HTML tags check in C - c

Question

I'm writing a little program in C, for check the HTML file have the right open and close tags?
but i've got some issues...
i have a file what contains the all possible tags, named tags.txt(those are only the first ones):
<a>
</a>
<abbr>
</abbr>
<area>
</area>
<aside>
</aside>
and i have the htmlfile.html, what I have to check:
<!--#echo var="date" -->
<area>
</area>
<area>
</area>
secondly, i want to replace the comments like this to the sysdate
like , the format is OK i can do it, but the prog puts in the file
this
my code:
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <time.h>
#define MAX_SIZE 512
void menu();
void check();
void datumos();
int main(int argc,char *argv[])
{
menu();
return 0;
}
void menu()
{
char menu[MAX_SIZE];
while(1 < 2)
{
printf("\npress a button:\n\n");
printf("\tFile HTML check..............:c\n");
printf("\t<!--#echo var="date" -->...........:d\n");
printf("\tExit:\tCTRL + C\n");
scanf("%s",menu);
if( strcmp(menu,"c") == 0 )
{
check();
}
else if( strcmp(menu,"d") == 0 )
{
datumos();
}
}
}
void check()
{
FILE *htmlfile;
FILE *checkfile;
htmlfile = fopen("htmlfile.html","w");
checkfile = fopen("tags.txt","r");
char line[MAX_SIZE];
char htmlline[MAX_SIZE];
char tags[189][30];
int i=0;
printf("\tcheck__1\n");
while(fgets(line,sizeof(line),checkfile) != NULL)
{
int j;
for(j=0; j<sizeof(line); ++j)
{
tags[i][j]=line[j];
}
++i;
}
printf("\tcheck__2\n");
int k=0; char htmlfiletags[MAX_SIZE][30];
while(fgets(htmlline,sizeof(htmlline),htmlfile) != NULL)
{
char currentline[sizeof(htmlline)];
int j=0;
if( currentline[j]=="<" )
{
while(currentline[j]!=">")
{
htmlfiletags[k][j]=currentline[j];
++j;
}
strcat(htmlfiletags[k][j+1],">");
++k;
}
}
printf("\tcheck__3\n");
int n;
for(n=0; n<sizeof(htmlfiletags); ++n)
{
int j; int howmanytimesnot=0;
for(j=0; j<sizeof(tags); ++j)
{
printf("\tcheck__3/1\n");
if(strcmp(htmlfiletags[n],tags[j])==0)
{
printf("\t%d\n", howmanytimesnot);
++howmanytimesnot;
}
}
printf("\tcheck__3/3\n");
if(!(howmanytimesnot<sizeof(tags)))
{
printf("\tcheck__3/4\n");
printf("the file is not wellformed");
exit (1);
}
}
printf("\tcheck__4\n");
}
void copy_file(const char *from,const char *to)
{
FILE *fr;
FILE *t;
fr = fopen(from,"r");
t = fopen(to,"w");
char line[MAX_SIZE];
char row[MAX_SIZE];
while(fgets(line,sizeof(line),fr) != NULL)
{
sscanf(line,"%s",row);
fprintf(t,"%s\n",row);
}
fclose(fr);
fclose(t);
remove("tempfile.html");
}
void datumos()
{
time_t now = time(NULL);
struct tm *t = localtime(&now);
char date_time[30];
strftime( date_time, sizeof(date_time), "%x_%X", t );
FILE *htmlfile;
FILE *tempfile;
htmlfile = fopen("htmlfile.html","r");
tempfile = fopen("tempfile.html","w");
char line[MAX_SIZE];
//char datecomment[]="<!--#echo var=date -->";
while(fgets(line,sizeof(line),htmlfile) != NULL)
{
if( strcmp(line,"<!--#echo var="date" -->") == 0 )
{
char row[40];
strcpy(row,"<!--");
strcat(row, date_time);
strcat(row,"-->");
printf("%s",row);
fputs(row,tempfile);
}
else
{
fputs(line,tempfile);
}
}
fclose(htmlfile);
fclose(tempfile);
copy_file("tempfile.html","htmlfile.html");
}
it dies in this, in the inner for loop, in the if at the 200th check... i dont know why...
int n;
for(n=0; n<sizeof(htmlfiletags); ++n)
{
int j; int howmanytimesnot=0;
for(j=0; j<sizeof(tags); ++j)
{
printf("\tcheck__3/1\n");
if(strcmp(htmlfiletags[n],tags[j])==0)
{
printf("\t%d\n", howmanytimesnot);
++howmanytimesnot;
}
}
printf("\tcheck__3/3\n");
if(!(howmanytimesnot<sizeof(tags)))
{
printf("\tcheck__3/4\n");
printf("the file is not wellformed");
exit (1);
}
}
Thanks for all reply!!
G

Answer 1

Your code is very complicated, it has several issues.
Here's one:
for(j=0; j<sizeof(tags); ++j)
this will not do what I believe you expect; sizeof(tags) is not the array length of tags (which is declared as char tags[189][30];), it's the total size of the variable. So, this loop will go from 0 to 189 * 30 - 1, i.e. 5669, and thus index way out beyond the end of array.
Also, the idea to use sizeof here in any way is wrong, since the content of tags comes from a file and it thus impossible for the compiler to know. Remember that sizeof is evaluated at compile-time, for expressions like these.
You need to have a variable (e.g. size_t num_tags) that you increment for each line parsed from the tags file, and that you later use to iterate over tags.

