Nomad changes

docs/snippets/swift/config/advanced_config.md

@@ -0,0 +1,59 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+// Build a fully-featured `ExtractionConfig` via JSON. ExtractionConfig has
+// 30+ fields, so JSON is the ergonomic path for non-trivial configs.
+let configJson = """
+{
+    "use_cache": true,
+    "enable_quality_processing": true,
+    "ocr": {
+        "backend": "tesseract",
+        "language": "eng"
+    },
+    "chunking": {
+        "max_characters": 1000,
+        "overlap": 200,
+        "embedding": {
+            "model": {"preset": {"name": "balanced"}},
+            "batch_size": 32,
+            "normalize": true,
+            "show_download_progress": false
+        }
+    },
+    "language_detection": {
+        "enabled": true,
+        "min_confidence": 0.8,
+        "detect_multiple": false
+    },
+    "keywords": {
+        "algorithm": "yake",
+        "max_keywords": 10,
+        "min_score": 0.1,
+        "ngram_range": [1, 3],
+        "language": "en"
+    },
+    "token_reduction": {
+        "mode": "moderate",
+        "preserve_important_words": true
+    },
+    "postprocessor": {
+        "enabled": true
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+print("Content: \(result.content().toString())")
+if let languages = result.detected_languages() {
+    let langs = languages.map { $0.toString() }
+    print("Languages: \(langs)")
+}
+if let chunks = result.chunks() {
+    print("Chunks: \(chunks.count)")
+}
+```

docs/snippets/swift/config/chunking_config.md

@@ -0,0 +1,27 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "chunking": {
+        "max_characters": 1000,
+        "overlap": 100,
+        "chunker_type": "markdown",
+        "prepend_heading_context": true
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.md", nil, config)
+
+if let chunks = result.chunks() {
+    print("Chunks: \(chunks.count)")
+    for chunk in chunks {
+        let content = chunk.content().toString()
+        print("Length: \(content.count)")
+    }
+}
+```

docs/snippets/swift/config/config_basic.md

@@ -0,0 +1,17 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "use_cache": true,
+    "enable_quality_processing": true
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+print(result.content().toString())
+```

docs/snippets/swift/config/config_discover.md

@@ -0,0 +1,21 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+// Swift bindings build configs via JSON. To honor an on-disk
+// `kreuzberg.{toml,yaml,json}`, load the file and pass its JSON
+// representation to `extractionConfigFromJson`. Unknown formats
+// can be normalized to JSON on the caller side.
+let configJson: String
+if let data = try? Data(contentsOf: URL(fileURLWithPath: "kreuzberg.json")),
+   let text = String(data: data, encoding: .utf8) {
+    configJson = text
+} else {
+    configJson = "{}"
+}
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+print(result.content().toString())
+```

docs/snippets/swift/config/config_ocr.md

@@ -0,0 +1,20 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "ocr": {
+        "backend": "tesseract",
+        "language": "eng"
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("scanned.pdf", nil, config)
+
+print("Content length: \(result.content().toString().count)")
+print("Tables detected: \(result.tables().count)")
+```

docs/snippets/swift/config/config_programmatic.md

@@ -0,0 +1,28 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "use_cache": true,
+    "enable_quality_processing": true,
+    "ocr": {
+        "backend": "tesseract",
+        "language": "eng+deu",
+        "tesseract_config": {
+            "psm": 6
+        }
+    },
+    "chunking": {
+        "max_characters": 1000,
+        "overlap": 200
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+print("Content length: \(result.content().toString().count)")
+```

docs/snippets/swift/config/document_structure_config.md

@@ -0,0 +1,18 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "include_document_structure": true
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+if let document = result.document() {
+    print("Document nodes: \(document.nodes().count)")
+}
+```

docs/snippets/swift/config/element_based_output.md

@@ -0,0 +1,22 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "result_format": "element_based"
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+if let elements = result.elements() {
+    print("Elements: \(elements.count)")
+    for element in elements {
+        print("Type: \(element.element_type().toString())")
+        print("Text: \(element.text().toString().prefix(100))")
+    }
+}
+```

