Files
MoneyMgr/moneymgr_mobile/lib/utils/ocr_utils.dart
Pierre HUBERT 23cc189e53
All checks were successful
continuous-integration/drone/push Build is passing
continuous-integration/drone/tag Build is passing
Fix date extraction
2025-07-20 19:17:47 +02:00

87 lines
2.5 KiB
Dart

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({
required Uint8List imgBuff,
required bool extractDates,
}) 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: false,
caseSensitive: false,
);
final currDate = DateTime.now();
DateTime? newest;
for (final match in dateRegexp.allMatches(extractionResult.text)) {
if (match.groupCount < 6) continue;
int year = int.tryParse(match.group(5)!) ?? currDate.year;
try {
final date = DateTime(
year > 99 ? year : (2000 + 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: extractDates && (newest?.isBefore(currDate) ?? false)
? newest!
: currDate,
);
}