Support not finding the Pages dictionary in lenient mode (#897)

* Support not finding the Pages dictionary in lenient mode and support Kids object not referencing a page object in lenient mode

---------

Co-authored-by: Arnaud TAMAILLON <arnaud.tamaillon@younited-credit.fr>

src/UglyToad.PdfPig.Tests/Integration/CatGeneticsTests.cs

@@ -1,6 +1,7 @@
 namespace UglyToad.PdfPig.Tests.Integration
 {
     using Annotations;
+    using PdfPig.Core;
 
     public class CatGeneticsTests
     {
@@ -39,5 +40,39 @@
                 }
             }
         }
+
+        [Fact]
+        public void CanSupportPageInformationNotFoundInLenientMode()
+        {
+            var path = IntegrationHelpers.GetSpecificTestDocumentPath("pages-indirect-to-null.pdf");
+            // Lenient Parsing On -> can process
+            using (var document = PdfDocument.Open(path))
+            {
+                // unable to parse
+                Assert.Equal(1, document.NumberOfPages);
+                Assert.NotNull(document.GetPage(1));
+            }
+
+            // Lenient Parsing Off -> throws
+            var ex = Assert.Throws<PdfDocumentFormatException>(() => PdfDocument.Open(path, ParsingOptions.LenientParsingOff));
+            Assert.Equal("Pages entry is null", ex.Message);
+        }
+        
+        [Fact]
+        public void CanSupportPageKidsObjectNotBeingAPage()
+        {
+            var path = IntegrationHelpers.GetSpecificTestDocumentPath("pages-kids-not-page.pdf");
+
+            using (var document = PdfDocument.Open(path))
+            {
+                // unable to parse
+                Assert.Equal(1, document.NumberOfPages);
+                Assert.NotNull(document.GetPage(1));
+            }
+
+            // Lenient Parsing Off -> throws
+            var ex = Assert.Throws<PdfDocumentFormatException>(() => PdfDocument.Open(path, ParsingOptions.LenientParsingOff));
+            Assert.Equal("Could not find dictionary associated with reference in pages kids array: 3 0.", ex.Message);
+        }
     }
 }

src/UglyToad.PdfPig.Tests/Integration/SpecificTestDocuments/pages-indirect-to-null.pdf

@@ -0,0 +1,35 @@
+%PDF-1.0
+%µ¶
+
+1 0 obj
+<</Type/Catalog/Pages 4 0 R>>
+endobj
+
+2 0 obj
+<</Kids[3 0 R]/Count 1/Type/Pages/MediaBox[0 0 595 792]>>
+endobj
+
+3 0 obj
+<</Length 58>>
+stream
+q
+BT
+/ 96 Tf
+1 0 0 1 36 684 Tm
+(Hello World!) Tj
+ET
+Q
+
+endstream
+endobj
+xref
+0 4
+0000000000 65536 f 
+0000000019 00000 n 
+0000000069 00000 n 
+0000000147 00000 n
+trailer
+<</Size 6/Root 1 0 R>>
+startxref
+262
+%%EOF

src/UglyToad.PdfPig.Tests/Integration/SpecificTestDocuments/pages-kids-not-page.pdf

@@ -0,0 +1,35 @@
+%PDF-1.0
+%µ¶
+
+1 0 obj
+<</Type/Catalog/Pages 2 0 R>>
+endobj
+
+2 0 obj
+<</Kids[3 0 R]/Count 1/Type/Pages/MediaBox[0 0 595 792]>>
+endobj
+
+3 0 obj
+<</Length 58>>
+stream
+q
+BT
+/ 96 Tf
+1 0 0 1 36 684 Tm
+(Hello World!) Tj
+ET
+Q
+
+endstream
+endobj
+xref
+0 4
+0000000000 65536 f 
+0000000019 00000 n 
+0000000069 00000 n 
+0000000147 00000 n
+trailer
+<</Size 6/Root 1 0 R>>
+startxref
+262
+%%EOF

src/UglyToad.PdfPig/Content/PagesFactory.cs

@@ -143,15 +143,19 @@
 
                 foreach (var kid in kids.Data)
                 {
+                    DictionaryToken? kidDictionaryToken = null;
                     if (!(kid is IndirectReferenceToken kidRef))
                     {
                         throw new PdfDocumentFormatException($"Kids array contained invalid entry (must be indirect reference): {kid}.");
                     }
-
+                    if (!DirectObjectFinder.TryGet(kidRef, pdfTokenScanner, out kidDictionaryToken))
-                    if (!DirectObjectFinder.TryGet(kidRef, pdfTokenScanner, out DictionaryToken? kidDictionaryToken))
+                    {
+                        if (!isLenientParsing)
                         {
                             throw new PdfDocumentFormatException($"Could not find dictionary associated with reference in pages kids array: {kidRef}.");
                         }
+                    }
+                    kidDictionaryToken ??= new DictionaryToken(new Dictionary<NameToken, IToken>());
 
                     bool isChildPage = CheckIfIsPage(kidDictionaryToken, current.reference, false, pdfTokenScanner, isLenientParsing);
 

src/UglyToad.PdfPig/Parser/CatalogFactory.cs

@@ -46,6 +46,18 @@
                 pagesDictionary = DirectObjectFinder.Get<DictionaryToken>(value, scanner);
             }
 
+            if (pagesDictionary == null)
+            {
+                if (isLenientParsing)
+                {
+                    pagesDictionary = new DictionaryToken(new Dictionary<NameToken, IToken>());
+                }
+                else
+                {
+                    throw new PdfDocumentFormatException($"Pages entry is null");
+                }
+            }
+
             var pages = PagesFactory.Create(pagesReference, pagesDictionary, scanner, pageFactory, log, isLenientParsing);
             var namedDestinations = NamedDestinationsProvider.Read(dictionary, scanner, pages, null);