MarkdownToIHKChemnits/markdown_parser.go

package main

import (
	"bytes"
	"compress/zlib"
	"crypto/sha256"
	"encoding/base64"
	"fmt"
	"github.com/yuin/goldmark"
	"github.com/yuin/goldmark-meta"
	"github.com/yuin/goldmark/ast"
	"github.com/yuin/goldmark/extension"
	extast "github.com/yuin/goldmark/extension/ast"
	"github.com/yuin/goldmark/parser"
	"github.com/yuin/goldmark/text"
	"gopkg.in/yaml.v3"
	"io"
	"net/http"
	"os"
	"path/filepath"
	"strings"
)

func ParseMarkdown(mdPath string) (Config, ast.Node, []byte, error) {
	content, err := os.ReadFile(mdPath)
	if err != nil {
		return Config{}, nil, nil, err
	}

	md := goldmark.New(
		goldmark.WithExtensions(meta.Meta, extension.Table),
	)

	context := parser.NewContext()
	doc := md.Parser().Parse(text.NewReader(content), parser.WithContext(context))

	metaData := meta.Get(context)

	// Convert metaData map to Config struct
	var config Config
	yamlData, _ := yaml.Marshal(metaData)
	err = yaml.Unmarshal(yamlData, &config)
	if err != nil {
		return Config{}, nil, nil, fmt.Errorf("error parsing metadata: %v", err)
	}

	return config, doc, content, nil
}

type parserState struct {
	nextCodeIsAppendix bool
	appendixTitle      string
}

func RenderAST(doc ast.Node, content []byte, r *IHKRenderer) error {
	r.StartFrontMatter()
	state := &parserState{}

	return ast.Walk(doc, func(n ast.Node, entering bool) (ast.WalkStatus, error) {
		if !entering {
			return ast.WalkContinue, nil
		}

		switch node := n.(type) {
		case *ast.Heading:
			if node.Level == 1 && r.numType == NumRoman {
				title := extractText(node, content)
				if title != "Vorwort" && title != "Abkürzungsverzeichnis" {
					r.StartMainBody()
				}
			}
			title := extractText(node, content)
			r.RenderHeader(node.Level, title)
			return ast.WalkSkipChildren, nil
		case *ast.Paragraph:
			text := extractText(node, content)
			lines := strings.Split(text, "\n")
			isMeta := false
			for _, line := range lines {
				line = strings.TrimSpace(line)
				if strings.HasPrefix(line, "@Quelle:") {
					r.AddSource(strings.TrimPrefix(line, "@Quelle:"))
					isMeta = true
				} else if strings.HasPrefix(line, "@Anhang:") {
					r.AddAppendix(strings.TrimPrefix(line, "@Anhang:"))
					isMeta = true
				} else if strings.HasPrefix(line, "@AnhangUML:") {
					state.nextCodeIsAppendix = true
					state.appendixTitle = strings.TrimSpace(strings.TrimPrefix(line, "@AnhangUML:"))
					isMeta = true
				}
			}
			if isMeta {
				return ast.WalkSkipChildren, nil
			}
			r.RenderParagraph(text)
			return ast.WalkSkipChildren, nil
		case *ast.FencedCodeBlock:
			lang := string(node.Language(content))
			code := extractCode(node, content)
			if lang == "mermaid" || lang == "plantuml" || lang == "puml" {
				imgPath, err := RenderDiagramViaKroki(lang, code)
				if err == nil {
					if state.nextCodeIsAppendix {
						r.AddAppendix(state.appendixTitle + " | " + imgPath)
						state.nextCodeIsAppendix = false
					} else {
						r.RenderImage(imgPath, "Diagramm: "+lang)
					}
					return ast.WalkSkipChildren, nil
				}
			}
		case *ast.Image:
			imgPath := string(node.Destination)
			title := string(node.Title)
			r.RenderImage(imgPath, title)
			return ast.WalkSkipChildren, nil
		case *ast.Blockquote:
			// Check if first paragraph starts with "Quelle:"
			first := node.FirstChild()
			if first != nil {
				if para, ok := first.(*ast.Paragraph); ok {
					pText := extractText(para, content)
					if strings.HasPrefix(pText, "Quelle:") || strings.HasPrefix(pText, "Source:") {
						sourceText := strings.TrimPrefix(pText, "Quelle:")
						sourceText = strings.TrimPrefix(sourceText, "Source:")
						r.AddSource(sourceText)
						return ast.WalkSkipChildren, nil
					}
				}
			}
		case *ast.List:
			// Items will be handled by ListItem
		case *ast.ListItem:
			text := extractText(node, content)
			r.RenderListItem(text, true, 0) // Basic bullet point for now
			return ast.WalkSkipChildren, nil
		case *extast.Table:
			var tableData [][]string
			for row := node.FirstChild(); row != nil; row = row.NextSibling() {
				var rowData []string
				for cell := row.FirstChild(); cell != nil; cell = cell.NextSibling() {
					rowData = append(rowData, extractText(cell, content))
				}
				tableData = append(tableData, rowData)
			}
			r.RenderTable(tableData)
			return ast.WalkSkipChildren, nil
		}
		return ast.WalkContinue, nil
	})
}

func extractText(n ast.Node, content []byte) string {
	var textStr string
	for child := n.FirstChild(); child != nil; child = child.NextSibling() {
		if textNode, ok := child.(*ast.Text); ok {
			textStr += string(textNode.Segment.Value(content))
			if textNode.HardLineBreak() || textNode.SoftLineBreak() {
				textStr += "\n"
			}
		} else {
			textStr += extractText(child, content)
		}
	}
	if textStr == "" {
		// Fallback for simple nodes
		return string(n.Text(content))
	}
	return textStr
}

func extractCode(n *ast.FencedCodeBlock, content []byte) string {
	var buf bytes.Buffer
	for i := 0; i < n.Lines().Len(); i++ {
		line := n.Lines().At(i)
		buf.Write(line.Value(content))
	}
	return buf.String()
}

func RenderDiagramViaKroki(lang string, code string) (string, error) {
	if lang == "puml" {
		lang = "plantuml"
	}

	// Kroki encoding: zlib + base64url
	var b bytes.Buffer
	w := zlib.NewWriter(&b)
	w.Write([]byte(code))
	w.Close()

	encoded := base64.URLEncoding.EncodeToString(b.Bytes())
	url := fmt.Sprintf("https://kroki.io/%s/png/%s", lang, encoded)

	// Cache based on hash
	hash := fmt.Sprintf("%x", sha256.Sum256([]byte(code)))
	cachePath := filepath.Join(os.TempDir(), "ihk_cache_"+hash+".png")

	if _, err := os.Stat(cachePath); err == nil {
		return cachePath, nil
	}

	resp, err := http.Get(url)
	if err != nil {
		return "", err
	}
	defer resp.Body.Close()

	if resp.StatusCode != 200 {
		return "", fmt.Errorf("kroki error: %d", resp.StatusCode)
	}

	out, err := os.Create(cachePath)
	if err != nil {
		return "", err
	}
	defer out.Close()

	_, err = io.Copy(out, resp.Body)
	return cachePath, err
}