7 Commits

Author SHA1 Message Date
71d32d72ef Can extract date of expenses
All checks were successful
continuous-integration/drone/push Build is passing
2025-07-20 18:07:22 +02:00
28f61a3099 Improve regular expression
All checks were successful
continuous-integration/drone/push Build is passing
2025-07-20 17:44:46 +02:00
f61e3541fb Perform first OCR extraction 2025-07-20 17:25:52 +02:00
fb7891d913 Merge pull request 'Update Rust crate serde_json to 1.0.141' (#63) from renovate/serde_json-1.x into main
Some checks failed
continuous-integration/drone/push Build is failing
2025-07-20 00:15:22 +00:00
d9ede224cf Update Rust crate serde_json to 1.0.141
Some checks failed
continuous-integration/drone/push Build is failing
continuous-integration/drone/pr Build is failing
2025-07-20 00:15:16 +00:00
fc9334b20b Merge pull request 'Update dependency dart to ^3.8.2' (#62) from renovate/dart-3.x into main
Some checks failed
continuous-integration/drone/push Build is failing
2025-07-19 00:18:13 +00:00
c4cbd7ec8b Update dependency dart to ^3.8.2
Some checks failed
renovate/artifacts Artifact file update failure
continuous-integration/drone/push Build is failing
continuous-integration/drone/pr Build is failing
2025-07-19 00:17:10 +00:00
7 changed files with 145 additions and 42 deletions

View File

@@ -3232,9 +3232,9 @@ dependencies = [
[[package]] [[package]]
name = "serde_json" name = "serde_json"
version = "1.0.140" version = "1.0.141"
source = "registry+https://github.com/rust-lang/crates.io-index" source = "registry+https://github.com/rust-lang/crates.io-index"
checksum = "20068b6e96dc6c9bd23e01df8827e6c7e1f2fddd43c21810382803c136b99373" checksum = "30b9eff21ebe718216c6ec64e1d9ac57087aad11efc64e32002bce4a0d4c03d3"
dependencies = [ dependencies = [
"itoa", "itoa",
"memchr", "memchr",

View File

@@ -22,7 +22,7 @@ rust-s3 = "0.36.0-beta.2"
thiserror = "2.0.12" thiserror = "2.0.12"
tokio = "1.45.1" tokio = "1.45.1"
futures-util = "0.3.31" futures-util = "0.3.31"
serde_json = "1.0.140" serde_json = "1.0.141"
light-openid = "1.0.4" light-openid = "1.0.4"
rand = "0.9.1" rand = "0.9.1"
ipnet = { version = "2.11.0", features = ["serde"] } ipnet = { version = "2.11.0", features = ["serde"] }

