Dokumenten-Scanner mit OCR-Unterstützung
All checks were successful
Deploy bericht / deploy (push) Successful in 1s

Neuer "Dokument scannen" Button in der Mobile-Upload-Seite:
- Eigene Scanner-View für mehrseitige Dokumente
- Seiten-Vorschau-Leiste mit Thumbnails
- Seiten hinzufügen/löschen vor dem Upload
- Backend-Verarbeitung zu PDF

OCR-Verarbeitung (wenn verfügbar):
- ocrmypdf für durchsuchbares PDF (deu+eng)
- Automatische Rotation und Schräglagekorrektur
- Fallback auf ImageMagick oder TCPDF

Server-Konfiguration (optional für OCR):
  apt install tesseract-ocr tesseract-ocr-deu ocrmypdf

[deploy]

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
Eduard Wisch 2026-04-13 09:59:14 +02:00
parent 04fdd87ffd
commit aa58b5692c
2 changed files with 682 additions and 0 deletions

249
ajax/process_document.php Normal file
View file

@ -0,0 +1,249 @@
<?php
/* Verarbeitet gescannte Dokumentseiten zu einem durchsuchbaren PDF.
* KEIN Dolibarr-Login nötig Authentifizierung über Token in der URL.
* Verwendet ocrmypdf für OCR (optional), Fallback auf einfaches Bild-PDF.
*
* POST: token, pages[] (multipart)
*/
if (!defined('NOLOGIN')) define('NOLOGIN', '1');
if (!defined('NOCSRFCHECK')) define('NOCSRFCHECK', '1');
if (!defined('NOTOKENRENEWAL')) define('NOTOKENRENEWAL', '1');
if (!defined('NOREQUIREMENU')) define('NOREQUIREMENU', '1');
if (!defined('NOREQUIREHTML')) define('NOREQUIREHTML', '1');
if (!defined('NOREQUIREAJAX')) define('NOREQUIREAJAX', '1');
$res = 0;
if (!$res && !empty($_SERVER["CONTEXT_DOCUMENT_ROOT"])) $res = @include $_SERVER["CONTEXT_DOCUMENT_ROOT"]."/main.inc.php";
$tmp = empty($_SERVER['SCRIPT_FILENAME']) ? '' : $_SERVER['SCRIPT_FILENAME']; $tmp2 = realpath(__FILE__); $i = strlen($tmp) - 1; $j = strlen($tmp2) - 1;
while ($i > 0 && $j > 0 && isset($tmp[$i]) && isset($tmp2[$j]) && $tmp[$i] == $tmp2[$j]) { $i--; $j--; }
if (!$res && $i > 0 && file_exists(substr($tmp, 0, ($i + 1))."/main.inc.php")) $res = @include substr($tmp, 0, ($i + 1))."/main.inc.php";
if (!$res && $i > 0 && file_exists(dirname(substr($tmp, 0, ($i + 1)))."/main.inc.php")) $res = @include dirname(substr($tmp, 0, ($i + 1)))."/main.inc.php";
if (!$res && file_exists("../../main.inc.php")) $res = @include "../../main.inc.php";
if (!$res && file_exists("../../../main.inc.php")) $res = @include "../../../main.inc.php";
if (!$res) die("Include of main fails");
require_once DOL_DOCUMENT_ROOT.'/core/lib/files.lib.php';
require_once __DIR__.'/../class/upload_token.class.php';
header('Content-Type: application/json; charset=utf-8');
function fail($msg, $code = 400)
{
http_response_code($code);
echo json_encode(array('success' => false, 'error' => $msg));
exit;
}
// Token validieren
$token = (string) ($_REQUEST['token'] ?? '');
$tok = BerichtUploadToken::fetchValid($db, $token);
if (!$tok) {
fail('Token ungültig oder abgelaufen', 403);
}
// Upload-Ziel ermitteln
$upload_dir = $tok->getUploadDir();
if (!$upload_dir) {
fail('Objekt nicht gefunden', 404);
}
// Seiten prüfen
if (empty($_FILES['pages']) || !is_array($_FILES['pages']['tmp_name'])) {
fail('Keine Seiten hochgeladen');
}
$pages = $_FILES['pages'];
$page_count = count($pages['tmp_name']);
if ($page_count === 0) {
fail('Keine Seiten hochgeladen');
}
// Temp-Verzeichnis für Verarbeitung
$tmpdir = sys_get_temp_dir().'/bericht_scan_'.uniqid();
