Fix: keep trailing content out of an unmatched emphasis closer (#743) (#950)

* fix: keep trailing content out of unmatched emphasis closer (#743)

When EnableTrackTrivia() is enabled, an emphasis closer whose delimiter
run is only partially consumed (its leftover dropping below the minimum
count) kept the inlines that follow it nested inside the emphasis. The
roundtrip renderer then emitted the closing delimiters after that
trailing content, corrupting the output, for example a grid table
separator row using the "marked" (==) delimiter from EmphasisExtras.

Move the trailing content out to the outermost enclosing emphasis when
the closer can no longer be matched, mirroring the existing handling for
a fully consumed opener. The leftover closer keeps its position, so
cases without trailing content are unaffected. Adds a regression test.

* Fix unmatched emphasis closer cleanup order (#743)

---------

Co-authored-by: Alexandre Mutel <alexandre_mutel@live.com>

site/docs/extensions/emphasis-extras.md

@@ -60,6 +60,11 @@ This is ==highlighted== text.
 
 This is ==highlighted== text.
 
+An unmatched remainder of a closing delimiter run must not pull subsequent text
+into the emphasis. For example, `===a===== tail` renders as
+`=<mark>a===</mark> tail`. This example also round-trips unchanged with
+`EnableTrackTrivia()` and `RoundtripRenderer`.
+
 ## HTML output
 
 {.table}

src/Markdig.Tests/TestEmphasisRoundtrip.cs

@@ -0,0 +1,83 @@
+// Copyright (c) Alexandre Mutel. All rights reserved.
+// This file is licensed under the BSD-Clause 2 license.
+// See the license.txt file in the project root for more information.
+
+using Markdig.Renderers.Roundtrip;
+using Markdig.Syntax;
+
+namespace Markdig.Tests;
+
+[TestFixture]
+public class TestEmphasisRoundtrip
+{
+    private static string RoundTrip(string markdown, MarkdownPipeline pipeline)
+    {
+        MarkdownDocument document = Markdown.Parse(markdown, pipeline);
+        var writer = new StringWriter();
+        var renderer = new RoundtripRenderer(writer);
+        pipeline.Setup(renderer);
+        renderer.Write(document);
+        return writer.ToString();
+    }
+
+    // Leftover delimiter characters below the minimum count used to stay nested
+    // inside the emphasis they closed, moving the following text to the end of
+    // the block. https://github.com/xoofx/markdig/issues/743
+    [Test]
+    public void GridTableSeparatorRoundtripsWithTrackTrivia()
+    {
+        string markdown =
+            "+---------------+---------------+--------------------+\n" +
+            "| Fruit         | Price         | Advantages         |\n" +
+            "+===============+===============+====================+\n" +
+            "| Bananas       | first line    | first line         |\n" +
+            "|               | next line     | next line          |\n" +
+            "+---------------+---------------+--------------------+\n";
+
+        var pipeline = new MarkdownPipelineBuilder()
+            .UseAutoLinks()
+            .UseEmphasisExtras()
+            .UseListExtras()
+            .EnableTrackTrivia()
+            .Build();
+
+        Assert.That(RoundTrip(markdown, pipeline), Is.EqualTo(markdown));
+    }
+
+    // Same defect, minimal: leftovers (15 = 7 * 2 + 1) below the "==" minimum.
+    [Test]
+    public void UnbalancedMarkedRunsRoundtrip()
+    {
+        string markdown = "+===============+===============+\n";
+
+        var pipeline = new MarkdownPipelineBuilder()
+            .UseEmphasisExtras()
+            .EnableTrackTrivia()
+            .Build();
+
+        Assert.That(RoundTrip(markdown, pipeline), Is.EqualTo(markdown));
+    }
+
+    [TestCase("===a===== tail\n", "<p>=<mark>a===</mark> tail</p>")]
+    [TestCase("===a==== tail\n", "<p>=<mark>a==</mark> tail</p>")]
+    [TestCase("=====a======= tail\n", "<p>=<mark><mark>a===</mark></mark> tail</p>")]
+    [TestCase("+++a+++++ tail\n", "<p>+<ins>a+++</ins> tail</p>")]
+    [TestCase("*===a===== tail*\n", "<p><em>=<mark>a===</mark> tail</em></p>")]
+    [TestCase("===a===== tail ==next==\n", "<p>=<mark>a===</mark> tail <mark>next</mark></p>")]
+    public void UnmatchedCloserPreservesTrailingContent(string markdown, string expectedHtml)
+    {
+        var pipeline = new MarkdownPipelineBuilder().UseEmphasisExtras().EnableTrackTrivia().Build();
+
+        Assert.That(RoundTrip(markdown, pipeline), Is.EqualTo(markdown));
+        Assert.That(Markdown.ToHtml(markdown, pipeline).Trim(), Is.EqualTo(expectedHtml));
+        Assert.That(Markdown.ToHtml(markdown, new MarkdownPipelineBuilder().UseEmphasisExtras().Build()).Trim(), Is.EqualTo(expectedHtml));
+    }
+
+    // The fix must not prevent balanced "marked" emphasis from being detected.
+    [Test]
+    public void BalancedMarkedEmphasisStillParses()
+    {
+        var pipeline = new MarkdownPipelineBuilder().UseEmphasisExtras().Build();
+        Assert.That(Markdown.ToHtml("==bold==", pipeline).Trim(), Is.EqualTo("<p><mark>bold</mark></p>"));
+    }
+}

src/Markdig/Parsers/Inlines/EmphasisInlineParser.cs

@@ -407,18 +407,33 @@ public class EmphasisInlineParser : InlineParser, IPostInlineProcessor
                             i--;
                         }
                     }
-                    else if ((closeDelimiter.Type & DelimiterType.Open) == 0)
-                    {
-                        closeDelimiter.ReplaceBy(closeDelimiter.AsLiteralInline());
-                        delimiters.RemoveAt(i);
-                        i--;
-                        break;
-                    }
                     else
                     {
+                        // Keep unmatched closers attached until their trailing content
+                        // has been moved out below.
                         break;
                     }
                 }
+
+                // Leftover delimiter characters keep the following inlines nested inside
+                // the emphasis; move them out so they keep their original position.
+                if (closeDelimiter.DelimiterCount > 0 && closeDelimiter.Parent is EmphasisInline closerParentEmphasis)
+                {
+                    var outermostEmphasis = closerParentEmphasis;
+                    while (outermostEmphasis.Parent is EmphasisInline ancestorEmphasis)
+                    {
+                        outermostEmphasis = ancestorEmphasis;
+                    }
+
+                    closeDelimiter.MoveChildrenAfter(outermostEmphasis);
+                }
+
+                if (closeDelimiter.DelimiterCount > 0 && (closeDelimiter.Type & DelimiterType.Open) == 0)
+                {
+                    closeDelimiter.ReplaceBy(closeDelimiter.AsLiteralInline());
+                    delimiters.RemoveAt(i);
+                    i--;
+                }
             }
         }