Nomad changes

docs/snippets/swift/metadata/language_detection.md

@@ -0,0 +1,25 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "language_detection": {
+        "enabled": true,
+        "min_confidence": 0.8,
+        "detect_multiple": false
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+if let languages = result.detected_languages() {
+    let langs = languages.map { $0.toString() }
+    print("Detected languages: \(langs)")
+} else {
+    print("No languages detected")
+}
+```

docs/snippets/swift/metadata/language_detection_multilingual.md

@@ -0,0 +1,23 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "language_detection": {
+        "enabled": true,
+        "min_confidence": 0.8,
+        "detect_multiple": true
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("multilingual_document.pdf", nil, config)
+
+if let languages = result.detected_languages() {
+    let langs = languages.map { $0.toString() }
+    print("Detected languages: \(langs)")
+}
+```

docs/snippets/swift/metadata/metadata.md

@@ -0,0 +1,43 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let config = try extractionConfigFromJson("{}")
+let result = try extractFileSync("document.pdf", nil, config)
+
+let metadata = result.metadata()
+
+if let title = metadata.title() {
+    print("Title: \(title.toString())")
+}
+if let subject = metadata.subject() {
+    print("Subject: \(subject.toString())")
+}
+if let language = metadata.language() {
+    print("Language: \(language.toString())")
+}
+if let createdAt = metadata.created_at() {
+    print("Created at: \(createdAt.toString())")
+}
+if let modifiedAt = metadata.modified_at() {
+    print("Modified at: \(modifiedAt.toString())")
+}
+if let createdBy = metadata.created_by() {
+    print("Created by: \(createdBy.toString())")
+}
+if let authors = metadata.authors() {
+    let names = authors.map { $0.toString() }
+    print("Authors: \(names)")
+}
+if let keywords = metadata.keywords() {
+    let words = keywords.map { $0.toString() }
+    print("Keywords: \(words)")
+}
+if let duration = metadata.extraction_duration_ms() {
+    print("Extraction duration (ms): \(duration)")
+}
+if let pages = metadata.pages() {
+    print("Page count: \(pages.total_count())")
+}
+```

docs/snippets/swift/metadata/page_boundaries.md

@@ -0,0 +1,35 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let config = try extractionConfigFromJson("{}")
+let result = try extractFileSync("document.pdf", nil, config)
+
+let content = result.content().toString()
+let utf8 = Array(content.utf8)
+
+guard let pageStructure = result.metadata().pages() else {
+    print("No page structure available")
+    exit(0)
+}
+guard let boundaries = pageStructure.boundaries() else {
+    print("No page boundaries available")
+    exit(0)
+}
+
+for (index, boundary) in boundaries.enumerated() {
+    if index >= 3 { break }
+
+    let byteStart = boundary.byte_start()
+    let byteEnd = boundary.byte_end()
+    let pageBytes = Array(utf8[byteStart..<byteEnd])
+    let pageText = String(bytes: pageBytes, encoding: .utf8) ?? ""
+    let previewEnd = min(100, pageText.count)
+    let preview = String(pageText.prefix(previewEnd))
+
+    print("Page \(boundary.page_number()):")
+    print("  Byte range: \(byteStart)-\(byteEnd)")
+    print("  Preview: \(preview)...")
+}
+```

docs/snippets/swift/metadata/page_tracking_basic.md

@@ -0,0 +1,28 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "pages": {
+        "extract_pages": true
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+if let pages = result.pages() {
+    for page in pages {
+        let pageContent = page.content().toString()
+        print("Page \(page.page_number()):")
+        print("  Content: \(pageContent.count) chars")
+        print("  Tables: \(page.tables().count)")
+        print("  Images: \(page.images().count)")
+    }
+} else {
+    print("No per-page content available")
+}
+```

docs/snippets/swift/metadata/tables.md

@@ -0,0 +1,20 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let config = try extractionConfigFromJson("{}")
+let result = try extractFileSync("document.pdf", nil, config)
+
+let tables = result.tables()
+print("Tables: \(tables.count)")
+
+for (index, table) in tables.enumerated() {
+    print("Table \(index) on page \(table.page_number())")
+    print(table.markdown().toString())
+
+    if let bbox = table.bounding_box() {
+        print("  Bounding box: \(bbox.toString())")
+    }
+}
+```

docs/snippets/swift/metadata/vector_database_integration.md

@@ -0,0 +1,55 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+struct VectorRecord {
+    let id: String
+    let content: String
+    let embedding: [Float]
+    let metadata: [String: String]
+}
+
+let configJson = """
+{
+    "chunking": {
+        "max_characters": 512,
+        "overlap": 50,
+        "embedding": {
+            "model": {"preset": {"name": "balanced"}},
+            "batch_size": 32,
+            "normalize": true
+        }
+    }
+}
+"""
+
+let documentId = "doc_001"
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+var records: [VectorRecord] = []
+
+if let chunks = result.chunks() {
+    for (index, chunk) in chunks.enumerated() {
+        guard let embedding = chunk.embedding() else { continue }
+
+        let content = chunk.content().toString()
+        let vector = embedding.map { $0 }
+
+        var metadata: [String: String] = [:]
+        metadata["document_id"] = documentId
+        metadata["chunk_index"] = String(index)
+        metadata["content_length"] = String(content.count)
+
+        records.append(VectorRecord(
+            id: "\(documentId)_chunk_\(index)",
+            content: content,
+            embedding: vector,
+            metadata: metadata
+        ))
+    }
+}
+
+print("Generated \(records.count) vector records")
+```