Introduce ParsingOptions.FilterProvider and BaseFilterProvider and ma…

…ke CcittFaxCompressionType a byte

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

@@ -0,0 +1,126 @@
+namespace UglyToad.PdfPig.Tests.Integration
+{
+    using PdfPig.Filters;
+    using PdfPig.Tokens;
+    using System;
+    using System.Collections.Generic;
+    using System.Linq;
+
+    public class FilterTests
+    {
+        private static readonly Lazy<string> DocumentFolder = new Lazy<string>(() => Path.GetFullPath(Path.Combine(AppDomain.CurrentDomain.BaseDirectory, "..", "..", "..", "Integration", "Documents")));
+        private static readonly HashSet<string> _documentsToIgnore =
+        [
+            "issue_671.pdf",
+            "GHOSTSCRIPT-698363-0.pdf",
+            "ErcotFacts.pdf"
+        ];
+
+        [Theory]
+        [MemberData(nameof(GetAllDocuments))]
+        public void NoImageDecoding(string documentName)
+        {
+            // Add the full path back on, we removed it so we could see it in the test explorer.
+            documentName = Path.Combine(DocumentFolder.Value, documentName);
+
+            var parsingOptions = new ParsingOptions
+            {
+                UseLenientParsing = true,
+                FilterProvider = MyFilterProvider.Instance
+            };
+
+            using (var document = PdfDocument.Open(documentName, parsingOptions))
+            {
+                for (var i = 0; i < document.NumberOfPages; i++)
+                {
+                    var page = document.GetPage(i + 1);
+
+                    foreach (var pdfImage in page.GetImages())
+                    {
+                        if (pdfImage.ImageDictionary.TryGet(NameToken.Filter, out NameToken filter))
+                        {
+                            if (filter.Data.Equals(NameToken.FlateDecode.Data) ||
+                                filter.Data.Equals(NameToken.FlateDecodeAbbreviation.Data) || 
+                                filter.Data.Equals(NameToken.LzwDecode.Data) ||
+                                filter.Data.Equals(NameToken.LzwDecodeAbbreviation.Data))
+                            {
+                                continue;
+                            }
+                        }
+                        else
+                        {
+                            continue;
+                        }
+
+                        Assert.False(pdfImage.TryGetPng(out _));
+                    }
+                }
+            }
+        }
+
+        public sealed class NoFilter : IFilter
+        {
+            public bool IsSupported => false;
+
+            public ReadOnlyMemory<byte> Decode(ReadOnlySpan<byte> input, DictionaryToken streamDictionary, int filterIndex)
+            {
+                throw new NotImplementedException();
+            }
+        }
+
+        public class MyFilterProvider : BaseFilterProvider
+        {
+            /// <summary>
+            /// The single instance of this provider.
+            /// </summary>
+            public static readonly IFilterProvider Instance = new MyFilterProvider();
+
+            /// <inheritdoc/>
+            protected MyFilterProvider() : base(GetDictionary())
+            {
+            }
+
+            private static Dictionary<string, IFilter> GetDictionary()
+            {
+                var ascii85 = new Ascii85Filter();
+                var asciiHex = new AsciiHexDecodeFilter();
+                var flate = new FlateFilter();
+                var runLength = new RunLengthFilter();
+                var lzw = new LzwFilter();
+
+                var noFilter = new NoFilter();
+
+                return new Dictionary<string, IFilter>
+                {
+                    { NameToken.Ascii85Decode.Data, ascii85 },
+                    { NameToken.Ascii85DecodeAbbreviation.Data, ascii85 },
+                    { NameToken.AsciiHexDecode.Data, asciiHex },
+                    { NameToken.AsciiHexDecodeAbbreviation.Data, asciiHex },
+                    { NameToken.CcittfaxDecode.Data, noFilter },
+                    { NameToken.CcittfaxDecodeAbbreviation.Data, noFilter },
+                    { NameToken.DctDecode.Data, noFilter },
+                    { NameToken.DctDecodeAbbreviation.Data, noFilter },
+                    { NameToken.FlateDecode.Data, flate },
+                    { NameToken.FlateDecodeAbbreviation.Data, flate },
+                    { NameToken.Jbig2Decode.Data, noFilter },
+                    { NameToken.JpxDecode.Data, noFilter },
+                    { NameToken.RunLengthDecode.Data, runLength },
+                    { NameToken.RunLengthDecodeAbbreviation.Data, runLength },
+                    {NameToken.LzwDecode, lzw },
+                    {NameToken.LzwDecodeAbbreviation, lzw }
+                };
+            }
+        }
+
+        public static IEnumerable<object[]> GetAllDocuments
+        {
+            get
+            {
+                var files = Directory.GetFiles(DocumentFolder.Value, "*.pdf");
+
+                // Return the shortname so we can see it in the test explorer.
+                return files.Where(x => !_documentsToIgnore.Any(i => x.EndsWith(i))).Select(x => new object[] { Path.GetFileName(x) });
+            }
+        }
+    }
+}

