Test netstandard (#915)

* Add GH Action to test netstandard 2.0 and 2.1

* Account for TFM changes in tests project

.github/workflows/test-netstandard.yml

@@ -0,0 +1,44 @@
+name: Test netstandard
+
+on: pull_request
+
+jobs:
+  test-netstandard:
+    runs-on: ubuntu-latest
+
+    strategy:
+      matrix:
+        netstandard-version: ['netstandard2.0', 'netstandard2.1']
+
+    steps:
+    - uses: actions/checkout@v4
+
+    - name: Setup .NET
+      uses: actions/setup-dotnet@v4
+      with:
+        dotnet-version: |
+          8.0.x
+          9.0.x
+
+    - name: Patch build to test ${{ matrix.netstandard-version }}
+      run: |
+        cd src
+        sed -i 's/<TargetFrameworks>.*<\/TargetFrameworks>/<TargetFrameworks>${{ matrix.netstandard-version }}<\/TargetFrameworks>/' Markdig/Markdig.targets
+        sed -i 's/<TargetFrameworks>.*<\/TargetFrameworks>/<TargetFrameworks>net8.0;net9.0<\/TargetFrameworks>/' Markdig.Tests/Markdig.Tests.csproj
+        echo "Markdig.targets TFMs:"
+        grep "TargetFrameworks" Markdig/Markdig.targets
+        echo "Markdig.Tests.csproj TFMs:"
+        grep "TargetFrameworks" Markdig.Tests/Markdig.Tests.csproj
+
+    - name: Restore dependencies
+      run: dotnet restore src/Markdig.Tests/Markdig.Tests.csproj
+
+    - name: Test Debug
+      run: |
+        dotnet build src/Markdig.Tests/Markdig.Tests.csproj -c Debug --no-restore
+        dotnet test src/Markdig.Tests/Markdig.Tests.csproj -c Debug --no-build
+
+    - name: Test Release
+      run: |
+        dotnet build src/Markdig.Tests/Markdig.Tests.csproj -c Release --no-restore
+        dotnet test src/Markdig.Tests/Markdig.Tests.csproj -c Release --no-build

src/Directory.Packages.props

@@ -0,0 +1,23 @@
+<Project>
+  <PropertyGroup>
+    <ManagePackageVersionsCentrally>true</ManagePackageVersionsCentrally>
+    <CentralPackageTransitivePinningEnabled>false</CentralPackageTransitivePinningEnabled>
+  </PropertyGroup>
+  <ItemGroup>
+    <PackageVersion Include="BenchmarkDotNet" Version="0.14.0" />
+    <PackageVersion Include="BenchmarkDotNet.Diagnostics.Windows" Version="0.14.0" />
+    <PackageVersion Include="CommonMark.NET" Version="0.15.1" />
+    <PackageVersion Include="Markdown" Version="2.2.1" />
+    <PackageVersion Include="MarkdownSharp" Version="2.0.5" />
+    <PackageVersion Include="Microsoft.ApplicationInsights.AspNetCore" Version="2.23.0" />
+    <PackageVersion Include="Microsoft.Diagnostics.Runtime" Version="3.1.512801" />
+    <PackageVersion Include="Microsoft.NET.Test.Sdk" Version="18.0.1" />
+    <PackageVersion Include="MinVer" Version="6.0.0" />
+    <PackageVersion Include="NUnit" Version="4.4.0" />
+    <PackageVersion Include="NUnit3TestAdapter" Version="5.2.0" />
+    <PackageVersion Include="SharpFuzz" Version="2.2.0" />
+  </ItemGroup>
+  <ItemGroup Condition=" '$(TargetFramework)' == 'net462' OR '$(TargetFramework)' == 'netstandard2.0'">
+    <PackageVersion Include="System.Memory" Version="4.6.3" />
+  </ItemGroup>
+</Project>

src/Markdig.Benchmarks/Markdig.Benchmarks.csproj

@@ -19,12 +19,12 @@
     </Content>
   </ItemGroup>
   <ItemGroup>
-    <PackageReference Include="BenchmarkDotNet" Version="0.14.0" />
-    <PackageReference Include="BenchmarkDotNet.Diagnostics.Windows" Version="0.14.0" />
-    <PackageReference Include="CommonMark.NET" Version="0.15.1" />
-    <PackageReference Include="Markdown" Version="2.2.1" />
-    <PackageReference Include="MarkdownSharp" Version="2.0.5" />
-    <PackageReference Include="Microsoft.Diagnostics.Runtime" Version="3.1.512801" />
+    <PackageReference Include="BenchmarkDotNet" />
+    <PackageReference Include="BenchmarkDotNet.Diagnostics.Windows" />
+    <PackageReference Include="CommonMark.NET" />
+    <PackageReference Include="Markdown" />
+    <PackageReference Include="MarkdownSharp" />
+    <PackageReference Include="Microsoft.Diagnostics.Runtime" />
   </ItemGroup>
   <ItemGroup>
     <ProjectReference Include="..\Markdig\Markdig.csproj" />