View File

@@ -1,45 +1,23 @@
import 'dart:io';
import 'dart:typed_data'; import 'dart:typed_data';
import 'package:flutter/material.dart'; import 'package:flutter/material.dart';
import 'package:hooks_riverpod/hooks_riverpod.dart'; import 'package:hooks_riverpod/hooks_riverpod.dart';
import 'package:logging/logging.dart'; import 'package:logging/logging.dart';
import 'package:moneymgr_mobile/services/storage/expenses.dart'; import 'package:moneymgr_mobile/services/storage/expenses.dart';
import 'package:moneymgr_mobile/utils/ocr_utils.dart';
import 'package:moneymgr_mobile/utils/pdf_utils.dart';
import 'package:moneymgr_mobile/widgets/expense_editor.dart'; import 'package:moneymgr_mobile/widgets/expense_editor.dart';
import 'package:riverpod_annotation/riverpod_annotation.dart'; import 'package:riverpod_annotation/riverpod_annotation.dart';
import 'package:scanbot_sdk/scanbot_sdk.dart';
import 'package:scanbot_sdk/scanbot_sdk_ui_v2.dart' hide IconButton, EdgeInsets;
part 'scan_screen.g.dart'; part 'scan_screen.g.dart';
/// Scan a document & return generated PDF as byte file /// Scan a document & return generated PDF as byte file
@riverpod @riverpod
Future<Uint8List?> _scanDocument(Ref ref) async { Future<(Uint8List?, BaseExpenseInfo?)> _scanDocument(Ref ref) async {
var configuration = DocumentScanningFlow( final pdf = await scanDocAsPDF();
appearance: DocumentFlowAppearanceConfiguration( final img = await renderPdf(pdfBytes: pdf);
statusBarMode: StatusBarMode.DARK, final amount = await extractInfoFromBill(img);
), return (pdf, amount);
cleanScanningSession: true,
outputSettings: DocumentScannerOutputSettings(pagesScanLimit: 1),
screens: DocumentScannerScreens(
review: ReviewScreenConfiguration(enabled: false),
),
);
var documentResult = await ScanbotSdkUiV2.startDocumentScanner(configuration);
if (documentResult.status != OperationStatus.OK) {
throw Exception("Scanner failed with status ${documentResult.status}");
}
// Convert result to PDF
var result = await ScanbotSdk.document.createPDFForDocument(
PDFFromDocumentParams(
documentID: documentResult.data!.uuid,
pdfConfiguration: PdfConfiguration(),
),
);
final pdfPath = result.pdfFileUri.replaceFirst("file://", "");
return File(pdfPath).readAsBytes();
} }
class ScanScreen extends HookConsumerWidget { class ScanScreen extends HookConsumerWidget {
@@ -62,12 +40,13 @@ class ScanScreen extends HookConsumerWidget {
return Padding( return Padding(
padding: const EdgeInsets.all(8.0), padding: const EdgeInsets.all(8.0),
child: switch (scanDocProvider) { child: switch (scanDocProvider) {
AsyncData(:final value) when value != null => ExpenseEditor( AsyncData(:final value) when value.$1 != null => ExpenseEditor(
file: value, file: value.$1!,
initialData: value.$2,
onFinished: (expense) async { onFinished: (expense) async {
await expenses.add( await expenses.add(
info: expense, info: expense,
fileContent: value, fileContent: value.$1!,
fileMimeType: "application/pdf", fileMimeType: "application/pdf",
); );
restartScan(); restartScan();
@@ -76,7 +55,7 @@ class ScanScreen extends HookConsumerWidget {
), ),
// No data // No data
AsyncData(:final value) when value == null => ScanErrorScreen( AsyncData(:final value) when value.$1 == null => ScanErrorScreen(
message: "No document scanned!", message: "No document scanned!",
onTryAgain: restartScan, onTryAgain: restartScan,
), ),

View File

@@ -0,0 +1,79 @@
import 'dart:math';
import 'dart:typed_data';
import 'dart:ui' as ui;
import 'package:flutter/material.dart';
import 'package:google_mlkit_text_recognition/google_mlkit_text_recognition.dart';
import 'package:logging/logging.dart';
import 'package:moneymgr_mobile/services/storage/expenses.dart';
/// Attempt to extract information from invoice image
Future<BaseExpenseInfo?> extractInfoFromBill(Uint8List imgBuff) async {
final decodedImage = await decodeImageFromList(imgBuff);
final byteData = await decodedImage.toByteData(
format: ui.ImageByteFormat.rawRgba,
);
final image = InputImage.fromBitmap(
bitmap: byteData!.buffer.asUint8List(),
width: decodedImage.width,
height: decodedImage.height,
);
final textRecognizer = TextRecognizer(script: TextRecognitionScript.latin);
final extractionResult = await textRecognizer.processImage(image);
Logger.root.fine("Expense text: ${extractionResult.text}");
// Check for highestCost amount on invoice
final costRegexp = RegExp(
r'([0-9]+([ ]*(\\.|,)[ ]*[0-9]{1,2}){0,1})([ \\t\\n]*(EUR|eur|€)|E)',
multiLine: true,
caseSensitive: false,
);
var highestCost = 0.0;
for (final match in costRegexp.allMatches(extractionResult.text)) {
if (match.groupCount == 0) continue;
// Process only numeric value
final value = (match.group(1) ?? "").replaceAll(",", ".");
highestCost = max(highestCost, double.tryParse(value) ?? 0.0);
}
// Check for highestCost amount on invoice
final dateRegexp = RegExp(
r'([0-3][0-9])(\/|-)([0-1][0-9])(\/|-)(20[0-9]{2})',
multiLine: true,
caseSensitive: false,
);
final currDate = DateTime.now();
DateTime? newest;
for (final match in dateRegexp.allMatches(extractionResult.text)) {
if (match.groupCount < 5) continue;
try {
final date = DateTime(
int.tryParse(match.group(5)!) ?? currDate.year,
int.tryParse(match.group(3)!) ?? currDate.month,
int.tryParse(match.group(1)!) ?? currDate.day,
);
if (newest == null) {
newest = date;
} else {
newest = DateTime.fromMillisecondsSinceEpoch(
max(newest.millisecondsSinceEpoch, date.millisecondsSinceEpoch),
);
}
} catch (e, s) {
Logger.root.warning("Failed to parse date! $e$s");
}
}
return BaseExpenseInfo(
label: null,
cost: highestCost,
time: newest?.isBefore(currDate) ?? false ? newest! : currDate,
);
}