src/UglyToad.PdfPig.Tests/PublicApiScannerTests.cs

@@ -97,6 +97,7 @@ public void OnlyExposedApiIsPublic()
                 "UglyToad.PdfPig.CrossReference.CrossReferenceType",
                 "UglyToad.PdfPig.CrossReference.TrailerDictionary",
                 "UglyToad.PdfPig.Exceptions.PdfDocumentEncryptedException",
+                "UglyToad.PdfPig.Filters.BaseFilterProvider",
                 "UglyToad.PdfPig.Filters.DefaultFilterProvider",
                 "UglyToad.PdfPig.Filters.IFilter",
                 "UglyToad.PdfPig.Filters.IFilterProvider",

src/UglyToad.PdfPig/Filters/BaseFilterProvider.cs

@@ -0,0 +1,96 @@
+namespace UglyToad.PdfPig.Filters
+{
+    using Core;
+    using System;
+    using System.Collections.Generic;
+    using System.Linq;
+    using Tokens;
+    using Util;
+
+    /// <summary>
+    /// Base abstract class for FilterProvider.
+    /// </summary>
+    public abstract class BaseFilterProvider : IFilterProvider
+    {
+        /// <summary>
+        /// Dictionary of filters.
+        /// </summary>
+        protected readonly IReadOnlyDictionary<string, IFilter> FilterInstances;
+
+        /// <summary>
+        /// Create a new <see cref="BaseFilterProvider"/> with the given filters.
+        /// </summary>
+        /// <param name="filterInstances"></param>
+        protected BaseFilterProvider(IReadOnlyDictionary<string, IFilter> filterInstances)
+        {
+            FilterInstances = filterInstances;
+        }
+
+        /// <inheritdoc />
+        public IReadOnlyList<IFilter> GetFilters(DictionaryToken dictionary)
+        {
+            if (dictionary is null)
+            {
+                throw new ArgumentNullException(nameof(dictionary));
+            }
+
+            var token = dictionary.GetObjectOrDefault(NameToken.Filter, NameToken.F);
+            if (token is null)
+            {
+                return Array.Empty<IFilter>();
+            }
+
+            switch (token)
+            {
+                case ArrayToken filters:
+                    var result = new IFilter[filters.Data.Count];
+                    for (var i = 0; i < filters.Data.Count; i++)
+                    {
+                        var filterToken = filters.Data[i];
+                        var filterName = ((NameToken)filterToken).Data;
+                        result[i] = GetFilterStrict(filterName);
+                    }
+
+                    return result;
+                case NameToken name:
+                    return new[] { GetFilterStrict(name.Data) };
+                default:
+                    throw new PdfDocumentFormatException($"The filter for the stream was not a valid object. Expected name or array, instead got: {token}.");
+            }
+        }
+
+        /// <inheritdoc />
+        public IReadOnlyList<IFilter> GetNamedFilters(IReadOnlyList<NameToken> names)
+        {
+            if (names is null)
+            {
+                throw new ArgumentNullException(nameof(names));
+            }
+
+            var result = new List<IFilter>();
+
+            foreach (var name in names)
+            {
+                result.Add(GetFilterStrict(name));
+            }
+
+            return result;
+        }
+
+        private IFilter GetFilterStrict(string name)
+        {
+            if (!FilterInstances.TryGetValue(name, out var factory))
+            {
+                throw new NotSupportedException($"The filter with the name {name} is not supported yet. Please raise an issue.");
+            }
+
+            return factory;
+        }
+
+        /// <inheritdoc />
+        public IReadOnlyList<IFilter> GetAllFilters()
+        {
+            return FilterInstances.Values.Distinct().ToList();
+        }
+    }
+}

src/UglyToad.PdfPig/Filters/CcittFax/CcittFaxCompressionType.cs

@@ -3,7 +3,7 @@
     /// <summary>
     /// Specifies the compression type to use with <see cref="T:UglyToad.PdfPig.Filters.CcittFaxDecoderStream" />.
     /// </summary>
-    internal enum CcittFaxCompressionType
+    internal enum CcittFaxCompressionType : byte
     {
         /// <summary>
         /// Modified Huffman (MH) - Group 3 variation (T2)

src/UglyToad.PdfPig/Filters/DefaultFilterProvider.cs

@@ -1,25 +1,24 @@
 namespace UglyToad.PdfPig.Filters
 {
-    using System;
     using System.Collections.Generic;
-    using System.Linq;
-    using Core;
     using Tokens;
-    using Util;
 
     /// <summary>
     /// The default implementation of the <see cref="T:UglyToad.PdfPig.Filters.IFilterProvider" />.
     /// </summary>
-    public class DefaultFilterProvider : IFilterProvider
+    public sealed class DefaultFilterProvider : BaseFilterProvider
     {
-        private readonly IReadOnlyDictionary<string, IFilter> filterInstances; 
-
         /// <summary>
         /// The single instance of this provider.
         /// </summary>
         public static readonly IFilterProvider Instance = new DefaultFilterProvider();
 
