Add basic static analysis for PDF content streams

This is pretty basic as a proof of concept. It can currently show the
following issues:
* Array/Dictionary/String object was not closed.
* Unnecessary whitespace at the end of lines.
* Unexpected tokens.
* Operand count and type for path construction operators.

Author: Eswcvlad

src/main/java/com/itextpdf/rups/model/contentstream/ParseTreeNode.java

@@ -269,6 +269,50 @@ public int getTextCount() {
         return textCount;
     }
 
+    /**
+     * Returns the start offset for the node. If this is a primitive node,
+     * then it is equivalent to calling {@link #getTextOffset()}. But if it is
+     * a composite node, it returns the text offset of the leftmost
+     * primitive descendant.
+     *
+     * @return The start offset for the node.
+     */
+    public int getStartOffset() {
+        if (textArray != null) {
+            return textOffset;
+        }
+        ParseTreeNode child = getFirstChild();
+        while (child != null) {
+            if (child.textArray != null) {
+                return child.textOffset;
+            }
+            child = child.getFirstChild();
+        }
+        return 0;
+    }
+
+    /**
+     * Returns the end offset for the node. If this is a primitive node, then
+     * it is equivalent to summing {@link #getTextOffset()} and
+     * {@link #getTextCount()}. But if it is a composite node, it returns the
+     * end offset of the leftmost primitive descendant.
+     *
+     * @return The start offset for the node.
+     */
+    public int getEndOffset() {
+        if (textArray != null) {
+            return textOffset + textCount;
+        }
+        ParseTreeNode child = getLastChild();
+        while (child != null) {
+            if (child.textArray != null) {
+                return child.textOffset + child.textCount;
+            }
+            child = child.getLastChild();
+        }
+        return 0;
+    }
+
     /**
      * Returns the first child of a node, or null, if it is a leaf.
      *

src/main/java/com/itextpdf/rups/view/Language.java

@@ -197,6 +197,21 @@ public enum Language {
     PAGE_NUMBER,
     PAGES,
     PAGES_TABLE_OBJECT,
+
+    PARSER_NOT_CLOSED_ARRAY,
+    PARSER_NOT_CLOSED_DICTIONARY,
+    PARSER_NOT_CLOSED_STRING_HEX,
+    PARSER_NOT_CLOSED_STRING_LITERAL,
+    PARSER_OPERAND_TYPES_C,
+    PARSER_OPERAND_TYPES_H,
+    PARSER_OPERAND_TYPES_L,
+    PARSER_OPERAND_TYPES_M,
+    PARSER_OPERAND_TYPES_RE,
+    PARSER_OPERAND_TYPES_V,
+    PARSER_OPERAND_TYPES_Y,
+    PARSER_UNEXPECTED_TOKEN,
+    PARSER_WASTEFUL_WHITESPACE,
+
     PDF_READING,
     PDF_OBJECT_TREE,
     PLAINTEXT,

src/main/java/com/itextpdf/rups/view/itext/StreamTextEditorPane.java

@@ -58,6 +58,7 @@ This file is part of the iText (R) project.
 import com.itextpdf.rups.view.contextmenu.StreamPanelContextMenu;
 import com.itextpdf.rups.view.itext.editor.Latin1Filter;
 import com.itextpdf.rups.view.itext.editor.PdfFoldParser;
+import com.itextpdf.rups.view.itext.editor.PdfParser;
 import com.itextpdf.rups.view.itext.editor.PdfTokenMaker;
 import com.itextpdf.rups.view.itext.editor.PdfTokenPainterFactory;
 import com.itextpdf.rups.view.itext.treenodes.PdfObjectTreeNode;
@@ -73,6 +74,7 @@ This file is part of the iText (R) project.
 import org.fife.ui.rsyntaxtextarea.DefaultTokenPainterFactory;
 import org.fife.ui.rsyntaxtextarea.RSyntaxDocument;
 import org.fife.ui.rsyntaxtextarea.RSyntaxTextArea;
+import org.fife.ui.rsyntaxtextarea.SyntaxConstants;
 import org.fife.ui.rsyntaxtextarea.TokenMakerFactory;
 import org.fife.ui.rsyntaxtextarea.folding.FoldParserManager;
 import org.fife.ui.rtextarea.ExpandedFoldRenderStrategy;
@@ -82,11 +84,7 @@ public final class StreamTextEditorPane extends RTextScrollPane implements IRups
     /**
      * MIME type for a PDF content stream.
      */
-    private static final String MIME_PDF = "application/pdf";
-    /**
-     * MIME type for plain text.
-     */
-    private static final String MIME_PLAIN_TEXT = "plain/text";
+    private static final String SYNTAX_STYLE_PDF = "application/pdf";
 
