14 Commits

Author SHA1 Message Date
Adrian Zürcher
d80764b943 update function convert html 2026-01-06 10:03:20 +01:00
Adrian Zürcher
bd07add64e add function convert html 2026-01-06 09:54:47 +01:00
Adrian Zürcher
7fec8e1101 fix orphan instances 2025-12-29 17:15:52 +01:00
Adrian Zürcher
5b20d4406c fix canelation 2025-12-29 16:48:07 +01:00
Adrian Zürcher
40f683a917 fix hid terminal on windows 2025-12-29 14:59:24 +01:00
Adrian Zürcher
ff460fffa1 add new flags 2025-12-29 12:13:12 +01:00
Adrian Zürcher
70102e991a add new flags 2025-12-29 11:44:07 +01:00
Adrian Zürcher
39fb913f87 fix compile error for other os than windows 2025-12-29 11:20:50 +01:00
Adrian Zürcher
8df18a243f add console hide for windows 2025-12-29 11:07:36 +01:00
Adrian Zürcher
613b74ee55 add new feature to set progress callback 2025-12-25 01:14:03 +01:00
Adrian Zürcher
edb8d7e830 change chrome path priority close #1 2025-12-24 12:15:50 +01:00
Adrian Zürcher
537179af03 fix test model 2025-12-24 12:09:34 +01:00
Adrian Zürcher
48793c65bc add new model 2025-12-24 12:08:43 +01:00
Adrian Zürcher
5985854092 add new instance and provide multiple converting files close #2 2025-12-24 12:06:45 +01:00
6 changed files with 340 additions and 59 deletions

View File

@@ -2,41 +2,39 @@ package converter
import (
"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)
}
func NewConverter(chromePath string) *Converter {
return &Converter{chromePath: chromePath}
}
// NewConverter starts a new converter instance with a chrome headless shell executable
func NewConverter(chromePath string) (*Converter, error) {
var err error
func (c *Converter) Convert(imputFile, outputFile string) error {
chromePath := c.getChromePath()
htmlURL := "file://"
switch runtime.GOOS {
case "windows":
htmlURL += "/"
}
// Convert to absolute path
absPath, err := filepath.Abs(imputFile)
c := &Converter{chromePath: chromePath}
chromePath, err = c.getChromePath()
if err != nil {
return err
return nil, err
}
htmlURL += filepath.ToSlash(absPath)
opts := append(chromedp.DefaultExecAllocatorOptions[:],
chromedp.ExecPath(chromePath),
chromedp.NoSandbox,
@@ -44,17 +42,61 @@ func (c *Converter) Convert(imputFile, outputFile string) error {
chromedp.DisableGPU,
)
allocCtx, cancel := chromedp.NewExecAllocator(context.Background(), opts...)
defer cancel()
ctx, cancel := chromedp.NewContext(allocCtx)
defer cancel()
opts = append(opts, platformOptions())
ctx, cancel = context.WithTimeout(ctx, 60*time.Second)
defer cancel()
c.allocCtx, c.allocCancel = chromedp.NewExecAllocator(context.Background(), opts...)
c.browserCtx, c.browserCancel = chromedp.NewContext(c.allocCtx)
var pdf []byte
err = chromedp.Run(ctx,
chromedp.Navigate(htmlURL),
// 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) Convert(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().
@@ -65,26 +107,106 @@ func (c *Converter) Convert(imputFile, outputFile string) error {
if err != nil {
return err
}
pdf = buf
pdfData = buf
return nil
}),
)
timeoutCancel()
taskCancel()
if err != nil {
return err
}
// Save PDF to file
if err := os.WriteFile(outputFile, pdf, 0644); err != nil {
if err := os.WriteFile(f.Output, pdfData, 0644); err != nil {
return err
}
}
return nil
}
// Convert converts all given input files
func (c *Converter) ConvertHtmls(htmls ...models.Html) error {
for _, h := range htmls {
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
}
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 {
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{
@@ -108,14 +230,9 @@ func (c *Converter) getChromePath() string {
// Check system paths first
for _, path := range candidates {
if _, err := os.Stat(path); err == nil {
return path
return path, nil
}
}
// Fallback: use bundled headless shell
chromeExec := "chrome-headless-shell"
if runtime.GOOS == "windows" {
chromeExec += ".exe"
}
return filepath.Join(c.chromePath, chromeExec)
return "", errors.New("chrome path not found")
}