-        private DefaultFilterProvider()
+        /// <inheritdoc/>
+        private DefaultFilterProvider() : base(GetDictionary())
+        {
+        }
+
+        private static Dictionary<string, IFilter> GetDictionary()
         {
             var ascii85 = new Ascii85Filter();
             var asciiHex = new AsciiHexDecodeFilter();
@@ -31,7 +30,7 @@ private DefaultFilterProvider()
             var runLength = new RunLengthFilter();
             var lzw = new LzwFilter();
 
-            filterInstances = new Dictionary<string, IFilter>
+            return new Dictionary<string, IFilter>
             {
                 { NameToken.Ascii85Decode.Data, ascii85 },
                 { NameToken.Ascii85DecodeAbbreviation.Data, ascii85 },
@@ -47,77 +46,9 @@ private DefaultFilterProvider()
                 { NameToken.JpxDecode.Data, jpx },
                 { NameToken.RunLengthDecode.Data, runLength },
                 { NameToken.RunLengthDecodeAbbreviation.Data, runLength },
-                {NameToken.LzwDecode, lzw },
-                {NameToken.LzwDecodeAbbreviation, lzw }
+                { NameToken.LzwDecode.Data, lzw },
+                { NameToken.LzwDecodeAbbreviation.Data, lzw }
             };
         }
-
-        /// <inheritdoc />
-        public IReadOnlyList<IFilter> GetFilters(DictionaryToken dictionary)
-        {
-            if (dictionary is null)
-            {
-                throw new ArgumentNullException(nameof(dictionary));
-            }
-
-            var token = dictionary.GetObjectOrDefault(NameToken.Filter, NameToken.F);
-            if (token is null)
-            {
-                return Array.Empty<IFilter>();
-            }
-
-            switch (token)
-            {
-                case ArrayToken filters:
-                    var result = new IFilter[filters.Data.Count];
-                    for (var i = 0; i < filters.Data.Count; i++)
-                    {
-                        var filterToken = filters.Data[i];
-                        var filterName = ((NameToken) filterToken).Data;
-                        result[i] = GetFilterStrict(filterName);
-                    }
-
-                    return result;
-                case NameToken name:
-                    return new[] { GetFilterStrict(name.Data) };
-                default:
-                    throw new PdfDocumentFormatException($"The filter for the stream was not a valid object. Expected name or array, instead got: {token}.");
-            }
-        }
-
-        /// <inheritdoc />
-        public IReadOnlyList<IFilter> GetNamedFilters(IReadOnlyList<NameToken> names)
-        {
-            if (names is null)
-            {
-                throw new ArgumentNullException(nameof(names));
-            }
-
-            var result = new List<IFilter>();
-
-            foreach (var name in names)
-            {
-                result.Add(GetFilterStrict(name));
-            }
-
-            return result;
-        }
-
-        private IFilter GetFilterStrict(string name)
-        {
-            if (!filterInstances.TryGetValue(name, out var factory))
-            {
-                throw new NotSupportedException($"The filter with the name {name} is not supported yet. Please raise an issue.");
-            }
-
-            return factory;
-        }
-
-        /// <inheritdoc />
-        public IReadOnlyList<IFilter> GetAllFilters()
-        {
-            return filterInstances.Values.Distinct().ToList();
-        }
-
     }
 }

src/UglyToad.PdfPig/Filters/FlateFilter.cs

@@ -85,6 +85,7 @@ private static byte[] Decompress(byte[] input)
             }
         }
 
+        /// <inheritdoc />
         public byte[] Encode(Stream input, DictionaryToken streamDictionary, int index)
         {
             const int headerLength = 2;

src/UglyToad.PdfPig/Parser/PdfDocumentFactory.cs

@@ -106,7 +106,7 @@ private static PdfDocument OpenDocument(
             ISeekableTokenScanner scanner,
             ParsingOptions parsingOptions)
         {
-            var filterProvider = new FilterProviderWithLookup(DefaultFilterProvider.Instance);
+            var filterProvider = new FilterProviderWithLookup(parsingOptions.FilterProvider ?? DefaultFilterProvider.Instance);
 
             CrossReferenceTable? crossReferenceTable = null;
 

src/UglyToad.PdfPig/ParsingOptions.cs

@@ -1,5 +1,6 @@
 namespace UglyToad.PdfPig
 {
+    using Filters;
     using System.Collections.Generic;
     using Logging;
 
@@ -50,5 +51,10 @@ public sealed class ParsingOptions
         /// forms and images when missing.
         /// </summary>
         public bool SkipMissingFonts { get; set; } = false;
+
+        /// <summary>
+        /// Filter provider to use while parsing the document. The <see cref="DefaultFilterProvider"/> will be used if set to <c>null</c>.
+        /// </summary>
+        public IFilterProvider? FilterProvider { get; set; } = null;
     }
 }