     /**
      * Char buffer with a single LF character.
@@ -115,8 +113,8 @@ public final class StreamTextEditorPane extends RTextScrollPane implements IRups
          */
         final AbstractTokenMakerFactory tokenMakerFactory =
                 (AbstractTokenMakerFactory) TokenMakerFactory.getDefaultInstance();
-        tokenMakerFactory.putMapping(MIME_PDF, PdfTokenMaker.class.getName());
-        FoldParserManager.get().addFoldParserMapping(MIME_PDF, new PdfFoldParser());
+        tokenMakerFactory.putMapping(SYNTAX_STYLE_PDF, PdfTokenMaker.class.getName());
+        FoldParserManager.get().addFoldParserMapping(SYNTAX_STYLE_PDF, new PdfFoldParser());
         /*
          * There doesn't seem to be a good way to detect, whether you can call
          * setData on a PdfStream or not in advance. It cannot be called if a
@@ -187,7 +185,7 @@ public void render(PdfObjectTreeNode target) {
         // Assuming that this will stop parsing for a moment...
         getTextArea().setVisible(false);
         String textToSet;
-        String mimeToSet;
+        String styleToSet;
         boolean editableToSet;
         /*
          * TODO: Differentiate between different content. See below.
@@ -207,21 +205,21 @@ public void render(PdfObjectTreeNode target) {
         try {
             if (isFont(stream) || isImage(stream)) {
                 textToSet = getText(stream, false);
-                mimeToSet = MIME_PLAIN_TEXT;
+                styleToSet = SyntaxConstants.SYNTAX_STYLE_NONE;
                 editableToSet = false;
             } else {
                 textToSet = prepareContentStreamText(getText(stream, true));
-                mimeToSet = MIME_PDF;
+                styleToSet = SYNTAX_STYLE_PDF;
                 editableToSet = true;
             }
             setTextEditableRoutine(true);
         } catch (RuntimeException e) {
             LoggerHelper.error(Language.ERROR_UNEXPECTED_EXCEPTION.getString(), e, getClass());
             textToSet = "";
-            mimeToSet = MIME_PLAIN_TEXT;
+            styleToSet = SyntaxConstants.SYNTAX_STYLE_NONE;
             editableToSet = false;
         }
-        setContentType(mimeToSet);
+        setContentType(styleToSet);
         getTextArea().setText(textToSet);
         getTextArea().setCaretPosition(0);
         setTextEditableRoutine(editableToSet);
@@ -316,8 +314,8 @@ private void clearPane() {
         setTextEditableRoutine(false);
     }
 
-    private void setContentType(String mime) {
-        setContentType(getTextArea(), mime);
+    private void setContentType(String style) {
+        setContentType(getTextArea(), style);
     }
 
     private void setUndoEnabled(boolean enabled) {
@@ -404,7 +402,8 @@ private static RSyntaxTextArea createTextArea() {
          * metadata we should just use the regular XML editor available. But
          * by default we will just assume a PDF content stream.
          */
-        setContentType(textArea, MIME_PDF);
+        setContentType(textArea, SYNTAX_STYLE_PDF);
+        textArea.addParser(new PdfParser());
         // This will allow to fold code blocks (like BT/ET blocks)
         textArea.setCodeFoldingEnabled(true);
         // This will automatically add tabulations, when you enter a new line
@@ -417,15 +416,15 @@ private static RSyntaxTextArea createTextArea() {
         return textArea;
     }
 
-    private static void setContentType(RSyntaxTextArea textArea, String mime) {
-        if (MIME_PDF.equals(mime)) {
+    private static void setContentType(RSyntaxTextArea textArea, String style) {
+        if (SYNTAX_STYLE_PDF.equals(style)) {
             getDocument(textArea).setDocumentFilter(new Latin1Filter());
             textArea.setTokenPainterFactory(new PdfTokenPainterFactory());
         } else {
             getDocument(textArea).setDocumentFilter(null);
             textArea.setTokenPainterFactory(new DefaultTokenPainterFactory());
         }
-        textArea.setSyntaxEditingStyle(mime);
+        textArea.setSyntaxEditingStyle(style);
     }
 
     private static String getText(PdfStream stream, boolean decoded) {