Searching a particular string in a large file - c

Question

I am making program in C which can search for a specific string in a large .txt file and count it and then print it out. But it seems that something have go wrong, cause the output of my program is different from that of the two text editor. According to the text editor, there are totally 3000 words,in this case I search for the word "make", in that .txt file. But the output of my program is just 2970.
I cannot find out the problem of my program. So I am curios about how could a text editor search for a specific string so accurately? How do people implement that? Can any people show me some code in C?
To make things clear: that is a large .txt file, 20M or so, containing lots of characters. So I think it's not so good to read it into memory all at once. I have implement my program by splitting my program in to pieces and then scan all of those for parsing. However, it fail some way.
Maybe I should put the code here. Wait a minute please.
The code is kinda long, 70 lines or so. I have put it on my github, if you have any interest, please help. https://github.com/walkerlala/searchText
note that the only related file is wordCount.c and testfile.txt which goes like:
#include<stdio.h>
#include<stdlib.h>
#include<stdbool.h>
char arr[51];
int flag=0;
int flag2=0;
int flag3=0;
int flag4=0;
int pieceCount(FILE*);
int main()
{
//the file in which I want to search the word is testfile.txt
//I have formatted the file so that it contain no newlins any more
FILE* fs=fopen("testfile.txt","r");
int n=pieceCount(fs);
printf("%d\n",n);
rewind(fs); //refresh the file...
static bool endOfPiece1=false,endOfPiece2=false,endOfPiece3=false;
bool begOfPiece1,begOfPiece2,begOfPiece3;
for(int start=0;start<n;++start){
fgets(arr,sizeof(arr),fs);
for(int i=0;i<=46;++i){
if((arr[i]=='M'||arr[i]=='m')&&(arr[i+1]=='A'||arr[i+1]=='a')&&(arr[i+2]=='K'||arr[i+2]=='k')&&(arr[i+3]=='E'||arr[i+3]=='e')){
flag+=1;
//continue;
}
}
//check the border
begOfPiece1=((arr[1]=='e'||arr[1]=='E'));
if(begOfPiece1==true&&endOfPiece1==true)
flag2+=1;
endOfPiece1=((arr[47]=='m'||arr[47]=='M')&&(arr[48]=='a'||arr[48]=='A')&&(arr[49]=='k'||arr[49]=='K'));
begOfPiece2=((arr[1]=='k'||arr[1]=='K')&&(arr[2]=='e'||arr[2]=='E'));
if(begOfPiece2==true&&endOfPiece2==true)
flag3+=1;
endOfPiece2=((arr[48]=='m'||arr[48]=='M')&&(arr[49]=='a'||arr[49]=='A'));
begOfPiece3=((arr[1]=='a'||arr[1]=='A')&&(arr[2]=='k'||arr[2]=='K')&&(arr[3]=='e'||arr[3]=='E'));
if(begOfPiece3==true&&endOfPiece3==true)
flag4+=1;
endOfPiece3=(arr[49]=='m'||arr[49]=='M');
}
printf("%d\n%d\n%d\n%d\n",flag,flag2,flag3,flag4);
getchar();
return 0;
}
//the function counts how many pieces have I split the file into
int pieceCount(FILE* file){
static int count=0;
char arr2[51]={'\0'};
while(fgets(arr2,sizeof(arr),file)){
count+=1;
continue;
}
return count;
}

Answer 1

You can do this quite simply just by having a rolling buffer. You don't need to break the file into sections.
#include <stdio.h>
#include <string.h>
int main(void) {
char buff [4]; // word buffer
int count = 0; // occurrences
FILE* fs=fopen("test.txt","r"); // open the file
if (fs != NULL) { // if the file opened
if (4 == fread(buff, 1, 4, fs)) { // fill the buffer
do { // if it worked
if (strnicmp(buff, "make", 4) == 0) // check for target word
count++; // tally
memmove(buff, buff+1, 3); // shift the buffer down
} while (1 == fread(buff+3, 1, 1, fs)); // fill the last position
} // end of file
fclose(fs); // close the file
}
printf("%d\n", count); // report the result
return 0;
}
For simplicity I stopped short of making the search word "softer" and allocating the correct buffer and various sizes, since that wasn't in the question. And I have to leave something for OP to do.