Nomad changes

docs/snippets/ruby/metadata/language_detection.md

@@ -0,0 +1,11 @@
+```ruby title="Ruby"
+require 'kreuzberg'
+
+config = Kreuzberg::ExtractionConfig.new(
+  language_detection: Kreuzberg::LanguageDetectionConfig.new(
+    enabled: true,
+    min_confidence: 0.9,
+    detect_multiple: true
+  )
+)
+```

docs/snippets/ruby/metadata/language_detection_multilingual.md

@@ -0,0 +1,16 @@
+```ruby title="Ruby"
+require 'kreuzberg'
+
+config = Kreuzberg::ExtractionConfig.new(
+  language_detection: Kreuzberg::LanguageDetectionConfig.new(
+    enabled: true,
+    min_confidence: 0.8,
+    detect_multiple: true
+  )
+)
+
+result = Kreuzberg.extract_file_sync('multilingual_document.pdf', config: config)
+
+puts "Detected languages: #{result.detected_languages}"
+# Output: ['eng', 'fra', 'deu']
+```

docs/snippets/ruby/metadata/metadata.md

@@ -0,0 +1,74 @@
+```ruby title="Ruby"
+require 'kreuzberg'
+
+result = Kreuzberg.extract_file_sync('document.pdf')
+
+# Metadata is flat — format-specific fields are at the top level
+metadata = result.metadata
+if metadata['page_count']
+  puts "Pages: #{metadata['page_count']}"
+end
+if metadata['title']
+  puts "Title: #{metadata['title']}"
+end
+if metadata['authors']
+  puts "Authors: #{metadata['authors'].join(', ')}"
+end
+
+# Access HTML metadata
+html_result = Kreuzberg.extract_file_sync('page.html')
+metadata = html_result.metadata
+if metadata['title']
+  puts "Title: #{metadata['title']}"
+end
+if metadata['description']
+  puts "Description: #{metadata['description']}"
+end
+
+# Access keywords as array
+if metadata['keywords']
+  puts "Keywords: #{metadata['keywords'].join(', ')}"
+end
+
+# Access canonical URL (renamed from canonical)
+puts "Canonical URL: #{metadata['canonical_url']}" if metadata['canonical_url']
+
+# Access Open Graph fields from map
+open_graph = metadata['open_graph'] || {}
+puts "Open Graph Image: #{open_graph['image']}" if open_graph['image']
+puts "Open Graph Title: #{open_graph['title']}" if open_graph['title']
+puts "Open Graph Type: #{open_graph['type']}" if open_graph['type']
+
+# Access Twitter Card fields from map
+twitter_card = metadata['twitter_card'] || {}
+puts "Twitter Card Type: #{twitter_card['card']}" if twitter_card['card']
+puts "Twitter Creator: #{twitter_card['creator']}" if twitter_card['creator']
+
+# Access new fields
+puts "Language: #{metadata['language']}" if metadata['language']
+puts "Text Direction: #{metadata['text_direction']}" if metadata['text_direction']
+
+# Access headers
+if metadata['headers']
+  puts "Headers: #{metadata['headers'].map { |h| h['text'] }.join(', ')}"
+end
+
+# Access links
+if metadata['links']
+  metadata['links'].each do |link|
+    puts "Link: #{link['href']} (#{link['text']})"
+  end
+end
+
+# Access images
+if metadata['images']
+  metadata['images'].each do |image|
+    puts "Image: #{image['src']}"
+  end
+end
+
+# Access structured data
+if metadata['structured_data']
+  puts "Structured data items: #{metadata['structured_data'].length}"
+end
+```

docs/snippets/ruby/metadata/page_boundaries.md

@@ -0,0 +1,18 @@
+```ruby title="Ruby"
+require 'kreuzberg'
+
+result = Kreuzberg.extract_file_sync('document.pdf')
+
+if result.metadata.pages&.boundaries
+  content_bytes = result.content.bytes
+
+  result.metadata.pages.boundaries.take(3).each do |boundary|
+    page_bytes = content_bytes[boundary.byte_start...boundary.byte_end]
+    page_text = page_bytes.pack('C*').force_encoding('UTF-8')
+
+    puts "Page #{boundary.page_number}:"
+    puts "  Byte range: #{boundary.byte_start}-#{boundary.byte_end}"
+    puts "  Preview: #{page_text[0..100]}..."
+  end
+end
+```

docs/snippets/ruby/metadata/page_tracking_basic.md

@@ -0,0 +1,16 @@
+Require 'Kreuzberg'
+
+Config = Kreuzberg::ExtractionConfig.new(
+pages: Kreuzberg::PageConfig.new(
+extract_pages: true
+)
+)
+
+Result = Kreuzberg.extract_file_sync("document.pdf", config: config)
+
+Result.pages&.each do |page|
+puts "Page #{page.page_number}:"
+puts " Content: #{page.content.length} chars"
+puts " Tables: #{page.tables.length}"
+puts " Images: #{page.images.length}"
+end

docs/snippets/ruby/metadata/tables.md

@@ -0,0 +1,16 @@
+```ruby title="Ruby"
+require 'kreuzberg'
+
+result = Kreuzberg.extract_file_sync('document.pdf')
+
+# Iterate over tables
+result.tables.each do |table|
+  puts "Table with #{table['cells'].length} rows"
+  puts table['markdown']  # Markdown representation
+
+  # Access cells
+  table['cells'].each do |row|
+    puts row
+  end
+end
+```

docs/snippets/ruby/metadata/vector_database_integration.md

@@ -0,0 +1,26 @@
+```ruby title="Ruby"
+require 'kreuzberg'
+
+config = Kreuzberg::ExtractionConfig.new(
+  chunking: Kreuzberg::ChunkingConfig.new(
+    max_characters: 512,
+    overlap: 50,
+    embedding: Kreuzberg::EmbeddingConfig.new(
+      model: Kreuzberg::EmbeddingModelType.new(
+        type: 'preset',
+        name: 'balanced'
+      ),
+      normalize: true
+    )
+  )
+)
+
+result = Kreuzberg.extract_file_sync('document.pdf', config: config)
+
+result.chunks.each_with_index do |chunk, i|
+  if chunk.embedding
+    puts "Chunk #{i}: #{chunk.embedding.length} dimensions"
+    # Store in vector database
+  end
+end
+```