HTML to TXT

Hello Everybody,
I have a question regarding the conversion of files from HTML to TXT. Im fairly new to the C language but understand the basics. Im a little confused with what to do.I have to write a program that asks the user to input a html file and then the program converts it to txt.
Thanks-

You may have a question, but you failed to ask it. :)

huh?

You would want to open the HTML file for reading. Read it and output the contents to another file with a .txt extension.

You can find out how to do that here:
http://faq.cprogramming.com/cgi-bin/...&id=1043284392

Quote:

---

Originally Posted by eobergfell

huh?

---

That's the first question you actually asked.

Look at your first post. There is no question. You just said you wanted to make a new txt file from an HTML file. Someone concluded that meant you just want a copy of the same file with a different file extension. I think it could mean you want to strip all the tags and create a new txt file with just straight text from the original file.

You were first too ambiguous in your initial description of what you wanted to do. Your second fault was to not even ask a question, and somehow hope we would just know what to say.

Have a look at this: http://www.catb.org/~esr/faqs/smart-questions.html

:)

Quote:

---

Originally Posted by eobergfell

Hello Everybody,
I have a question regarding the conversion of files from HTML to TXT. Im fairly new to the C language but understand the basics. Im a little confused with what to do.I have to write a program that asks the user to input a html file and then the program converts it to txt.
Thanks-

---

Your program will be very similar to one's that count up words. You'll need to keep track of the current position in the file, and whether that position is INSIDE an html tag, (in which case the text char's found there do NOT need to be copied to the text file you're program will make), or that position is OUTSIDE any html tag, and therefore is just text char's that DO need to be copied into the text file.

So your program will scan the html file, and keep an integer flag variable "Inside", equal to 1 (yes, it's inside currently), or equal to 0, (no, it is not inside, currently).

Now all you have to do is learn what are the border char's, that always distinguish html tags?

The more specific your questions are, the better we can answer your questions. Be sure to post up some code of yours as you ask, as well. That will also help guide our communications, and show that you are working on it, also.

Thanks mike and adak and to MacGyver keep being a sweetheart okay?

Yea I have to strip off all the tags <> and make it into a readable plan TXT file

If you want to strip XML (or HTML, same thing basically) properly, you need a finite state machine (never mind, it's just a term) with three states.

• Outside any tag.
• Inside a tag.
• Inside an attribute.

If you don't have the last state, your code would be fooled by

Code:

---

<tag data=">" />

---

Anyway, the logic is quite simple. If you're outside a tag, look for a '<', which would involve changing to mode inside-a-tag. If you're inside a tag, look for a ' or ", which would make the mode inside-an-attribute. Ditto for the closing characters.

Code:

---

#include<stdio.h>
int main()
{
        int a, b;
        int input;
        FILE* outfile;
        char filename[80];
        printf("Enter Filename: ");
        gets(filename);
        outfile = fopen(filename, "w");
        scanf("&#37;d", &input);

        fprintf(outfile, "%d\n",input );
        fclose(outfile);
}

---

thats what ive done so far...I understand the logic behind this "If you're outside a tag, look for a '<', which would involve changing to mode inside-a-tag. If you're inside a tag, look for a ' or ", which would make the mode inside-an-attribute" But I don't know how to write it out.
Help, Anybody?

Don't use gets(). It's bad. See the FAQ.

How are you going to read from the file if you open it in "w" mode? Or are you reading from stdin?

Regardless, the logic goes something like this.

Code:

---

int mode = outside;

while(get a character) {
    if(mode == outside) {
        if(char == '<') mode = inside;
        else write character
    }
    else if(mode == inside) {
        if(char == '>') mode = outside;
        else if(char == '"' || char == '\'') mode = attribute;
    }
    else {  /* mode = attribute */
        /* ... */
    }
}

---

An enum would be useful here. http://www.cprogramming.com/tutorial/enum.html

Code:

---

enum {
    MODE_OUTSIDE,
    MODE_INTAG,
    MODE_INATTRIBUTE
} mode = MODE_OUTSIDE;

---

But you can use ordinary variables too of course.

Using gets() on a C board is sort of like putting a match out on a burn victim - very touchy about that. Check the faq on how to use fgets() which is safe, for input from stdin (when needed in your program). You won't get much help beyond "don't use gets()!" until that's done.

You'll definitely want a variable named "InsideTag" (or OutsideTag), and InsideAttrib, (or OutsideAttrib). Please get those added and initialized to out.

