kreuzberg 4.8.5-aarch64-linux → 4.9.0-aarch64-linux

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 3a8ebc29c703cfb07252f68d86a5dd37f17eb06d79bccce78f82eada2d732e9c
-  data.tar.gz: e711b683c3bdfa37ea1b43bc9e50bc7f21a4e2fca9e4dee4e8c2a2c2664e1bc5
+  metadata.gz: c7a0c15c0ef0421fe0b2fe83a5feba779f2aa18a1be817b5c7b2f52323601531
+  data.tar.gz: 8553c7bcee1466b95ead2688b219d8953ddd941590e55a589f9c7b98317a45a5
 SHA512:
-  metadata.gz: a50806a6cee6edfcf55db3a62d748c293c7b01f20f2744540d520489ca6bbbfe89d65fb3d0132e5e3f723121b835fdc99ac3856559910f227f708e43931070d9
-  data.tar.gz: 26c2f577491b2be1f89e7f49832e308bdc7aabb7453b761a7a72692316a8fce84268d346a3adc545e1308a1b7f3b471d4b671b3f9f4a9ac1ef65dd07139ddb62
+  metadata.gz: 5b3e598e8cdecc91dbe6d6e68772f9fc3ceac7688d87d50b70e5120fb8957dafd4abe47dec6961b3d2046741b3a1bf0d8d618e0f8a36c255e687bbb290d2bd20
+  data.tar.gz: 8e82431165e64e89141d3077d2068a8435a08daf9d4708499121936493995dd874ba561a6c4b791f14be746bb91bd4d15ef444d6163a754ab5b555bc87c1f2eb

data/README.md

@@ -22,7 +22,7 @@
     <img src="https://img.shields.io/maven-central/v/dev.kreuzberg/kreuzberg?label=Java&color=007ec6" alt="Java">
   </a>
   <a href="https://github.com/kreuzberg-dev/kreuzberg/releases">
-    <img src="https://img.shields.io/github/v/tag/kreuzberg-dev/kreuzberg?label=Go&color=007ec6&filter=v4.8.5" alt="Go">
+    <img src="https://img.shields.io/github/v/tag/kreuzberg-dev/kreuzberg?label=Go&color=007ec6&filter=v4.9.0" alt="Go">
   </a>
   <a href="https://www.nuget.org/packages/Kreuzberg/">
     <img src="https://img.shields.io/nuget/v/Kreuzberg?label=C%23&color=007ec6" alt="C#">

data/lib/kreuzberg/errors.rb

@@ -116,5 +116,8 @@ module Kreuzberg
 
     # Raised when embedding fails
     class EmbeddingError < Error; end
+
+    # Raised when an extraction is cancelled via a cancellation token
+    class CancelledError < Error; end
   end
 end

data/lib/kreuzberg/result.rb

@@ -15,7 +15,7 @@ module Kreuzberg
     attr_reader :content, :mime_type, :metadata, :metadata_json, :tables,
                 :detected_languages, :chunks, :images, :pages, :elements, :ocr_elements, :djot_content,
                 :document, :extracted_keywords, :quality_score, :processing_warnings, :annotations,
-                :uris, :children
+                :uris, :children, :structured_output
 