src/Markdig.Fuzzing/Markdig.Fuzzing.csproj

@@ -1,4 +1,4 @@
-<Project Sdk="Microsoft.NET.Sdk">
+<Project Sdk="Microsoft.NET.Sdk">
 
   <PropertyGroup>
     <OutputType>Exe</OutputType>
@@ -9,7 +9,7 @@
   </PropertyGroup>
 
   <ItemGroup>
-    <PackageReference Include="SharpFuzz" Version="2.2.0" />
+    <PackageReference Include="SharpFuzz" />
   </ItemGroup>
 
   <ItemGroup>

src/Markdig.Tests/Markdig.Tests.csproj

@@ -1,7 +1,7 @@
 <Project Sdk="Microsoft.NET.Sdk">
 
   <PropertyGroup>
-    <TargetFrameworks>net6.0;net8.0;net9.0</TargetFrameworks>
+    <TargetFrameworks>net8.0;net9.0</TargetFrameworks>
     <OutputType>Exe</OutputType>
     <IsPackable>false</IsPackable>
     <ImplicitUsings>enable</ImplicitUsings>
@@ -13,9 +13,9 @@
   </PropertyGroup>
 
   <ItemGroup>
-    <PackageReference Include="Microsoft.NET.Test.Sdk" Version="17.12.0" />
-    <PackageReference Include="NUnit" Version="4.3.2" />
-    <PackageReference Include="NUnit3TestAdapter" Version="4.6.0" />
+    <PackageReference Include="Microsoft.NET.Test.Sdk" />
+    <PackageReference Include="NUnit" />
+    <PackageReference Include="NUnit3TestAdapter" />
   </ItemGroup>
 
   <ItemGroup>
@@ -36,10 +36,10 @@
     <InputSpecFiles Remove="Specs\readme.md" />
     <!-- Allow Visual Studio up-to-date check to verify that nothing has changed - https://github.com/dotnet/project-system/blob/main/docs/up-to-date-check.md -->
     <UpToDateCheckInput Include="@(InputSpecFiles)" />
-    <OutputSpecFiles Include="@(InputSpecFiles->'%(RelativeDir)%(Filename).generated.cs')" />
+    <OutputSpecFiles Include="@(InputSpecFiles-&gt;'%(RelativeDir)%(Filename).generated.cs')" />
   </ItemGroup>
 
-  <Target Name="GeneratedSpecsFile" BeforeTargets="BeforeCompile;CoreCompile" Inputs="@(ItemSpecExecutable);@(InputSpecFiles)" Outputs="@(ItemSpecExecutable->'%(RelativeDir)%(Filename).timestamp');@(InputSpecFiles->'%(RelativeDir)%(Filename).generated.cs')">
+  <Target Name="GeneratedSpecsFile" BeforeTargets="BeforeCompile;CoreCompile" Inputs="@(ItemSpecExecutable);@(InputSpecFiles)" Outputs="@(ItemSpecExecutable-&gt;'%(RelativeDir)%(Filename).timestamp');@(InputSpecFiles-&gt;'%(RelativeDir)%(Filename).generated.cs')">
     <Message Importance="high" Text="Regenerating Specs Files" />
     <Exec Command="dotnet $(SpecExecutable)" />
     <WriteLinesToFile File="$(SpecTimestamp)" Lines="$([System.DateTime]::Now)" />

src/Markdig.Tests/TestLinkHelper.cs

@@ -327,8 +327,8 @@ public class TestLinkHelper
         Assert.AreEqual(expectedResult, LinkHelper.Urilize(input, true));
     }
 
-    [TestCase("bær", "br")]
-    [TestCase("bør", "br")]
+    [TestCase("bær", "baer")]
+    [TestCase("bør", "boer")]
     [TestCase("bΘr", "br")]
     [TestCase("四五", "")]
     public void TestUrilizeOnlyAscii_NonAscii(string input, string expectedResult)
@@ -343,6 +343,75 @@ public class TestLinkHelper
         Assert.AreEqual(expectedResult, LinkHelper.Urilize(input, true));
     }
 
