kreuzberg 4.2.3 → 4.2.5

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 2c6fc44b151014f7e56c82bd191f55244a4294a259b24b95fc494dba6f8eaba6
-  data.tar.gz: 6e40a732814ff3e2a164e718cdb1c7a6ae838b2b2210a66b232f8675c7f79a80
+  metadata.gz: 800ea3cd581a353a00fd4c67025b004ad40a15cc5dc656d04e05c60544387b53
+  data.tar.gz: b9939af20686f24d25b983946c6c024da2def226c3a19ae46f60182981930ddd
 SHA512:
-  metadata.gz: f9c3a45f31c3ad9e3857872d8705b397b40c4317844ef421f4da4c2918e57411f5a626df4f6706d7db4916f33b8644c736e7b41508b398fd0197f1a87170fa3c
-  data.tar.gz: 8b05a75be261dbe583c4873d9d21079efff97d6c9c0340bbd8a73a43c9d15955431f4de20cd8b4a8b7956872f52e4467c253f5da03177a1e7d3b6a10d202b59d
+  metadata.gz: ad5fc359c85e886364cca94497635b8659d363ad2732e7eb5f67d1bc55951d413a6cec201bd16d76c1b826d982881a9f8b028c5d5608be63c30528d83ccafb13
+  data.tar.gz: 23d95fb583c36dd9c1bb060a4f5f43bec540165853ac2007903941463d12fa83d10e805f2cd9b4f5859eeda4465a2ffe3e289e7d884f313867b3611e445fb7f9

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    kreuzberg (4.2.3)
+    kreuzberg (4.2.5)
 
 GEM
   remote: https://rubygems.org/
@@ -97,7 +97,7 @@ GEM
     rspec-mocks (3.13.7)
       diff-lcs (>= 1.2.0, < 2.0)
       rspec-support (~> 3.13.0)
-    rspec-support (3.13.6)
+    rspec-support (3.13.7)
     rubocop (1.84.0)
       json (~> 2.3)
       language_server-protocol (~> 3.17.0.2)
@@ -121,7 +121,7 @@ GEM
       rubocop (~> 1.81)
     ruby-progressbar (1.13.0)
     securerandom (0.4.1)
-    sorbet-runtime (0.6.12903)
+    sorbet-runtime (0.6.12908)
     steep (1.10.0)
       activesupport (>= 5.1)
       concurrent-ruby (>= 1.1.10)
@@ -207,7 +207,7 @@ CHECKSUMS
   i18n (1.14.8) sha256=285778639134865c5e0f6269e0b818256017e8cde89993fdfcbfb64d088824a5
   io-console (0.8.2) sha256=d6e3ae7a7cc7574f4b8893b4fca2162e57a825b223a177b7afa236c5ef9814cc
   json (2.18.0) sha256=b10506aee4183f5cf49e0efc48073d7b75843ce3782c68dbeb763351c08fd505
-  kreuzberg (4.2.3)
+  kreuzberg (4.2.5)
   language_server-protocol (3.17.0.5) sha256=fd1e39a51a28bf3eec959379985a72e296e9f9acfce46f6a79d31ca8760803cc
   lint_roller (1.1.0) sha256=2c0c845b632a7d172cb849cc90c1bce937a28c5c8ccccb50dfd46a485003cc87
   listen (3.10.0) sha256=c6e182db62143aeccc2e1960033bebe7445309c7272061979bb098d03760c9d2
@@ -235,14 +235,14 @@ CHECKSUMS
   rspec-core (3.13.6) sha256=a8823c6411667b60a8bca135364351dda34cd55e44ff94c4be4633b37d828b2d
   rspec-expectations (3.13.5) sha256=33a4d3a1d95060aea4c94e9f237030a8f9eae5615e9bd85718fe3a09e4b58836
   rspec-mocks (3.13.7) sha256=0979034e64b1d7a838aaaddf12bf065ea4dc40ef3d4c39f01f93ae2c66c62b1c
-  rspec-support (3.13.6) sha256=2e8de3702427eab064c9352fe74488cc12a1bfae887ad8b91cba480ec9f8afb2
+  rspec-support (3.13.7) sha256=0640e5570872aafefd79867901deeeeb40b0c9875a36b983d85f54fb7381c47c
   rubocop (1.84.0) sha256=88dec310153bb685a879f5a7cdb601f6287b8f0ee675d9dc63a17c7204c4190a
   rubocop-ast (1.49.0) sha256=49c3676d3123a0923d333e20c6c2dbaaae2d2287b475273fddee0c61da9f71fd
   rubocop-performance (1.26.1) sha256=cd19b936ff196df85829d264b522fd4f98b6c89ad271fa52744a8c11b8f71834
   rubocop-rspec (3.9.0) sha256=8fa70a3619408237d789aeecfb9beef40576acc855173e60939d63332fdb55e2
   ruby-progressbar (1.13.0) sha256=80fc9c47a9b640d6834e0dc7b3c94c9df37f08cb072b7761e4a71e22cff29b33
   securerandom (0.4.1) sha256=cc5193d414a4341b6e225f0cb4446aceca8e50d5e1888743fac16987638ea0b1
-  sorbet-runtime (0.6.12903) sha256=c23968c0dcf5a5db57f32c003fe3db7fb588c168cdd57d92ea4dceaba063118a
+  sorbet-runtime (0.6.12908) sha256=229f43e76527b01c5291c00c43cc29ccebe437a87b34925c3ee250ebf23d328e
   steep (1.10.0) sha256=1b295b55f9aaff1b8d3ee42453ee55bc2a1078fda0268f288edb2dc014f4d7d1
   strscan (3.1.7) sha256=5f76462b94a3ea50b44973225b7d75b2cb96d4e1bee9ef1319b99ca117b72c8c
   terminal-table (4.0.0) sha256=f504793203f8251b2ea7c7068333053f0beeea26093ec9962e62ea79f94301d2

