Updated pdf cache clear steps consistently with current pdfbox version

- Removed calls to no more existing clearResources functions (on PDFont
class and its children) since upgrade to pdfbox 2.n.n
- Removed hacky usage of protected internal ClassLoader function. This
removes the warnings displayed when running with JDK9 or JDK10 :

     [java] WARNING: Illegal reflective access by
net.yacy.document.parser.pdfParser$ResourceCleaner (file:<path>) to
method java.lang.ClassLoader.findLoadedClass(java.lang.String)
     [java] WARNING: Please consider reporting this to the maintainers
of net.yacy.document.parser.pdfParser$ResourceCleaner
     [java] WARNING: Use --illegal-access=warn to enable warnings of
further illegal reflective access operations
     [java] WARNING: All illegal access operations will be denied in a
future release

Crawling thousands of pdf documents from various sources after
modifications applied, revealed no new memory leak related to pdfbox
(measurements done with JVisualVM).

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

@@ -32,14 +32,14 @@
 import java.io.FileNotFoundException;
 import java.io.IOException;
 import java.io.InputStream;
-import java.lang.reflect.Method;
 import java.nio.charset.StandardCharsets;
 import java.util.ArrayList;
 import java.util.Collection;
 import java.util.Date;
 import java.util.HashSet;
 import java.util.List;
 
+import org.apache.pdfbox.cos.COSName;
 import org.apache.pdfbox.io.MemoryUsageSetting;
 import org.apache.pdfbox.pdmodel.PDDocument;
 import org.apache.pdfbox.pdmodel.PDDocumentInformation;
@@ -81,10 +81,6 @@ public pdfParser() {
         this.SUPPORTED_MIME_TYPES.add("text/x-pdf");
     }
 
-    static {
-        clean_up_idiotic_PDFParser_font_cache_which_eats_up_tons_of_megabytes(); // must be called here to get that into the class loader; it will block other threads otherwise;
-    }
-
     @Override
     public Document[] parse(
             final DigestURL location,
@@ -249,18 +245,9 @@ public void run() {
             try {pdfDoc.close();} catch (final Throwable e) {}
         }
 
-        // clear resources in pdfbox. they say that is resolved but it's not. see:
-        // https://issues.apache.org/jira/browse/PDFBOX-313
-        // https://issues.apache.org/jira/browse/PDFBOX-351
-        // https://issues.apache.org/jira/browse/PDFBOX-441
-        // the pdfbox still generates enormeous number of object allocations and don't delete these
-        // the following Object are statically stored and never flushed:
-        // COSFloat, COSArray, COSInteger, COSObjectKey, COSObject, COSDictionary,
-        // COSStream, COSString, COSName, COSDocument, COSInteger[], COSNull
-        // the great number of these objects can easily be seen in Java Visual VM
-        // we try to get this shit out of the memory here by forced clear calls, hope the best the rubbish gets out.
+        // clear cached resources in pdfbox.
         pdfDoc = null;
-        clean_up_idiotic_PDFParser_font_cache_which_eats_up_tons_of_megabytes();
+        clearPdfBoxCaches();
 
         return result;
     }
@@ -295,55 +282,35 @@ private List<Collection<AnchorURL>> extractPdfLinks(final PDDocument pdf) {
         return linkCollections;
     }
 
