Fix XMP metadata parser to be able to parse not finished ASCII control characters

DEVSIX-9085

Autoported commit.
Original commit hash: [1f333ee6d]
Manual files:
commons/src/sharpenconfig/java/com/itextpdf/commons/SharpenConfigMapping.java
pom.xml

Author: vitali-pr

itext.tests/itext.kernel.tests/itext/kernel/pdf/XMPMetadataTest.cs

@@ -32,9 +32,9 @@ namespace iText.Kernel.Pdf {
     [NUnit.Framework.Category("IntegrationTest")]
     public class XMPMetadataTest : ExtendedITextTest {
         public static readonly String SOURCE_FOLDER = iText.Test.TestUtil.GetParentProjectDirectory(NUnit.Framework.TestContext
-            .CurrentContext.TestDirectory) + "/resources/itext/kernel/pdf/XmpWriterTest/";
+            .CurrentContext.TestDirectory) + "/resources/itext/kernel/pdf/XMPMetadataTest/";
 
-        public static readonly String DESTINATION_FOLDER = TestUtil.GetOutputPath() + "/kernel/pdf/XmpWriterTest/";
+        public static readonly String DESTINATION_FOLDER = TestUtil.GetOutputPath() + "/kernel/pdf/XMPMetadataTest/";
 
         [NUnit.Framework.OneTimeSetUp]
         public static void BeforeClass() {
@@ -289,5 +289,42 @@ public virtual void ListParsingTest() {
             NUnit.Framework.Assert.Catch(typeof(XMPException), () => XMPMetaFactory.ParseFromBuffer(xmp.GetBytes(System.Text.Encoding
                 .UTF8)));
         }
+
+        [NUnit.Framework.Test]
+        public virtual void ReadDocumentWithControlCharactersInXMPMetadata() {
+            String src = SOURCE_FOLDER + "docWithControlCharactersInXmp.pdf";
+            using (PdfDocument document = new PdfDocument(new PdfReader(src), new PdfWriter(new MemoryStream()), new StampingProperties
+                ())) {
+                NUnit.Framework.Assert.AreEqual(PdfConformance.PDF_A_3A, document.GetConformance());
+            }
+        }
+
+        [NUnit.Framework.Test]
+        public virtual void ReadDocumentWithBrokenControlCharactersInXMPMetadata() {
+            String src = SOURCE_FOLDER + "docWithBrokenControlCharactersInXmp.pdf";
+            using (PdfDocument document = new PdfDocument(new PdfReader(src), new PdfWriter(new MemoryStream()), new StampingProperties
+                ())) {
+                NUnit.Framework.Assert.AreEqual(PdfConformance.PDF_A_3A, document.GetConformance());
+            }
+        }
+
+        [NUnit.Framework.Test]
+        public virtual void ReadDocumentWithInvalidConformance() {
+            String src = SOURCE_FOLDER + "docWithInvalidConformance.pdf";
+            using (PdfDocument document = new PdfDocument(new PdfReader(src), new PdfWriter(new MemoryStream()), new StampingProperties
+                ())) {
+                NUnit.Framework.Assert.AreEqual(PdfConformance.PDF_NONE_CONFORMANCE, document.GetConformance());
+            }
+        }
+
+        [LogMessage(iText.IO.Logs.IoLogMessageConstant.EXCEPTION_WHILE_UPDATING_XMPMETADATA)]
+        [NUnit.Framework.Test]
+        public virtual void ReadDocumentWithInvalidXMPMetadata() {
+            String src = SOURCE_FOLDER + "docWithInvalidMetadata.pdf";
+            using (PdfDocument document = new PdfDocument(new PdfReader(src), new PdfWriter(new MemoryStream()), new StampingProperties
+                ())) {
+                NUnit.Framework.Assert.AreEqual(PdfConformance.PDF_NONE_CONFORMANCE, document.GetConformance());
+            }
+        }
     }
 }

itext.tests/itext.kernel.tests/resources/itext/kernel/pdf/XMPMetadataTest/docWithBrokenControlCharactersInXmp.pdf

@@ -29,7 +29,6 @@ You should have received a copy of the GNU Affero General Public License
 using iText.Kernel.XMP;
 using iText.Kernel.XMP.Options;
 using iText.Test;
-using iText.Test.Attributes;
 
 namespace iText.Pdfa {
     [NUnit.Framework.Category("IntegrationTest")]
@@ -113,34 +112,6 @@ public virtual void SaveAndReadDocumentWithCanonicalXmpMetadata() {
             }
         }
 