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=v4.2.3" alt="Go">
+    <img src="https://img.shields.io/github/v/tag/kreuzberg-dev/kreuzberg?label=Go&color=007ec6&filter=v4.2.5" 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/config.rb

@@ -877,7 +877,7 @@ module Kreuzberg
           chunking: @chunking&.to_h,
           language_detection: @language_detection&.to_h,
           pdf_options: @pdf_options&.to_h,
-          images: @images&.to_h,
+          image_extraction: @images&.to_h,
           postprocessor: @postprocessor&.to_h,
           token_reduction: @token_reduction&.to_h,
           keywords: @keywords&.to_h,

data/lib/kreuzberg/result.rb

@@ -103,13 +103,48 @@ module Kreuzberg
     #   @return [Array<Table>] Tables on this page
     # @!attribute [r] images
     #   @return [Array<Image>] Images on this page
-    PageContent = Struct.new(:page_number, :content, :tables, :images, keyword_init: true) do
+    # @!attribute [r] text
+    #   @return [String] The text content of this block
+    # @!attribute [r] font_size
+    #   @return [Float] The font size of the text
+    # @!attribute [r] level
+    #   @return [String] The hierarchy level (h1-h6 or body)
+    # @!attribute [r] bbox
+    #   @return [Array<Float>, nil] Bounding box (left, top, right, bottom)
+    HierarchicalBlock = Struct.new(:text, :font_size, :level, :bbox, keyword_init: true) do
+      def to_h
+        { text: text, font_size: font_size, level: level, bbox: bbox }
+      end
+    end
+
+    # @!attribute [r] block_count
+    #   @return [Integer] Number of hierarchy blocks
+    # @!attribute [r] blocks
+    #   @return [Array<HierarchicalBlock>] Hierarchical blocks
+    PageHierarchy = Struct.new(:block_count, :blocks, keyword_init: true) do
+      def to_h
+        { block_count: block_count, blocks: blocks.map(&:to_h) }
+      end
+    end
+
+    # @!attribute [r] page_number
+    #   @return [Integer] Page number (1-indexed)
+    # @!attribute [r] content
+    #   @return [String] Text content for this page
+    # @!attribute [r] tables
+    #   @return [Array<Table>] Tables on this page
+    # @!attribute [r] images
+    #   @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, keyword_init: true) do
       def to_h
         {
           page_number: page_number,
           content: content,
           tables: tables.map(&:to_h),
-          images: images.map(&:to_h)
+          images: images.map(&:to_h),
+          hierarchy: hierarchy&.to_h
         }
       end
     end
@@ -397,9 +432,28 @@ module Kreuzberg
           page_number: page_hash['page_number'],
           content: page_hash['content'],
           tables: parse_tables(page_hash['tables']),
-          images: parse_images(page_hash['images'])
+          images: parse_images(page_hash['images']),
+          hierarchy: parse_page_hierarchy(page_hash['hierarchy'])
+        )
+      end
+    end
+
+    def parse_page_hierarchy(hierarchy_data)
+      return nil if hierarchy_data.nil?
+
+      blocks = (hierarchy_data['blocks'] || []).map do |block_hash|
+        HierarchicalBlock.new(
+          text: block_hash['text'],
+          font_size: block_hash['font_size']&.to_f,
+          level: block_hash['level'],
+          bbox: block_hash['bbox']
         )
       end
+
+      PageHierarchy.new(
+        block_count: hierarchy_data['block_count'] || 0,
+        blocks: blocks
+      )
     end
 