-    public static void clean_up_idiotic_PDFParser_font_cache_which_eats_up_tons_of_megabytes() {
-        // thank you very much, PDFParser hackers, this font cache will occupy >80MB RAM for a single pdf and then stays forever
-        // AND I DO NOT EVEN NEED A FONT HERE TO PARSE THE TEXT!
-        // Don't be so ignorant, just google once "PDFParser OutOfMemoryError" to feel the pain.
-        ResourceCleaner cl = new ResourceCleaner();
-        cl.clearClassResources("org.apache.pdfbox.cos.COSName");
-        cl.clearClassResources("org.apache.pdfbox.pdmodel.font.PDFont");
-        cl.clearClassResources("org.apache.pdfbox.pdmodel.font.PDType1Font");
-        cl.clearClassResources("org.apache.pdfbox.pdmodel.font.PDTrueTypeFont");
-        cl.clearClassResources("org.apache.pdfbox.pdmodel.font.PDType0Font");
-        cl.clearClassResources("org.apache.pdfbox.pdmodel.font.PDType1AfmPfbFont");
-        cl.clearClassResources("org.apache.pdfbox.pdmodel.font.PDType3Font");
-        cl.clearClassResources("org.apache.pdfbox.pdmodel.font.PDType1CFont");
-        cl.clearClassResources("org.apache.pdfbox.pdmodel.font.PDCIDFont");
-        cl.clearClassResources("org.apache.pdfbox.pdmodel.font.PDCIDFontType0Font");
-        cl.clearClassResources("org.apache.pdfbox.pdmodel.font.PDCIDFontType2Font");
-        cl.clearClassResources("org.apache.pdfbox.pdmodel.font.PDMMType1Font");
-        cl.clearClassResources("org.apache.pdfbox.pdmodel.font.PDSimpleFont");
+    /**
+     * Clean up cache resources allocated by PDFBox that would otherwise not be released.
+     */
+    public static void clearPdfBoxCaches() {
+		/*
+		 * Prior to pdfbox 2.0.0 font cache occupied > 80MB RAM for a single pdf and
+		 * then stayed forever (detected in YaCy with pdfbox version 1.2.1). The
+		 * situation is now from far better, but one (unnecessary?) cache structure in
+		 * the COSName class still needs to be explicitely cleared.
+		 */
+
+		// History of related issues :
+    	// http://markmail.org/thread/quk5odee4hbsauhu
+		// https://issues.apache.org/jira/browse/PDFBOX-313 
+		// https://issues.apache.org/jira/browse/PDFBOX-351
+		// https://issues.apache.org/jira/browse/PDFBOX-441
+    	// https://issues.apache.org/jira/browse/PDFBOX-2200
+    	// https://issues.apache.org/jira/browse/PDFBOX-2149
+
+        COSName.clearResources();
+
+		/*
+		 * Prior to PDFBox 2.0.0, clearResources() function had to be called on the
+		 * org.apache.pdfbox.pdmodel.font.PDFont class and its children. After version
+		 * 2.0.0, there is no more such a function in PDFont class as font cache is
+		 * handled differently and hopefully more properly.
+		 */
     }
 
-    @SuppressWarnings({ "unchecked", "rawtypes" })
-    private static class ResourceCleaner {
-        Method findLoadedClass;
-        private ClassLoader sys;
-        public ResourceCleaner() {
-            try {
-                this.findLoadedClass = ClassLoader.class.getDeclaredMethod("findLoadedClass", new Class[] { String.class });
-                this.findLoadedClass.setAccessible(true);
-                this.sys = ClassLoader.getSystemClassLoader();
-            } catch (Throwable e) {
-                e.printStackTrace();
-                this.findLoadedClass = null;
-                this.sys = null;
-            }
-        }
-        public void clearClassResources(String name) {
-            if (this.findLoadedClass == null) return;
-            try {
-                Object pdfparserpainclass = this.findLoadedClass.invoke(this.sys, name);
-                if (pdfparserpainclass != null) {
-                    Method clearResources = ((Class) pdfparserpainclass).getDeclaredMethod("clearResources", new Class[] {});
-                    if (clearResources != null) clearResources.invoke(null);
-                }
-            } catch (Throwable e) {
-                //e.printStackTrace();
-            }
-        }
-    }
-
     /**
      * test
      * @param args

source/net/yacy/search/Switchboard.java

@@ -2454,7 +2454,7 @@ public boolean surrogateProcess() {
 
     public static void clearCaches() {
         // flush caches in used libraries
-        pdfParser.clean_up_idiotic_PDFParser_font_cache_which_eats_up_tons_of_megabytes(); // eats up megabytes, see http://markmail.org/thread/quk5odee4hbsauhu
+        pdfParser.clearPdfBoxCaches();
 
         // clear caches
         if (WordCache.sizeCommonWords() > 1000) WordCache.clearCommonWords();