While you're trying to code this, I'd temporarily /* remark out */ the file names, etc., until you get a handle on the logic. It will greatly reduce your frustration, and speed everything up.

Quote:

---

You'll definitely want a variable named "InsideTag" (or OutsideTag), and InsideAttrib, (or OutsideAttrib). Please get those added and initialized to out.

---

It makes more sense to have a single variable to hold three states, since the program can only be in one state at once. But two variables would work as well.

Hey everybody thanks, im having trouble with using the "enum function"

Code:

---

enum Status{OUTSIDE = ?, INSIDE = ?);

---

I dont know what to set OUTSIDE and INSIDE equals to.
If anyone could help I would really appreciate it. Thanks-

The first value in an enum is automatically set to zero. Succeeding values are automatically set to one more than the previous value. But you can override this if you like.

Code:

---

enum numbers {
    minusone = -1,
    zero,
    three = 3,
    four,
    five
};

---

But the point of enums is that the values don't matter. You just have something like

Code:

---

enum {
    MODE_OUTSIDE,
    MODE_INTAG,
    MODE_INATTRIBUTE
} mode = MODE_OUTSIDE;

---

and you always compare the enum to those values.

Code:

---

if(mode == MODE_OUTSIDE) mode = MODE_INTAG;
if(mode != MODE_INTAG) puts("not in tag");

---

Using enums the way they are intended to be used, it should not matter what the values of the enumerations are.

Quote:

---

Originally Posted by eobergfell

Hey everybody thanks, im having trouble with using the "enum function"

Code:

---

enum Status{OUTSIDE = ?, INSIDE = ?);

---

I dont know what to set OUTSIDE and INSIDE equals to.
If anyone could help I would really appreciate it. Thanks-

---

In C++, enumerated types can be used to good advantage - not so with most C compilers.
Read the tutorial link in the post up above, for crying out loud. Especially the "Note to C programmers: ..."

All you need is just:

int OutText = 1;
int OutAttrb = 1;

as global variables, preferrably.

Now as you scan through your document, anytime you are Outside ANY html tag, AND outside any html attribute, you're ready to copy the text out of the html file, and put it into the new text file your program is creating.

Don't make this harder than it needs to be, please! :D

im sorry you guys I know im retarded. I just got my feet wet in C a couple of months ago. Thank you adak and dwks for trying but i just dont get it. dwks i tried the enum out but didn't get it to work and now im tryin out adak's idea. here is my crappy code---

Code:

---

#include <stdio.h>
void main()
{
        FILE *fopen(),*fp;
        char c;
        int InsideTag =1;
        int InsideAttrib =1;


        printf("Enter Filename:");
        scanf("&#37;s",&fp);
        fp = fopen("home.html","r");

        while(!feof(fp))
        {
        if(fp==InsideTag)
        {
                printf("%s", InsideTag); /* just to test it out */
        }


                putchar(c);
                c=getc(fp);

        }
        fclose(fp);

}

---

Thanks to everybody though for trying:)

Code:

---

scanf("%s",&fp);
fp = fopen("home.html","r");

---

This is wrong. You are trying to store the filename in a pointer to a file! You have to ask for the filename and store it in a char *. By example:

Code:

---

FILE *fopen(), *fp;
char filename[100];
fgets(filename, 100, stdin);
fp = fopen(filename, "r");

---

Thanks- That wasnt the only error right because it still doesnt do anything?

Code:

---

#include <stdio.h>
void main()
{

        char c;
        int InsideTag =1;
        int InsideAttrib =1;

        FILE *open(), *fp;
        char filename[100];
        fgets(filename, 100, stdin);
        fp = fopen(filename, "r");

        fp = fopen("home.html","r");

        while(!feof(fp))
        {
        if(filename==InsideTag)
        {
                printf("&#37;s", InsideTag);
        }


                putchar(c);
                c=getc(fp);

        }
        fclose(fp);

}

---

Quote:

---

Originally Posted by eobergfell

Thanks- That wasnt the only error right because it still doesnt do anything?

Code:

---

#include <stdio.h>
void main()
{

        char c;
        int InsideTag =1;
        int InsideAttrib =1;

        FILE *open(), *fp;
        char filename[100];
        fgets(filename, 100, stdin);
        fp = fopen(filename, "r");

        fp = fopen("home.html","r");

        while(!feof(fp))
        {
        if(filename==InsideTag)
        {
                printf("&#37;s", InsideTag);
        }


                putchar(c);
                c=getc(fp);

        }
        fclose(fp);

}

---

---

OK, first things first - **FORGET THE FILE ACCESS, FOR NOW, ALREADY**. Just use a copy of a few lines of html text, for now, and put it right into your char array:

char array[] = "< /center line > <font size: 12> This is text... ";

After you get the test line or two or three to work, then you can start getting the file access (which is really just a detail, which is why we're putting it off), to work. It will go much faster.

