b1tsblog/internal/article/load.go

package article

import (
	"bytes"
	"encoding/json"
	"fmt"
	"html/template"
	"io/fs"
	"os"
	"path/filepath"
	"sort"
	"strings"
	"time"

	md "github.com/gomarkdown/markdown"
	"github.com/gomarkdown/markdown/html"
)

type LoaderOptions struct {
	// TrustedHTML erlaubt rohe .html-Inhalte aus deinem Content-Repository.
	// Achtung: Nur aktivieren, wenn du dem Repository vertraust.
	TrustedHTML bool
}

type frontMatter struct {
	Title       string `json:"title"`
	Date        string `json:"date"`
	Slug        string `json:"slug"`
	Cover       string `json:"cover"`
	Description string `json:"description"`
	Format      string `json:"format"`
}

func LoadDir(root string, opts LoaderOptions) ([]Article, error) {
	var out []Article
	seen := make(map[string]bool)

	err := filepath.WalkDir(root, func(path string, d fs.DirEntry, walkErr error) error {
		if walkErr != nil {
			return walkErr
		}
		if d.IsDir() {
			return nil
		}

		ext := strings.ToLower(filepath.Ext(path))
		if ext != ".md" && ext != ".markdown" && ext != ".html" && ext != ".htm" {
			return nil
		}

		meta, body, format, err := readContentFile(path, opts)
		if err != nil {
			return err
		}

		if meta.Slug == "" {
			meta.Slug = strings.TrimSuffix(filepath.Base(path), ext)
		}
		if seen[meta.Slug] {
			return fmt.Errorf("%s: duplicate slug %q", path, meta.Slug)
		}
		seen[meta.Slug] = true

		date, err := time.Parse("2006-01-02", meta.Date)
		if err != nil {
			return fmt.Errorf("%s: parse date: %w", path, err)
		}

		out = append(out, Article{
			Title:       meta.Title,
			Slug:        meta.Slug,
			Date:        date,
			Cover:       meta.Cover,
			Description: meta.Description,
			Body:        body,
			Format:      format,
			SourcePath:  path,
		})
		return nil
	})
	if err != nil {
		return nil, err
	}

	sort.Slice(out, func(i, j int) bool { return out[i].Date.After(out[j].Date) })
	return out, nil
}

func LoadStatic(dir string, opts LoaderOptions) (map[string]StaticPage, error) {
	pages := make(map[string]StaticPage)

	entries, err := os.ReadDir(dir)
	if err != nil {
		return nil, err
	}

	for _, e := range entries {
		if e.IsDir() {
			continue
		}
		ext := strings.ToLower(filepath.Ext(e.Name()))
		if ext != ".md" && ext != ".markdown" && ext != ".html" && ext != ".htm" {
			continue
		}

		path := filepath.Join(dir, e.Name())
		meta, body, format, err := readContentFile(path, opts)
		if err != nil {
			return nil, err
		}

		if meta.Slug == "" {
			meta.Slug = strings.TrimSuffix(e.Name(), ext)
		}
		if _, dup := pages[meta.Slug]; dup {
			return nil, fmt.Errorf("%s: duplicate slug %q", path, meta.Slug)
		}

		pages[meta.Slug] = StaticPage{
			Title:       meta.Title,
			Slug:        meta.Slug,
			Description: meta.Description,
			Body:        body,
			Format:      format,
			SourcePath:  path,
		}
	}

	return pages, nil
}

func readContentFile(path string, opts LoaderOptions) (frontMatter, template.HTML, Format, error) {
	raw, err := os.ReadFile(path)
	if err != nil {
		return frontMatter{}, "", "", fmt.Errorf("read %s: %w", path, err)
	}

	parts := bytes.SplitN(raw, []byte("\n"), 2)
	if len(parts) < 2 {
		return frontMatter{}, "", "", fmt.Errorf("%s: missing front-matter", path)
	}

	headerJSON, err := extractJSONHeader(parts[0])
	if err != nil {
		return frontMatter{}, "", "", fmt.Errorf("%s: %w", path, err)
	}

	var meta frontMatter
	if err := json.Unmarshal(headerJSON, &meta); err != nil {
		return frontMatter{}, "", "", fmt.Errorf("%s: %w", path, err)
	}

	ext := strings.ToLower(filepath.Ext(path))
	format := detectFormat(ext, meta.Format)
	body := renderBody(parts[1], format, opts)

	return meta, body, format, nil
}

func detectFormat(ext, explicit string) Format {
	switch strings.ToLower(strings.TrimSpace(explicit)) {
	case "html":
		return FormatHTML
	case "md", "markdown":
		return FormatMarkdown
	}

	switch ext {
	case ".html", ".htm":
		return FormatHTML
	default:
		return FormatMarkdown
	}
}

func renderBody(body []byte, format Format, opts LoaderOptions) template.HTML {
	if format == FormatMarkdown {
		renderer := html.NewRenderer(html.RendererOptions{
			Flags: html.CommonFlags | html.HrefTargetBlank,
		})
		return template.HTML(md.ToHTML(body, nil, renderer))
	}

	if opts.TrustedHTML {
		return template.HTML(body)
	}

	// Sicherheits-Fallback: HTML-Dateien werden escaped, solange BLOG_TRUSTED_HTML=false ist.
	return template.HTML(template.HTMLEscapeString(string(body)))
}

func extractJSONHeader(line []byte) ([]byte, error) {
	s := strings.TrimSpace(string(line))
	s = strings.TrimPrefix(s, "<!--")
	if idx := strings.Index(s, "-->"); idx != -1 {
		s = s[:idx]
	}
	s = strings.TrimSpace(s)
	if !strings.HasPrefix(s, "{") {
		return nil, fmt.Errorf("no JSON front-matter found")
	}
	return []byte(s), nil
}