optimize pdfParser

skip starting reader thread if all content already read

source/net/yacy/document/parser/pdfParser.java

@@ -143,6 +143,7 @@ public class pdfParser extends AbstractParser implements Parser {
             writer.append(stripper.getText(pdfDoc));
             contentBytes = writer.getBytes(); // remember text in case of interrupting thread
 
+            if (pdfDoc.getNumberOfPages() > 3) { // spare creating/starting thread if all pages read
                 stripper.setStartPage(4); // continue with page 4 (terminated, resulting in no text)
                 stripper.setEndPage(Integer.MAX_VALUE); // set to default
                 // we start the pdf parsing in a separate thread to ensure that it can be terminated
@@ -159,7 +160,7 @@ public class pdfParser extends AbstractParser implements Parser {
                 t.start();
                 t.join(3000);
                 if (t.isAlive()) t.interrupt();
-            pdfDoc.close();
+            }
             contentBytes = writer.getBytes(); // get final text before closing writer
         } catch (final Throwable e) {
             // close the writer