Nomad changes

docs/snippets/dart/advanced/chunk_page_mapping.md

@@ -0,0 +1,44 @@
+```dart title="Dart"
+import 'package:kreuzberg/kreuzberg.dart';
+
+Future<void> main() async {
+  final config = ExtractionConfig(
+    useCache: true,
+    enableQualityProcessing: true,
+    forceOcr: false,
+    disableOcr: false,
+    chunking: const ChunkingConfig(
+      maxCharacters: 500,
+      overlap: 50,
+      trim: true,
+      chunkerType: ChunkerType.text,
+      sizing: ChunkSizing.characters(),
+      prependHeadingContext: false,
+    ),
+    pages: const PageConfig(
+      extractPages: true,
+      insertPageMarkers: false,
+      markerFormat: '',
+    ),
+    resultFormat: ResultFormat.unified,
+    outputFormat: OutputFormat.plain(),
+    includeDocumentStructure: false,
+    maxArchiveDepth: 3,
+    useLayoutForMarkdown: false,
+  );
+
+  final result = await KreuzbergBridge.extractFile('document.pdf', null, config);
+  final chunks = result.chunks ?? const [];
+  for (final chunk in chunks) {
+    final first = chunk.metadata.firstPage;
+    final last = chunk.metadata.lastPage;
+    if (first != null && last != null) {
+      final preview = chunk.content.length > 50
+          ? chunk.content.substring(0, 50)
+          : chunk.content;
+      final pageRange = first == last ? 'Page $first' : 'Pages $first-$last';
+      print('Chunk: $preview... ($pageRange)');
+    }
+  }
+}
+```

docs/snippets/dart/advanced/chunking_config.md

@@ -0,0 +1,29 @@
+```dart title="Dart"
+import 'package:kreuzberg/kreuzberg.dart';
+
+Future<void> main() async {
+  final config = ExtractionConfig(
+    useCache: true,
+    enableQualityProcessing: true,
+    forceOcr: false,
+    disableOcr: false,
+    chunking: const ChunkingConfig(
+      maxCharacters: 1000,
+      overlap: 200,
+      trim: true,
+      chunkerType: ChunkerType.markdown,
+      sizing: ChunkSizing.characters(),
+      prependHeadingContext: true,
+    ),
+    resultFormat: ResultFormat.unified,
+    outputFormat: OutputFormat.plain(),
+    includeDocumentStructure: false,
+    maxArchiveDepth: 3,
+    useLayoutForMarkdown: false,
+  );
+
+  final result = await KreuzbergBridge.extractFile('document.pdf', null, config);
+  final chunks = result.chunks ?? const [];
+  print('Chunks: ${chunks.length}');
+}
+```

docs/snippets/dart/advanced/chunking_rag.md

@@ -0,0 +1,50 @@
+```dart title="Dart"
+import 'package:kreuzberg/kreuzberg.dart';
+
+Future<void> main() async {
+  final config = ExtractionConfig(
+    useCache: true,
+    enableQualityProcessing: true,
+    forceOcr: false,
+    disableOcr: false,
+    chunking: const ChunkingConfig(
+      maxCharacters: 500,
+      overlap: 50,
+      trim: true,
+      chunkerType: ChunkerType.text,
+      sizing: ChunkSizing.characters(),
+      prependHeadingContext: false,
+      embedding: EmbeddingConfig(
+        model: EmbeddingModelType.preset(name: 'balanced'),
+        normalize: true,
+        batchSize: 32,
+        showDownloadProgress: false,
+      ),
+    ),
+    resultFormat: ResultFormat.unified,
+    outputFormat: OutputFormat.plain(),
+    includeDocumentStructure: false,
+    maxArchiveDepth: 3,
+    useLayoutForMarkdown: false,
+  );
+
+  final result = await KreuzbergBridge.extractFile('research_paper.pdf', null, config);
+  final chunks = result.chunks ?? const [];
+  for (final chunk in chunks) {
+    final index = chunk.metadata.chunkIndex;
+    final total = chunk.metadata.totalChunks;
+    final start = chunk.metadata.byteStart;
+    final end = chunk.metadata.byteEnd;
+    final preview = chunk.content.length > 100
+        ? chunk.content.substring(0, 100)
+        : chunk.content;
+    print('Chunk ${index + 1}/$total');
+    print('Position: $start-$end');
+    print('Content: $preview...');
+    final embedding = chunk.embedding;
+    if (embedding != null) {
+      print('Embedding: ${embedding.length} dimensions');
+    }
+  }
+}
+```

