- Optimize performance

- Disable commons-logging in Apache PDFBox
hwding · Aug 27, 2017 · fb7cf0e · fb7cf0e
1 parent 6ab6df4
commit fb7cf0e
Show file tree

Hide file tree

Showing 5 changed files with 48 additions and 23 deletions.
diff --git a/pom.xml b/pom.xml
@@ -6,7 +6,7 @@
     <modelVersion>4.0.0</modelVersion>
     <groupId>com.amastigote</groupId>
     <artifactId>unstamper</artifactId>
-    <version>0.1.0</version>
+    <version>0.1.1</version>
     <description>text stamp remover for PDF files</description>
     <name>pdf-unstamper</name>
     <url>https://github.com/hwding/pdf-unstamper</url>

diff --git a/script/install b/script/install
@@ -7,7 +7,7 @@
 user_bin=`echo ~`"/bin/"
 jar_name="pdf-unstamper.jar"
 exe_name="unstamp"
-_version="0.1.0"
+_version="0.1.1"
 jar_durl="https://github.com/hwding/pdf-unstamper/releases/download/$_version/$jar_name"
 
 function chk_f() {

diff --git a/src/com/amastigote/unstamper/Main.java b/src/com/amastigote/unstamper/Main.java
@@ -16,6 +16,12 @@
 
 public class Main {
 
+    static {
+        /* Disable Logging in Apache PDFBox */
+        System.setProperty("org.apache.commons.logging.Log",
+                "org.apache.commons.logging.impl.NoOpLog");
+    }
+
     public static void main(String[] args) {
         CommandLine commandLine = null;
         try {

diff --git a/src/com/amastigote/unstamper/core/Processor.java b/src/com/amastigote/unstamper/core/Processor.java
@@ -19,6 +19,7 @@
 import java.io.File;
 import java.io.IOException;
 import java.io.OutputStream;
+import java.util.Collections;
 import java.util.HashSet;
 import java.util.List;
 import java.util.Set;
@@ -37,33 +38,51 @@ public static void process(File file, String[] strings) {
                     /* START: loading font resources for further parsing */
                     PDFStreamParser pdfStreamParser = new PDFStreamParser(pdPage);
                     pdfStreamParser.parse();
-                    List<Object> objects = pdfStreamParser.getTokens();
-                    List<Object> cosNames = objects.parallelStream()
-                            .filter(e -> e instanceof COSName)
-                            .collect(Collectors.toList());
-                    Set<PDFont> pdFonts = new HashSet<>();
-                    cosNames.forEach(e -> {
-                        try {
-                            PDFont pdFont = pdPage.getResources().getFont(((COSName) e));
-                            if (pdFont != null)
-                                pdFonts.add(pdFont);
-                        } catch (IOException ignored) {
-                        }
-                    });
+
+                    List<Object> objects =
+                            Collections.synchronizedList(pdfStreamParser.getTokens());
+
+                    List<Object> cosNames =
+                            objects.parallelStream()
+                                    .filter(e -> e instanceof COSName)
+                                    .collect(Collectors.toList());
+
+                    Set<PDFont> pdFonts =
+                            Collections.synchronizedSet(new HashSet<>());
+
+                    cosNames.parallelStream()
+                            .forEach(e -> {
+                                /* Ignore Any Exception During Parallel Processing */
+                                try {
+                                    PDFont pdFont = pdPage.getResources().getFont(((COSName) e));
+                                    if (pdFont != null)
+                                        pdFonts.add(pdFont);
+                                } catch (Exception ignored) {
+                                }
+                            });
                     /* END */
-                    for (Object o : objects) {
-                        if (o instanceof COSString) {
-                            if (TextStampRecognizer.recognize(strings, ((COSString) o).getBytes(), pdFonts))
-                                ((COSString) o).setValue(new byte[0]);
-                        }
-                    }
+                    objects
+                            .parallelStream()
+                            .forEach(e -> {
+                                        if (e instanceof COSString) {
+                                    /* Ignore Any Exception During Parallel Processing */
+                                            try {
+                                                if (TextStampRecognizer.recognize(strings, ((COSString) e).getBytes(), pdFonts))
+                                                    ((COSString) e).setValue(new byte[0]);
+                                            } catch (Exception ignored) {
+                                            }
+                                        }
+                                    }
+                            );
+
                     PDStream newContents = new PDStream(pdDocument);
                     OutputStream out = newContents.createOutputStream();
                     ContentStreamWriter writer = new ContentStreamWriter(out);
                     writer.writeTokens(objects);
                     out.close();
+
                     pdPage.setContents(newContents);
-                } catch (IOException e) {
+                } catch (Exception e) {
                     GeneralLogger.Processor.errorProcess(file.getName());
                 }
             });

diff --git a/src/com/amastigote/unstamper/log/GeneralLogger.java b/src/com/amastigote/unstamper/log/GeneralLogger.java
@@ -10,7 +10,7 @@
 public class GeneralLogger {
     public static class Help {
         private static final String usage =
-                "\nPDF-UnStamper ver. 0.1.0 by hwding@GitHub\n" +
+                "\nPDF-UnStamper ver. 0.1.1 by hwding@GitHub\n" +
                         "\nUsage: " +
                         "\n   [OPTION] -i [INPUT PDF] -k [KEYWORDS...] (-o [OUTPUT PDF])" +
                         "\n   [OPTION] -I [INPUT DIR] -k [KEYWORDS...] (-O [OUTPUT DIR])\n" +