File buffering read stream investigation (#1140)

* add test for filebufferingreadstream

* #1124 do not trust reported stream length if bytes can be read at end

the filebufferingreadstream input stream does not report more than the read
length. the change to seek the xref  in a sliding window from the end broke
with the assumption that the reported length was correct. here we switch to
reading the window or continue reading if we can read beyond the stream's
initially reported length while seeking the startxref marker

* remove rogue newlines

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

@@ -1,5 +1,9 @@
 namespace UglyToad.PdfPig.Tests.Integration
 {
+#if NET9_0_OR_GREATER
+    using Microsoft.AspNetCore.WebUtilities;
+#endif
+
     public class SinglePageLibreOfficeImages
     {
         private static string GetFilePath() => IntegrationHelpers.GetDocumentPath(@"Single Page Images - from libre office.pdf");
@@ -17,6 +21,23 @@
             }
         }
 
+        #if NET9_0_OR_GREATER
+        [Fact]
+        public void CanUseFileBufferingReadStream()
+        {
+            var bytes = File.ReadAllBytes(GetFilePath());
+
+            using var mem = new MemoryStream(bytes);
+            using var fbrs = new FileBufferingReadStream(mem, 256);
+
+            using var doc = PdfDocument.Open(fbrs);
+
+            var page = doc.GetPage(1);
+
+            Assert.NotEmpty(page.Text);
+        }
+        #endif
+
         [Fact]
         public void ImagesHaveCorrectDimensionsAndLocations()
         {

src/UglyToad.PdfPig.Tests/UglyToad.PdfPig.Tests.csproj

@@ -208,4 +208,9 @@
   <ItemGroup>
     <Using Include="Xunit" />
   </ItemGroup>
+  <ItemGroup Condition="'$(TargetFramework)' == 'net9.0'">
+    <PackageReference Include="Microsoft.AspNetCore.WebUtilities">
+      <Version>9.0.8</Version>
+    </PackageReference>
+  </ItemGroup>
 </Project>