zona-go/internal/builder/convert.go

package builder

import (
	"bytes"
	"fmt"
	"io"
	"os"
	"path/filepath"

	"github.com/ficcdaf/zona/internal/util"
	"github.com/gomarkdown/markdown"
	"github.com/gomarkdown/markdown/ast"
	"github.com/gomarkdown/markdown/html"
	"github.com/gomarkdown/markdown/parser"
)

// This function takes a Markdown document and returns an HTML document.
func MdToHTML(md []byte) []byte {
	// create parser with extensions
	extensions := parser.CommonExtensions | parser.AutoHeadingIDs | parser.NoEmptyLineBeforeBlock
	p := parser.NewWithExtensions(extensions)
	doc := p.Parse(md)

	// build HTML renderer
	htmlFlags := html.CommonFlags | html.HrefTargetBlank
	opts := html.RendererOptions{Flags: htmlFlags}
	renderer := newZonaRenderer(opts)

	return markdown.Render(doc, renderer)
}

// PathIsValid checks if a path is valid.
// If requireFile is set, directories are not considered valid.
func PathIsValid(path string, requireFile bool) bool {
	s, err := os.Stat(path)
	if os.IsNotExist(err) {
		return false
	} else if requireFile {
		// fmt.Printf("Directory status: %s\n", strconv.FormatBool(s.IsDir()))
		return !s.IsDir()
	}
	return err == nil
}

func processLink(p string) string {
	// fmt.Println("Processing link...")
	ext := filepath.Ext(p)
	// Only process if it points to an existing, local markdown file
	if ext == ".md" && filepath.IsLocal(p) {
		// fmt.Println("Markdown link detected...")
		return util.ChangeExtension(p, ".html")
	} else {
		return p
	}
}

// renderImage outputs an ast Image node as HTML string.
func renderImage(w io.Writer, node *ast.Image, entering bool, next *ast.Text) {
	// we add image-container div tag
	// here before the opening img tag
	if entering {
		fmt.Fprintf(w, "<div class=\"image-container\">\n")
		fmt.Fprintf(w, `<img src="%s" title="%s"`, node.Destination, node.Title)
		if next != nil && len(next.Literal) > 0 {
			md := []byte(next.Literal)
			html, doc := convertEmbedded(md)
			altText := extractPlainText(md, doc)
			fmt.Fprintf(w, ` alt="%s">`, altText)
			// TODO: render inside a special div?
			// is this necessary since this is all inside image-container anyways?
			fmt.Fprintf(w, `<small>%s</small>`, html)
		} else {
			//
			io.WriteString(w, ">")
		}
	} else {
		// if it's the closing img tag
		// we close the div tag *after*
		fmt.Fprintf(w, `</div>`)
		fmt.Println("Image node not entering??")
	}
}

func renderLink(w io.Writer, l *ast.Link, entering bool) {
	if entering {
		destPath := processLink(string(l.Destination))
		fmt.Fprintf(w, `<a href="%s"`, destPath)
		for _, attr := range html.BlockAttrs(l) {
			fmt.Fprintf(w, ` %s`, attr)
		}
		io.WriteString(w, ">")
	} else {
		io.WriteString(w, "</a>")
	}
}

func htmlRenderHookNoImage(w io.Writer, node ast.Node, entering bool) (ast.WalkStatus, bool) {
	if link, ok := node.(*ast.Link); ok {
		renderLink(w, link, entering)
		return ast.GoToNext, true
	} else if _, ok := node.(*ast.Image); ok {
		// we do not render images
		return ast.GoToNext, true
	}
	return ast.GoToNext, false
}

// htmlRenderHook hooks the HTML renderer and overrides the rendering of certain nodes.
func htmlRenderHook(w io.Writer, node ast.Node, entering bool) (ast.WalkStatus, bool) {
	if link, ok := node.(*ast.Link); ok {
		renderLink(w, link, entering)
		return ast.GoToNext, true
	} else if image, ok := node.(*ast.Image); ok {
		var nextNode *ast.Text
		if entering {
			nextNodes := node.GetChildren()
			fmt.Println("img next node len:", len(nextNodes))
			if len(nextNodes) == 1 {
				if textNode, ok := nextNodes[0].(*ast.Text); ok {
					nextNode = textNode
				}
			}
		}
		renderImage(w, image, entering, nextNode)
		// Skip rendering of `nextNode` explicitly
		if nextNode != nil {
			return ast.SkipChildren, true
		}
		return ast.GoToNext, true
	}
	return ast.GoToNext, false
}

// convertEmbedded renders markdown as HTML
// but does NOT render images
// also returns document AST
func convertEmbedded(md []byte) ([]byte, *ast.Node) {
	p := parser.NewWithExtensions(parser.CommonExtensions)
	doc := p.Parse(md)
	htmlFlags := html.CommonFlags | html.HrefTargetBlank
	opts := html.RendererOptions{Flags: htmlFlags}
	opts.RenderNodeHook = htmlRenderHookNoImage
	r := html.NewRenderer(opts)
	html := markdown.Render(doc, r)
	return html, &doc
}

func newZonaRenderer(opts html.RendererOptions) *html.Renderer {
	opts.RenderNodeHook = htmlRenderHook
	return html.NewRenderer(opts)
}

// ExtractPlainText walks the AST and extracts plain text from the Markdown input.
func extractPlainText(md []byte, doc *ast.Node) string {
	var buffer bytes.Buffer

	// Walk the AST and extract text nodes
	ast.WalkFunc(*doc, func(node ast.Node, entering bool) ast.WalkStatus {
		if textNode, ok := node.(*ast.Text); ok && entering {
			buffer.Write(textNode.Literal) // Append the text content
		}
		return ast.GoToNext
	})

	return buffer.String()
}