Add custom DisableHtmlWithExceptions extension

Author: reakaleek

src/Elastic.Markdown/Myst/Extensions/DisableHtmlWithExceptionsExtension.cs

@@ -0,0 +1,25 @@
+// Licensed to Elasticsearch B.V under one or more agreements.
+// Elasticsearch B.V licenses this file to you under the Apache 2.0 License.
+// See the LICENSE file in the project root for more information
+
+using Elastic.Markdown.Myst.InlineParsers;
+using Markdig;
+using Markdig.Parsers;
+using Markdig.Parsers.Inlines;
+
+namespace Elastic.Markdown.Myst.Extensions;
+
+public static class MarkdownPipelineBuilderExtensions
+{
+	public static MarkdownPipelineBuilder DisableHtmlWithExceptions(this MarkdownPipelineBuilder pipeline, HashSet<string> exceptions)
+	{
+		var parser = pipeline.BlockParsers.Find<HtmlBlockParser>();
+        if (parser != null)
+        {
+            pipeline.BlockParsers.Remove(parser);
+        }
+
+        pipeline.InlineParsers.ReplaceOrAdd<AutolinkInlineParser>(new RestrictedAutolinkInlineParser { AllowedTags = exceptions });
+		return pipeline;
+	}
+}

src/Elastic.Markdown/Myst/InlineParsers/RestrictedAutolinkInlineParser.cs

@@ -0,0 +1,54 @@
+// Licensed to Elasticsearch B.V under one or more agreements.
+// Elasticsearch B.V licenses this file to you under the Apache 2.0 License.
+// See the LICENSE file in the project root for more information
+
+using System.Collections.ObjectModel;
+using Markdig;
+using Markdig.Helpers;
+using Markdig.Parsers;
+using Markdig.Parsers.Inlines;
+using Markdig.Syntax;
+using Markdig.Syntax.Inlines;
+
+namespace Elastic.Markdown.Myst.InlineParsers;
+
+// This class is a copy of the AutolinkInlineParser with the addition of a check for allowed tags
+public class RestrictedAutolinkInlineParser : InlineParser
+{
+
+    public RestrictedAutolinkInlineParser() => OpeningCharacters = ['<'];
+
+    public required IReadOnlySet<string> AllowedTags { get; init; }
+
+    public override bool Match(InlineProcessor processor, ref StringSlice slice)
+    {
+	    var saved = slice;
+	    int line;
+	    int column;
+	    if (LinkHelper.TryParseAutolink(ref slice, out var link, out var isEmail))
+	    {
+		    processor.Inline = new AutolinkInline(link)
+		    {
+			    IsEmail = isEmail,
+			    Span = new SourceSpan(processor.GetSourcePosition(saved.Start, out line, out column), processor.GetSourcePosition(slice.Start - 1)),
+			    Line = line,
+			    Column = column
+		    };
+	    }
+	    else
+	    {
+		    slice = saved;
+		    if (!HtmlHelper.TryParseHtmlTag(ref slice, out var htmlTag))
+			    return false;
+		    if (!AllowedTags.Contains(htmlTag))
+			    return false;
+		    processor.Inline = new HtmlInline(htmlTag)
+		    {
+			    Span = new SourceSpan(processor.GetSourcePosition(saved.Start, out line, out column), processor.GetSourcePosition(slice.Start - 1)),
+			    Line = line,
+			    Column = column
+		    };
+	    }
+	    return true;
+    }
+}

src/Elastic.Markdown/Myst/MarkdownParser.cs

@@ -9,6 +9,7 @@
 using Elastic.Markdown.Myst.CodeBlocks;
 using Elastic.Markdown.Myst.Comments;
 using Elastic.Markdown.Myst.Directives;
+using Elastic.Markdown.Myst.Extensions;
 using Elastic.Markdown.Myst.FrontMatter;
 using Elastic.Markdown.Myst.InlineParsers;
 using Elastic.Markdown.Myst.Substitution;
@@ -26,16 +27,6 @@ public class MarkdownParser(
 {
 	public IDirectoryInfo SourcePath { get; } = sourcePath;
 
-	private BuildContext Context { get; } = context;
-
-	public static MarkdownPipeline MinimalPipeline { get; } =
-		new MarkdownPipelineBuilder()
-			.UseYamlFrontMatter()
-			.UseInlineAnchors()
-			.UseHeadingsWithSlugs()
-			.UseDirectives()
-			.Build();
-
 	public static MarkdownPipeline Pipeline { get; } =
 		new MarkdownPipelineBuilder()
 			.EnableTrackTrivia()
@@ -52,7 +43,17 @@ public class MarkdownParser(
 			.UsePipeTables()
 			.UseDirectives()
 			.UseEnhancedCodeBlocks()
-			.DisableHtml()
+			.DisableHtmlWithExceptions(["<br>"])
+			.Build();
+
+	private BuildContext Context { get; } = context;
+
+	public static MarkdownPipeline MinimalPipeline { get; } =
+		new MarkdownPipelineBuilder()
+			.UseYamlFrontMatter()
+			.UseInlineAnchors()
+			.UseHeadingsWithSlugs()
+			.UseDirectives()
 			.Build();
 
 	public ConfigurationFile Configuration { get; } = configuration;
@@ -107,6 +108,4 @@ public MarkdownDocument Parse(string yaml, IFileInfo parent, YamlFrontMatter? ma
 		var markdownDocument = Markdig.Markdown.Parse(yaml, Pipeline, context);
 		return markdownDocument;
 	}
-
-
 }

tests/Elastic.Markdown.Tests/Elastic.Markdown.Tests.csproj

@@ -28,4 +28,8 @@
     <Using Include="Xunit"/>
   </ItemGroup>
 
+  <ItemGroup>
+    <Folder Include="Extensions\" />
+  </ItemGroup>
+
 </Project>

tests/Elastic.Markdown.Tests/Inline/RestrictedAutolinkinlineTests.cs

@@ -0,0 +1,31 @@
+// Licensed to Elasticsearch B.V under one or more agreements.
+// Elasticsearch B.V licenses this file to you under the Apache 2.0 License.
+// See the LICENSE file in the project root for more information
+
+using FluentAssertions;
+using Xunit.Abstractions;
+
+namespace Elastic.Markdown.Tests.Inline;
+
+public class AllowBRTagTest(ITestOutputHelper output)
+	: InlineTest(output,
+		"Hello,<br>World!")
+{
+	[Fact]
+	public void GeneratesHtml() =>
+		Html.Should().Contain(
+			"<p>Hello,<br>World!</p>"
+		);
+}
+
+public class DisallowSpanTag(ITestOutputHelper output)
+	: InlineTest(output,
+		"Hello,<span>World!</span>")
+{
+	[Fact]
+	// span tag is rendered as text
+	public void GeneratesHtml() =>
+		Html.Should().Contain(
+			"<p>Hello,&lt;span&gt;World!&lt;/span&gt;</p>"
+		);
+}