docs/snippets/dart/advanced/embedding_with_chunking.md

@@ -0,0 +1,35 @@
+```dart title="Dart"
+import 'package:kreuzberg/kreuzberg.dart';
+
+Future<void> main() async {
+  final config = ExtractionConfig(
+    useCache: true,
+    enableQualityProcessing: true,
+    forceOcr: false,
+    disableOcr: false,
+    chunking: const ChunkingConfig(
+      maxCharacters: 1024,
+      overlap: 100,
+      trim: true,
+      chunkerType: ChunkerType.text,
+      sizing: ChunkSizing.characters(),
+      prependHeadingContext: false,
+      embedding: EmbeddingConfig(
+        model: EmbeddingModelType.preset(name: 'balanced'),
+        normalize: true,
+        batchSize: 32,
+        showDownloadProgress: false,
+      ),
+    ),
+    resultFormat: ResultFormat.unified,
+    outputFormat: OutputFormat.plain(),
+    includeDocumentStructure: false,
+    maxArchiveDepth: 3,
+    useLayoutForMarkdown: false,
+  );
+
+  final result = await KreuzbergBridge.extractFile('document.pdf', null, config);
+  final chunks = result.chunks ?? const [];
+  print('Chunks with embeddings: ${chunks.length}');
+}
+```

docs/snippets/dart/advanced/keyword_extraction_config.md

@@ -0,0 +1,29 @@
+```dart title="Dart"
+import 'package:flutter_rust_bridge/flutter_rust_bridge.dart' show Int64List;
+
+import 'package:kreuzberg/kreuzberg.dart';
+
+Future<void> main() async {
+  final config = ExtractionConfig(
+    useCache: true,
+    enableQualityProcessing: true,
+    forceOcr: false,
+    disableOcr: false,
+    keywords: KeywordConfig(
+      algorithm: KeywordAlgorithm.yake,
+      maxKeywords: 10,
+      minScore: 0.3,
+      ngramRange: Int64List.fromList(<int>[1, 3]),
+      language: 'en',
+    ),
+    resultFormat: ResultFormat.unified,
+    outputFormat: OutputFormat.plain(),
+    includeDocumentStructure: false,
+    useLayoutForMarkdown: false,
+    maxArchiveDepth: 3,
+  );
+
+  final result = await KreuzbergBridge.extractFile('document.pdf', null, config);
+  print('Keywords: ${result.extractedKeywords}');
+}
+```

docs/snippets/dart/advanced/keyword_extraction_example.md

@@ -0,0 +1,33 @@
+```dart title="Dart"
+import 'package:flutter_rust_bridge/flutter_rust_bridge.dart' show Int64List;
+
+import 'package:kreuzberg/kreuzberg.dart';
+
+Future<void> main() async {
+  final config = ExtractionConfig(
+    useCache: true,
+    enableQualityProcessing: true,
+    forceOcr: false,
+    disableOcr: false,
+    keywords: KeywordConfig(
+      algorithm: KeywordAlgorithm.yake,
+      maxKeywords: 10,
+      minScore: 0.3,
+      ngramRange: Int64List.fromList(<int>[1, 3]),
+    ),
+    resultFormat: ResultFormat.unified,
+    outputFormat: OutputFormat.plain(),
+    includeDocumentStructure: false,
+    useLayoutForMarkdown: false,
+    maxArchiveDepth: 3,
+  );
+
+  final result = await KreuzbergBridge.extractFile('research_paper.pdf', null, config);
+  final keywords = result.extractedKeywords;
+  if (keywords != null) {
+    for (final keyword in keywords) {
+      print('${keyword.text} (score: ${keyword.score})');
+    }
+  }
+}
+```

