Reading tab separated file using getchar() - c

Question

I am trying to read a text file using input redirection, ./program < file.txt, and the file I made looks like this:
Country\tSport\tGender\tMedal\n
America\tCycling\tMens\tGold\n
New Zealand\tSwimming\tWomens\tSilver\n
India\tBadminton\tMens\tbronze\n
Which just reads some random data, in according to the first row with 4 columns, headed Country, Sport, Gender and Medal.
I also inserted \t and \n to make the file more readable, but the file actually has tabs and newlines in it.
I am trying to read each line of that file, and store them in an array of strings, which I have declared as:
char *records[ROWS][COLUMNS];
I would like the array of strings, records to look something like:
{{"Country", "Sport", "Gender", "Medal"}, {"America", "Cycling", "Mens", "Gold"},
{"New Zealand", "Swimming", "Womens", "Silver"}, {"India", "Badminton", "Mens", "Bronze"}}
So far I have just been using scanf to read the lines, like this:
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#define ROWS 1000
#define COLS 30
#define MAX_CH 50
int
main(int argc, char *argv[]) {
char *records[ROWS][COLS];
char country[MAX_CH];
char sport[MAX_CH];
char gender[MAX_CH];
char medal[MAX_CH];
while (scanf("%s\t%s\t%s\t%s\n", country, sport, gender, medal) == 4) {
printf("%s %s %s %s\n", country, sport, gender, medal);
}
return 0;
}
I know this will not work as the country name New Zealand has a space between both strings, and my scanf wil only read the first four characters. My method of scanf will also not be effective because it only works for 4 columns.
Is there a way I can use getchar() instead to do this? I'm just not sure how to use getchar to analyse each character in the input stream, and convert the necessary characters to strings depending on the tabs and newlines.

Answer 1

Pseudo code with getChar():
while (char = getChar()) is not 'EOF': // EOF = End of file
if char is not '\t' and char is not '\n'
save into current string
else if char is '\t'
terminate current string
increment column index
else if char is '\n'
terminate current string
increment row index
Edit:
The problem with getChar() is that you only know how long the string will be once you reach the next tab.
So you either have to iterate a first time to know the length of the string and then allocate an appropriate amount of memory, or you need to always allocate a safe amount of memory (your max string length).
Then in both options you can use strcat() to concatenate strings, but you can also access a char in a char* or char[] (String) by its index:
char string[] = "MINE"; // string[0] -> 'M'
string[0] = 'N'; // string -> "NINE"
// with dynamic memory allocation
char *string = (char*) malloc(5*sizeof(char));
string[0] = 'N'; // string -> "N"

