Merge pull request xoofx#139 from kenmuse/enhance/yaml-frontmatter

Implement Pandoc compatible front matter parsing

src/Markdig.Tests/Specs/Specs.cs

@@ -20492,6 +20492,124 @@ public void Example003()
             Console.WriteLine("Example {0}" + Environment.NewLine + "Section: {0}" + Environment.NewLine, 3, "Extensions YAML frontmatter discard");
 			TestParser.TestSpec("----\nthis: is a frontmatter\n----\nThis is a text", "<hr />\n<h2>this: is a frontmatter</h2>\n<p>This is a text</p>", "yaml");
         }
+    }
+        // It can end with three dots `...`:
+    [TestFixture]
+    public partial class TestExtensionsYAMLfrontmatterdiscard
+    {
+        [Test]
+        public void Example004()
+        {
+            // Example 4
+            // Section: Extensions YAML frontmatter discard
+            //
+            // The following CommonMark:
+            //     ---
+            //     this: is a frontmatter
+            //     ...
+            //     This is a text
+            //
+            // Should be rendered as:
+            //     <p>This is a text</p>
+
+            Console.WriteLine("Example {0}" + Environment.NewLine + "Section: {0}" + Environment.NewLine, 4, "Extensions YAML frontmatter discard");
+			TestParser.TestSpec("---\nthis: is a frontmatter\n...\nThis is a text", "<p>This is a text</p>", "yaml");
+        }
+    }
+        // It expects exactly three dots `...`:
+    [TestFixture]
+    public partial class TestExtensionsYAMLfrontmatterdiscard
+    {
+        [Test]
+        public void Example005()
+        {
+            // Example 5
+            // Section: Extensions YAML frontmatter discard
+            //
+            // The following CommonMark:
+            //     ---
+            //     this: is a frontmatter
+            //     ....
+            //     This is a text
+            //
+            // Should be rendered as:
+            //     
+
+            Console.WriteLine("Example {0}" + Environment.NewLine + "Section: {0}" + Environment.NewLine, 5, "Extensions YAML frontmatter discard");
+			TestParser.TestSpec("---\nthis: is a frontmatter\n....\nThis is a text", "", "yaml");
+        }
+    }
+        // Front matter ends with the first line containing three dots `...` or three dashes `...`:
+    [TestFixture]
+    public partial class TestExtensionsYAMLfrontmatterdiscard
+    {
+        [Test]
+        public void Example006()
+        {
+            // Example 6
+            // Section: Extensions YAML frontmatter discard
+            //
+            // The following CommonMark:
+            //     ---
+            //     this: is a frontmatter
+            //     ....
+            //     
+            //     Hello
+            //     ---
+            //     This is a text
+            //
+            // Should be rendered as:
+            //     <p>This is a text</p>
+
+            Console.WriteLine("Example {0}" + Environment.NewLine + "Section: {0}" + Environment.NewLine, 6, "Extensions YAML frontmatter discard");
+			TestParser.TestSpec("---\nthis: is a frontmatter\n....\n\nHello\n---\nThis is a text", "<p>This is a text</p>", "yaml");
+        }
+    }
+        // It expects whitespace can exist after the leading characters
+    [TestFixture]
+    public partial class TestExtensionsYAMLfrontmatterdiscard
+    {
+        [Test]
+        public void Example007()
+        {
+            // Example 7
+            // Section: Extensions YAML frontmatter discard
+            //
+            // The following CommonMark:
+            //     ---   
+            //     this: is a frontmatter
+            //     ...
+            //     This is a text
+            //
+            // Should be rendered as:
+            //     <p>This is a text</p>
+
+            Console.WriteLine("Example {0}" + Environment.NewLine + "Section: {0}" + Environment.NewLine, 7, "Extensions YAML frontmatter discard");
+			TestParser.TestSpec("---   \nthis: is a frontmatter\n...\nThis is a text", "<p>This is a text</p>", "yaml");
+        }
+    }
+        // It expects whitespace can exist after the trailing characters
+    [TestFixture]
+    public partial class TestExtensionsYAMLfrontmatterdiscard
+    {
+        [Test]
+        public void Example008()
+        {
+            // Example 8
+            // Section: Extensions YAML frontmatter discard
+            //
+            // The following CommonMark:
+            //     ---
+            //     this: is a frontmatter
+            //     ...     
+            //     This is a text
+            //
+            // Should be rendered as:
+            //     <p>This is a text</p>
+
+            Console.WriteLine("Example {0}" + Environment.NewLine + "Section: {0}" + Environment.NewLine, 8, "Extensions YAML frontmatter discard");
+			TestParser.TestSpec("---\nthis: is a frontmatter\n...     \nThis is a text", "<p>This is a text</p>", "yaml");
+        }
     }
         // # Extensions
         //

