markdown

Markdown Parser Module

This module provides a production-grade Markdown parser for the mdpress project, a Markdown-to-PDF book converter.

Overview

The markdown package wraps goldmark with carefully selected extensions to provide:

• GFM Extensions: Tables, strikethrough, task lists, and autolinks
• Code Syntax Highlighting: Via goldmark-highlighting
• Footnotes: Full footnote support
• Custom Heading IDs: Automatic ID generation for cross-referencing
• Heading Collection: Automatic gathering of heading information for TOC generation
• Cross-Reference Links: Support for internal document references

Features

Supported Markdown Extensions

1. Tables (GFM)

   • Standard GitHub Flavored Markdown tables
   • Column alignment support

2. Strikethrough (GFM)

   • ~~text~~ syntax for strikethrough

3. Task Lists (GFM)

   • - [ ] unchecked and - [x] checked syntax

4. Autolinks (GFM)

   • <https://example.com> and <user@example.com> formats

5. Footnotes

   • [^1] references with [^1]: content definitions

6. Code Highlighting

   • Syntax highlighting for code blocks
   • Multiple theme support (github, monokai, dracula, etc.)

7. Custom Heading IDs

   • Automatic ID generation for headings
   • Support for manual ID specification via attributes

Installation

This module is part of the mdpress project and requires:

go get github.com/yuin/goldmark
go get github.com/yuin/goldmark-highlighting/v2

Usage

Basic Parsing

package main

import (
    "fmt"
    "github.com/yeasy/mdpress/internal/markdown"
)

func main() {
    parser := markdown.NewParser()

    source := []byte(`# Hello World
This is **bold** text.`)

    html, headings, err := parser.Parse(source)
    if err != nil {
        panic(err)
    }

    fmt.Println("HTML:", html)
    fmt.Println("Headings:", headings)
}

With Custom Code Theme

parser := markdown.NewParser(
    markdown.WithCodeTheme("monokai"),
)

html, headings, err := parser.Parse(source)

Collecting Headings for TOC

parser := markdown.NewParser()
html, headings, _ := parser.Parse(source)

// headings is []HeadingInfo with Level, Text, and ID
for _, h := range headings {
    fmt.Printf("%d. %s -> #%s\n", h.Level, h.Text, h.ID)
}

Changing Code Theme After Creation

parser := markdown.NewParser()
parser.SetCodeTheme("github")
html, _, _ := parser.Parse(source)

API Reference

Parser

The main parser struct that holds the goldmark instance and manages parsing state.

type Parser struct {
    // ... private fields
}

NewParser

Creates and returns a new Markdown parser instance.

func NewParser(opts ...ParserOption) *Parser

Parameters:

• opts: Variable number of ParserOption functions for customization

Returns:

• *Parser: Initialized parser instance

Example:

parser := markdown.NewParser(
    markdown.WithCodeTheme("github"),
)

Parse

Parses Markdown source code and returns HTML string and heading information.

func (p *Parser) Parse(source []byte) (string, []HeadingInfo, error)

Parameters:

• source: Markdown source code as byte slice

Returns:

• string: Generated HTML content
• []HeadingInfo: Slice of collected heading information
• error: Any parsing errors

Example:

html, headings, err := parser.Parse([]byte("# Title\nContent"))
if err != nil {
    log.Fatal(err)
}

SetCodeTheme

Sets the code syntax highlighting theme.

func (p *Parser) SetCodeTheme(theme string)

Parameters:

• theme: Theme name (e.g., "github", "monokai", "dracula")

Note: Reinitializes the parser with the new theme. Invalid theme names will fall back to the default style gracefully.

Supported Themes:

• github (default)
• monokai
• dracula
• solarized-dark
• solarized-light
• And others supported by goldmark-highlighting

HeadingInfo

Structure containing information about a heading.

type HeadingInfo struct {
    Level  int    // Heading level (1-6)
    Text   string // Heading text content
    ID     string // Custom heading ID for cross-referencing
    Line   int    // Line number of the heading
    Column int    // Column number of the heading
}

ParserOption

Functional option type for customizing parser behavior.

type ParserOption func(*Parser)

Built-in Options:

• WithCodeTheme(theme string): Set code highlighting theme

Examples

Complete Document Parsing

package main

import (
    "fmt"
    "github.com/yeasy/mdpress/internal/markdown"
)

func main() {
    parser := markdown.NewParser(
        markdown.WithCodeTheme("github"),
    )

    md := []byte(`# Go Tutorial

## Chapter 1

Here's a code example:

\`\`\`go
func main() {
    fmt.Println("Hello")
}
\`\`\`

| Feature | Support |
|---------|---------|
| Tables  | Yes     |
| Code    | Yes     |

- [x] Implemented
- [ ] TODO
`)

    html, headings, err := parser.Parse(md)
    if err != nil {
        panic(err)
    }

    // Print TOC
    for _, h := range headings {
        fmt.Printf("%s# %s\n",
            repeatString("  ", h.Level-1), h.Text)
    }
}

func repeatString(s string, count int) string {
    result := ""
    for i := 0; i < count; i++ {
        result += s
    }
    return result
}

Multi-Document Processing

parser := markdown.NewParser()

documents := [][]byte{
    []byte("# Document 1\nContent..."),
    []byte("# Document 2\nContent..."),
}

for _, doc := range documents {
    html, headings, _ := parser.Parse(doc)
    // Process html and headings
}

Implementation Details

Heading ID Generation

IDs are automatically generated from heading text by:

1. Converting to lowercase
2. Removing special characters
3. Replacing spaces with hyphens
4. Trimming leading/trailing hyphens

Example: "Hello World" → "hello-world"

Heading Collection

The parser uses a custom AST transformer to collect heading information during parsing. This is done in a thread-safe manner using mutexes.

Extensions Architecture

The module uses goldmark's extensibility properly:

• Built-in GFM extensions for standard features
• Custom transformers for heading ID generation
• Cross-reference link resolver for internal references

Thread Safety

• The Parser instance is thread-safe for concurrent parsing operations
• Heading collection uses internal synchronization
• Recommended: Create one Parser instance and reuse it

Performance

Benchmarks on typical documents:

• Simple documents: ~1-2ms
• Complex documents with multiple features: ~5-10ms

For optimal performance:

• Reuse Parser instances
• Parse in parallel for multiple documents
• Consider caching results for frequently parsed content

Error Handling

The Parse method returns detailed errors:

• Syntax errors in the markdown (rare with goldmark)
• Rendering errors (usually configuration issues)

if html, headings, err := parser.Parse(source); err != nil {
    fmt.Printf("Parse error: %v\n", err)
    // Handle error
}

Testing

The module includes comprehensive tests:

• Unit tests for basic functionality
• Integration tests for extension support
• Benchmark tests for performance measurement

Run tests with:

go test ./internal/markdown -v

Future Enhancements

Potential improvements for future versions:

• Custom renderer for PDF-specific formatting
• Custom CSS class insertion for styling
• Metadata/front-matter extraction

Troubleshooting

Issue: Code blocks not highlighted

• Solution: Ensure language tag is specified in code fence (e.g., ```go)

Issue: Table not rendering

• Solution: Ensure table format follows GFM specification with proper separators

Issue: Heading IDs have conflicts

• Solution: The parser automatically appends numbers to duplicate IDs (e.g., "hello-world-1")

Contributing

When extending this module:

1. Maintain backward compatibility
2. Add tests for new features
3. Follow the existing code style
4. Update this documentation
5. Ensure thread safety for concurrent operations

License

This module is part of the mdpress project.