docs/snippets/dart/advanced/language_detection_config.md

@@ -0,0 +1,25 @@
+```dart title="Dart"
+import 'package:kreuzberg/kreuzberg.dart';
+
+Future<void> main() async {
+  final config = ExtractionConfig(
+    useCache: true,
+    enableQualityProcessing: true,
+    forceOcr: false,
+    disableOcr: false,
+    languageDetection: const LanguageDetectionConfig(
+      enabled: true,
+      minConfidence: 0.8,
+      detectMultiple: false,
+    ),
+    resultFormat: ResultFormat.unified,
+    outputFormat: OutputFormat.plain(),
+    includeDocumentStructure: false,
+    maxArchiveDepth: 3,
+    useLayoutForMarkdown: false,
+  );
+
+  final result = await KreuzbergBridge.extractFile('document.pdf', null, config);
+  print('Detected languages: ${result.detectedLanguages}');
+}
+```

docs/snippets/dart/advanced/language_detection_multilingual.md

@@ -0,0 +1,25 @@
+```dart title="Dart"
+import 'package:kreuzberg/kreuzberg.dart';
+
+Future<void> main() async {
+  final config = ExtractionConfig(
+    useCache: true,
+    enableQualityProcessing: true,
+    forceOcr: false,
+    disableOcr: false,
+    languageDetection: const LanguageDetectionConfig(
+      enabled: true,
+      minConfidence: 0.8,
+      detectMultiple: true,
+    ),
+    resultFormat: ResultFormat.unified,
+    outputFormat: OutputFormat.plain(),
+    includeDocumentStructure: false,
+    maxArchiveDepth: 3,
+    useLayoutForMarkdown: false,
+  );
+
+  final result = await KreuzbergBridge.extractFile('multilingual_document.pdf', null, config);
+  print('Detected languages: ${result.detectedLanguages}');
+}
+```

docs/snippets/dart/advanced/quality_processing_config.md

@@ -0,0 +1,20 @@
+```dart title="Dart"
+import 'package:kreuzberg/kreuzberg.dart';
+
+Future<void> main() async {
+  final config = ExtractionConfig(
+    useCache: true,
+    enableQualityProcessing: true,
+    forceOcr: false,
+    disableOcr: false,
+    resultFormat: ResultFormat.unified,
+    outputFormat: OutputFormat.plain(),
+    includeDocumentStructure: false,
+    maxArchiveDepth: 3,
+    useLayoutForMarkdown: false,
+  );
+
+  final result = await KreuzbergBridge.extractFile('document.pdf', null, config);
+  print('Quality score: ${result.qualityScore}');
+}
+```

docs/snippets/dart/advanced/quality_processing_example.md

@@ -0,0 +1,30 @@
+```dart title="Dart"
+import 'package:kreuzberg/kreuzberg.dart';
+
+Future<void> main() async {
+  final config = ExtractionConfig(
+    useCache: true,
+    enableQualityProcessing: true,
+    forceOcr: false,
+    disableOcr: false,
+    resultFormat: ResultFormat.unified,
+    outputFormat: OutputFormat.plain(),
+    includeDocumentStructure: false,
+    maxArchiveDepth: 3,
+    useLayoutForMarkdown: false,
+  );
+
+  final result = await KreuzbergBridge.extractFile('scanned_document.pdf', null, config);
+  final score = result.qualityScore;
+  if (score != null) {
+    if (score < 0.5) {
+      print('Warning: Low quality extraction (${score.toStringAsFixed(2)})');
+    } else {
+      print('Quality score: ${score.toStringAsFixed(2)}');
+    }
+  }
+  for (final warning in result.processingWarnings) {
+    print('Warning: $warning');
+  }
+}
+```