+    // Tests for NormalizeScandinavianOrGermanChar method mappings
+    // These special characters are always normalized (both allowOnlyAscii=true and false)
+    // 
+    // Note: When allowOnlyAscii=true, NFD (Canonical Decomposition) is applied first:
+    // - German umlauts ä,ö,ü decompose to base letter + combining mark (ü -> u + ¨)
+    //   The combining mark is then stripped, leaving just the base letter (ü -> u)
+    // - å decomposes similarly (å -> a + ˚ -> a)
+    // - But ø, æ, ß, þ, ð do NOT decompose, so they use NormalizeScandinavianOrGermanChar
+    //
+    // When allowOnlyAscii=false, NormalizeScandinavianOrGermanChar is used for ALL special chars
+    
+    // German ß (Eszett/sharp s) - does NOT decompose with NFD
+    [TestCase("Straße", "strasse")]    // ß -> ss (both allowOnlyAscii=true and false)
+    
+    // Scandinavian æ, ø - do NOT decompose with NFD
+    [TestCase("æble", "aeble")]        // æ -> ae (both modes)
+    [TestCase("Ærø", "aeroe")]         // Æ -> Ae, ø -> oe (both modes, then lowercase)
+    [TestCase("København", "koebenhavn")] // ø -> oe (both modes)
+    [TestCase("Øresund", "oeresund")]  // Ø -> Oe (both modes, then lowercase)
+    
+    // Icelandic þ, ð - do NOT decompose with NFD
+    [TestCase("þing", "thing")]        // þ (thorn) -> th (both modes)
+    [TestCase("bað", "bad")]           // ð (eth) -> d (both modes)
+    
+    // Mixed special characters (only chars that behave same in both modes)
+    [TestCase("øst-æble", "oest-aeble")] // ø->oe, æ->ae (both modes)
+    public void TestUrilizeScandinavianGermanChars(string input, string expectedResult)
+    {
+        // These transformations apply regardless of allowOnlyAscii flag
+        Assert.AreEqual(expectedResult, LinkHelper.Urilize(input, true));
+        Assert.AreEqual(expectedResult, LinkHelper.Urilize(input, false));
+    }
+    
+    // Tests specific to allowOnlyAscii=true behavior
+    // German umlauts (ä, ö, ü) and å decompose with NFD, so they become base letter only
+    [TestCase("schön", "schon")]       // ö decomposes to o (NFD strips combining mark)
+    [TestCase("Mädchen", "madchen")]   // ä decomposes to a
+    [TestCase("Übung", "ubung")]       // Ü decomposes to U (then lowercase to u)
+    [TestCase("Düsseldorf", "dusseldorf")] // ü decomposes to u
+    [TestCase("Käse", "kase")]         // ä decomposes to a
+    [TestCase("gå", "ga")]             // å decomposes to a
+    [TestCase("Ålesund", "alesund")]   // Å decomposes to A (then lowercase)
+    [TestCase("grüßen", "grussen")]    // ü decomposes to u, ß -> ss
+    [TestCase("Þór", "thor")]          // Þ -> Th, ó decomposes to o (then lowercase)
+    [TestCase("Íslandsbanki", "islandsbanki")] // Í decomposes to I (then lowercase)
+    public void TestUrilizeOnlyAscii_GermanUmlautsDecompose(string input, string expectedResult)
+    {
+        // With allowOnlyAscii=true, these characters decompose via NFD and lose their diacritics
+        Assert.AreEqual(expectedResult, LinkHelper.Urilize(input, true));
+    }
+    
+    // Tests specific to allowOnlyAscii=false behavior
+    // All special chars use NormalizeScandinavianOrGermanChar (including ä, ö, ü, å)
+    [TestCase("schön", "schoen")]      // ö -> oe (NormalizeScandinavianOrGermanChar)
+    [TestCase("Mädchen", "maedchen")]  // ä -> ae
+    [TestCase("Übung", "uebung")]      // Ü -> Ue (then lowercase)
+    [TestCase("Düsseldorf", "duesseldorf")] // ü -> ue
+    [TestCase("Käse", "kaese")]        // ä -> ae
+    [TestCase("gå", "gaa")]            // å -> aa
+    [TestCase("Ålesund", "aalesund")]  // Å -> Aa (then lowercase)
+    [TestCase("grüßen", "gruessen")]   // ü -> ue, ß -> ss
+    [TestCase("Þór", "thór")]          // Þ -> Th (then lowercase 'th'), ó is kept as-is
+    [TestCase("Íslandsbanki", "íslandsbanki")] // í is kept as-is when allowOnlyAscii=false
+    public void TestUrilizeNonAscii_GermanUmlautsExpanded(string input, string expectedResult)
+    {
+        // With allowOnlyAscii=false, these characters use NormalizeScandinavianOrGermanChar
+        Assert.AreEqual(expectedResult, LinkHelper.Urilize(input, false));
+    }
+    
     [TestCase("123", "")]
     [TestCase("1,-b", "b")]
     [TestCase("b1,-", "b1")] // Not Pandoc equivalent: b1-
@@ -360,11 +429,11 @@ public class TestLinkHelper
         Assert.AreEqual(expectedResult, LinkHelper.Urilize(input, false));
     }
 
