Julmar.GenMarkdown 1.0.7

GenMarkdown

.NET Core library to make it easy to generate Markdown with an object graph.

Supported elements

The library supports all standard Markdown and several extensions including:

• Bold, Italic, BoldItalic
• Highlight text
• Superscript text
• Subscript text
• Strikethrough text
• Pipe tables
• Grid tables
• Definition lists
• Task lists

Basic example

Markdown is constructed with blocks and inlines. The MarkdownDocument object holds blocks, where each block can contain inlines. All the block objects implement collection semantics so you can use standard Add methods, or even constructor initializers.

Example code

Here's an example document with Headings and paragraphs.

var doc = new MarkdownDocument
{
    // Level 1 Heading (# Example title)
    new Heading(1, "Example title"),
    
    // Paragraph with some inline bold and italic text
    new Paragraph {
        "This is some text with ", Text.Bold("some inline bold text"), " and ", Text.Italic("some inline italics"),
        ". Here is some ", Text.BoldItalic("bold and italic"), " text. ",
        "There's an implicit converter from strings to create plain text.",
    },
    
    "Implicit converter works in the initializer too.",
    new Paragraph("Can also have a single paragraph as part of the constructor."),
    new Paragraph { "Can inline code with the Text.Code helper - for example: ", Text.Code("Program"), " object." }
};

This will generate the following output:

# Example title

This is some text with **some inline bold text** and _some inline italics_. Here is some ***bold and italic*** text. There's an implicit converter from strings to create plain text.

Implicit converter works in the initializer too.

Can also have a single paragraph as part of the constructor.

Can inline code with the Text.Code helper - for example: `Program` object.

The same document can be created through methods:

var doc = new MarkdownDocument();
// Level 1 Heading (# Example title)
doc.Add(new Heading(1, "Example title")); // Paragraph with some inline bold and italic text
doc.Add(new Paragraph {
    "This is some text with ", Text.Bold("some inline bold text"), " and ",
    Text.Italic("some inline italics"),
    ". There's an implicit converter from strings to create plain text.",
});
doc.Add("Implicit converter works in the initializer too.");
doc.Add(new Paragraph("Can also have a single paragraph as part of the constructor."));
doc.Add(new Paragraph {
    "Can inline code with the Text.Code helper - for example: ", Text.Code("Program"), " object."
});

Generating Markdown

Once you have created the object graph, you can use the Write method to generate the text Markdown content.

void Write(TextWriter writer, MarkdownFormatting formatting);

The MarkdownFormatting object has options to control how Markdown is generated - for example whether to use _ or * for emphasis (italic) inlines.

You can also use the ToString() method on any object to generate Markdown, however this approach does not support customization of the markdown through formatting options.

var doc = new MarkdownDocument();
// Add things..

// Write to the specified file.
doc.Write(File.OpenWrite("somefile.md"));

// Customize the output
doc.Write(Console.Out, 
    new MarkdownFormatting {
        UseAsterisksForBullets = true,    // Use '*'' on unordered lists instead of '-'
        UseAsterisksForEmphasis = true,   // Use '*' to emphasize text instead of '_'
        OrderedListUsesSequence = true,  // Always use correct sequence on ordered lists instead of '1.'
        UseAlternateHeadingSyntax = true   // For H1/H2, use older Heading syntax (text + underline)
    });

Modifying the document

The document itself is a writable collection of MarkdownBlock objects - you can modify the document at any time by adjusting these objects.

var doc = new MarkdownDocument
{
    // Level 1 Heading (# Example title)
    new Heading(1, "Example title"), // Paragraph with some inline bold and italic text
    new Paragraph
    {
        "This is some text with ", Text.Bold("some inline bold text"), " and ",
        Text.Italic("some inline italics"),
        ". There's an implicit converter from strings to create plain text.",
    }
};

// Modify the Heading to be ## and add an {#main-Heading} identifier.
var Heading = (Heading) doc[0];
Heading.Level = 2;
Heading.Id = "main-Heading";

This produces:

## Example title {#main-Heading}

This is some text with **some inline bold text** and _some inline italics_. There's an implicit converter from strings to create plain text.

Inlines

The library supports several inlines, along with helper methods on the Text object to generate the most common types.

Type	Description
BoldText	Creates bolded text.
BoldItalicText	Creates bold + italicized text.
Emoji	Creates an emoji (:xyz:)
HighlightText	Highlights the specified text with ==. This is an extension and not supported by all Markdown parsers.
InlineCode	Displays inline code surrounded by tilde characters.
InlineLink	Creates an inline link.
ItalicText	Creates emphasized text.
LineBreak	Generates a line-break (back slash at end of line).
StrikethroughText	Strikes the specified text with ~~. This is an extension and not supported by all Markdown parsers.
SubscriptText	Subscript text specified with ~. This is an extension and not supported by all Markdown parsers.
SuperscriptText	Superscript text specified with ^. This is an extension and not supported by all Markdown parsers.
Text	Base text element. There is an implicit converter from string to this type.