Answer 2

Do not use regex, or some kind of string parsing, to parse HTML. Instead search the web, or this site, for a c library to parse html. Then check the parsed HTML file for the tags. This will ease the development a lot as you don't have to parse the files yourself.

Answer 3

i've fixed some things, but
- i still cant check the file's htmltags, dies at the same loop, i've fixed allocation of the tags array
- when in the htmlfile are 2 or more different comments and i'm replacing the comment the program replaces it with the sysdate, but the program copy the another comments badly, like =>
the code is now:
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include <time.h>
#define MAX_SIZE 512
void menu();
void check();
void datumos();
int main(int argc,char *argv[])
{
menu();
return 0;
}
void menu()
{
char menu[MAX_SIZE];
while(1 < 2)
{
printf("\npress a button:\n\n");
printf("\tFile HTML check..............:c\n");
printf("\t<!--#echo var="date" -->...........:d\n");
printf("\tExit:\tCTRL + C\n");
scanf("%s",menu);
if( strcmp(menu,"c") == 0 )
{
check();
}
else if( strcmp(menu,"d") == 0 )
{
datumos();
}
}
}
void check()
{
FILE *htmlfile;
FILE *checkfile;
htmlfile = fopen("htmlfile.html","r");
checkfile = fopen("tags.txt","r");
char line[MAX_SIZE];
char htmlline[MAX_SIZE];
int i2=0;
printf("\tcheck__1\n");
while(fgets(line,sizeof(line),checkfile) != NULL)
{
++i2;
}
char tags[i2][20];
int i=0;
printf("\tcheck__11\n");
while(fgets(line,sizeof(line),checkfile) != NULL)
{
int j;
for(j=0; j<sizeof(line); ++j)
{
tags[i][j]=line[j];
}
++i;
}
printf("\tcheck__2\n");
int k=0; char htmlfiletags[MAX_SIZE][30];
while(fgets(htmlline,sizeof(htmlline),htmlfile) != NULL)
{
char currentline[sizeof(htmlline)];
int j=0;
if( currentline[j]=="<" )
{
while(currentline[j]!=">")
{
htmlfiletags[k][j]=currentline[j];
++j;
}
strcat(htmlfiletags[k][j+1],">");
++k;
}
}
printf("\tcheck__3\n");
int n;
for(n=0; n<sizeof(htmlfiletags); ++n)
{
int j; int howmanytimesnot=0;
for(j=0; j<sizeof(tags); ++j)
{
//printf("\tcheck__3/1\n");
if(strcmp(htmlfiletags[n],tags[j])==0)
{
// printf("\t%d\n", howmanytimesnot);
++howmanytimesnot;
}
}
printf("\tcheck__3/3\n");
if(!(howmanytimesnot<sizeof(tags)))
{
printf("\tcheck__3/4\n");
printf("the file is not wellformed");
exit (1);
}
}
printf("\tcheck__4\n");
}
void copy_file(const char *from,const char *to)
{
FILE *fr;
FILE *t;
fr = fopen(from,"r");
t = fopen(to,"w");
char line[MAX_SIZE];
char row[MAX_SIZE];
while(fgets(line,sizeof(line),fr) != NULL)
{
sscanf(line,"%s",row);
fprintf(t,"%s\n",row);
}
fclose(fr);
fclose(t);
remove("tempfile.html");
}
void datumos()
{
time_t now = time(NULL);
struct tm *t = localtime(&now);
char date_time[30];
strftime( date_time, sizeof(date_time), "%x_%X", t );
FILE *htmlfile;
FILE *tempfile;
htmlfile = fopen("htmlfile.html","r");
tempfile = fopen("tempfile.html","w");
char line[MAX_SIZE];
char* datecomment="<!--#echo var=\"date\" -->";
while(fgets(line,sizeof(line),htmlfile) != NULL)
{
int i3; int db=0;
for(i3=0; i3<strlen(datecomment); ++i3)
{
if(line[i3]==datecomment[i3])
{
++db;
}
}
if(db==strlen(datecomment))
{
char row[30];
strcpy(row,"<!--");
strcat(row, date_time);
strcat(row,"-->\n");
fputs(row,tempfile);
}
else
{
fputs(line,tempfile);
}
}
fclose(htmlfile);
fclose(tempfile);
copy_file("tempfile.html","htmlfile.html");
}

Answer 4

the currentline it's not necessary,and i've fixed the compares too
while(fgets(htmlline,sizeof(htmlline),htmlfile) != NULL)
{
int j=0;
if( htmlline[j]=='<' )
{
while(htmlline[j]!='>')
{
htmlfiletags[k][j]=htmlline[j];
++j;
}
strcat(htmlfiletags[k][j+1],">");
++k;
}
}
-in addition, the another problem to replace only the suitable comments, and dont hurt the different ones still dont work
"so it replaces
<!--#echo var="date" --> to the sysdate, it's ok, but when there are different comments like
<!--#include something -->, it wont be copied back well, in the htmlfile will be only <!--#include"
ideas?