<?php
 
//1-download fichier à http://www.pdflib.com/download/tet/ (version windows xp 32 bits, TET-4.0p2-MSWin32.msi)
//2-installer sur un ordi, pas obliger d'être le serveur
//3-aller C:\Program Files (x86)\PDFlib\TET 4.0p2 32-bit\bind\php, il y a les versions de php
//4-prendre libtet_php.dll dans le bon répertoire représentant la version de php installé
//5-mettre ce fichier dans le répertoire de php\ext et modifier php.ini pour mettre extension=libtet_php.dll
// il doit y avoir une section TET lorsqu'on fait php phpinfo();  
 
 
// ***   $Id: extractor.php,v 1.11 2010/07/21 19:58:35 rjs Exp $ Simple PDF text extractor based on PDFlib TET   ***
$globaloptlist = "searchpath={../data ../../data ../../../resource/cmap}"; //global option list
$docoptlist = "";       // document-specific option list
$pageoptlist = "granularity=page";     // page-specific option list
$separator = "\n";    // separator to emit after each chunk of text. This depends on the application's needs; for granularity=word a space character may be useful
 
$pageno = 0;
 
try {
 
    $tet = new TET();
 
//mettre ici les infos nécessaire
$filepdf = file_get_contents('http://www.irs.gov/pub/irs-pdf/fw4.pdf'); //adresse web du fichier PDF
$pvf_filename = "/pvf/virt.pdf";     //nom du fichier virtuel, peux laisser tel quel
$tet->create_pvf($pvf_filename, $filepdf,"" );    //ici c'est la création dufichier virtuel, donc pas touche
$infilename = $pvf_filename;
$outfilename = "fichier.txt";      //nom du fichier de sortie
 
 
    if (!$outfp = fopen("$outfilename", "wb" )) {
 die("Couldn't open output file '" . $outfilename . "'\n" );
    }
 
 
    $tet->set_option($globaloptlist);
 
    $doc = $tet->open_document($infilename, $docoptlist);
 
    if ($doc == -1) {
 die("Error ". $tet->get_errnum() . " in " . $tet->get_apiname()
     . "(): " . $tet->get_errmsg() . "\n" );
    }
 
    /* get number of pages in the document */
    $n_pages = $tet->pcos_get_number($doc, "length:pages" );
 
    /* loop over pages in the document */
    for ($pageno = 1; $pageno <= $n_pages; ++$pageno) {
 
 $page = $tet->open_page($doc, $pageno, $pageoptlist);
 
 if ($page == -1) {
     print("Error ". $tet->get_errnum() ." in ". $tet->get_apiname()
  . "(): " . $tet->get_errmsg() . "\n" );
     next;                        /* try next page */
 }
 
 /* Retrieve all text fragments; This is actually not required
  * for granularity=page, but must be used for other granularities.
  */
 while (($text = $tet->get_text($page)) != "" ) {
 
     fwrite($outfp, $text);  /* print the retrieved text */
 
     /* print a separator between chunks of text */
     fwrite($outfp, $separator);
 }
 
 
 if ($tet->get_errnum() != 0) {
     print("Error ". $tet->get_errnum() . " in " .  
      $tet->get_apiname() . "(): on page $pageno"  
      . $tet->get_errmsg() . "\n" );
 }
 
 $tet->close_page($page);
    }
 
    fclose($outfp);
 
    $tet->close_document($doc);
}
catch (TETException $e) {
    if ($pageno == 0) {
 die("TET exception occurred in extractor sample:\n" .
     "[" . $e->get_errnum() . "] " . $e->get_apiname() . ": " .
     $e->get_errmsg() . "\n" );
    } else {
 die("TET exception occurred in extractor sample:\n" .
     "[" . $e->get_errnum() . "] " . $e->get_apiname() .
     "on page $pageno: " .  $e->get_errmsg() . "\n" );
    }
}
catch (Exception $e) {
    die($e);
}
 
$tet = 0;
 
?>