if (!mkdir($tmpdir, 0755, true)) {
fail('Temp-Verzeichnis konnte nicht erstellt werden');
}
$cleanup = function() use ($tmpdir) {
// Temp-Dateien aufräumen
array_map('unlink', glob($tmpdir.'/*'));
@rmdir($tmpdir);
};
// Seiten speichern und optimieren
$image_files = array();
for ($i = 0; $i < $page_count; $i++) {
if (empty($pages['tmp_name'][$i])) continue;
$ext = strtolower(pathinfo($pages['name'][$i], PATHINFO_EXTENSION));
if (!in_array($ext, array('jpg', 'jpeg', 'png'))) {
$cleanup();
fail('Nur JPG/PNG erlaubt');
}
// Seite nummeriert speichern
$page_file = sprintf('%s/page_%03d.jpg', $tmpdir, $i + 1);
// Mit ImageMagick optimieren (Kontrast, Größe)
$tmp_file = $pages['tmp_name'][$i];
// Prüfen ob convert (ImageMagick) verfügbar ist
$convert_bin = '/usr/bin/convert';
if (!file_exists($convert_bin)) {
$convert_bin = trim(shell_exec('which convert 2>/dev/null'));
}
if ($convert_bin && file_exists($convert_bin)) {
// Bild optimieren: Kontrast erhöhen, Größe begrenzen, als JPG speichern
$cmd = escapeshellcmd($convert_bin).' '
.escapeshellarg($tmp_file)
.' -resize 2400x2400\> ' // Max 2400px, Seitenverhältnis beibehalten
.' -normalize ' // Kontrast optimieren
.' -quality 92 '
.escapeshellarg($page_file).' 2>&1';
exec($cmd, $output, $ret);
if ($ret !== 0 || !file_exists($page_file)) {
// Fallback: einfach kopieren
copy($tmp_file, $page_file);
}
} else {
// Kein ImageMagick: einfach kopieren
copy($tmp_file, $page_file);
}
if (file_exists($page_file)) {
$image_files[] = $page_file;
}
}
if (empty($image_files)) {
$cleanup();
fail('Keine gültigen Seiten verarbeitet');
}
// Ziel-Dateiname
if (!is_dir($upload_dir)) dol_mkdir($upload_dir);
$timestamp = dol_print_date(dol_now(), '%Y%m%d_%H%M%S');
$pdf_filename = 'scan_'.$timestamp.'_'.uniqid().'.pdf';
$pdf_path = $upload_dir.'/'.$pdf_filename;
// PDF erstellen
$pdf_created = false;
// Methode 1: ocrmypdf (bevorzugt - erzeugt durchsuchbares PDF)
$ocrmypdf_bin = getDolGlobalString('BERICHT_OCRMYPDF_BIN', '/usr/bin/ocrmypdf');
if (!file_exists($ocrmypdf_bin)) {
$ocrmypdf_bin = trim(shell_exec('which ocrmypdf 2>/dev/null'));
}
if ($ocrmypdf_bin && file_exists($ocrmypdf_bin)) {
// Erst Bilder zu PDF zusammenfügen mit img2pdf oder convert
$img2pdf_bin = trim(shell_exec('which img2pdf 2>/dev/null'));
$temp_pdf = $tmpdir.'/temp.pdf';
if ($img2pdf_bin && file_exists($img2pdf_bin)) {
// img2pdf ist schneller und verlustfrei
$cmd = escapeshellcmd($img2pdf_bin).' -o '.escapeshellarg($temp_pdf);
foreach ($image_files as $f) {
$cmd .= ' '.escapeshellarg($f);
}
exec($cmd.' 2>&1', $output, $ret);
} elseif ($convert_bin && file_exists($convert_bin)) {
// Fallback: ImageMagick convert
$cmd = escapeshellcmd($convert_bin);
foreach ($image_files as $f) {
$cmd .= ' '.escapeshellarg($f);
}
$cmd .= ' '.escapeshellarg($temp_pdf);
exec($cmd.' 2>&1', $output, $ret);
}
if (file_exists($temp_pdf)) {
// OCR mit ocrmypdf
$lang = getDolGlobalString('BERICHT_OCR_LANG', 'deu+eng');
$cmd = escapeshellcmd($ocrmypdf_bin)
.' -l '.escapeshellarg($lang)
.' --rotate-pages ' // Automatische Rotation
.' --deskew ' // Schräglage korrigieren
.' --clean ' // Hintergrund bereinigen
.' --optimize 1 ' // Leichte Optimierung
.' '.escapeshellarg($temp_pdf)
.' '.escapeshellarg($pdf_path)
.' 2>&1';
exec($cmd, $output, $ret);
if ($ret === 0 && file_exists($pdf_path)) {
$pdf_created = true;
}
}
}
