html2pdf/converter/converter.go

package converter

import (
	"archive/zip"
	"bytes"
	"context"
	"errors"
	"fmt"
	"os"
	"path/filepath"
	"runtime"
	"strings"
	"time"

	"gitea.tecamino.com/paadi/html2pdf/models"
	"github.com/chromedp/cdproto/page"
	"github.com/chromedp/chromedp"
)

// html to pdf converter structure for
type Converter struct {
	chromePath    string
	allocCtx      context.Context
	allocCancel   context.CancelFunc // Cancels the whole Chrome process manager
	browserCtx    context.Context    // The specific browser instance
	browserCancel context.CancelFunc // Closes the browser
	progress      func(progress int)
}

// NewConverter starts a new converter instance with a chrome headless shell executable
func NewConverter(chromePath string) (*Converter, error) {
	var err error

	c := &Converter{chromePath: chromePath}
	chromePath, err = c.getChromePath()
	if err != nil {
		return nil, err
	}

	opts := append(chromedp.DefaultExecAllocatorOptions[:],
		chromedp.ExecPath(chromePath),
		chromedp.NoSandbox,
		chromedp.Headless,
		chromedp.DisableGPU,
	)

	opts = append(opts, platformOptions())

	c.allocCtx, c.allocCancel = chromedp.NewExecAllocator(context.Background(), opts...)
	c.browserCtx, c.browserCancel = chromedp.NewContext(c.allocCtx)

	// 5. "Warm up" the browser to ensure the executable actually runs
	// This catches "file not found" or permission errors immediately
	err = chromedp.Run(c.browserCtx)
	if err != nil {
		c.Close() // Cleanup if start fails
		return nil, fmt.Errorf("failed to start chrome: %w", err)
	}

	return c, nil
}

func (c *Converter) SetProgressCallback(cb func(progress int)) {
	c.progress = cb
}

// Convert converts all given input files
func (c *Converter) ConvertToPdf(files ...models.File) error {
	for i, f := range files {
		if c.progress != nil {
			c.progress(i + 1)
		}

		if f.Input == "" || filepath.Ext(f.Input) != ".html" {
			return fmt.Errorf("no .html input file path provided: %s", f.Input)
		} else if f.Output == "" || filepath.Ext(f.Output) != ".pdf" {
			return fmt.Errorf("no .pdf output file path provided: %s", f.Output)
		}

		var htmlURL strings.Builder
		htmlURL.WriteString("file://")

		switch runtime.GOOS {
		case "windows":
			htmlURL.WriteString("/")
		}

		// Convert to absolute path
		absPath, err := filepath.Abs(f.Input)
		if err != nil {
			return err
		}

		htmlURL.WriteString(filepath.ToSlash(absPath))

		taskCtx, taskCancel := chromedp.NewContext(c.browserCtx)
		timeoutCtx, timeoutCancel := context.WithTimeout(taskCtx, 60*time.Second)

		var pdfData []byte
		err = chromedp.Run(timeoutCtx,
			chromedp.Navigate(htmlURL.String()),
			chromedp.WaitReady("body", chromedp.ByQuery),
			chromedp.ActionFunc(func(ctx context.Context) error {
				buf, _, err := page.PrintToPDF().
					WithPrintBackground(true).
					WithPaperWidth(8.27).
					WithPaperHeight(11.69).
					Do(ctx)
				if err != nil {
					return err
				}
				pdfData = buf
				return nil
			}),
		)

		timeoutCancel()
		taskCancel()
		if err != nil {
			return err
		}

		// Save PDF to file
		if err := os.WriteFile(f.Output, pdfData, 0644); err != nil {
			return err
		}
	}
	return nil
}

// Convert converts all given input files
func (c *Converter) ConvertHtmlsToPdf(htmls ...models.Html) error {
	for i, h := range htmls {
		if c.progress != nil {
			c.progress(i + 1)
		}
		if len(h.Html) == 0 {
			return fmt.Errorf("no .html input provided")
		} else if h.Output == "" || filepath.Ext(h.Output) != ".pdf" {
			return fmt.Errorf("no .pdf output file path provided: %s", h.Output)
		}

		taskCtx, taskCancel := chromedp.NewContext(c.browserCtx)
		timeoutCtx, timeoutCancel := context.WithTimeout(taskCtx, 60*time.Second)

		var pdfData []byte
		err := chromedp.Run(timeoutCtx,
			// Start with a blank page
			chromedp.Navigate("about:blank"),
			// Inject HTML directly
			chromedp.ActionFunc(func(ctx context.Context) error {
				frameTree, err := page.GetFrameTree().Do(ctx)
				if err != nil {
					return err
				}

				return page.SetDocumentContent(frameTree.Frame.ID, string(h.Html)).Do(ctx)
			}),
			chromedp.WaitReady("body", chromedp.ByQuery),
			chromedp.ActionFunc(func(ctx context.Context) error {
				buf, _, err := page.PrintToPDF().
					WithPrintBackground(true).
					WithPaperWidth(8.27).
					WithPaperHeight(11.69).
					Do(ctx)
				if err != nil {
					return err
				}
				pdfData = buf
				return nil
			}),
		)

		timeoutCancel()
		taskCancel()
		if err != nil {
			return err
		}

		// Save PDF to file
		if err := os.WriteFile(h.Output, pdfData, 0644); err != nil {
			return err
		}
	}
	return nil
}