src/Markdig.Tests/Specs/YamlSpecs.md

@@ -42,3 +42,62 @@ This is a text
 <p>This is a text</p>
 ````````````````````````````````
 
+It can end with three dots `...`:
+
+```````````````````````````````` example
+---
+this: is a frontmatter
+...
+This is a text
+.
+<p>This is a text</p>
+````````````````````````````````
+
+It expects exactly three dots `...`:
+
+```````````````````````````````` example
+---
+this: is a frontmatter
+....
+This is a text
+.
+````````````````````````````````
+
+Front matter ends with the first line containing three dots `...` or three dashes `...`:
+
+```````````````````````````````` example
+---
+this: is a frontmatter
+....
+
+Hello
+---
+This is a text
+.
+<p>This is a text</p>
+````````````````````````````````
+
+It expects whitespace can exist after the leading characters
+
+```````````````````````````````` example
+---   
+this: is a frontmatter
+...
+This is a text
+.
+<p>This is a text</p>
+````````````````````````````````
+
+It expects whitespace can exist after the trailing characters
+
+```````````````````````````````` example
+---
+this: is a frontmatter
+...     
+This is a text
+.
+<p>This is a text</p>
+````````````````````````````````
+
+
+

src/Markdig/Extensions/Yaml/YamlFrontMatterBlock.cs

@@ -10,7 +10,7 @@ namespace Markdig.Extensions.Yaml
     /// A YAML frontmatter block.
     /// </summary>
     /// <seealso cref="Markdig.Syntax.CodeBlock" />
-    public class YamlFrontMatterBlock : CodeBlock, IFencedBlock
+    public class YamlFrontMatterBlock : CodeBlock
     {
         /// <summary>
         /// Initializes a new instance of the <see cref="YamlFrontMatterBlock"/> class.
@@ -19,13 +19,5 @@ public class YamlFrontMatterBlock : CodeBlock, IFencedBlock
         public YamlFrontMatterBlock(BlockParser parser) : base(parser)
         {
         }
-
-        public string Info { get; set; }
-
-        public string Arguments { get; set; }
-
-        public int FencedCharCount { get; set; }
-
-        public char FencedChar { get; set; }
     }
 }

src/Markdig/Extensions/Yaml/YamlFrontMatterParser.cs

@@ -1,44 +1,128 @@
 // 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.Helpers;
 using Markdig.Parsers;
+using Markdig.Syntax;
 
 namespace Markdig.Extensions.Yaml
 {
     /// <summary>
     /// Block parser for a YAML frontmatter. 
     /// </summary>
-    /// <seealso cref="Markdig.Parsers.FencedBlockParserBase{YamlFrontMatterBlock}" />
-    public class YamlFrontMatterParser : FencedBlockParserBase<YamlFrontMatterBlock>
+    /// <seealso cref="YamlFrontMatterBlock" />
+    public class YamlFrontMatterParser : BlockParser
     {
         // We reuse a FencedCodeBlock parser to grab a frontmatter, only active if it happens on the first line of the document.
 
         /// <summary>
-        /// Initializes a new instance of the <see cref="FencedCodeBlockParser"/> class.
+        /// Initializes a new instance of the <see cref="YamlFrontMatterParser"/> class.
         /// </summary>
         public YamlFrontMatterParser()
         {
-            OpeningCharacters = new[] { '-' };
-            InfoPrefix = null;
-            // We expect only 3 --- at the beginning of the file no more, no less
-            MinimumMatchCount = 3;
-            MaximumMatchCount = 3;
+            this.OpeningCharacters = new[] { '-' };
         }
 
-        protected override YamlFrontMatterBlock CreateFencedBlock(BlockProcessor processor)
+        /// <summary>
+        /// Creates the front matter block.
+        /// </summary>
+        /// <param name="processor">The block processor</param>
+        /// <returns>The front matter block</returns>