-    [TestCase("bær", "bær")]
-    [TestCase("æ5el", "æ5el")]
-    [TestCase("-æ5el", "æ5el")]
-    [TestCase("-frø-", "frø")]
-    [TestCase("-fr-ø", "fr-ø")]
+    [TestCase("bær", "baer")]
+    [TestCase("æ5el", "ae5el")]
+    [TestCase("-æ5el", "ae5el")]
+    [TestCase("-frø-", "froe")]
+    [TestCase("-fr-ø", "fr-oe")]
     public void TestUrilizeNonAscii_Simple(string input, string expectedResult)
     {
         Assert.AreEqual(expectedResult, LinkHelper.Urilize(input, false));
@@ -393,4 +462,4 @@ public class TestLinkHelper
     {
         TestParser.TestSpec("[Foo]\n\n[Foo]: http://ünicode.com", "<p><a href=\"http://xn--nicode-2ya.com\">Foo</a></p>");
     }
-}
+}

src/Markdig.WebApp/Markdig.WebApp.csproj

@@ -14,7 +14,7 @@
   </PropertyGroup>
 
   <ItemGroup>
-    <PackageReference Include="Microsoft.ApplicationInsights.AspNetCore" Version="2.22.0" />
+    <PackageReference Include="Microsoft.ApplicationInsights.AspNetCore" />
   </ItemGroup>
 
   <ItemGroup>

src/Markdig/Helpers/LinkHelper.cs

@@ -2,11 +2,13 @@
 // This file is licensed under the BSD-Clause 2 license. 
 // See the license.txt file in the project root for more information.
 
+using Markdig.Syntax;
 using System.Buffers;
 using System.Diagnostics;
 using System.Diagnostics.CodeAnalysis;
+using System.Globalization;
 using System.Runtime.CompilerServices;
-using Markdig.Syntax;
+using System.Text;
 
 namespace Markdig.Helpers;
 
@@ -30,11 +32,38 @@ public static class LinkHelper
         var headingBuffer = new ValueStringBuilder(stackalloc char[ValueStringBuilder.StackallocThreshold]);
         bool hasLetter = keepOpeningDigits && headingText.Length > 0 && char.IsLetterOrDigit(headingText[0]);
         bool previousIsSpace = false;
-        for (int i = 0; i < headingText.Length; i++)
+
+        // First normalize the string to decompose characters if allowOnlyAscii is true
+        string normalizedString = string.Empty;
+        if (allowOnlyAscii)
         {
-            var c = headingText[i];
-            var normalized = allowOnlyAscii ? CharNormalizer.ConvertToAscii(c) : null;
-            for (int j = 0; j < (normalized?.Length ?? 1); j++)
+            normalizedString = headingText.ToString().Normalize(NormalizationForm.FormD);
+        }
+
+        var textToProcess = string.IsNullOrEmpty(normalizedString) ? headingText : normalizedString.AsSpan();
+
+        for (int i = 0; i < textToProcess.Length; i++)
+        {
+            var c = textToProcess[i];
+
+            // Skip combining diacritical marks when normalized
+            if (allowOnlyAscii && CharUnicodeInfo.GetUnicodeCategory(c) == UnicodeCategory.NonSpacingMark)
+            {
+                continue;
+            }
+
+            // Handle German umlauts and Norwegian/Danish characters explicitly (they don't decompose properly)
+            ReadOnlySpan<char> normalized;
+            if (IsSpecialScandinavianOrGermanChar(c))
+            {
+                normalized = NormalizeScandinavianOrGermanChar(c);
+            }
+            else
+            {
+                normalized = allowOnlyAscii ? CharNormalizer.ConvertToAscii(c) : null;
+            }
+
+            for (int j = 0; j < (normalized.Length < 1 ? 1 : normalized.Length); j++)
             {
                 if (normalized != null)
                 {
@@ -101,6 +130,50 @@ public static class LinkHelper
         return headingBuffer.ToString();
     }
 
