Reading in strings, char, int -

Imagine I have a txt file, with different text "types" in it, for example a long paragraph, then a few words, then some numbers, etc. :

-----------------
Hello, this is the first part of my text file, and it's a paragraph. Therefore in this area, I blabber on and on, endlessly, or almost, because it's gotta be a long paragraph, quite a few lines. My next area will be numbers.

183751 59329 1534.07

Now one line, next will be one word.



WORD.
----------------

What can I use to make sure that I can read :

1. ALL of it, in ONE variable
2. Only the first paragraph ( I don't know it's length )
3. Only numbers
4. Read everything but the spaces, in ONE variable
5. Fit each line in a different variable
6. Have a variable which will contain only the numbers, but each number skips one character, which would give "135 539 13.7"



I'm not sure what to use for any of these. If I use

Code:

---

while(fin.get(stuff))
{
cout << stuff
}

---

which wouldn't give me everything in one variable, etc..

Thanks for any help.

Homework.

Face, because I know this guy and he doesn't get homework!

I have no homework, I don't take classes. You're really helpful.

OK, let me narrow it down a bit.

Would this work ?


NUMBER 1

Code:

---

fin.open("textfile.txt", ios::in);
fin >> myarray;

---

NUMBER 6

Code:

---

fin.open("textfile.txt", ios::in);
fin >> myarray;
for(int i = 0; i < strlen(myarray); i++)
{
cout << array[i];
}

---

I skip from 1 to 6 because I have no idea on how to do the others. Also, 6 isn't correct, because it would take in text too, and it wouldn't fit the output into a variable.

Dude, I don't know exactly but there are some nice functions in the string.h header file.... maybe you can use those....

Use google to find out more about them

1. ALL of it, in ONE variable

Code:

---

...
char document[32000];
char letter;
ifstream infile(...);
while(infile.getline(letter))
{
  strcat(document,letter);
}
...

---

2. Only the first paragraph ( I don't know it's length )

Code:

---

...
char paragraph[1000];
ifstream infile(...);
infile.getline(paragraph,1000,'\n');
...

---

3. Only numbers

Code:

---

...
ifstream infile(...);
char letter;
char output[1000];
infile>>letter;
if(isdigit(letter))
  strcat(output,letter);
...

---

4. Read everything but the spaces, in ONE variable

Code:

---

...
ifstream infile(...);
char word[20];
char everything[32000];
while(!infile.eof())
{
  infile.getline(word,20,' ');
  strcat(everything,word);
}
...

---

5. Fit each line in a different variable
look up linked lists... (too much code for now)

6. Have a variable which will contain only the numbers, but each number skips one character, which would give "135 539 13.7"

Code:

---

...
ifstream infile(...);
char letter;
char output[1000];
infile>>letter;
if(isdigit(letter))
{
  strcat(output,letter);
  infile.ignore(1);
}
...

---

note: these only work for the file you gave an example of... well the last one atleast... actually i'm not sure if they work at all because i didn't compile them...

edit: that last one won't get the decimal either...

Also check for a leading + or - before numbers.

Might want to use std::string instead of char[] as well. Its safer, and easier to deal with.

Thanks a lot for all that, Major Small, your help was very useful. What library is strcat in ? Also, would you use infile.getline or infile.get ( in a while loop ) ? Finally, Zach, how would you check for leading + or -, and why std::string ?

strcat is in <cstring>, but the member functions of std::string from <string> are better.

Quote:

---

Originally posted by Korhedron
Also, would you use infile.getline or infile.get ( in a while loop ) ?

---

you would use infile.getline() to make sure the program stops reading and destroys the character (a space, in this case)... and you would use it in a while loop so that it would run more than once, because the way i have it set up, it takes it in one token (word, in this case) at a time...

>>actually i'm not sure if they work at all because i didn't compile them...
At first sight, 4 out of 5 won't even compile.

Code:

---

char letter;
....
strcat(document,letter);

---

Wrong.

Quote:

---

char* strcat(
char *strDest, //Null-terminated destination string
const char* strSource //Null terminated source string
)

---

Is char document[3000] null-terminated?
Is char letter null-terminated,or a string?

Code:

---

infile>>letter;
if(isdigit(letter))
{
  strcat(output,letter);
  infile.ignore(1);
}

---

Read one,skip one.

>>> Finally, Zach, how would you check for leading + or -, and why std::string ?

std::string (from <string>) because you don't have to deal with directly allocating memory for the character array (it'll do that for you... even resizes). Also, instead of strcat et al, it has nicely defined +, =, and += operators to use.

i.e.
std::string document;
char letter;
...
document += letter;

To check for a leading +/-, just keep track of one extra character - the character preceding the current one. Then, when you find a digit, you check the preceding character to see if it is a +/-, and if it is, you know its part of the number.

>Is char document[3000] null-terminated?
change it to copy the first letter...

>Is char letter null-terminated,or a string?
change it to use tokens instead

Read one,skip one.
>that's the way it was designed...

i didn't catch the first two because i wasn't sure about the prototype for strcat