docs/snippets/swift/config/embedding_config.md

@@ -0,0 +1,27 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "chunking": {
+        "max_characters": 1000,
+        "overlap": 200,
+        "embedding": {
+            "model": {"preset": {"name": "balanced"}},
+            "batch_size": 16,
+            "normalize": true,
+            "show_download_progress": true
+        }
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+if let chunks = result.chunks() {
+    print("Chunks with embeddings: \(chunks.count)")
+}
+```

docs/snippets/swift/config/html_output.md

@@ -0,0 +1,19 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "output_format": "html",
+    "html_output": {
+        "theme": "github"
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+print(result.content().toString()) // HTML with kb-* classes
+```

docs/snippets/swift/config/keyword_extraction_config.md

@@ -0,0 +1,23 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "keywords": {
+        "algorithm": "yake",
+        "max_keywords": 10,
+        "min_score": 0.1,
+        "ngram_range": [1, 3],
+        "language": "en"
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+print("Keywords extracted from document")
+print("Content length: \(result.content().toString().count)")
+```

docs/snippets/swift/config/language_detection_config.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("document.pdf", nil, config)
+
+if let languages = result.detected_languages() {
+    let langs = languages.map { $0.toString() }
+    print("Detected languages: \(langs)")
+}
+```

docs/snippets/swift/config/ocr_dpi_config.md

@@ -0,0 +1,23 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "images": {
+        "extract_images": true,
+        "target_dpi": 300,
+        "max_image_dimension": 4096,
+        "auto_adjust_dpi": true,
+        "min_dpi": 150,
+        "max_dpi": 600
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+print("Content length: \(result.content().toString().count)")
+```

docs/snippets/swift/config/pdf_config.md

@@ -0,0 +1,20 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "pdf_options": {
+        "extract_images": true,
+        "passwords": ["password123"],
+        "extract_metadata": true
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("encrypted.pdf", nil, config)
+
+print("Content length: \(result.content().toString().count)")
+```

docs/snippets/swift/config/pdf_hierarchy_config.md

@@ -0,0 +1,24 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "pdf_options": {
+        "hierarchy": {
+            "enabled": true,
+            "detection_threshold": 0.75,
+            "ocr_coverage_threshold": 0.8,
+            "min_level": 1,
+            "max_level": 5
+        }
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+print("Content length: \(result.content().toString().count)")
+```

docs/snippets/swift/config/postprocessor_config.md

@@ -0,0 +1,19 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "postprocessor": {
+        "enabled": true,
+        "enabled_processors": ["whitespace_normalizer", "unicode_normalizer"]
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+print("Processed content: \(result.content().toString())")
+```

docs/snippets/swift/config/quality_processing_config.md

@@ -0,0 +1,18 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "enable_quality_processing": true,
+    "use_cache": true
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+print("Content length: \(result.content().toString().count)")
+print("Tables: \(result.tables().count)")
+```

docs/snippets/swift/config/tesseract_config.md

@@ -0,0 +1,23 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "ocr": {
+        "backend": "tesseract",
+        "language": "eng+deu",
+        "tesseract_config": {
+            "psm": 6,
+            "oem": 3
+        }
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("scanned.pdf", nil, config)
+
+print("OCR text: \(result.content().toString())")
+```

docs/snippets/swift/config/token_reduction_config.md

@@ -0,0 +1,19 @@
+```swift title="Swift"
+import Foundation
+import Kreuzberg
+import RustBridge
+
+let configJson = """
+{
+    "token_reduction": {
+        "mode": "moderate",
+        "preserve_important_words": true
+    }
+}
+"""
+
+let config = try extractionConfigFromJson(configJson)
+let result = try extractFileSync("document.pdf", nil, config)
+
+print("Reduced content length: \(result.content().toString().count)")
+```