improve parsing 2

2026-09-02 12:35:41 +02:00
parent e98fb0e7f4
commit 98574af903
9 changed files with 1721 additions and 0 deletions
@@ -0,0 +1,66 @@
+/// Extract positioned OCR elements from ML Kit structured output.
+library;
+
+import 'package:google_mlkit_text_recognition/google_mlkit_text_recognition.dart';
+
+import '../models/ocr_element.dart';
+
+class ReceiptOcr {
+  ReceiptOcr._();
+
+  /// Reads blocks → lines → elements. Falls back to line-level boxes when a
+  /// line has no element children.
+  static OcrDocument fromRecognizedText(
+    RecognizedText recognized, {
+    required double imageWidth,
+    required double imageHeight,
+  }) {
+    final elements = <OcrElement>[];
+    var maxRight = 0.0;
+    var maxBottom = 0.0;
+
+    for (final block in recognized.blocks) {
+      for (final line in block.lines) {
+        if (line.elements.isNotEmpty) {
+          for (final element in line.elements) {
+            final box = element.boundingBox;
+            elements.add(
+              OcrElement(
+                text: element.text,
+                left: box.left,
+                top: box.top,
+                right: box.right,
+                bottom: box.bottom,
+              ),
+            );
+            if (box.right > maxRight) maxRight = box.right;
+            if (box.bottom > maxBottom) maxBottom = box.bottom;
+          }
+        } else {
+          final box = line.boundingBox;
+          elements.add(
+            OcrElement(
+              text: line.text,
+              left: box.left,
+              top: box.top,
+              right: box.right,
+              bottom: box.bottom,
+            ),
+          );
+          if (box.right > maxRight) maxRight = box.right;
+          if (box.bottom > maxBottom) maxBottom = box.bottom;
+        }
+      }
+    }
+
+    final width = imageWidth > 0 ? imageWidth : maxRight;
+    final height = imageHeight > 0 ? imageHeight : maxBottom;
+
+    return OcrDocument(
+      elements: elements,
+      imageWidth: width,
+      imageHeight: height,
+      flattenedText: recognized.text.trim(),
+    );
+  }
+}