10
converter/hideOthers.go Normal file
View File

@@ -0,0 +1,10 @@
//go:build !windows
package converter
import "github.com/chromedp/chromedp"
func platformOptions() chromedp.ExecAllocatorOption {
// Returns a no-op option for Mac/Linux
return chromedp.NoSandbox
}

20
converter/hideWindows.go Normal file
View File

@@ -0,0 +1,20 @@
//go:build windows
package converter
import (
"os/exec"
"syscall"
"github.com/chromedp/chromedp"
)
func platformOptions() chromedp.ExecAllocatorOption {
return chromedp.ModifyCmdFunc(func(cmd *exec.Cmd) {
if cmd.SysProcAttr == nil {
cmd.SysProcAttr = &syscall.SysProcAttr{}
}
cmd.SysProcAttr.HideWindow = true
cmd.SysProcAttr.CreationFlags = 0x08000000 // CREATE_NO_WINDOW
})
}

View File

@@ -1,8 +1,26 @@
package html2pdf
import "gitea.tecamino.com/paadi/html2pdf/converter"
import (
"gitea.tecamino.com/paadi/html2pdf/converter"
"gitea.tecamino.com/paadi/html2pdf/models"
)
// Convert converts one .html file to .pdf
func Convert(chromePath, inputFile, outputFile string) error {
c := converter.NewConverter(chromePath)
return c.Convert(inputFile, outputFile)
input := models.File{
Input: inputFile,
Output: outputFile,
}
c, err := converter.NewConverter(chromePath)
if err != nil {
return err
}
defer c.Close()
return c.Convert(input)
}
// NewConverterInstance start new chrome headless shell instance
func NewConverterInstance(chromePath string) (*converter.Converter, error) {
return converter.NewConverter(chromePath)
}

View File

@@ -1,12 +1,119 @@
package html2pdf
import "testing"
import (
"bytes"
"fmt"
"os"
"path/filepath"
"strings"
"testing"
"gitea.tecamino.com/paadi/html2pdf/converter"
"gitea.tecamino.com/paadi/html2pdf/models"
)
func TestConvert(t *testing.T) {
t.Log("start test")
t.Log("start test convert one file")
err := Convert("./assets", "dst/test.html", "out.pdf")
if err != nil {
t.Fatal(err)
}
}
func TestConvertFiles(t *testing.T) {
t.Log("start test convert files")
rootPath := "dst"
files, err := os.ReadDir(rootPath)
if err != nil {
t.Fatal(err)
}
var input []models.File
for _, f := range files {
ext := filepath.Ext(f.Name())
if ext != ".html" {
continue
}
input = append(input, models.File{
Input: filepath.Join(rootPath, f.Name()),
Output: strings.Replace(f.Name(), ext, ".pdf", 1),
})
}
c, err := converter.NewConverter("assets/chrome-headless-shell/win64")
if err != nil {
t.Fatal(err)
}
defer c.Close()
c.SetProgressCallback(func(progress int) {
fmt.Println(progress)
})
if err := c.Convert(input...); err != nil {
t.Fatal(err)
}
t.Log("test successfull")
}
func TestConvertHtml(t *testing.T) {
t.Log("start test ConvertHtml")
// Skip in short mode (useful for CI)
if testing.Short() {
t.Skip("skipping ConvertHtml integration test in short mode")
}
html := []byte(`
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<title>Test PDF</title>
<style>
body { font-family: Arial, sans-serif; }
h1 { color: #333; }
</style>
</head>
<body>
<h1>Hello PDF</h1>
<p>This is a test.</p>
</body>
</html>
`)
tmpDir := t.TempDir()
outputPath := filepath.Join(tmpDir, "test.pdf")
c, err := converter.NewConverter("assets/chrome-headless-shell/win64")
if err != nil {
t.Fatal(err)
}
defer c.Close()
err = c.ConvertHtmls(models.Html{Html: html, Output: outputPath})
if err != nil {
t.Fatalf("ConvertHtml failed: %v", err)
}
// Assert PDF exists
data, err := os.ReadFile(outputPath)
if err != nil {
t.Fatalf("PDF not created: %v", err)
}
// Assert non-empty
if len(data) == 0 {
t.Fatal("PDF file is empty")
}
// Assert valid PDF header
if !bytes.HasPrefix(data, []byte("%PDF-")) {
t.Fatalf("output is not a valid PDF (missing %%PDF- header)")
}
t.Log("ConvertHtml test successful")
}

9
models/file.go Normal file
View File

@@ -0,0 +1,9 @@
package models
type Files []File
// input model for converter
type File struct {
Input string
Output string
}