Answer 2

getchar can be used to read into the rows and columns.
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>
#define ROWS 1000
#define COLS 30
#define MAX_CH 50
void print_array(char str[][COLS][MAX_CH], int nrows, int ncols, int col[], int reorder);
void sort_array(char str[][COLS][MAX_CH], int nrows, int ncols, int col[]);
int main( int argc, char *argv[])
{
char records[ROWS][COLS][MAX_CH] = { { { '\0'}}};//set each element
char item[MAX_CH] = { '\0'};//buffer for each item
//char yn[40];
int row = 0;
int col = 0;
int usedcol = 0;
int sortby[COLS] = { 0};
int ch = 0;
int each = 0;
int loop = 0;
int result = 0;
if ( argc < 2 || argc > COLS + 1) {
fprintf ( stderr, "syntax is\n%s column0 [column1]...[column%d] < inputfile\n", argv[0], COLS - 1);
exit(EXIT_FAILURE);
}
for ( each = 1; each <= COLS + 1; each++) {// +1 to get to extra element
sortby[each - 1] = -1;//extra element will be -1
if ( each < argc) {
if ( ( result = sscanf ( argv[each], "%d", &sortby[each - 1])) != 1 || sortby[each - 1] < 0 || sortby[each - 1] >= COLS) {
fprintf ( stderr, "syntax is\n%s column0 [column1]...[column%d] < inputfile\n", argv[0], COLS - 1);
fprintf ( stderr, "column%d must be 0 to %d\n", each - 1, COLS - 1);
exit(EXIT_FAILURE);
}
}
}
for ( each = 0; each < argc - 1; each++) {
for ( loop = 0; loop < argc - 1; loop++) {
if ( loop != each && sortby[each] == sortby[loop]) {
fprintf ( stderr, "found duplicate columns in args\n");
exit(EXIT_FAILURE);
}
}
}
//printf ( "\n");
if ( isatty ( fileno ( stdin))) {
printf ( "expected that a file would be redirected to this program\n");
printf ( "syntax is\n%s column0 [column1]...[column%d] < inputfile\n", argv[0], COLS - 1);
//printf ( "enter y to continue WITHOUT the redirected file?\n");
//if ( ( fgets ( yn, sizeof ( yn), stdin))) {
//if ( strcmp ( yn, "y\n") != 0) {
exit(EXIT_FAILURE);
//}
//}
}
row = 0;
col = 0;
each = 0;
while ( ( ch = getchar ( )) != EOF) {//read each character
if ( ch == '\r') {//skip carriage return
continue;
}
if ( ch == '\t') {
strcpy ( records[row][col], item);
each = 0;//back to first character
col++;//next column
if ( col >= COLS) {
col = 0;
row++;
if ( row >= ROWS) {
fprintf ( stderr, "too many rows\n");
break;
}
}
continue;
}
if ( ch == '\n') {
strcpy ( records[row][col], item);
col++;
if ( col > usedcol) {
usedcol = col;
}
col = 0;//back to first column
each = 0;//back to first character
row++;//next row
if ( row >= ROWS) {
fprintf ( stderr, "too many rows\n");
break;
}
continue;
}
item[each] = ch;
each++;//next character
item[each] = '\0';//terminate with '\0'
if ( each >= MAX_CH - 1) {
fprintf ( stderr, "too many characters in item\n");
each = 0;
col++;
if ( col >= COLS) {
col = 0;
row++;
if ( row >= ROWS) {
fprintf ( stderr, "too many rows\n");
break;
}
}
continue;
}
}
print_array ( records, row, usedcol, sortby, 0);
sort_array ( records, row, usedcol, sortby);
print_array ( records, row, usedcol, sortby, 1);
return 0;
}
void print_array(char str[][COLS][MAX_CH], int nrows, int ncols, int col[], int reorder) {
int i, j;
int order[COLS] = { 0};
for ( i = 0; i < COLS; i++) {
if ( reorder) {
order[i] = col[i];
}
else {
order[i] = i;
}
}
if ( reorder) {
for ( i = 0; i < COLS; i++) {
for ( j = 0; j < COLS; j++) {
if ( order[j] == i) {
break;
}
if ( order[j] == -1) {
order[j] = i;
break;
}
}
}
}
for (i = 0; i < nrows; i++) {
for (j = 0; j < ncols; j++) {
printf("%-12s ", str[i][order[j]]);
}
printf("\n");
}
printf("\n");
}
void sort_array(char str[][COLS][MAX_CH], int nrows, int ncols, int col[]) {
int i = 0, j = 0, swap = 0, each = 0;
char temp[MAX_CH] = { '\0'};
do {
swap = 0;
for ( i = 1; i < nrows - 1; i++) {//iterate through rows. i=1 skip first row
for ( each = 0; col[each] != -1; each++) {//col[] has last element of -1
if ( strcmp( str[i][col[each]], str[i + 1][col[each]]) < 0) {
break;
}
if ( strcmp( str[i][col[each]], str[i + 1][col[each]]) == 0) {
continue;
}
for ( j = 0; j < ncols; j++) {//iterate through cols and swap rows
strcpy ( temp, str[i][j]);
strcpy ( str[i][j], str[i + 1][j]);
strcpy ( str[i + 1][j], temp);
}
swap = 1;
break;
}
}
} while ( swap);//loop until no swaps
}

Answer 3

You can read by lines and then apply strtok() using '\t' as delimiter.
Reference for strtok() (it's from cplusplus.com, but strtok is in string.h, so it works also in c)

Answer 4

The posted code contains several problems:
unused stack variable records[][]
unused parameter: argc
unused parameter: argv
places all input data into first entries in the arrays: country[], sport[], gender[], medal[] I.E. all the other entries are not used
does not handle 'columns' that contain any white space
includes a header file those contents are not used: string.h
The following code cleanly compiles and performs the desired function:
#include <stdio.h>
#include <stdlib.h>
//#include <string.h>
#define ROWS 1000
//#define COLS 30
#define MAX_CH 49
struct record
{
char country[ MAX_CH+1 ];
char sport [ MAX_CH+1 ];
char gender [ MAX_CH+1 ];
char medal [ MAX_CH+1 ];
};
int main( void )
{
struct record records[ROWS];
for( size_t i=0; i< ROWS; i++)
{
if( scanf(" %" MAX_CH "[^\t] %" MAX_CH "[^\t] %" MAX_CH "[^\t] %" MAX_CH "[^\n]",
records[i].country,
records[i].sport,
records[i].gender,
records[i].medal) == 4)
{
printf("%s %s %s %s\n",
records[i].country,
records[i].sport,
records[i].gender,
records[i].medal);
}
else
{
break;
}
}
return 0;
} // end function: main