Reading Binary files in Perl

[pranjali_malode]

I have a binary file which contains the data about the drive past errors and failure reasons for the same stored in the binary format. I also have the structure available which tells me what the first 8 bytes are, then the next 4 bytes and so on. I require to get the binary data parse the binary file and display drive related all errors / reasons in meaningful format to the user. I do not understand what to use - arrays, hashes in perl! I tried doing it through unpack function available in perl but i am stuck on how to get the first 8 bytes then the next 8 bytes and so on.. I will appreciate if someone can help me in this.

[moritz]

You will probably need open, binmode, read and unpack.

If you don't get along well with pack and unpack, I can highly recommend perlpacktut, which is a gentle introduction to using those two functions.

[BrowserUk]

For us to be able to help you, you'll have to tell us more.

I also have the structure available which tells me what the first 8 bytes are, then the next 4 bytes and so on.

A good start would be to show us what that "structure" looks like. What form is it in? How do you get it into your perl program?

I tried doing it through unpack function available in perl but i am stuck on how to get the first 8 bytes then the next 8 bytes and so on..

If you show us what you tried (post your non-working code here (don't forget the <code></code> tags), then it serves as a starting point from which we can guide you.

[pranjali_malode]

Below is the structure or format of data in binary file: <structure> Offset Field Name size 0x0 magic_numbe 8 0x8 retry_count 1 0x9 number_record 1 0x10 dc_timestamp 8 <\structure> and the code is:

open(FH, "<DCdata.bin") or die "Can't open DCdata.bin file for reading
+!";
binmode(FH); 

while (<FH>)
{
    $index = 0;
    
    foreach $field_name (keys (%disk_structure)) {
        
        $value = read (FH, $buffer, ($field_name * 8), 0);
        print tell (FH);
        print "\n";
    }
    
    foreach $f_name (keys (%disk_structure_record)) {
        
        $value_record = read (FH, $buffer, ($f_name * 8), 0);
        #print tell (FH)\n;
    }
}

close(FILE);

[BrowserUk]

If your file consists of 18 character records, then the easiest way to read it is to set the input record separator ($/) to a reference to the number of bytes per record: $/ = \18;. This causes readline() (<FH>) to read the file as a series of fixed length records. You can then unpack the fields within each record very easily:

#! perl -slw
use strict;
 
=comment
Offset Field Name    size 
0x0    magic_numbe   8 
0x8    retry_count   1 
0x9    number_record 1 
0x10   dc_timestamp  8
=cut

open( FH, '<', DCdata.bin" ) 
    or die "Can't open DCdata.bin file for reading!";
binmode(FH); 

$/ = \18; ## set record size

while( <FH> ) {
    my( $magic, $retry, $recnum, $timestamp ) = unpack 'A8 C C A8', $_
+;
    ## Do something with the information??
}
close FH;

[Corion]

A hash is not ordered, so it makes no sense to read the (ordered) elements from a file in a foreach loop over the hash keys.