// Convert converts all given input files
func (c *Converter) ConvertHtmlsToBytes(htmls ...models.Html) ([][]byte, error) {
	var output [][]byte
	for i, h := range htmls {
		if c.progress != nil {
			c.progress(i + 1)
		}
		if len(h.Html) == 0 {
			return nil, fmt.Errorf("no .html input provided")
		} else if h.Output == "" || filepath.Ext(h.Output) != ".pdf" {
			return nil, fmt.Errorf("no .pdf output file path provided: %s", h.Output)
		}

		taskCtx, taskCancel := chromedp.NewContext(c.browserCtx)
		timeoutCtx, timeoutCancel := context.WithTimeout(taskCtx, 60*time.Second)

		var pdfData []byte
		err := chromedp.Run(timeoutCtx,
			// Start with a blank page
			chromedp.Navigate("about:blank"),
			// Inject HTML directly
			chromedp.ActionFunc(func(ctx context.Context) error {
				frameTree, err := page.GetFrameTree().Do(ctx)
				if err != nil {
					return err
				}

				return page.SetDocumentContent(frameTree.Frame.ID, string(h.Html)).Do(ctx)
			}),
			chromedp.WaitReady("body", chromedp.ByQuery),
			chromedp.ActionFunc(func(ctx context.Context) error {
				buf, _, err := page.PrintToPDF().
					WithPrintBackground(true).
					WithPaperWidth(8.27).
					WithPaperHeight(11.69).
					Do(ctx)
				if err != nil {
					return err
				}
				pdfData = buf
				return nil
			}),
		)

		timeoutCancel()
		taskCancel()
		if err != nil {
			return nil, err
		}

		// add to array of bytes
		output = append(output, pdfData)
	}
	return output, nil
}

// Convert html to pdf and return zip as bytes
func (c *Converter) ConvertHtmlsToZip(htmls ...models.Html) ([]byte, error) {
	zipBuf := new(bytes.Buffer)
	zipWriter := zip.NewWriter(zipBuf)

	for i, h := range htmls {
		if c.progress != nil {
			c.progress(i + 1)
		}
		if len(h.Html) == 0 {
			return nil, fmt.Errorf("no .html input provided")
		} else if h.Output == "" || filepath.Ext(h.Output) != ".pdf" {
			return nil, fmt.Errorf("no .pdf output file path provided: %s", h.Output)
		}

		taskCtx, taskCancel := chromedp.NewContext(c.browserCtx)
		timeoutCtx, timeoutCancel := context.WithTimeout(taskCtx, 60*time.Second)

		var pdfData []byte
		err := chromedp.Run(timeoutCtx,
			// Start with a blank page
			chromedp.Navigate("about:blank"),
			// Inject HTML directly
			chromedp.ActionFunc(func(ctx context.Context) error {
				frameTree, err := page.GetFrameTree().Do(ctx)
				if err != nil {
					return err
				}

				return page.SetDocumentContent(frameTree.Frame.ID, string(h.Html)).Do(ctx)
			}),
			chromedp.WaitReady("body", chromedp.ByQuery),
			chromedp.ActionFunc(func(ctx context.Context) error {
				buf, _, err := page.PrintToPDF().
					WithPrintBackground(true).
					WithPaperWidth(8.27).
					WithPaperHeight(11.69).
					Do(ctx)
				if err != nil {
					return err
				}
				pdfData = buf
				return nil
			}),
		)

		timeoutCancel()
		taskCancel()
		if err != nil {
			return nil, err
		}

		// add to zip
		f, err := zipWriter.Create(h.Output)
		if err != nil {
			return nil, err
		}
		f.Write(pdfData)
	}
	zipWriter.Close()

	return zipBuf.Bytes(), nil
}

func (c *Converter) Close() {
	// Close browser first, then allocator
	if c.browserCancel != nil {
		c.browserCancel()
	}
	if c.allocCancel != nil {
		c.allocCancel()
	}
}

// getChromePath checks for system Chrome, else falls back to bundled headless shell
func (c *Converter) getChromePath() (string, error) {

	chromeExec := "chrome-headless-shell"
	if runtime.GOOS == "windows" {
		chromeExec += ".exe"
	}

	path := filepath.Join(c.chromePath, chromeExec)
	if _, err := os.Stat(path); err == nil {
		return path, nil
	}

	// Candidate paths for system Chrome
	candidates := []string{}

	// Fallback:
	switch runtime.GOOS {
	case "windows":
		candidates = []string{
			`C:\Program Files\Google\Chrome\Application\chrome.exe`,
			`C:\Program Files (x86)\Google\Chrome\Application\chrome.exe`,
			`C:\Program Files\Chromium\Application\chrome.exe`,
		}
	case "darwin":
		candidates = []string{
			"/Applications/Google Chrome.app/Contents/MacOS/Google Chrome",
			"/Applications/Chromium.app/Contents/MacOS/Chromium",
		}
	default: // Linux
		candidates = []string{
			"/usr/bin/google-chrome",
			"/usr/bin/chromium-browser",
			"/usr/bin/chromium",
		}
	}

	// Check system paths first
	for _, path := range candidates {
		if _, err := os.Stat(path); err == nil {
			return path, nil
		}
	}

	return "", errors.New("chrome path not found")
}