Add checks that Pdf Document has been completely read to the PdfReader.

DEVSIX-3139

Author: EvgenyB1001

kernel/src/main/java/com/itextpdf/kernel/PdfException.java

@@ -142,6 +142,7 @@ public class PdfException extends RuntimeException {
     public static final String DocumentDoesntContainStructTreeRoot = "Document doesn't contain StructTreeRoot.";
     public static final String DocumentHasNoPages = "Document has no pages.";
     public static final String DocumentHasNoPdfCatalogObject = "Document has no PDF Catalog object.";
+    public static final String DocumentHasNotBeenReadYet = "The PDF document has not been read yet. Document reading occurs in PdfDocument class constructor";
     public static final String DocumentMustBePreClosed = "Document must be preClosed.";
     public static final String DocumentForCopyToCannotBeNull = "Document for copyTo cannot be null.";
     public static final String DuringDecompressionMultipleStreamsInSumOccupiedMoreMemoryThanAllowed = "During decompression multiple streams in sum occupied more memory than allowed. Please either check your pdf or increase the allowed single decompressed pdf stream maximum size value by setting the appropriate parameter of ReaderProperties's MemoryLimitsAwareHandler.";

kernel/src/main/java/com/itextpdf/kernel/pdf/PdfReader.java

@@ -225,8 +225,13 @@ public void setCloseStream(boolean closeStream) {
      * If any exception generated while reading XRef section, PdfReader will try to rebuild it.
      *
      * @return true, if PdfReader rebuilt Cross-Reference section.
+     * @throws PdfException, if the method has been invoked before the PDF document was read.
      */
     public boolean hasRebuiltXref() {
+        if (pdfDocument == null || !pdfDocument.getXref().isReadingCompleted()) {
+            throw new PdfException(PdfException.DocumentHasNotBeenReadYet);
+        }
+
         return rebuiltXref;
     }
 
@@ -235,35 +240,57 @@ public boolean hasRebuiltXref() {
      * That Do Not Support Compressed Reference Streams" in PDF 32000-1:2008 spec.
      *
      * @return true, if the document has hybrid Cross-Reference section.
+     * @throws PdfException, if the method has been invoked before the PDF document was read.
      */
     public boolean hasHybridXref() {
+        if (pdfDocument == null || !pdfDocument.getXref().isReadingCompleted()) {
+            throw new PdfException(PdfException.DocumentHasNotBeenReadYet);
+        }
+
         return hybridXref;
     }
 
     /**
      * Indicates whether the document has Cross-Reference Streams.
      *
      * @return true, if the document has Cross-Reference Streams.
+     * @throws PdfException, if the method has been invoked before the PDF document was read.
      */
     public boolean hasXrefStm() {
+        if (pdfDocument == null || !pdfDocument.getXref().isReadingCompleted()) {
+            throw new PdfException(PdfException.DocumentHasNotBeenReadYet);
+        }
+
         return xrefStm;
     }
 
     /**
      * If any exception generated while reading PdfObject, PdfReader will try to fix offsets of all objects.
-     *
+     * <p>
+     * This method's returned value might change over time, because PdfObjects reading
+     * can be postponed even up to document closing.
      * @return true, if PdfReader fixed offsets of PdfObjects.
+     * @throws PdfException, if the method has been invoked before the PDF document was read.
      */
     public boolean hasFixedXref() {
+        if (pdfDocument == null || !pdfDocument.getXref().isReadingCompleted()) {
+            throw new PdfException(PdfException.DocumentHasNotBeenReadYet);
+        }
+
         return fixedXref;
     }
 
     /**
      * Gets position of the last Cross-Reference table.
      *
      * @return -1 if Cross-Reference table has rebuilt, otherwise position of the last Cross-Reference table.
+     * @throws PdfException, if the method has been invoked before the PDF document was read.
      */
     public long getLastXref() {
+        if (pdfDocument == null || !pdfDocument.getXref().isReadingCompleted()) {
+            throw new PdfException(PdfException.DocumentHasNotBeenReadYet);
+        }
+
         return lastXref;
     }
 
@@ -478,8 +505,13 @@ public long getFileLength() throws IOException {
      *
      * @return {@code true} if the document was opened with the owner password or if it's not encrypted,
      * {@code false} if the document was opened with the user password.
+     * @throws PdfException, if the method has been invoked before the PDF document was read.
      */
     public boolean isOpenedWithFullPermission() {
+        if (pdfDocument == null || !pdfDocument.getXref().isReadingCompleted()) {
+            throw new PdfException(PdfException.DocumentHasNotBeenReadYet);
+        }
+
         return !encrypted || decrypt.isOpenedWithFullPermission() || unethicalReading;
     }
 