     # @!attribute [r] cells
     #   @return [Array<Array<String>>] Table cells (2D array)
@@ -145,7 +145,7 @@ module Kreuzberg
     #   @return [Array<Image>] Images on this page
     # @!attribute [r] hierarchy
     #   @return [PageHierarchy, nil] Hierarchy information for the page
-    PageContent = Struct.new(:page_number, :content, :tables, :images, :hierarchy, :is_blank) do
+    PageContent = Struct.new(:page_number, :content, :tables, :images, :hierarchy, :is_blank, :layout_regions) do
       def to_h
         {
           page_number: page_number,
@@ -153,7 +153,27 @@ module Kreuzberg
           tables: tables.map(&:to_h),
           images: images.map(&:to_h),
           hierarchy: hierarchy&.to_h,
-          is_blank: is_blank
+          is_blank: is_blank,
+          layout_regions: layout_regions&.map(&:to_h)
+        }
+      end
+    end
+
+    # @!attribute [r] class_name
+    #   @return [String] Layout class name (e.g. "picture", "table", "text")
+    # @!attribute [r] confidence
+    #   @return [Float] Detection confidence score (0.0 to 1.0)
+    # @!attribute [r] bounding_box
+    #   @return [ElementBoundingBox] Bounding box in document coordinate space
+    # @!attribute [r] area_fraction
+    #   @return [Float] Fraction of page area covered (0.0 to 1.0)
+    LayoutRegion = Struct.new(:class_name, :confidence, :bounding_box, :area_fraction) do
+      def to_h
+        {
+          class: class_name,
+          confidence: confidence,
+          bounding_box: bounding_box&.to_h,
+          area_fraction: area_fraction
         }
       end
     end
@@ -342,6 +362,7 @@ module Kreuzberg
       @annotations = parse_annotations(get_value(hash, 'annotations'))
       @uris = parse_uris(get_value(hash, 'uris'))
       @children = parse_children(get_value(hash, 'children'))
+      @structured_output = get_value(hash, 'structured_output')
     end
     # rubocop:enable Metrics/AbcSize, Metrics/MethodLength
 
@@ -369,7 +390,8 @@ module Kreuzberg
         processing_warnings: @processing_warnings.map(&:to_h),
         annotations: @annotations&.map(&:to_h),
         uris: @uris&.map(&:to_h),
-        children: @children&.map(&:to_h)
+        children: @children&.map(&:to_h),
+        structured_output: @structured_output
       }
     end
     # rubocop:enable Metrics/CyclomaticComplexity, Metrics/MethodLength
@@ -568,11 +590,36 @@ module Kreuzberg
           tables: parse_tables(page_hash['tables']),
           images: parse_images(page_hash['images']),
           hierarchy: parse_page_hierarchy(page_hash['hierarchy']),
-          is_blank: page_hash['is_blank']
+          is_blank: page_hash['is_blank'],
+          layout_regions: parse_layout_regions(page_hash['layout_regions'])
         )
       end
     end
 
+    def parse_layout_regions(regions_data)
+      return nil if regions_data.nil?
+
+      regions_data.map do |region_hash|
+        LayoutRegion.new(
+          class_name: region_hash['class'],
+          confidence: region_hash['confidence']&.to_f,
+          bounding_box: parse_element_bounding_box(region_hash['bounding_box']),
+          area_fraction: region_hash['area_fraction']&.to_f
+        )
+      end
+    end
+
+    def parse_element_bounding_box(bounding_box_data)
+      return nil if bounding_box_data.nil?
+
+      ElementBoundingBox.new(
+        x0: bounding_box_data['x0'].to_f,
+        y0: bounding_box_data['y0'].to_f,
+        x1: bounding_box_data['x1'].to_f,
+        y1: bounding_box_data['y1'].to_f
+      )
+    end
+
     def parse_page_hierarchy(hierarchy_data)
       return nil if hierarchy_data.nil?
 

data/lib/kreuzberg/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Kreuzberg
-  VERSION = '4.8.5'
+  VERSION = '4.9.0'
 end

data/sig/kreuzberg.rbs

@@ -596,13 +596,15 @@ module Kreuzberg
       attr_reader batch_size: Integer?
       attr_reader show_download_progress: bool?
       attr_reader cache_dir: String?
+      attr_reader acceleration: Acceleration?
 
       def initialize: (
         ?model: Hash[Symbol, untyped],
         ?normalize: bool?,
         ?batch_size: Integer?,
         ?show_download_progress: bool?,
-        ?cache_dir: String?
+        ?cache_dir: String?,
+        ?acceleration: (Acceleration | Hash[Symbol, untyped])?
       ) -> void
       def to_h: () -> Hash[Symbol, untyped]
     end
@@ -799,8 +801,9 @@ module Kreuzberg
       attr_reader confidence_threshold: Float?
       attr_reader apply_heuristics: bool
       attr_reader table_model: String?
+      attr_reader acceleration: Acceleration?
 