// Methode 2: Fallback mit ImageMagick (kein OCR)
if (!$pdf_created && $convert_bin && file_exists($convert_bin)) {
$cmd = escapeshellcmd($convert_bin);
foreach ($image_files as $f) {
$cmd .= ' '.escapeshellarg($f);
}
$cmd .= ' -quality 92 '.escapeshellarg($pdf_path).' 2>&1';
exec($cmd, $output, $ret);
if ($ret === 0 && file_exists($pdf_path)) {
$pdf_created = true;
}
}
// Methode 3: Fallback mit TCPDF (PHP-only)
if (!$pdf_created) {
// TCPDF sollte in Dolibarr verfügbar sein
if (file_exists(DOL_DOCUMENT_ROOT.'/includes/tecnickcom/tcpdf/tcpdf.php')) {
require_once DOL_DOCUMENT_ROOT.'/includes/tecnickcom/tcpdf/tcpdf.php';
$pdf = new TCPDF('P', 'mm', 'A4', true, 'UTF-8', false);
$pdf->SetCreator('Dolibarr Bericht-Modul');
$pdf->SetAuthor('Dokumenten-Scanner');
$pdf->SetTitle('Gescanntes Dokument');
$pdf->SetMargins(0, 0, 0);
$pdf->SetAutoPageBreak(false);
$pdf->setPrintHeader(false);
$pdf->setPrintFooter(false);
foreach ($image_files as $img) {
$pdf->AddPage();
$size = getimagesize($img);
if ($size) {
$w = $size[0];
$h = $size[1];
// Bild auf A4 skalieren (210x297mm)
$ratio = min(210 / ($w * 0.264583), 297 / ($h * 0.264583));
$pdf->Image($img, 0, 0, $w * 0.264583 * $ratio, $h * 0.264583 * $ratio);
}
}
$pdf->Output($pdf_path, 'F');
if (file_exists($pdf_path)) {
$pdf_created = true;
}
}
}
// Aufräumen
$cleanup();
if (!$pdf_created) {
fail('PDF konnte nicht erstellt werden. Bitte prüfen Sie die Server-Konfiguration (ImageMagick/ocrmypdf).');
}
// Token-Counter erhöhen
$tok->incrementCount();
// Erfolg
echo json_encode(array(
'success' => true,
'filename' => $pdf_filename,
'pages' => count($image_files),
'ocr' => ($ocrmypdf_bin && file_exists($ocrmypdf_bin)) ? true : false,
));

View file

@ -323,6 +323,197 @@ body {
color: #fff; color: #fff;
} }
.btn-save:active { background: #4cae4c; } .btn-save:active { background: #4cae4c; }
/* Dokumenten-Scanner Styles */
.scanner-view {
position: fixed;
inset: 0;
background: #1a1a1f;
z-index: 1000;
display: flex;
flex-direction: column;
}
.scanner-view.hidden { display: none; }
.scanner-header {
display: flex;
justify-content: space-between;
align-items: center;
padding: 12px 16px;
background: #222;
border-bottom: 1px solid #333;
}
.scanner-header h2 {
margin: 0;
font-size: 16px;
color: #5cb85c;
}
.scanner-header .page-count {
font-size: 14px;
color: #888;
}
.scanner-camera-area {
flex: 1;
position: relative;
background: #000;
display: flex;
align-items: center;
justify-content: center;
}
.scanner-camera-area video {
width: 100%;
height: 100%;
object-fit: cover;
}
/* Seiten-Vorschau-Leiste am unteren Rand */
.scanner-pages-strip {
background: #222;
padding: 8px;
display: flex;
gap: 8px;
overflow-x: auto;
min-height: 80px;
align-items: center;
}
.scanner-page-thumb {
width: 60px;
height: 80px;
border-radius: 4px;
object-fit: cover;
border: 2px solid #444;
flex-shrink: 0;
position: relative;
}
.scanner-page-thumb.selected {
border-color: #5cb85c;
}
.scanner-page-wrapper {
position: relative;
flex-shrink: 0;
}
.scanner-page-delete {
position: absolute;
top: -6px;
right: -6px;
width: 20px;
height: 20px;
border-radius: 50%;
background: #d9534f;
color: #fff;
border: none;
font-size: 12px;
cursor: pointer;
display: flex;
align-items: center;
justify-content: center;
}
.scanner-page-number {
position: absolute;
bottom: 2px;
left: 2px;
background: rgba(0,0,0,0.7);
