Reading data from a .dat binary file

Question

I have a .dat file needs to be read and prints records for people that have the attributes specified as command line arguments. The format of the file is as follows. The records for each person begins with an unsigned 32-bit integer that contains various information about the person:

Bits

0-4 : length of first name (key=first)

5-9 : length of middle name (key=middle)

10-14 : length of last name (key=last)

15-21 : age (key=age)

22 : sex (0=male, 1=female) (key=sex)

23-28 : state (0 to 49 in alphabetical order) (key=state)

29 : currently married (0=false, 1=true) (key=married)

30 : employed full time (0=false, 1=true) (key=employed)

31 : attended college (0=false, 1=true) (key=college)

I think I suppose to read the whole 32-bit(4 bytes) first, then read bit by bit from the 32-bit integer. I am new to the fread and fseek so I really don't know if I am on the right track, any help will be appreciated. Here are my codes so far.

  int main(int argc, char *argv[]) {

  char *buf;
  long lSize;
  size_t result;

  FILE *fp;
  fp = fopen("/u1/junk/people.dat","r");
  if(fp == NULL) {
    printf("Error: can't open file to read\n");
    return -1;
  }
  else {
    printf("File people.dat opened successfully to read\n");
  }


  //obtian file size
  fseek(fp, 0, SEEK_END);
  lSize = ftell(fp);
  rewind(fp);

  //allocate memory to contain the whole file
  buf = (char*) malloc (sizeof(char)*lSize);

  while (!feof(fp)) {
    fread(buf, 4, 1, fp);
    fseek(fp, i, SEEK_CUR);
    fread(buf, 32, 1, fp);
    printf("%s\n", buf);
    i+=32;
  }

  fclose(fp);
  return 0;  
}

Answer 1

This is untested but hopefully gives you the idea of the "mask and shift" technique. You can read more about it: What are bitwise shift (bit-shift) operators and how do they work? and Bitfield manipulation in C

Let's say you read the unsigned 32-bit integer:

unsigned long myint;
fread(&myint, sizeof(myint), 1, fid);

Now, shift and mask to pull values out.

// bits 0-4 (5 bits)
unsigned long firstNameLength = myint & 0xF1;

// bit 31
unsigned long attendedCollege = (myint & 0x0000000E) << 31;

Answer 2

#include <stdio.h>

// define first 32 bits of record
struct personRecord
{
    unsigned first    :5;
    unsigned middle   :5;
    unsigned last     :5;
    unsigned age      :7;
    unsigned sex      :1;
    unsigned state    :6;
    unsigned married  :1;
    unsigned employed :1;
    unsigned college  :1;
};

// prototypes
void processRecord(FILE* fp, char* buf);


int main(int argc, char *argv[])
{

    struct personRecord key;

    FILE *fp;
    fp = fopen("/u1/junk/people.dat","r");
    if(fp == NULL)
    {
        printf("Error: can't open file to read\n");
        return -1;
    }

    // implied else, fopen successful

    printf("File people.dat opened successfully to read\n");



    while ( 1 == fread( &key, 4, 1, fp ) )
    {
        processRecord(fp, key);
    }

    fclose(fp);
    return 0;
} // end function: main



void processRecord(FILE* fp, personRecord key)
{
      int result;
      int bufFirst[key.first] = {'\0'};
      ...

      // use  lengths to determine how much more  bytes to read for each field
      result = fread( bufFirst, key.first, 1, fp );
      if ( result == key.first )
      { // then successful acquire first name
          ...
      }
      ...
      // first, middle, last,

      // use lookup table to get state name from state field
      char * pState = alphabeticalState[key.state];

      // do something with the extracted info
      ...
} // end function: processRecord


const char * alphabeticalStates[] =
{
    "alabama";
    ...
    "washington";
    "wisconsin";
};