Examples

new Paragraph 
{
    "Implicit converter to Text() or can use ",
    new Text("Normal text. Or can use inline types like "),
    new BoldText("BoldText"), " or ", new ItalicText("ItalicText"), "."
};

This generates:

Implicit converter to Text() or can use Normal text. Or can use inline types like **BoldText** or _ItalicText_.

Helpers

The Text type has several helpers to generate the most common types - this removes the need to call new inline:

Helper	Creates
Text.Bold method	BoldText object.
Text.BoldItalic method	BoldItalicText object.
Text.Italic method	ItalicText object.
Text.Code method	InlineCode object.
Text.Link method	InlineLink object.
Text.LineBreak property	LineBreak object.

new Paragraph
{
    "This is an example with ", Text.Bold("bold"), " and ", Text.Italic("italic"), " text.",
    " You can also have inline ", Text.Code("Code"), " or ", Text.Link("links", "https://www.msn.com"), "."
}

This is an example with **bold** and _italic_ text. You can also have inline `Code` or [links](https://www.msn.com).

Supported blocks

Here's examples of each supported block type.

Paragraph

The Paragraph is the main block type and represents a paragraph of Markdown text. It can contain any inline content and has implicit conversions from strings.

var doc = new MarkdownDocument
{
    "Direct text is turned into a paragraph.",

    new Paragraph
    {
        "This is some text with ", Text.Bold("some inline bold text"), " and ",
        Text.Italic("some inline italics"), ". Now a line break ->",
        Text.LineBreak,
        "There's an explicit converter from strings to create plain text.",
        "You can also escape ", Text.Code("any `code` items used in the file"), "."
    },

    new Paragraph("Can also have a single paragraph as part of the constructor."),

    new Paragraph
    {
        new Text("Or use the explicit Text objects to add plain text or "),
        new BoldText("BoldText"), new Text(" or "), 
        new ItalicText("ItalicText"), new Text(" objects.")
    },

    new Paragraph
    {
        "Can inline code with the Text.Code helper - for example: ",
        Text.Code("Program"), " object.", "Or use the explicit ",
        new InlineCode("InlineCode"), " object."
    },
};

This generates:

Direct text is turned into a paragraph.

This is some text with **some inline bold text** and _some inline italics_. Now a line break ->\
There's an explicit converter from strings to create plain text.You can also escape ``any `code` items used in the file``.

Can also have a single paragraph as part of the constructor.

Or use the explicit Text objects to add plain text or **BoldText** or _ItalicText_ objects.

Can inline code with the Text.Code helper - for example: `Program` object.Or use the explicit `InlineCode` object.

Heading

The Heading object takes a numeric level (1-5) and generates a Markdown Heading. It allows inline Markdown as part of the creation and also supports an optional Id property which will create the identifier extension on the Heading ({#id}).

new Heading("Creates an H1 by default"),
"Some text",
new Heading(1, "Example title"),
"Here some example paragraph text under the title.",
new Heading(2) { "Headings can have ", Text.Bold("inline elements"), " too." },
"With some more text.",
new Heading(3, "Or identifiers") { Id = "level3-hdr" },
"Fini."

Generates:

# Creates an H1 by default

Some text

# Example title

Here some example paragraph text under the title.

## Headings can have **inline elements** too.

With some more text.

### Or identifiers {#level3-hdr}

Fini.

Image

The Image object creates a Markdown image.

new Image("alt-text goes here", "https://avatars.githubusercontent.com/u/5099741?v=4")

![alt-text goes here](https://avatars.githubusercontent.com/u/5099741?v=4)

BlockQuote

BlockQuote creates quotes (>) in the document. It can contain one or more blocks.

new BlockQuote("This is a quote.");

// Can contain multiple blocks.
new BlockQuote
{
    "This is also a quote.",
    new Image("Image", "https://www.nuget.org/Content/gallery/img/logo-Heading.svg")
};

new BlockQuote
{
    "This is a quote.",
    "With multiple lines",
    "And an embedded\r\ncarriage return!"
}

> This is a quote.

> This is also a quote.
> ![Image](https://www.nuget.org/Content/gallery/img/logo-Heading.svg)

> This is a quote.
> With multiple lines
> And an embedded
> carriage return!