-        [NUnit.Framework.Test]
-        public virtual void ReadDocumentWithControlCharactersInXMPMetadata() {
-            String src = sourceFolder + "pdfs/docWithControlCharactersInXmp.pdf";
-            using (PdfADocument document = new PdfADocument(new PdfReader(src), new PdfWriter(new MemoryStream()), new 
-                StampingProperties())) {
-                NUnit.Framework.Assert.AreEqual(PdfConformance.PDF_A_3A, document.GetConformance());
-            }
-        }
-
-        [NUnit.Framework.Test]
-        public virtual void ReadDocumentWithInvalidConformance() {
-            String src = sourceFolder + "pdfs/docWithInvalidConformance.pdf";
-            using (PdfDocument document = new PdfDocument(new PdfReader(src), new PdfWriter(new MemoryStream()), new StampingProperties
-                ())) {
-                NUnit.Framework.Assert.AreEqual(PdfConformance.PDF_NONE_CONFORMANCE, document.GetConformance());
-            }
-        }
-
-        [LogMessage(iText.IO.Logs.IoLogMessageConstant.EXCEPTION_WHILE_UPDATING_XMPMETADATA)]
-        [NUnit.Framework.Test]
-        public virtual void ReadDocumentWithInvalidXMPMetadata() {
-            String src = sourceFolder + "pdfs/docWithInvalidMetadata.pdf";
-            using (PdfDocument document = new PdfDocument(new PdfReader(src), new PdfWriter(new MemoryStream()), new StampingProperties
-                ())) {
-                NUnit.Framework.Assert.AreEqual(PdfConformance.PDF_NONE_CONFORMANCE, document.GetConformance());
-            }
-        }
-
         [NUnit.Framework.Test]
         public virtual void TestPdfUAExtensionMetadata() {
             String outFile = destinationFolder + "testPdfUAExtensionMetadata.pdf";

itext.tests/itext.pdfa.tests/resources/itext/pdfa/pdfs/docWithControlCharactersInXmp.pdf renamed to itext.tests/itext.kernel.tests/resources/itext/kernel/pdf/XMPMetadataTest/docWithControlCharactersInXmp.pdf

@@ -32,7 +32,7 @@
 
 namespace iText.Kernel.XMP.Impl {
     /// <since>22.08.2006</since>
-    public class FixASCIIControlsReader : PushbackReader {
+    public class FixASCIIControlsReader : TextReader {
         private const int STATE_START = 0;
 
         private const int STATE_AMP = 1;
@@ -45,8 +45,6 @@ public class FixASCIIControlsReader : PushbackReader {
 
         private const int STATE_ERROR = 5;
 
-        private const int BUFFER_SIZE = 8;
-
         /// <summary>the state of the automaton</summary>
         private int state = STATE_START;
 
@@ -56,56 +54,48 @@ public class FixASCIIControlsReader : PushbackReader {
         /// <summary>count the digits of the sequence</summary>
         private int digits = 0;
 
-        /// <summary>The look-ahead size is 6 at maximum (&amp;#xAB;)</summary>
-        /// <seealso cref="iText.Commons.Utils.PushbackReader.PushbackReader(System.IO.TextReader, int)"/>
+        private TextReader @in;
+
+        /// <summary>A wrapper xmp reader to handle control characters (&amp;#xAB;)</summary>
         /// <param name="input">a Reader</param>
-        public FixASCIIControlsReader(TextReader input)
-            : base(input, BUFFER_SIZE) {
+        public FixASCIIControlsReader(TextReader input) {
+            @in = input;
         }
 
         /// <seealso cref="System.IO.TextReader.Read(char[], int, int)"/>
         public override int Read(char[] cbuf, int off, int len) {
-            int readAhead = 0;
             int read = 0;
             int pos = off;
-            char[] readAheadBuffer = new char[BUFFER_SIZE];
+            char[] readAheadBuffer = new char[1];
             bool available = true;
             while (available && read < len) {
-                available = base.Read(readAheadBuffer, readAhead, 1) == 1;
+                available = @in.Read(readAheadBuffer, 0, 1) == 1;
                 if (available) {
-                    char c = ProcessChar(readAheadBuffer[readAhead]);
+                    char c = ProcessChar(readAheadBuffer[0]);
                     if (state == STATE_START) {
                         // replace control chars with space
-                        if (iText.Kernel.XMP.Impl.Utils.IsControlChar(c)) {
+                        if (Utils.IsControlChar(c)) {
                             c = ' ';
                         }
                         cbuf[pos++] = c;
-                        readAhead = 0;
                         read++;
                     }
                     else {
                         if (state == STATE_ERROR) {
-                            Unread(readAheadBuffer, 0, readAhead + 1);
-                            readAhead = 0;
-                        }
-                        else {
-                            readAhead++;
                         }
                     }
                 }
-                else {
-                    if (readAhead > 0) {
-                        // handles case when file ends within excaped sequence
-                        Unread(readAheadBuffer, 0, readAhead);
-                        state = STATE_ERROR;
-                        readAhead = 0;
-                        available = true;
-                    }
-                }
             }
+            // It's broken ASCII character sequence, let's just skip them
+            // If we try to preserve them, SAX parser will throw later on anyway
             return read > 0 || available ? read : XMPUtilsImpl.EofReadBytesValue();
         }
 
+        /// <summary><inheritDoc/></summary>
+        public override void Close() {
+            @in.Close();
+        }
+
         /// <summary>Processes numeric escaped chars to find out if they are a control character.</summary>
         /// <param name="ch">a char</param>
         /// <returns>Returns the char directly or as replacement for the escaped sequence.</returns>