+        protected virtual YamlFrontMatterBlock CreateFrontMatterBlock(BlockProcessor processor)
         {
             return new YamlFrontMatterBlock(this);
         }
 
+        /// <summary>
+        /// Tries to match a block opening.
+        /// </summary>
+        /// <param name="processor">The parser processor.</param>
+        /// <returns>The result of the match</returns>
         public override BlockState TryOpen(BlockProcessor processor)
         {
+            // We expect no indentation for a fenced code block.
+            if (processor.IsCodeIndent)
+            {
+                return BlockState.None;
+            }
+
             // Only accept a frontmatter at the beginning of the file
-            if (processor.LineIndex != 0)
+            if (processor.Start != 0)
             {
                 return BlockState.None;
             }
+
+            int count = 0;
+            var line = processor.Line;
+            char c = line.CurrentChar;
+
+            // Must consist of exactly three dashes
+            while (c == '-' && count < 4)
+            {
+                count++;
+                c = line.NextChar();
+            }
+
+            // If three dashes (optionally followed by whitespace)
+            // this is a YAML front matter blcok
+            if (count == 3 && (c == '\0' || c.IsWhitespace()) && line.TrimEnd())
+            {
+                // Create a front matter block
+                var block = this.CreateFrontMatterBlock(processor);
+                block.Column = processor.Column;
+                block.Span.Start = 0;
+                block.Span.End = line.Start;
+
+                // Store the number of matched string into the context
+                processor.NewBlocks.Push(block);
+
+                // Discard the current line as it is already parsed
+                return BlockState.ContinueDiscard;
+            }
+
+            return BlockState.None;
+        }
+
+        /// <summary>
+        /// Tries to continue matching a block already opened.
+        /// </summary>
+        /// <param name="processor">The parser processor.</param>
+        /// <param name="block">The block already opened.</param>
+        /// <returns>The result of the match. By default, don't expect any newline</returns>
+        public override BlockState TryContinue(BlockProcessor processor, Block block)
+        {
+            char matchChar;
+            int count = 0;
+            var c = processor.CurrentChar;
+
+            // Determine if we have a closing fence.
+            // It can start or end with either <c>---</c> or <c>...</c>
+            var line = processor.Line;
+            if (processor.Column == 0 && (c == '-' || c == '.'))
+            {
+                matchChar = c;
+
+                while (c == matchChar)
+                {
+                    c = line.NextChar();
+                    count++;
+                }
+
+                // If we have a closing fence, close it and discard the current line
+                // The line must contain only fence characters and optional following whitespace.
+                if (count == 3 && !processor.IsCodeIndent && (c == '\0' || c.IsWhitespace()) && line.TrimEnd())
+                {
+                    block.UpdateSpanEnd(line.Start - 1);
+
+                    // Don't keep the last line
+                    return BlockState.BreakDiscard;
+                }
+            }
+
+            // Reset the indentation to the column before the indent
+            processor.GoToColumn(processor.ColumnBeforeIndent);
 
-            return base.TryOpen(processor);
+            return BlockState.Continue;
         }
     }
 }