docs/snippets/dart/advanced/token_reduction_config.md

@@ -0,0 +1,24 @@
+```dart title="Dart"
+import 'package:kreuzberg/kreuzberg.dart';
+
+Future<void> main() async {
+  final config = ExtractionConfig(
+    useCache: true,
+    enableQualityProcessing: true,
+    forceOcr: false,
+    disableOcr: false,
+    tokenReduction: const TokenReductionOptions(
+      mode: 'moderate',
+      preserveImportantWords: true,
+    ),
+    resultFormat: ResultFormat.unified,
+    outputFormat: OutputFormat.plain(),
+    includeDocumentStructure: false,
+    maxArchiveDepth: 3,
+    useLayoutForMarkdown: false,
+  );
+
+  final result = await KreuzbergBridge.extractFile('document.pdf', null, config);
+  print('Reduced content length: ${result.content.length}');
+}
+```

docs/snippets/dart/advanced/token_reduction_example.md

@@ -0,0 +1,24 @@
+```dart title="Dart"
+import 'package:kreuzberg/kreuzberg.dart';
+
+Future<void> main() async {
+  final config = ExtractionConfig(
+    useCache: true,
+    enableQualityProcessing: true,
+    forceOcr: false,
+    disableOcr: false,
+    tokenReduction: const TokenReductionOptions(
+      mode: 'moderate',
+      preserveImportantWords: true,
+    ),
+    resultFormat: ResultFormat.unified,
+    outputFormat: OutputFormat.plain(),
+    includeDocumentStructure: false,
+    maxArchiveDepth: 3,
+    useLayoutForMarkdown: false,
+  );
+
+  final result = await KreuzbergBridge.extractFile('verbose_document.pdf', null, config);
+  print('Content length after reduction: ${result.content.length}');
+}
+```

docs/snippets/dart/advanced/vector_database_integration.md

@@ -0,0 +1,77 @@
+```dart title="Dart"
+import 'dart:typed_data';
+
+import 'package:kreuzberg/kreuzberg.dart';
+
+class VectorRecord {
+  VectorRecord({
+    required this.id,
+    required this.content,
+    required this.embedding,
+    required this.metadata,
+  });
+
+  final String id;
+  final String content;
+  final Float64List embedding;
+  final Map<String, String> metadata;
+}
+
+Future<List<VectorRecord>> extractAndVectorize(
+  String documentPath,
+  String documentId,
+) async {
+  final config = ExtractionConfig(
+    useCache: true,
+    enableQualityProcessing: true,
+    forceOcr: false,
+    disableOcr: false,
+    chunking: const ChunkingConfig(
+      maxCharacters: 512,
+      overlap: 50,
+      trim: true,
+      chunkerType: ChunkerType.text,
+      sizing: ChunkSizing.characters(),
+      prependHeadingContext: false,
+      embedding: EmbeddingConfig(
+        model: EmbeddingModelType.preset(name: 'balanced'),
+        normalize: true,
+        batchSize: 32,
+        showDownloadProgress: false,
+      ),
+    ),
+    resultFormat: ResultFormat.unified,
+    outputFormat: OutputFormat.plain(),
+    includeDocumentStructure: false,
+    maxArchiveDepth: 3,
+    useLayoutForMarkdown: false,
+  );
+
+  final result = await KreuzbergBridge.extractFile(documentPath, null, config);
+  final records = <VectorRecord>[];
+  final chunks = result.chunks ?? const [];
+  for (var index = 0; index < chunks.length; index++) {
+    final chunk = chunks[index];
+    final embedding = chunk.embedding;
+    if (embedding == null) {
+      continue;
+    }
+    records.add(VectorRecord(
+      id: '${documentId}_chunk_$index',
+      content: chunk.content,
+      embedding: embedding,
+      metadata: <String, String>{
+        'document_id': documentId,
+        'chunk_index': index.toString(),
+        'content_length': chunk.content.length.toString(),
+      },
+    ));
+  }
+  return records;
+}
+
+Future<void> main() async {
+  final records = await extractAndVectorize('document.pdf', 'doc-001');
+  print('Vector records: ${records.length}');
+}
+```