Why would "filename = InsideTag", make any sense at all??

You need program logic to handle 1) reading the char array, char by char, (I'd use the array instead of the file, for now, to make it easier) 2) If or switch statements to decide if the char being examined is text or not, and 3) output of text char's to your text file.

I see no trace of such logic statements in your code.

Its also worth remembering that fgets() puts a new line character at the end of the string. You will need to replace the newline char with a null char which is easy to do.

Something like this would work placed after the fgets() call. But it also require string.h to be included.

Code:

---

filename[strlen(filename)-1]='\0';

---

I don't feel like typing a lengthy explanation of fgets(), so have a look here: http://cboard.cprogramming.com/showp...2&postcount=16

Quote:

---

Originally Posted by Adak

You need program logic to handle 1) reading the char array, char by char, (I'd use the array instead of the file, for now, to make it easier) 2) If or switch statements to decide if the char being examined is text or not, and 3) output of text char's to your text file.

---

Thats what im stuck on. I dont know how I would utilize the if statement to check if its text or not. Help anyone?-Thanks

So you have step 1 down, processing each character individually from some location, like a file or a string? Post your code.

Okay, maybe thats not the only step i need. At least I thought it was reading each char.

Code:

---

#include <stdio.h>
#include<string.h>
void main()
{

        FILE *open(), *fp;
        char filename[100];
        char c, a;
        int InsideTag =1;
        int InsideAttrib =1;
        printf("Enter Filename: ");
        scanf("&#37;s", &a);
        fgets(a,100, stdin);
        filename[strlen(filename)-1]='\0';
        fp = fopen("home.html", "r");


        while(!feof(fp))
        {
        if(.....................)
        }
}

---

void main() is a bad idea. int main() is good -- see the FAQ. http://faq.cprogramming.com/cgi-bin/...&id=1043284376

while(!feof(fp)) is a semi-common idiom, but it doesn't work very well. [edit] http://faq.cprogramming.com/cgi-bin/...&id=1043284351 [/edit]

Code:

---

FILE *open(), *fp;

---

You realize that you're prototyping a function here, not declaring a variable? ;) Leave out the () (assuming you want to declare a variable).

You have two lines that would both read a string from stdin, if a was a string (it's a single character):

Code:

---

        scanf("%s", &a);
        fgets(a,100, stdin);

---

Use one or the other, not both. And you should probably be reading into filename, not a.

What's the point of getting a string from the user if you don't use it?

Code:

---

fp = fopen("home.html", "r");

---

Pass filename or whatever the string is called instead of "home.html".

Maybe you should look at this very basic program for some ideas.

Code:

---

#include <stdio.h>

int main() {
    int c;

    while((c = getchar()) != EOF) {
        putchar(c);
    }

    return 0;
}

---

getchar() is the same as [f]getc(stdin), and putchar(c) is the same as putc(c, stdout) or printf("%c", c). You can modify that for your loop. Just echo every character at first like this is doing.

[edit] Are you following me so far? Okay. Now, with the above program, instead of just calling putchar(c), check what c is first. If it's a '<', switch to "tag" mode. If it's a '>', switch to "outside" mode.

Then, if the program is in "outside" mode, print c.

Let me know how you're doing. [/edit]

Code:

---

#include <stdio.h>
#include<string.h>
int main()
{

        FILE *open, *fp;
        char filename[100];
        char c, a;
        int InsideTag =1;
        int InsideAttrib =1;
        printf("Enter Filename: ");
        scanf("&#37;s", &filename);

        filename[strlen(filename)-1]='\0';
        fp = fopen("filename", "r");

        while((filename=getchar())!=EOF){
                putchar(filename);
        }

        return 0;

}

---

Getting there?

Erhm . . . yeah, getting there.

But the variable you use in your while loop should be an int. getchar() etc all process a single character, so you'd think you'd use a char; but EOF has to be stored in an int, so an int it is. Not a char array of 100 elements.

Don't use scanf() -- that's why you have the newline-stripping code. Replace

Code:

---

scanf("%s", &filename);

---

with

Code:

---

fgets(filename, sizeof(filename), stdin);

---

For future reference, when reading strings with scanf(), you don't use an &. You use an & for every variable you can read, except strings. That's because strings are really arrays of characters and not a variable type unto themselves.