CodeBlock

The CodeBlock element generates a fenced codeblock with optional language.

new CodeBlock
{
    "using namespace System;\r\n",
    "\r\n",
    "namespace Test\r\n",
    "{\r\n",
    "   public static class Program\r\n",
    "   {\r\n",
    "       Console.WriteLine(\"Hello World\");\r\n",
    "   }\r\n",
    "}\r\n",
}

Generates

```
using namespace System;

namespace Test
{
   public static class Program
   {
       Console.WriteLine("Hello World");
   }
}
```

Adding a language will place that onto the code fence.

new CodeBlock("csharp")
{
    "using namespace System;\r\n",
    "\r\n",
    "namespace Test\r\n",
    "{\r\n",
    "   public static class Program\r\n",
    "   {\r\n",
    "       Console.WriteLine(\"Hello World\");\r\n",
    "   }\r\n",
    "}\r\n",
}

```csharp
    ... same code as above

HorizontalRule

The HorizontalRule block generates a horizontal rule comprised of three dashes.

"Starting paragraph",
new HorizontalRule(),
"Ending paragraph"

Starting paragraph

---

Ending paragraph

Image

The Image block creates an embedded image in the document, this includes an alt-tag for accessibility, the URL, and an optional description.

"Here's an image:",
new Image("An image representing the NuGet logo", "https://www.nuget.org/Content/gallery/img/logo-Heading.svg", "A description for the image")

Here's an image:

![An image representing the NuGet logo](https://www.nuget.org/Content/gallery/img/logo-Heading.svg "A description for the image")

Link

The Link block creates a link to some other content.

"An example link:",
new Link("www.microsoft.com", "https://microsoft.com"),

An example link:

[www.microsoft.com](https://microsoft.com)

Tables

The Table block generates standard Markdown tables, sometimes referred to as pipe tables. These are structured rows, each containing a specific number of columns. This table type does not support column or row spanning.

The table is comprised of TableRow objects, with each row having one or more TableCell objects. The TableCell is a block container - so it can contain one or more Markdown blocks. There's also an implicit converter to take a Paragraph object and turn it into a TableCell to simplify the inline collection syntax.

Here's a simple example of a 3x3 table:

new Table(3)
{
    new TableRow
    {
        new TableCell("Heading 0"),
        new TableCell("Heading 1"),
        new TableCell("Heading 2"),
    },
    new TableRow
    {
        new TableCell("(0,0)"),
        new TableCell("(0,1)"),
        new TableCell("(0,2)"),
    },
    new TableRow
    {
        new TableCell("(1,0)"),
        new TableCell("(1,1)"),
        new TableCell("(1,2)"),
    },
    new TableRow
    {
        new TableCell("(2,0)"),
        new TableCell("(2,1)"),
        new TableCell("(2,2)"),
    }
}

This generates:

|Heading 0|Heading 1|Heading 2|
|---|---|---|
|(0,0)|(0,1)|(0,2)|
|(1,0)|(1,1)|(1,2)|
|(2,0)|(2,1)|(2,2)|

A MarkdownFormatting option - PrettyPipeTables can be set to add spacing (up to 30 characters per column) to align the columns nicely.

Table table = new Table() { ... };
document.Add(table);

document.Write(Console.Out, new MarkdownFormatting { PrettyPipeTables = true });

The previous table will now render as:

| Heading 0 | Heading 1 | Heading 2 |
|-----------|-----------|-----------|
| (0,0)     | (0,1)     | (0,2)     |
| (1,0)     | (1,1)     | (1,2)     |
| (2,0)     | (2,1)     | (2,2)     |

You can simplify the creation by relying on the implicit TableCell conversion, and using the new targeted-type new support in C#9. This code generates the exact same Markdown:

new Table(3)
{
    new() {"Heading 0", "Heading 1", "Heading 2"},
    new() { "(0,0)", "(0,1)", "(0,2)" },
    new() { "(1,0)", "(1,1)", "(1,2)" },
    new() { "(2,0)", "(2,1)", "(2,2)" },
}

The TableCell can contain any type of Paragraph text:

new Table(3)
{
    new()
    {
        "", new Paragraph { Text.Bold("Math") }, new Paragraph { Text.Bold("Science") },
    },
    new()
    {
        new Paragraph { Text.Bold("John Smith") }, "A", "B"
    },
    new()
    {
        new Paragraph { Text.Bold("Susan Green") }, "C", "A"
    }
}

This generates (with pretty print):

|                 | **Math** | **Science** |
|-----------------|----------|-------------|
| **John Smith**  | A        | B           |
| **Susan Green** | C        | A           |

A second constructor allows you to control the column alignment by passing in an array of ColumnAlignment values:

new Table(new[] {ColumnAlignment.Default, ColumnAlignment.Center, ColumnAlignment.Right})
{
    new()
    {
        new TableCell(),
        new Paragraph {Text.Bold("Math")}, 
        new Paragraph {Text.Bold("Science")},
    },
    new()
    {
        new Paragraph {Text.Bold("John Smith")},
        "A",
        "B"
    },
    new()
    {
        new Paragraph {Text.Bold("Susan Green")},
        "C",
        "A"
    },
}

This changes the second row of the generated markdown to include justification hints:

|                 | **Math** | **Science** |
|-----------------|:--------:|------------:|
| **John Smith**  | A        | B           |
| **Susan Green** | C        | A           |

Grid table

A second table type (GridTable) conforms to the GridTableSpec for PanDoc.

Warning This is not a standard Markdown construct and cannot be handled by all Markdown parsers.

This type adds a few additional constructors and properties to control the shape of the produced table.

Property	Description
MaxWidth	This controls the max width of the generated table. It defaults to 80 characters. Note that the actual width might vary slightly by one or two characters depending on where text can be wrapped and how columns are split. But it will be close to this value.
HasHeader	This determines whether the first row is considered a header row and rendered as such. It defaults to false.

Both the TableRow and TableCell objects have a column span property which allow the content to span columns. These are ignored by the Table renderer, but supported by the GridTable renderer.

Note The spec allows for row spanning as well, but that feature is not implemented here as no parsers currently support it.

Finally, there is a new constructor which takes a GridColumnDefinition array to define each column. This allows column widths to be specified as percentages. For example, the following would create a table with three columns:

1. Left justified, sized to content or half remaining space, whichever is smaller.
2. Centered, sized to content or half remaining space, whichever is smaller,
3. Left justified, 50% of the space available.

var gridTable = new GridTable(new GridColumnDefinition(), new GridColumnDefinition(ColumnAlignment.Center), new GridColumnDefinition { Width = .5 });

Numeric Lists

The OrderedList block creates sequenced numeric lists of content. The list itself is a container and can contain different block types including other lists, tables, quotes, etc.

The simplest form of numbered list is text:

new OrderedList {"One", "Two", "Three" },
new OrderedList(4) {"Four", "Five", "Six"}, // Can start at a specific number.
new OrderedList {"One", "Two", "Three"}

This will generate three lists.

1. One
1. Two
1. Three

4. Four
1. Five
1. Six

1. One
1. Two
1. Three

Notice that the generator follows Markdown guidelines and emits the sequence "1" for each item unless it has a given starting number. This behavior can be controlled with the MarkdownFormatting.OrderedListUsesSequence property:

// Same list as above:
doc.Write(Console.Out, new MarkdownFormatting() { OrderedListUsesSequence = true });

Will now create:

1. One
2. Two
3. Three

4. Four
5. Five
6. Six

1. One
2. Two
3. Three

The items in the list are blocks themselves, for example you can have code blocks in the list:

new OrderedList
{
    "Do this:",
    {
        new Paragraph
        {
            "Use the following command to make a new directory named ",
            Text.Code("data"),
            "."
        },
        new CodeBlock("bash", "mkdir data")
    },
    {
        new Paragraph
        {
            "Use the ", Text.Code("wget"), " command to download the dataset.",
        },
        new CodeBlock("bash")
        {
            "wget -P data/ https://raw.githubusercontent.com/MicrosoftDocs/mslearn-data-wrangling-shell/main/NASA-logs-1995.txt\r\n",
            "wget - P data / https://raw.githubusercontent.com/MicrosoftDocs/mslearn-data-wrangling-shell/main/NASA-software-API.txt"
        }
    },
    {
        new Paragraph
        {
            "Change to the new directory by using the command ", Text.Code("cd"), "."
        },
        new CodeBlock("bash", "cd data")
    },
    {
        new Paragraph
        {
            "Verify that you have the correct files by using the command ", Text.Code("ls"), "."
        },
        new CodeBlock("bash", "ls")
    }
};

Which will generate the following Markdown - notice that the blocks are indented to keep the structure:

1. Do this:
1. Use the following command to make a new directory named `data`.

   ```bash
   mkdir data
   ```

1. Use the `wget` command to download the dataset.

   ```bash
   wget -P data/ https://raw.githubusercontent.com/MicrosoftDocs/mslearn-data-wrangling-shell/main/NASA-logs-1995.txt
   wget - P data / https://raw.githubusercontent.com/MicrosoftDocs/mslearn-data-wrangling-shell/main/NASA-software-API.txt
   ```

1. Change to the new directory by using the command `cd`.

   ```bash
   cd data
   ```

1. Verify that you have the correct files by using the command `ls`.

   ```bash
   ls
   ```

A note about indenting

Another way to keep bits of Markdown together is through the IndentLevel property. This is on every MarkdownBlock-derived object and allows you to force an element to be indented in the generated structure. This can be useful if an item is added into the document in between two lists which are sequentially tied together. For example, take the following code:

var doc = new MarkdownDocument
{
    new OrderedList
    {
        "Item #1",
        "Item #2"
    },

    new BlockQuote("A quote is here"),

    new OrderedList(3)
    {
        "Item #3",
        "Item #4"
    }
};

The intent here is to have a 1-4 sequence, however the block quote in the middle will break the sequence and not be technically part of the list itself. This will generate the following markdown:

1. Item #1
1. Item #2

> A quote is here

3. Item #3
1. Item #4

This isn't wrong, but it will not look quite right as the quote block won't be indented. One way to fix it would be to include it into the numbered list directly as a sibling of "Item #2":

new OrderedList
{
    "Item #1",
    {
        "Item #2",
        new BlockQuote("A quote is here")
    }
},

However, another way to solve this is to set the IndentLevel onto the block quote:

var doc = new MarkdownDocument
{
    new OrderedList
    {
        "Item #1",
        "Item #2"
    },

    // Set the indent level to '1' to push this under the list.
    new BlockQuote("A quote is here") { IndentLevel = 1},

    new OrderedList(3)
    {
        "Item #3",
        "Item #4"
    }
};

This will change the Markdown to essentially space in the quote block - giving us the syntax we want.

1. Item #1
1. Item #2

   > A quote is here

3. Item #3
1. Item #4

The IndentLevel property must be between 0 and 3 and each level will push the block further into the content. It's set to zero by default (no indentation).

Bullet lists

The List is identical to the OrderedList except it doesn't sequence items, but instead prefaces them with a dash to create a bulleted list. Here's an example:

new List
{
    new Paragraph { "Item #1 - ", Text.Link("google.com", "https://google.com"), "." },
    new Paragraph("Item #2"),
    {
        new Paragraph("Item #2.5"),
        new Paragraph("With some additional text"),
        new Image("and an image", "https://www.nuget.org/Content/gallery/img/logo-Heading.svg"),
        new CodeBlock
        {
            "using namespace System;\r\n",
            "\r\n",
            "namespace Test\r\n",
            "{\r\n",
            "   public static class Program\r\n",
            "   {\r\n",
            "       Console.WriteLine(\"Hello World\");\r\n",
            "   }\r\n",
            "}\r\n"
        }
    },
    {
        new Paragraph("Item #3"),
        new BlockQuote
        {
            "This is a quote.",
            "With multiple lines",
            "And an embedded\r\ncarriage return!"
        }
    },
},

This generates

- Item #1 - [google.com](https://google.com).
- Item #2
- Item #2.5

  With some additional text

  ![and an image](https://www.nuget.org/Content/gallery/img/logo-Heading.svg)