View File

@@ -6,14 +6,40 @@ import 'package:flutter/material.dart';
import 'package:path/path.dart' as p; import 'package:path/path.dart' as p;
import 'package:path_provider/path_provider.dart'; import 'package:path_provider/path_provider.dart';
import 'package:pdf_image_renderer/pdf_image_renderer.dart'; import 'package:pdf_image_renderer/pdf_image_renderer.dart';
import 'package:scanbot_sdk/scanbot_sdk.dart';
import 'package:scanbot_sdk/scanbot_sdk_ui_v2.dart' hide IconButton, EdgeInsets;
/// Scan document as PDF
Future<Uint8List> scanDocAsPDF() async {
var configuration = DocumentScanningFlow(
appearance: DocumentFlowAppearanceConfiguration(
statusBarMode: StatusBarMode.DARK,
),
cleanScanningSession: true,
outputSettings: DocumentScannerOutputSettings(pagesScanLimit: 1),
screens: DocumentScannerScreens(
review: ReviewScreenConfiguration(enabled: false),
),
);
var documentResult = await ScanbotSdkUiV2.startDocumentScanner(configuration);
if (documentResult.status != OperationStatus.OK) {
throw Exception("Scanner failed with status ${documentResult.status}");
}
// Convert result to PDF
var result = await ScanbotSdk.document.createPDFForDocument(
PDFFromDocumentParams(
documentID: documentResult.data!.uuid,
pdfConfiguration: PdfConfiguration(),
),
);
final pdfPath = result.pdfFileUri.replaceFirst("file://", "");
return File(pdfPath).readAsBytes();
}
/// Render PDF to image bits /// Render PDF to image bits
Future<Uint8List> renderPdf( Future<Uint8List> renderPdf({String? path, Uint8List? pdfBytes}) async {
{
String? path,
Uint8List? pdfBytes,
}) async {
assert(path != null || pdfBytes != null); assert(path != null || pdfBytes != null);
// Create temporary file if required // Create temporary file if required
@@ -62,4 +88,4 @@ Future<Uint8List> renderPdf(
await File(path).delete(); await File(path).delete();
} }
} }
} }

View File

@@ -488,6 +488,22 @@ packages:
url: "https://pub.dev" url: "https://pub.dev"
source: hosted source: hosted
version: "16.0.0" version: "16.0.0"
google_mlkit_commons:
dependency: transitive
description:
name: google_mlkit_commons
sha256: "8f40fbac10685cad4715d11e6a0d86837d9ad7168684dfcad29610282a88e67a"
url: "https://pub.dev"
source: hosted
version: "0.11.0"
google_mlkit_text_recognition:
dependency: "direct main"
description:
name: google_mlkit_text_recognition
sha256: "96173ad4dd7fd06c660e22ac3f9e9f1798a517fe7e48bee68eeec83853224224"
url: "https://pub.dev"
source: hosted
version: "0.15.0"
graphs: graphs:
dependency: transitive dependency: transitive
description: description:

View File

@@ -93,6 +93,9 @@ dependencies:
# PDF renderer # PDF renderer
pdf_image_renderer: ^1.0.1 pdf_image_renderer: ^1.0.1
# Text extraction
google_mlkit_text_recognition: ^0.15.0
dev_dependencies: dev_dependencies:
flutter_test: flutter_test:
sdk: flutter sdk: flutter