Printable View
There are actually two parts to this problem.
I've only begun to learn file handling, so as much help as possible would be appreciated.
What I need the program to first do is read the file until it finds the string "XMAP" and record its offset. Then I need it to count each occurance of the string "CMAP" before and after the offset in two different values.
How to go about it?
have an integer initially set to 0 for counter and flag.
read the file string by string
if the current string is "XMAP" and the flag is zero, increase the flag by one
while the flag > 0. if the current string is "CMAP", increment counter. Otherwise if it is XMAP, decrease the flag, return the counter.
I'd say that depends very much on how large that file is.
If it's small ( some megabytes ) then I would read it completely into memory( using fread() ), append a '\0' ( to make it a string ) and use strstr() to look for the strings.
If it's a very large file ( wouldn't fit into memory ) you would have to read the file line by line and check the line for both substrings. ( using strstr() ) until you have found the first search string.
Another possibility is to use a buffer of fixed size and use fread() to fill that buffer in a loop, but in this case you would have to take care about split seach strings at the end of the buffer.
Kurt
EDIT: Just noticed this is C++. getline() or fstream::read() and std::string functions would be better.
The files can vary from about 2 meg to 300 meg.
Well, you could keep the string in a FIFO queue... the queue is 4 chars long, search the queue for "XMAP", remove the first character from the queue and add the next character from the file to the back. You can use std::deque for a ready-made implementation.
Code:std::deque <char> buffer;
//read in the first 3 characters
//loop
while(!eof)
{
buffer.push_back(next_char)
if("XMAP" in buffer)
{
//do something
}
buffer.pop_front()
}
If the file has VERY LONG LINES you'd be in for some trouble with this one. Eg. trying to read a UNIX-generated file using Windows EOL conversions.Quote:
you would have to read the file line by line and check the line for both substrings.
So how could I get the FIFO method to record the offset of the 'X' in "XMAP"?