color: #fff;
font-size: 10px;
padding: 1px 4px;
border-radius: 2px;
}
.scanner-controls {
display: flex;
gap: 12px;
padding: 16px;
background: #222;
border-top: 1px solid #333;
}
.scanner-controls button {
flex: 1;
padding: 14px;
font-size: 15px;
font-weight: 600;
border-radius: 8px;
border: none;
cursor: pointer;
}
.scanner-btn-capture {
background: #5cb85c;
color: #fff;
}
.scanner-btn-capture:active { background: #4cae4c; }
.scanner-btn-done {
background: #337ab7;
color: #fff;
}
.scanner-btn-done:active { background: #2868a0; }
.scanner-btn-done:disabled {
background: #444;
color: #888;
cursor: not-allowed;
}
.scanner-btn-cancel {
background: #444;
color: #fff;
flex: 0.5;
}
/* Scanner Vorschau (einzelne Seite) */
.scanner-preview {
position: absolute;
inset: 0;
background: #000;
display: flex;
flex-direction: column;
z-index: 10;
}
.scanner-preview.hidden { display: none; }
.scanner-preview img {
flex: 1;
width: 100%;
object-fit: contain;
}
.scanner-preview-buttons {
display: flex;
gap: 12px;
padding: 16px;
background: #222;
}
.scanner-preview-buttons button {
flex: 1;
padding: 14px;
font-size: 15px;
font-weight: 600;
border-radius: 8px;
border: none;
cursor: pointer;
}
/* Upload-Fortschritt */
.scanner-uploading {
position: absolute;
inset: 0;
background: rgba(0,0,0,0.9);
display: flex;
flex-direction: column;
align-items: center;
justify-content: center;
z-index: 20;
}
.scanner-uploading.hidden { display: none; }
.scanner-uploading .spinner {
width: 48px;
height: 48px;
border: 4px solid #333;
border-top-color: #5cb85c;
border-radius: 50%;
animation: spin 1s linear infinite;
}
@keyframes spin { to { transform: rotate(360deg); } }
.scanner-uploading p {
margin-top: 16px;
color: #fff;
font-size: 14px;
}
</style> </style>
</head> </head>
<body> <body>
@ -336,6 +527,8 @@ body {
<div class="action-buttons"> <div class="action-buttons">
<button class="btn" id="open-camera-btn">Foto aufnehmen</button> <button class="btn" id="open-camera-btn">Foto aufnehmen</button>
<button class="btn" id="open-scanner-btn" style="background: #5cb85c;">📄 Dokument scannen</button>
<label class="btn btn-secondary" for="gallery-input">Aus Galerie wählen</label> <label class="btn btn-secondary" for="gallery-input">Aus Galerie wählen</label>
<input type="file" id="gallery-input" class="hidden-input" accept="image/*" multiple> <input type="file" id="gallery-input" class="hidden-input" accept="image/*" multiple>
</div> </div>
@ -365,6 +558,45 @@ body {
</div> </div>
</div> </div>
<!-- Dokumenten-Scanner Overlay -->
<div id="scanner-view" class="scanner-view hidden">
<div class="scanner-header">
<h2>📄 Dokument scannen</h2>
<span id="scanner-page-count" class="page-count">0 Seiten</span>
</div>
<div class="scanner-camera-area">
<video id="scanner-stream" autoplay playsinline muted></video>
<canvas id="scanner-canvas" style="display:none;"></canvas>
<!-- Vorschau einer gescannten Seite -->
<div id="scanner-preview" class="scanner-preview hidden">
<img id="scanner-preview-image" src="">
<div class="scanner-preview-buttons">
<button id="scanner-preview-discard" class="btn-discard"> Verwerfen</button>
<button id="scanner-preview-add" class="btn-save"> Seite hinzufügen</button>
</div>
</div>
<!-- Upload-Fortschritt -->
<div id="scanner-uploading" class="scanner-uploading hidden">
<div class="spinner"></div>