And this

Code:

---

fp = fopen("filename", "r");

---

should be

Code:

---

fp = fopen(filename, "r");

---

"home.html" and "filename" are string literals. filename by itself is a reference to the variable filename, and fopen() will be passed whatever is contained in filename.

Code:

---

FILE *open, *fp;

---

Why not think of better names? . . .

You do realize that getchar() and putchar() use stdin and stdout -- the keyboard and the screen? To use the file, go

Code:

---

getc(fp)

---

and

Code:

---

putc(character, fp);

---

except that you'd want to write to another file, not the one you're reading from. Come to think of it, why don't you just leave the putchar() as is right now, so you don't have to open a file to write to.

I'd use fgets(). your prog will crash using scanf("%s", filename); if there is a space in the filename.

With this:

Code:

---

        while((filename=getchar())!=EOF){
                putchar(filename);
        }

---

EOF is an integer value. filename is a character array.

Code:

---

#include <stdio.h>
int main(argc, argv)
int argc; char *argv;
{
  char inputchar;
  int count = 0;
  char filename[100];
        FILE *open(), *fp;
        printf("Enter Filename: ");
        fgets(filename, sizeof(filename), stdin);
        filename[strlen(filename)-1]='\0';
        fp = fopen(filename, "r");
  while ( ( inputchar= fgetc(stdin) )!=EOF )
  {  if ( inputchar != '<' && count == 0 )
    { fputc( inputchar,stdout );
    }
    else
    if ( inputchar == '<' )
    { count++;
    }
    else
    if (inputchar == '>' )
    { count--;
    }

  }
}

---

With the help of an engineer I got this so far.

Quote:

---

fgets(filename, sizeof(filename), stdin);
filename[strlen(filename)-1]='\0';

---

fgets will put the \n character in the buffer only if there is enough space...
So better search for the presence of the \n character and not replace the last one without check. (see example in th eFAQ)

Also 100 bytes buffer maybe not long enough to store the file name with path
In windows MAX_PATH constant is at least 256 characters

[edit]

Quote:

---

Also 100 bytes buffer maybe not long enough to store the file name with path
In windows MAX_PATH constant is at least 256 characters

---

But if MAX_PATH is anything like FILENAME_MAX, you shouldn't declare an array of that size. FILENAME_MAX is often 2^16-1 = 32767 on Linux systems or higher. I'd use BUFSIZ myself, which is at minumum 512 and often 8192 for 64-bit systems. [/edit]

Quote:

---

Originally Posted by eobergfell

Code:

---

#include <stdio.h>
int main(argc, argv)
int argc; char *argv;
{
  char inputchar;
  int count = 0;
  char filename[100];
        FILE *open(), *fp;
        printf("Enter Filename: ");
        fgets(filename, sizeof(filename), stdin);
        filename[strlen(filename)-1]='\0';
        fp = fopen(filename, "r");
  while ( ( inputchar= fgetc(stdin) )!=EOF )
  {  if ( inputchar != '<' && count == 0 )
    { fputc( inputchar,stdout );
    }
    else
    if ( inputchar == '<' )
    { count++;
    }
    else
    if (inputchar == '>' )
    { count--;
    }

  }
}

---

With the help of an engineer I got this so far.

---

You shouldn't declare main() like that. That's the old K&R style. You should use

Code:

---

int main(int argc, char *argv[]) {

---

At any rate, argv should be declared as **argv or *argv[], not *argv. It's an array of strings, not a single pointer.

There are also several other problems with your code. The most obvious is that HTML is never nested -- you don't get <html <something > >. So you only need to have count as 1 or 0, not zero or more -- though it works as you have it, it's just overkill.

Also, inputchar must be an int to store EOF, like I think I said earlier. You can't store EOF in a char (properly).

Code:

---

FILE *open(), *fp;

---

You're still prototyping a function here. Believe me, that's not what you want to do. Get rid of the ()s if you want to use the variable open, or get rid of it entirely if you don't intend to use it.

And your original problem stated that you had to read from a file. You're opening a file, but you're reading from stdin (the keyboard, usually). You might also want to open another file for writing and print to that instead of to stdout (the screen, usually).

You're also comparing inputchar against '<' twice. The logic would make more sense when written like this:

Code:

---

if(c == '<') count ++;
if(!count) putchar(c);
if(c == '>') count --;

---

The putchar() goes in between the if statements to avoid printing '<' or '>'.