-      def initialize: (?preset: String, ?confidence_threshold: Float?, ?apply_heuristics: bool, ?table_model: String?) -> void
+      def initialize: (?preset: String, ?confidence_threshold: Float?, ?apply_heuristics: bool, ?table_model: String?, ?acceleration: (Acceleration | Hash[Symbol, untyped])?) -> void
       def to_h: () -> Hash[Symbol, untyped]
     end
 
@@ -1009,7 +1012,8 @@ module Kreuzberg
     content: String,
     tables: Array[table_hash],
     images: Array[image_hash],
-    is_blank: bool?
+    is_blank: bool?,
+    layout_regions: Array[untyped]?
   }
 
   type djot_content_hash = {
@@ -1280,6 +1284,17 @@ module Kreuzberg
       def to_h: () -> image_hash
     end
 
+    # Layout detection region on a page (Struct from result.rb)
+    class LayoutRegion
+      attr_reader class_name: String
+      attr_reader confidence: Float
+      attr_reader bounding_box: ElementBoundingBox?
+      attr_reader area_fraction: Float
+
+      def initialize: (class_name: String, confidence: Float, bounding_box: ElementBoundingBox?, area_fraction: Float) -> void
+      def to_h: () -> Hash[Symbol, untyped]
+    end
+
     # Page content with text and extracted elements (Struct from result.rb)
     class PageContent
       attr_reader page_number: Integer
@@ -1288,8 +1303,9 @@ module Kreuzberg
       attr_reader images: Array[Image]?
       attr_reader hierarchy: PageHierarchy?
       attr_reader is_blank: bool?
+      attr_reader layout_regions: Array[LayoutRegion]?
 
-      def initialize: (page_number: Integer, content: String, tables: Array[Table], images: Array[Image]?, hierarchy: PageHierarchy?, is_blank: bool?) -> void
+      def initialize: (page_number: Integer, content: String, tables: Array[Table], images: Array[Image]?, hierarchy: PageHierarchy?, is_blank: bool?, layout_regions: Array[LayoutRegion]?) -> void
       def to_h: () -> Hash[Symbol, untyped]
     end
 
@@ -1548,6 +1564,7 @@ module Kreuzberg
     attr_reader uris: Array[uri_hash]?
     attr_reader children: Array[archive_entry_hash]?
     attr_reader llm_usage: Array[LlmUsage]?
+    attr_reader structured_output: (Hash[String, untyped] | Array[untyped] | Integer | Float | String | bool | nil)
 
     # PDF annotation extracted from a document page (Struct from result.rb)
     class PdfAnnotation
@@ -1581,6 +1598,8 @@ module Kreuzberg
     def parse_element: (Hash[String, untyped] element_hash) -> ElementStruct
     def parse_element_coordinates: (Hash[String, untyped]? coordinates_data) -> ElementBoundingBox?
     def parse_ocr_elements: (Array[ocr_element_hash]? ocr_elements_data) -> Array[OcrElement]?
+    def parse_layout_regions: (Array[untyped]? regions_data) -> Array[LayoutRegion]?
+    def parse_element_bounding_box: (Hash[String, untyped]? bounding_box_data) -> ElementBoundingBox?
     def parse_page_hierarchy: (Hash[String, untyped]? hierarchy_data) -> PageHierarchy?
     def parse_djot_content: (Hash[String, untyped]? djot_data) -> DjotContent?
     def parse_document_structure: (Hash[String, untyped]? document_data) -> DocumentStructure?
@@ -1812,6 +1831,9 @@ module Kreuzberg
 
     class EmbeddingError < Error
     end
+
+    class CancelledError < Error
+    end
   end
 
   # Internal modules (prepended to Kreuzberg singleton)

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: kreuzberg
 version: !ruby/object:Gem::Version
-  version: 4.8.5
+  version: 4.9.0
 platform: aarch64-linux
 authors:
 - Na'aman Hirschfeld
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-04-14 00:00:00.000000000 Z
+date: 2026-04-18 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -86,14 +86,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '3.0'
+        version: '4.0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '3.0'
+        version: '4.0'
 - !ruby/object:Gem::Dependency
   name: rubocop
   requirement: !ruby/object:Gem::Requirement
@@ -142,14 +142,14 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.8'
+        version: '2.0'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '1.8'
+        version: '2.0'
 - !ruby/object:Gem::Dependency
   name: yard
   requirement: !ruby/object:Gem::Requirement