
<snip>
zone "." {
type hint;
file "root.cache ";
};

zone "localhost" {
type master;
file "localhost" ;
};

zone "my.local" {
type master;
file "my.local";
allow-update { 127.0.0.1; 192.168.11.0/24; };
};
<snip> (here's the code:)
#include <stdio.h>
#include <string.h>
#include <strings.h>
The last of these is not Standard C; see my discussion later.

#define BUFSIZE 100

int main(int argc, char **argv)
{
FILE *f; /* original */
FILE *fn; /* new */
char buf[BUFSIZE] = { 0 };
register char *idx;
char sidx[BUFSIZE];
int found = 0; /* set if zone found */
int i = 0;

if (argc != 3) {
fprintf(stderr, "Usage: %s <zonename> <configfile>\n" , *argv);
return (0);
}

printf("%s %s\n", argv[1], argv[2]);

/* open original */
if ( (f = fopen(argv[2], "r")) == NULL )
return -1;
A return of -1 will have an implementation-defined meaning... slightly
better would be to #include <stdlib.h> and return EXIT_FAILURE; which
is guaranteed to have the obvious meaning on any current or future
implementation.

Also, it would really pay to emit some sort of message here. You'll
really thank yourself when you accidentally open a non-existant file,
or non-readable.... etc.
if ( (fn = fopen("named.co nf.tmp", "w+")) == NULL )
return -1;

while ( !feof(f) ) {
fgets(buf, BUFSIZE, f);

/* skip leading spaces */
for (idx = buf; *idx != '\0' && (*idx == ' ' || *idx == '\t'); idx++)
;

/* look for 'zone' token */
if ( strncasecmp(idx , "zone ", sizeof("zone ") - 1) != 0 ) {
Whenever I find myself having to use the exact same string literal
twice like this, it's time to farm it out. Using something like:

const char zone_token[] = "zone ";
...
if (strncasecmp(id x, zone_token, (sizeof zone_token)-1) != 0)

saves you from the problem of accidental typos. If the number of
characters between the two of those were to accidentally get out of
sync, it'd be a bug. And even though you have them exactly right now,
you might later need to change the code to use a different literal,
and accidentally change only one.

BTW, strncasecmp() is not a Standard C function (it's POSIX, which is
not on-topic here). The residents on this newsgroup insist that you do
not post code containing functions that are (a) not Standard C
functions and (b) not defined in the provided code. For the sake of
example code on this newsgroup, the easiest way for you to avoid
difficulties is probably to use strncmp() instead. Or, define your own
implementation of strncasecmp().. .
if (!found) {
fprintf(fn, buf);
fflush(fn);
}
continue;
}

idx += sizeof("zone ") - 1;
If you decide to follow my advice, this would become:

idx += (sizeof zone_array) - 1;

Actually, my own preference has been to define a macro such as:

#define ARY_STRLEN(a) (sizeof(a)-1)

which leads to a slightly better expression, in terms of
self-documentation:

idx += ARY_STRLEN(zone _array) - 1;
/* skip spaces inside of token */
for (; *idx != '\0' && (*idx == ' ' || *idx == '\t'); idx++)
;
This is the same thing as:

idx += strspn(idx, " \t");

/* extract token from commas */
if (*idx++ != '"')
continue;

/* put token into buffer to compare */
while (*idx != '\0' && *idx != '"')
sidx[i++] = *idx++;

sidx[i] = '\0'; i = 0;
The second statement on the line above is easy to miss. I'd recommend
giving each of these a separate line.
if (strcasecmp(sid x, argv[1]) != 0) {
found = 0;
fprintf(fn, buf);
fflush(fn);
}
else
found = 1;
}

fclose(f);
fclose(fn);

return 0;
}

The last pair of fclose()s are not strictly necessary, if you're not
going to check their return value. However, I'd recommend that you do
check their return value, so you can emit an error message if fn
didn't flush properly... actually, you should check the return code of
your fflush() calls, too.

I'll point out that the code above won't work properly if your config
file contains lines longer than 100 characters.

-Micah

[Micah Cowan]

boa <bo*****@gmail. com> writes:

Roman Mashak wrote:

/* skip leading spaces */
for (idx = buf; *idx != '\0' && (*idx == ' ' || *idx == '\t'); idx++)
;
/* look for 'zone' token */
if ( strncasecmp(idx , "zone ", sizeof("zone ") - 1) != 0 ) {

strlen("zone"), not sizeof().

You mean, strlen("zone ") (with the space); and there's nothing wrong
with sizeof: it's much more efficient.

if (!found) {
fprintf(fn, buf);
fflush(fn);

Why do you need to call fflush()?

Could be useful if he's using another program to view the temporary
file while it's being written (UNIX tail...).

Slightly better might be to set line buffering on fn with setvbuf().

-Micah

[Pedro Graca]

Micah Cowan wrote:

boa <bo*****@gmail. com> writes:

Roman Mashak wrote:

> /* look for 'zone' token */
> if ( strncasecmp(idx , "zone ", sizeof("zone ") - 1) != 0 ) {

strlen("zone"), not sizeof().

You mean, strlen("zone ") (with the space); and there's nothing wrong
with sizeof: it's much more efficient.

Hmmmm ???

What's the type of a literal string in code?
I thought it was `const char *'

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

int main(void) {
int a, b, c, d, e, f;
char * z1 = "forty two";
char z2[] = "forty two";

a = strlen("forty two");
b = (int)sizeof("fo rty two");
c = (int)sizeof(cha r*);
d = (int)sizeof(con st char*);
e = (int)sizeof z1;
f = (int)sizeof z2; /* Ah! A match */
printf("%d, %d, %d, %d, %d, %d\n",
a, b, c, d, e, f);
return 0;
}
No Micah, I wasn't doubting you (... well, maybe just a little)
I stand corrected.

--
If you're posting through Google read <http://cfaj.freeshell. org/google>

[pete]

Pedro Graca wrote:

What's the type of a literal string in code?

An array of some number of char.
No implict const qualifier,
just trouble if you attempt to modify the object.

--
pete

[Keith Thompson]

pete <pf*****@mindsp ring.com> writes:

Pedro Graca wrote:

What's the type of a literal string in code?

An array of some number of char.
No implict const qualifier,
just trouble if you attempt to modify the object.

"trouble" being undefined behavior.

--
Keith Thompson (The_Other_Keit h) ks***@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.

[Roman Mashak]

Hello, boa!
You wrote on Thu, 23 Feb 2006 20:16:14 +0100:

??>> #include <strings.h>

b> strings.h?
I should've apologized for using of non-standard stuffs :)
??>> #define BUFSIZE 100
??>>
??>> int main(int argc, char **argv)
??>> if ( (fn = fopen("named.co nf.tmp", "w+")) == NULL )

b> How about using tmpnam() instead?
Manual of my system strongly recommends never to use this function cause of
being unsafe.
??>> return -1;
??>>
??>> while ( !feof(f) ) {
??>> fgets(buf, BUFSIZE, f);

b> See http://c-faq.com/stdio/feof.html
Yeas, that was the problem I run into: printing '\n' twice, BUT it happened
only if I removed entry from the middle of file, otherwise everything came
right.
Thanks for hint.

??>> if (!found) {
??>> fprintf(fn, buf);
??>> fflush(fn);

b> Why do you need to call fflush()?
I view the result file in the next console via "tail -f" (I'm using linux
presently).
??>> }
??>> continue;
??>> }
??>>

With best regards, Roman Mashak. E-mail: mr*@tusur.ru