safariparse.php

<?php
// code to accompany this blog post http://objectivesea.tumblr.com/post/9033067018/safaribooks
// body of blog post is included in a separate file from this gist.
try 
{
  //create or open the database
  $database = new SQLite3('/Users/cory/Desktop/Safari.sqlite');
  if ($database) echo "Database connection open...\n";
  $booksQuery = 'SELECT * from ZBOOK where ZOFFLINESTATUS = "offline"';
  	if($result = $database->query($booksQuery)) {
	    echo "Found " . $result->numColumns() . " books stored offline\n";
	    while($row = $result->fetchArray()) {
			echo "Saving book " . $row['ZTITLE'] . "\n";
			save_a_book($row['ZTITLE'], $row['Z_PK'], $database, $row['ZFPID']);
		}
	} else echo "Database failed to return results for query: " . $booksQuery;
	
}
catch(Exception $e) 
{
  die($error);
}


function save_a_book($name, $book_id, $database, $fpid) {
	$query = 'SELECT ZHTMLBODY, Z_PK, ZORDERINDEX from ZPAGE where ZBOOK = "' . $book_id . '" ORDER BY ZORDERINDEX';
	  if($result = $database->query($query)) {
		// make dir
		mkdir_recursive(sanitize($name, TRUE));
	    $counter = 1;
	    while($row = $result->fetchArray()) {
	      $myFile = "./" . sanitize($name, TRUE) . "/pg_" . $counter . ".html";
	      $fh = fopen($myFile, 'w') or die("can't open file ". $row['Z_PK']);
	      $stringData = $row['ZHTMLBODY'];
	      fwrite($fh, $stringData);
	      fclose($fh);
	      $counter++;
	      print("Wrote page #: {$row['Z_PK']} to file {$myFile}\n");// . "Director: {$row['Director']} <br />" . "Year: {$row['Year']} <br /><br />");
	    }
	    mkdir_recursive(sanitize($name, TRUE) . "/images/" . $fpid);
	  }
}

function mkdir_recursive($pathname)
{
    is_dir(dirname($pathname)) || mkdir_recursive(dirname($pathname));
    return is_dir($pathname) || @mkdir($pathname);
}


function sanitize($string = '', $is_filename = FALSE)
{
 // Replace all weird characters with dashes
 $string = preg_replace('/[^\w\-'. ($is_filename ? '~_\.' : ''). ']+/u', '-', $string);

 // Only allow one dash separator at a time (and make string lowercase)
 return mb_strtolower(preg_replace('/--+/u', '-', $string), 'UTF-8');
}

?>