  ```
  using namespace System;

  namespace Test
  {
     public static class Program
     {
         Console.WriteLine("Hello World");
     }
  }
  ```

- Item #3

  > This is a quote.
  > With multiple lines
  > And an embedded
  > carriage return!

The default bullet character is the dash, however you can change that to an asterisk with the MarkdownFormatting.UseAsterisksForBullets property when using the formatted Write method.

TaskList

The TaskList block will create a checklist of tasks in Markdown. This is an extension and not supported by all Markdown parsers. Each task is represented by the TaskItem object and a list is simply a collection of these items. The TaskItem has a IsChecked property which can be set to indicate a checkmark. This value can also be passed as a constructor parameter as shown below.

new TaskList()
{
    new TaskItem("Item #1"),
    new TaskItem("Item #2", isChecked: true),
    new TaskItem()
    {
        "This is a ", Text.Bold("Bold"), " choice."
    },
    new TaskItem("Final shot!")
};

This generates:

- [ ] Item #1
- [x] Item #2
- [ ] This is a **Bold** choice.
- [ ] Final shot!

Definition lists

The DefinitionList block creates definition lists in Markdown. This is an extension and not supported by all Markdown parsers. It consists of a term and one or more definitions. The term and definitions can only be string types - inline and Markdown blocks are not supported for this element.

new DefinitionList
{
    new Definition("First term", "This is a definition"),
    new Definition("Second term")
    {
        "First definition",
        "Second definition"
    }
},

This generates:

First term
: This is a definition

Second term
: First definition
: Second definition