File Handling in C reading multiple chars - c++

Question

abort action islemi durdur(MS)
abort sequence durdurma dizisi(IBM)
I have a file.txt like above. I want to read this from the file.txt separately. Besides the file.txt I got 2 more turkce.txt and ingilizce.txt
Here is what I want to do :
I want to read from file.txt and separate the words English and Turkish. After that ingilizce.txt become like this
abort action
abort sequence
and turkce.txt like this
islemi durdur(MS)
durdurma dizisi(IBM)
Also, I have multiple columns and 5127 rows. Column numbers can changes each and every row.
Here is a pic of some part of my file.txt
http://i59.tinypic.com/33m0iu8.png
Thank you for your answers.
Update : I solved the problem. The difference between left column's starting of first letter and right column's starting of firs letter are same and it equals 37.
So I use
FILE* fp = fopen("file.txt","r");
char s[256];
fgets(s, 37 , "fp);

Answer 1

You don't say it explicitly, but your file has two fixed-width columns, which you want to separate.
A substring of a string str from a fixed index i to the end can be expressed with pointer arithmetic: str + i or &str[i]. Strings that are not zero-terminated (like your first column) can be printed by specifying a length with printfs precision field, e.g. printf("%.*s", len, str).
A quick and dirty way to print your two columns is:
char line[80];
int col = 36;
while (fgets(line, sizeof(line), in)) {
fprintf(en, "%.*s\n", col, line);
fprintf(tr, "%s", line + col);
printf("\n");
}
This method has some drawbacks: It will print garbage if the string is shorter than your separation width, i.e. if the right column is empty. It also prints the column padding spaces for the left column, which looks untidy. So let's write a function that splits the strings nicely, which we can call like so:
while (fgets(line, sizeof(line), in)) {
char *stren, *strtr;
split_at(line, &stren, &strtr, 36);
fprintf(en, "%s\n", stren);
fprintf(tr, "%s\n", strtr);
}
The function looks like this:
void split_at(char *line, char **left, char **right, int col)
{
char *trim = line;
char *p = line;
*left = line;
*right = line + col;
while (p < *right) {
if (*p == '\0') {
*right = p;
break;
}
if (!isspace(*p)) trim = p + 1;
p++;
}
*trim = '\0';
trim = p;
while (*p) {
if (!isspace(*p)) trim = p + 1;
p++;
}
if (trim) *trim = '\0';
}
This should work for your example data. It will also work for empty left or right columns. It will not work if there is no space between the left and right columns, i.e. when the left and right art are pasted together.
This method will also work only if the code points of the strings have the same length. You haven't said which encoding you use for your data. If you use ISO-8859-9, you will be okay. If you use UTF-8, all non ASCII-codepoints, i.e. the Turkish special characters, will be represented by more than one byte. What looks like a fixed-width column doesn't have a fixed width in its memory representation.
That said, you should be safe as long as your English text is in the left column. English text is made up of only ASCII characters unless you have fancy formatting with typographic quotation marks or some such.

Answer 2

There could be better solutions but here is simple one.
#include <iostream>
#include <fstream>
int main()
{
std::ifstream inFile("file.txt");
std::ofstream outFileT("turkce.txt", std::ios::app);
std::ofstream outFileE("ingilizce.txt", std::ios::app);
std::string a;
std::string b;
for (int i = 0; i < 2; i++) {
inFile >> a >> b;
outFileE << a + " " + b + "\n";
inFile >> a >> b;
outFileT << a + " " + b + "\n";
}
}
I assumed you have two lines but you can determine lines count in the file first.