unpack

Example 902. Display the ASCII character codes for an entire string

echo implode (' ',unpack ('C*', 'abcdef'));

Output:
97 98 99 100 101 102

Example 903. Show the format of a dbf database file

<pre>
<?php
# For more information on the format of dbf database files
# visit http://www.e-bachmann.dk/docs/xbase.htm

$file = "/tmp/sushi_eaten.dbf";

# Ensure that the file is big enough to be a db file
($filesize = filesize ($file)) > 68
    or die ("File <i>$file</i> is not large enough to be a dbf file.");

# Open a binary read-only connection to the file
$fp = fopen ($file, 'rb')
    or die ("File <i>$file</i> cannot be opened.");

# Get the top of the dbf file header
$data = fread ($fp, 32)
    or die ("Could not read data from file <i>$file</i>");

# Create the format for unpacking the header data
$header_format = 
    'H2id/' .           # Grab two Big-Endian hex digits
    'CYear/' .          # Grab an unsigned bit
    'CMonth/' .         # Grab an unsigned bit
    'CDay/' .           # Grab an unsigned bit
    'L# of Records/' .  # Grab an unsigned long (32 bit number)
    'SHeader Size/' .   # Grab an unsigned short (16 bit number)
    'SRecord Size';     # Grab an unsigned short (16 bit number)

# Unpack the header data
$header = unpack ($header_format, $data);

# Convert the year value to a full four digits
$header['Year'] += 1900;

# Display the data stored in $data
print_r ($header);

# Make sure that the file is the right size
if ($filesize != $size = $header['Header Size'] + ($header['Record Size'] * $header['# of Records']))
    die ("File <i>$file</i> is not a valid dbf file. Perhaps the file has been corrupted?");

# Get the rest of the dbf file header
$data = fread ($fp, $header['Header Size'] - 34)
    or die ("Could not read data from file <i>$file</i>");
    
# Create the format for unpacking the data that describes the format of the records in the file
$record_format = 
    'A11Field Name/' .  # Grab 11 alphanumeric characters
    'AField Type/' .    # Grab a single alphanumeric character
    'x4/' .             # Skip 4 bytes forward
    'CField Length/' .  # Grab an unsigned bit
    'CField Precision'; # Grab an unsigned bit

for ($offset = 0; $offset < strlen ($data); $offset += 32) {
    print_r (unpack ("@$offset/$record_format", $data));
}
?>
</pre>

Example 904. Create a read-only version of Perl's vec() function

<?php
function vector ($value, $offset, $bits) {
    # Ensure that the bit argument is 1, 2, 4, 8, 16, or 32
    if (! in_array ($bits, array (1,2,4,8,16,32))) {
        trigger_error ('<b>vector()</b> The bit argument must be one of the '
           . 'following values: 1, 2, 4, 8, 16, or 32');
        return FALSE;
    }

    # Cast $value to type string
    $value = (string) $value;

    # Convert the string to any array of ASCII character values
    # One odd behavior of unpack is that the array it makes will
    # have a starting index of 1, instead of 0.
    $str = unpack ('c*', $value);

    # Find the character in the string where the offset starts
    # Add 1 to compensate for the odd starting index in the $string array
    $chr = floor ($offset * $bits / 8) + 1;

    # For values less than 8 bits, we only need to work within a single character
    if ($bits < 8) {
        # Find the offset of the desired bytes within the character
        $bit_offset = $offset * $bits % 8;

        # Create a bit mask to use with & to help us extract our data
        $mask       = (pow (2, $bits) - 1) << $bit_offset;

        # Return the needed bytes from the character
        return ($str[$chr] & $mask) >> $bit_offset;
    }

    # If we need more than four bits, grab multiple characters
    for ($byte = 0; ($byte * 8) < $bits; ++$byte) {
        $output = ($output << 8) + $str[$chr + $byte];
    }

    return $output;
}
?>