update on the crackpdf feature

Author: driehle

Diff for: README.md

@@ -39,6 +39,28 @@ vi config.php
 
 Run the update script, ideally you would do this periodically e.g. by cron:
 
-```
+```bash
 php bin/update.php
 ```
+
+
+Configuration
+-------------
+
+### Credentials
+
+You need to specify the username and the passwort of your ZIV account in `username` and `password` setting in the `config.php`.
+
+### Courses
+
+Add all courses you want to crawl to the `config.php`, see `config.php.dist` for some examples. There are two types of courses in Learnweb, some use the course view to list files, some use the folder view of Moodle. Looking at the URL you can easily determine which type of course you are dealing with.
+
+### Cracking PDF files
+
+Sometimes PDF files have a user password which you need to enter all the time you open the PDF file. Learnweb-Crawler can convert the PDF file to PS and back to PDF which generates a PDF file which can be opened without password. This only works if you know the user password and the tools `pdftops` and `ps2pdf` are installed on your system. On Debian simply install these two packages:
+
+```bash
+apt-get install ghostscript poppler-utils
+```
+
+The actually executed commands may be specified as `pdf2PsCmd` and `ps2PdfCmd` in the `config.php`, see `config.php.dist` for an example.

Diff for: bin/update.php

@@ -18,7 +18,9 @@
 define('LEARNWEB_DEBUG', (bool) $config->debug);
 
 // create pdf cracker
-$pdf = new LearnwebCrawler\PdfCrack();
+$pdf2Ps = (isset($config->pdf2PsCmd) ? $config->pdf2PsCmd : null);
+$ps2Pdf = (isset($config->ps2Pdf) ? $config->ps2Pdf : null);
+$pdf = new LearnwebCrawler\PdfCrack($pdf2Ps, $ps2Pdf);
 
 // handle learnweb courses
 if (isset($config->learnwebcourses)) {

Diff for: config.php.dist

@@ -25,6 +25,7 @@ return array(
 		array(
 			'source' => '...',
 			'target' => '...',
+			// add the crackpdf key if you want to convert PDF files
 			'crackpdf' => '...'
 		)
 	),
@@ -46,8 +47,8 @@ return array(
 		)
 	),
 
-		// There is still support for uDoo courses, even though it is not used anymore
-		// You need to specify the URL of the course and the target folder where to store files
+	// There is still support for uDoo courses, even though it is not used anymore
+	// You need to specify the URL of the course and the target folder where to store files
 	'udoo' => array(
 		array(
 			'source' => 'http://www.wi.uni-muenster.de/...',
@@ -67,5 +68,15 @@ return array(
 				)
 			)
 		)
-	)
+	),
+	
+	// You man specify a custom pdf2ps command if you like, %1$s is the PDF passwort,
+	// %2$s is the source file (PDF file), %3$s is the target file (temporary PS file)
+	// simply remove this setting if you want to stick with the defaults
+	'pdf2PsCmd' => '/usr/bin/pdftops -upw %1$s %2$s %3$s',
+	
+	// You man specify a custom ps2pdf command if you like, %1$s is the source file
+	// (temporary PS file) and %1$s is the target file (PDF file)
+	// simply remove this setting if you want to stick with the defaults
+	'ps2PdfCmd' => '/usr/bin/ps2pdf %1$s %1$s'
 );

Diff for: library/LearnwebCrawler/PdfCrack.php

@@ -4,23 +4,44 @@
 
 class PdfCrack
 {
+	protected $pdf2PsCmd = null;
+	protected $ps2PdfCmd = null;
+	
+	public function __construct($pdf2PsCmd, $ps2PdfCmd)
+	{
+		if ($pdf2PsCmd != null) {
+			$this->pdf2PsCmd = (string) $pdf2PsCmd;
+		}
+		else {
+			$this->pdf2PsCmd = "pdftops -upw %s %s %s";
+		}
+		
+		if ($ps2PdfCmd != null) {
+			$this->ps2PdfCmd = (string) $ps2PdfCmd;
+		}
+		else {
+			$this->ps2PdfCmd = "ps2pdf %s %s";
+		}
+	}
+	
 	public function crackFile ($file, $password)
 	{
+		if ($password === false) return;
 		if (substr($file, -4) != '.pdf') return;
 
 		$psFile = substr($file, 0, -4) . '.tmp.ps';
 		$crackFile = substr($file, 0, -4) . '.cracked.pdf';
 
 		$cmd = sprintf(
-			'pdftops -upw %s %s %s',
+			$this->pdf2PsCmd,
 			escapeshellarg($password),
 			escapeshellarg($file),
 			escapeshellarg($psFile)
 		);
 		exec ($cmd);
 
 		$cmd = sprintf(
-			'ps2pdf %s %s',
+			$this->ps2PdfCmd,
 			escapeshellarg($psFile),
 			escapeshellarg($crackFile)
 		);