@@ -489,8 +521,19 @@ public boolean isOpenedWithFullPermission() {
      * See ISO 32000-1, Table 22 for more details.
      *
      * @return the encryption permissions, an unsigned 32-bit quantity.
+     * @throws PdfException, if the method has been invoked before the PDF document was read.
      */
     public long getPermissions() {
+
+        /* !pdfDocument.getXref().isReadingCompleted() can be used for encryption properties as well,
+         * because decrypt object is initialized in private readDecryptObj method which is called in our code
+         * in the next line after the setting isReadingCompleted line. This means that there's no way for users
+         * when this method would work incorrectly right now.
+         */
+        if (pdfDocument == null || !pdfDocument.getXref().isReadingCompleted()) {
+            throw new PdfException(PdfException.DocumentHasNotBeenReadYet);
+        }
+
         long perm = 0;
         if (encrypted && decrypt.getPermissions() != null) {
             perm = (long) decrypt.getPermissions();
@@ -502,8 +545,13 @@ public long getPermissions() {
      * Gets encryption algorithm and access permissions.
      *
      * @see EncryptionConstants
+     * @throws PdfException, if the method has been invoked before the PDF document was read.
      */
     public int getCryptoMode() {
+        if (pdfDocument == null || !pdfDocument.getXref().isReadingCompleted()) {
+            throw new PdfException(PdfException.DocumentHasNotBeenReadYet);
+        }
+
         if (decrypt == null)
             return -1;
         else
@@ -525,9 +573,17 @@ public PdfAConformanceLevel getPdfAConformanceLevel() {
      * Computes user password if standard encryption handler is used with Standard40, Standard128 or AES128 encryption algorithm.
      *
      * @return user password, or null if not a standard encryption handler was used or if ownerPasswordUsed wasn't use to open the document.
+     * @throws PdfException, if the method has been invoked before the PDF document was read.
      */
     public byte[] computeUserPassword() {
-        if (!encrypted || !decrypt.isOpenedWithFullPermission()) return null;
+        if (pdfDocument == null || !pdfDocument.getXref().isReadingCompleted()) {
+            throw new PdfException(PdfException.DocumentHasNotBeenReadYet);
+        }
+
+        if (!encrypted || !decrypt.isOpenedWithFullPermission()) {
+            return null;
+        }
+
         return decrypt.computeUserPassword(properties.password);
     }
 
@@ -540,8 +596,13 @@ public byte[] computeUserPassword() {
      *
      * @return byte array represents original file ID.
      * @see PdfDocument#getOriginalDocumentId()
+     * @throws PdfException, if the method has been invoked before the PDF document was read.
      */
     public byte[] getOriginalFileId() {
+        if (pdfDocument == null || !pdfDocument.getXref().isReadingCompleted()) {
+            throw new PdfException(PdfException.DocumentHasNotBeenReadYet);
+        }
+
         PdfArray id = trailer.getAsArray(PdfName.ID);
         if (id != null && id.size() == 2) {
             return ByteUtils.getIsoBytes(id.getAsString(0).getValue());
@@ -559,8 +620,13 @@ public byte[] getOriginalFileId() {
      *
      * @return byte array represents modified file ID.
      * @see PdfDocument#getModifiedDocumentId()
+     * @throws PdfException, if the method has been invoked before the PDF document was read.
      */
     public byte[] getModifiedFileId() {
+        if (pdfDocument == null || !pdfDocument.getXref().isReadingCompleted()) {
+            throw new PdfException(PdfException.DocumentHasNotBeenReadYet);
+        }
+
         PdfArray id = trailer.getAsArray(PdfName.ID);
         if (id != null && id.size() == 2) {
             return ByteUtils.getIsoBytes(id.getAsString(1).getValue());
@@ -569,7 +635,14 @@ public byte[] getModifiedFileId() {
         }
     }
 
+    /**
+     * @throws PdfException, if the method has been invoked before the PDF document was read.
+     */
     public boolean isEncrypted() {
+        if (pdfDocument == null || !pdfDocument.getXref().isReadingCompleted()) {
+            throw new PdfException(PdfException.DocumentHasNotBeenReadYet);
+        }
+
         return encrypted;
     }
 
@@ -737,7 +810,7 @@ protected PdfObject readObject(boolean readAsDirect, boolean objStm) throws IOEx
                 return new PdfNumber(tokens.getByteContent());
             case String: {
                 PdfString pdfString = new PdfString(tokens.getByteContent(), tokens.isHexString());
-                if (isEncrypted() && !decrypt.isEmbeddedFilesOnly() && !objStm) {
+                if (encrypted && !decrypt.isEmbeddedFilesOnly() && !objStm) {
                     pdfString.setDecryption(currentIndirectReference.getObjNumber(), currentIndirectReference.getGenNumber(), decrypt);
                 }
                 return pdfString;

kernel/src/test/java/com/itextpdf/kernel/pdf/PdfCopyTest.java

@@ -170,8 +170,8 @@ public void copying3() throws IOException {
         pdfDoc.close();
 
         PdfReader reader = new PdfReader(destinationFolder + "copying3_1.pdf");
-        assertEquals("Rebuilt", false, reader.hasRebuiltXref());
         pdfDoc = new PdfDocument(reader);
+        assertEquals("Rebuilt", false, reader.hasRebuiltXref());
 
         PdfDictionary dic0 = pdfDoc.getPage(1).getPdfObject().getAsDictionary(new PdfName("HelloWorld"));
         assertEquals(4, dic0.getIndirectReference().getObjNumber());