<p id="scanner-upload-status">Dokument wird verarbeitet…</p>
</div>
</div>
<!-- Seiten-Vorschau-Leiste -->
<div id="scanner-pages-strip" class="scanner-pages-strip">
<!-- Thumbnails werden hier eingefügt -->
</div>
<div class="scanner-controls">
<button id="scanner-cancel" class="scanner-btn-cancel"></button>
<button id="scanner-capture" class="scanner-btn-capture">📷 Seite aufnehmen</button>
<button id="scanner-done" class="scanner-btn-done" disabled>Fertig</button>
</div>
</div>
<div id="uploading-area"></div> <div id="uploading-area"></div>
<div class="uploaded-list"> <div class="uploaded-list">
@ -573,6 +805,207 @@ function updateCameraCounter() {
cameraCounter.textContent = text; cameraCounter.textContent = text;
} }
// === DOKUMENTEN-SCANNER ===
const openScannerBtn = document.getElementById('open-scanner-btn');
const scannerView = document.getElementById('scanner-view');
const scannerStream = document.getElementById('scanner-stream');
const scannerCanvas = document.getElementById('scanner-canvas');
const scannerPreview = document.getElementById('scanner-preview');
const scannerPreviewImage = document.getElementById('scanner-preview-image');
const scannerPreviewDiscard = document.getElementById('scanner-preview-discard');
const scannerPreviewAdd = document.getElementById('scanner-preview-add');
const scannerPagesStrip = document.getElementById('scanner-pages-strip');
const scannerPageCount = document.getElementById('scanner-page-count');
const scannerCapture = document.getElementById('scanner-capture');
const scannerDone = document.getElementById('scanner-done');
const scannerCancel = document.getElementById('scanner-cancel');
const scannerUploading = document.getElementById('scanner-uploading');
const scannerUploadStatus = document.getElementById('scanner-upload-status');
let scannerMediaStream = null;
let scannedPages = []; // Array von Blobs
let currentScanBlob = null;
// Scanner öffnen
openScannerBtn.addEventListener('click', openScannerView);
async function openScannerView() {
try {
const constraints = {
video: {
facingMode: 'environment',
width: { ideal: 2560 },
height: { ideal: 1920 }
},
audio: false
};
scannerMediaStream = await navigator.mediaDevices.getUserMedia(constraints);
scannerStream.srcObject = scannerMediaStream;
scannerView.classList.remove('hidden');
scannedPages = [];
updateScannerUI();
document.body.style.overflow = 'hidden';
} catch (err) {
console.error('Scanner-Fehler:', err);
toast('Kamera nicht verfügbar: ' + err.message, true);
}
}
// Scanner schließen
scannerCancel.addEventListener('click', closeScannerView);
function closeScannerView() {
if (scannerMediaStream) {
scannerMediaStream.getTracks().forEach(track => track.stop());
scannerMediaStream = null;
}
scannerStream.srcObject = null;
scannerView.classList.add('hidden');
scannerPreview.classList.add('hidden');
scannerUploading.classList.add('hidden');
// Alte Seiten-Blobs freigeben
scannedPages.forEach(p => URL.revokeObjectURL(p.url));
scannedPages = [];
document.body.style.overflow = '';
}
// Seite aufnehmen
scannerCapture.addEventListener('click', captureScanPage);
function captureScanPage() {
if (!scannerMediaStream) return;
const video = scannerStream;
const canvas = scannerCanvas;
canvas.width = video.videoWidth;
canvas.height = video.videoHeight;
const ctx = canvas.getContext('2d');
