Nomad changes

docs/snippets/ruby/utils/chunking.md

@@ -0,0 +1,16 @@
+```ruby title="Ruby"
+require 'kreuzberg'
+
+config = Kreuzberg::ExtractionConfig.new(
+  chunking: Kreuzberg::ChunkingConfig.new(
+    max_characters: 1500,
+    overlap: 200,
+    embedding: Kreuzberg::EmbeddingConfig.new(
+      model: Kreuzberg::EmbeddingModelType.new(
+        type: 'preset',
+        name: 'text-embedding-all-minilm-l6-v2'
+      )
+    )
+  )
+)
+```

docs/snippets/ruby/utils/chunking_rag.md

@@ -0,0 +1,26 @@
+```ruby title="Ruby"
+require 'kreuzberg'
+
+config = Kreuzberg::ExtractionConfig.new(
+  chunking: Kreuzberg::ChunkingConfig.new(
+    max_characters: 500,
+    overlap: 50,
+    embedding: Kreuzberg::EmbeddingConfig.new(
+      model: Kreuzberg::EmbeddingModelType.new(
+        type: 'preset',
+        name: 'balanced'
+      ),
+      normalize: true
+    )
+  )
+)
+
+result = Kreuzberg.extract_file_sync('research_paper.pdf', config: config)
+
+result.chunks.each_with_index do |chunk, i|
+  puts "Chunk #{i + 1}/#{result.chunks.length}"
+  puts "Position: #{chunk.metadata[:byte_start]}-#{chunk.metadata[:byte_end]}"
+  puts "Content: #{chunk.content[0..99]}..."
+  puts "Embedding: #{chunk.embedding.length} dimensions" if chunk.embedding
+end
+```

docs/snippets/ruby/utils/embedding_with_chunking.md

@@ -0,0 +1,19 @@
+```ruby title="Ruby"
+require 'kreuzberg'
+
+config = Kreuzberg::ExtractionConfig.new(
+  chunking: Kreuzberg::ChunkingConfig.new(
+    max_characters: 1024,
+    overlap: 100,
+    embedding: Kreuzberg::EmbeddingConfig.new(
+      model: Kreuzberg::EmbeddingModelType.new(
+        type: 'preset',
+        name: 'balanced'
+      ),
+      normalize: true,
+      batch_size: 32,
+      show_download_progress: false
+    )
+  )
+)
+```

docs/snippets/ruby/utils/keyword_extraction_example.md

@@ -0,0 +1,18 @@
+```ruby title="Ruby"
+require 'kreuzberg'
+
+config = Kreuzberg::ExtractionConfig.new(
+  keywords: Kreuzberg::KeywordConfig.new(
+    algorithm: Kreuzberg::KeywordAlgorithm::YAKE,
+    max_keywords: 10,
+    min_score: 0.3
+  )
+)
+
+result = Kreuzberg.extract_file_sync('research_paper.pdf', config: config)
+
+keywords = result.extracted_keywords
+keywords.each do |kw|
+  puts "#{kw['text']}: #{kw['score'].round(3)}"
+end
+```

docs/snippets/ruby/utils/quality_processing_example.md

@@ -0,0 +1,17 @@
+```ruby title="Ruby"
+require 'kreuzberg'
+
+config = Kreuzberg::ExtractionConfig.new(
+  enable_quality_processing: true
+)
+result = Kreuzberg.extract_file_sync('scanned_document.pdf', config: config)
+
+quality_score = result.quality_score
+
+if quality_score < 0.5
+  puts "Warning: Low quality extraction (#{quality_score.round(2)})"
+  puts "Consider re-scanning with higher DPI or adjusting OCR settings"
+else
+  puts "Quality score: #{quality_score.round(2)}"
+end
+```

docs/snippets/ruby/utils/standalone_embed.md

@@ -0,0 +1,15 @@
+```ruby title="Ruby"
+require "kreuzberg"
+
+config = { model: { type: "preset", name: "balanced" }, normalize: true }
+texts = ["Hello, world!", "Kreuzberg is fast"]
+
+# Synchronous
+embeddings = Kreuzberg.embed_sync(texts: texts, config: config)
+puts embeddings.length    # 2
+puts embeddings[0].length # 768
+
+# Async variant (uses same thread, returns when done)
+embeddings = Kreuzberg.embed(texts: texts, config: config)
+puts embeddings[0].length # 768
+```

docs/snippets/ruby/utils/token_reduction.md

@@ -0,0 +1,10 @@
+```ruby title="Ruby"
+require 'kreuzberg'
+
+config = Kreuzberg::ExtractionConfig.new(
+  token_reduction: Kreuzberg::TokenReductionConfig.new(
+    mode: 'moderate',
+    preserve_important_words: true
+  )
+)
+```

docs/snippets/ruby/utils/token_reduction_example.md

@@ -0,0 +1,20 @@
+```ruby title="Ruby"
+require 'kreuzberg'
+
+config = Kreuzberg::ExtractionConfig.new(
+  token_reduction: Kreuzberg::TokenReductionConfig.new(
+    mode: 'moderate',
+    preserve_markdown: true
+  )
+)
+
+result = Kreuzberg.extract_file_sync('verbose_document.pdf', config: config)
+
+# Check reduction statistics in metadata
+original_tokens = result.metadata['original_token_count']
+reduced_tokens = result.metadata['token_count']
+reduction_ratio = result.metadata['token_reduction_ratio']
+
+puts "Reduced from #{original_tokens} to #{reduced_tokens} tokens"
+puts "Reduction: #{reduction_ratio * 100}%"
+```