First off, you will need to add
#include <fstream> and
#include <sstream> to your header file.
fstream is the library for the file stream functions while
sstream is the library for the string stream functions. If you don't use the string functions you can drop the
sstream header.
Let's break this up into units. First, you need to go open the file and go through each line in the file. You say your file looks like this:
Code:
1 2 3
4 5 6
7 8 9
1 2 3
4 5 6
I called it TestFile.txt and put it into my C:\Temp directory
So now let's open the file via a file stream.
Code:
AnsiString File2Process = "C:\\Temp\\TestFile.txt";
std::ifstream FileStrm;
FileStrm.open(File2Process.c_str());
if (!FileStrm) // OH NO, an error!
{
// inform the user of the error then close the file
FileStrm.close();
return;
} // OH NO, an error
else // Process the file
{
...
} // Process the file
FileStrm.close() // Close the file
This opens the file
C:\Temp\TestFile.txt. Notice I prefer to use AnsiString and std::strings. That way I don't have to to worry buffer overflows. Next, I open the file stream for input using
ifstream. There are alternate ways to open a file stream for input but this will suffice. Then I call the open function for the file stream and immediately check if it opened, If it did not. I put some sort of message out and close the file and return from the function.
Now lets move through the lines in the file. I do this by defining a standard string and calling
getline in a while loop. This would go where the three dots (...) above.
Code:
std::string LineStdStr;
FileStrm.unsetf(std::ios::noskipws); // Do not skip whitespace
FileStrm.seekg(0); // Force to the beginning of the file (redundant)
while (std::getline(FileStrm, LineStdStr, '\n')) // Process as long as there are line to process
{
...
} // Process as long as there are line to process
The
getline will process each line in the file, looking for the newline character (/n) to determine where the end of the line is.
Here is where we have some options. You could use
seekp() to move to the correct position. I prefer to
get all the info in the standard string
LineStdStr and then process the data. For example,
if your data doesn't change much, you could use a substring funtion to get the data. For example, you could so the following in place of the three dots (...) above.
Code:
AnsiString SplitLineStr = LineStdStr.c_str(); // convert from std string to ansi
if (SplitLineStr.Length() != 0) // Data to process
{
// Collect the data from the line
AnsiString FirstStr = (SplitLineStr.SubString(1, 2)).Trim();
AnsiString SecondStr = (SplitLineStr.SubString(3, 2)).Trim();
}
However, there is another way that will work better if the length of the substrings change. For example, instead of the file above, suppose the file looked like
Code:
1 20 3
4 5 6
7 855 9
1 2 3
4 5 6
Now the substring approach would fail. So let's try using the string streams. Substitute the following for the above code.
Code:
// Get the parts
std::string FirstStdStr, SecondStdStr, RemainStdStr;
std::istringstream StrLine(LineStdStr);
std::getline(StrLine,FirstStdStr, ' '); // Get data to first blank
std::getline(StrLine,SecondStdStr, ' '); // Get data to second blank
std::getline(StrLine,RemainStdStr, '/n'); // Get the rest of the line
Play around a bit and see what works for you.
Disclaimer I did NOT test this out so you will have to debug it a bit (a lot?). All I did was take snippets from programs that work and put it here.
Good luck. Check back if you get stuck.
James P. Cottingham
-----------------------------------------
I'm number 1,229!
I'm number 1,229!