ctx.drawImage(video, 0, 0);
canvas.toBlob((blob) => {
if (blob) {
currentScanBlob = blob;
showScanPreview(blob);
}
}, 'image/jpeg', 0.92);
}
// Vorschau anzeigen
function showScanPreview(blob) {
const url = URL.createObjectURL(blob);
scannerPreviewImage.onload = () => URL.revokeObjectURL(url);
scannerPreviewImage.src = url;
scannerPreview.classList.remove('hidden');
}
// Seite verwerfen
scannerPreviewDiscard.addEventListener('click', () => {
currentScanBlob = null;
scannerPreview.classList.add('hidden');
});
// Seite hinzufügen
scannerPreviewAdd.addEventListener('click', () => {
if (!currentScanBlob) return;
const url = URL.createObjectURL(currentScanBlob);
scannedPages.push({ blob: currentScanBlob, url: url });
currentScanBlob = null;
scannerPreview.classList.add('hidden');
updateScannerUI();
});
// UI aktualisieren
function updateScannerUI() {
// Counter aktualisieren
const count = scannedPages.length;
scannerPageCount.textContent = count === 1 ? '1 Seite' : count + ' Seiten';
scannerDone.disabled = count === 0;
// Thumbnails rendern
scannerPagesStrip.innerHTML = '';
scannedPages.forEach((page, index) => {
const wrapper = document.createElement('div');
wrapper.className = 'scanner-page-wrapper';
const img = document.createElement('img');
img.className = 'scanner-page-thumb';
img.src = page.url;
const numLabel = document.createElement('span');
numLabel.className = 'scanner-page-number';
numLabel.textContent = index + 1;
const delBtn = document.createElement('button');
delBtn.className = 'scanner-page-delete';
delBtn.textContent = '✕';
delBtn.onclick = () => {
URL.revokeObjectURL(page.url);
scannedPages.splice(index, 1);
updateScannerUI();
};
wrapper.appendChild(img);
wrapper.appendChild(numLabel);
wrapper.appendChild(delBtn);
scannerPagesStrip.appendChild(wrapper);
});
}
// Dokument fertigstellen und hochladen
scannerDone.addEventListener('click', finishScanning);
async function finishScanning() {
if (scannedPages.length === 0) return;
scannerUploading.classList.remove('hidden');
scannerUploadStatus.textContent = 'Dokument wird hochgeladen…';
const fd = new FormData();
fd.append('token', TOKEN);
fd.append('action', 'scan_document');
// Alle Seiten hinzufügen
scannedPages.forEach((page, i) => {
fd.append('pages[]', page.blob, 'page_' + (i + 1) + '.jpg');
});
try {
const url = window.location.pathname.replace('mobile_upload.php', 'ajax/process_document.php');
scannerUploadStatus.textContent = 'OCR-Verarbeitung läuft…';
const r = await fetch(url + '?token=' + encodeURIComponent(TOKEN), {
method: 'POST',
body: fd
});
const data = await r.json();
if (data.success) {
uploadedCount++;
uploadedCountEl.textContent = '(' + uploadedCount + ')';
const item = document.createElement('div');
item.className = 'uploaded-item';
item.innerHTML = '<span class="check">✓</span><span class="name">📄 ' + escapeHtml(data.filename) + '</span>';
uploadedList.prepend(item);
toast('Dokument erstellt: ' + scannedPages.length + ' Seiten');
closeScannerView();
} else {
scannerUploading.classList.add('hidden');
toast('Fehler: ' + (data.error || 'unbekannt'), true);
}
} catch (e) {
console.error('Upload-Fehler:', e);
scannerUploading.classList.add('hidden');
toast('Netzwerkfehler', true);
}
}
// === UPLOAD-FUNKTIONEN === // === UPLOAD-FUNKTIONEN ===
async function uploadFile(file) { async function uploadFile(file) {