+    [MethodImpl(MethodImplOptions.AggressiveInlining)]
+    private static bool IsSpecialScandinavianOrGermanChar(char c)
+    {
+        // German umlauts and ß
+        // Norwegian/Danish/Swedish æ, ø, å
+        // Icelandic þ (thorn), ð (eth)
+        return c == 'ä' || c == 'ö' || c == 'ü' ||
+               c == 'Ä' || c == 'Ö' || c == 'Ü' ||
+               c == 'ß' ||
+               c == 'æ' || c == 'ø' || c == 'å' ||
+               c == 'Æ' || c == 'Ø' || c == 'Å' ||
+               c == 'þ' || c == 'ð' ||
+               c == 'Þ' || c == 'Ð';
+    }
+
+    [MethodImpl(MethodImplOptions.AggressiveInlining)]
+    private static ReadOnlySpan<char> NormalizeScandinavianOrGermanChar(char c)
+    {
+        return c switch
+        {
+            // German
+            'ä' => "ae",
+            'ö' => "oe",
+            'ü' => "ue",
+            'Ä' => "Ae",
+            'Ö' => "Oe",
+            'Ü' => "Ue",
+            'ß' => "ss",
+            // Norwegian/Danish/Swedish
+            'æ' => "ae",
+            'ø' => "oe",
+            'å' => "aa",
+            'Æ' => "Ae",
+            'Ø' => "Oe",
+            'Å' => "Aa",
+            // Icelandic
+            'þ' => "th",
+            'Þ' => "Th",
+            'ð' => "d",
+            'Ð' => "D",
+            _ => ReadOnlySpan<char>.Empty
+        };
+    }
+
     public static string UrilizeAsGfm(string headingText)
     {
         return UrilizeAsGfm(headingText.AsSpan());
@@ -218,7 +291,8 @@ public static class LinkHelper
                 }
                 state = 1;
                 break;
-            } else if (c == '@')
+            }
+            else if (c == '@')
             {
                 if (state > 0)
                 {
@@ -234,7 +308,7 @@ public static class LinkHelper
         }
 
         // append ':' or '@' 
-        builder.Append(c); 
+        builder.Append(c);
 
         if (state < 0)
         {

src/Markdig/Markdig.targets

@@ -25,17 +25,16 @@
   </PropertyGroup>
 
   <ItemGroup Condition=" '$(TargetFramework)' == 'net462' OR '$(TargetFramework)' == 'netstandard2.0'">
-    <PackageReference Include="System.Memory" Version="4.6.0" />
+    <PackageReference Include="System.Memory" />
   </ItemGroup>
 
   <ItemGroup>
     <None Include="../../img/markdig.png" Pack="true" PackagePath="" />
-    <None Include="../../readme.md" Pack="true" PackagePath="/"/>
-    <PackageReference Include="MinVer" Version="4.3.0">
+    <None Include="../../readme.md" Pack="true" PackagePath="/" />
+    <PackageReference Include="MinVer">
       <PrivateAssets>all</PrivateAssets>
       <IncludeAssets>runtime; build; native; contentfiles; analyzers; buildtransitive</IncludeAssets>
     </PackageReference>
-    <PackageReference Include="Microsoft.SourceLink.GitHub" Version="8.0.*" PrivateAssets="All"/>
   </ItemGroup>
 
   <Target Name="PatchVersion" AfterTargets="MinVer">

src/markdig.slnx

@@ -2,13 +2,17 @@
   <Folder Name="/Build/">
     <File Path="../.editorconfig" />
     <File Path="../.gitattributes" />
-    <File Path="../.github/workflows/ci.yml" />
     <File Path="../.gitignore" />
     <File Path="../changelog.md" />
     <File Path="../license.txt" />
     <File Path="../readme.md" />
+    <File Path="Directory.Packages.props" />
     <File Path="global.json" />
   </Folder>
+  <Folder Name="/Build/GitHub Actions/">
+    <File Path="../.github/workflows/ci.yml" />
+    <File Path="../.github/workflows/test-netstandard.yml" />
+  </Folder>
   <Project Path="Markdig.Benchmarks/Markdig.Benchmarks.csproj">
     <BuildDependency Project="Markdig/Markdig.csproj" />
   </Project>