     def parse_elements(elements_data)

data/lib/kreuzberg/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Kreuzberg
-  VERSION = '4.2.3'
+  VERSION = '4.2.5'
 end

data/sig/kreuzberg.rbs

@@ -195,7 +195,6 @@ module Kreuzberg
       attr_reader language_detection: LanguageDetection?
       attr_reader pdf_options: PDF?
       attr_reader image_extraction: ImageExtraction?
-      attr_reader image_preprocessing: ImagePreprocessing?
       attr_reader postprocessor: PostProcessor?
       attr_reader token_reduction: TokenReduction?
       attr_reader keywords: Keywords?
@@ -215,7 +214,6 @@ module Kreuzberg
         ?language_detection: (LanguageDetection | Hash[Symbol, untyped])?,
         ?pdf_options: (PDF | Hash[Symbol, untyped])?,
         ?image_extraction: (ImageExtraction | Hash[Symbol, untyped])?,
-        ?image_preprocessing: (ImagePreprocessing | Hash[Symbol, untyped])?,
         ?postprocessor: (PostProcessor | Hash[Symbol, untyped])?,
         ?token_reduction: (TokenReduction | Hash[Symbol, untyped])?,
         ?keywords: (Keywords | Hash[Symbol, untyped])?,

data/spec/unit/config/extraction_config_spec.rb

@@ -13,7 +13,6 @@ RSpec.describe Kreuzberg::Config::Extraction do
       expect(config.language_detection).to be_nil
       expect(config.pdf_options).to be_nil
       expect(config.image_extraction).to be_nil
-      expect(config.image_preprocessing).to be_nil
       expect(config.postprocessor).to be_nil
       expect(config.token_reduction).to be_nil
       expect(config.keywords).to be_nil
@@ -258,7 +257,6 @@ RSpec.describe Kreuzberg::Config::Extraction do
         language_detection: { enabled: true },
         pdf_options: { extract_images: true },
         image_extraction: { target_dpi: 600 },
-        image_preprocessing: { denoise: true },
         postprocessor: { enabled: true },
         token_reduction: { mode: 'light' },
         keywords: { algorithm: 'yake' },
@@ -401,7 +399,6 @@ RSpec.describe Kreuzberg::Config::Extraction do
         language_detection: { enabled: true, min_confidence: 0.9 },
         pdf_options: { extract_images: true, passwords: ['secret'] },
         image_extraction: { target_dpi: 600 },
-        image_preprocessing: { denoise: true, binarization_method: 'sauvola' },
         postprocessor: { enabled: true, enabled_processors: %w[quality] },
         token_reduction: { mode: 'light' },
         keywords: { algorithm: 'yake', max_keywords: 10 },
@@ -415,7 +412,6 @@ RSpec.describe Kreuzberg::Config::Extraction do
       expect(config.language_detection.enabled).to be true
       expect(config.pdf_options.extract_images).to be true
       expect(config.image_extraction.target_dpi).to eq 600
-      expect(config.image_preprocessing.denoise).to be true
       expect(config.postprocessor.enabled).to be true
       expect(config.token_reduction.mode).to eq 'light'
       expect(config.keywords.max_keywords).to eq 10

data/spec/unit/config/image_preprocessing_config_spec.rb

@@ -157,25 +157,6 @@ RSpec.describe Kreuzberg::Config::ImagePreprocessing do
   end
 
   describe 'nested config integration' do
-    it 'can be nested in Extraction config' do
-      preprocessing = described_class.new(target_dpi: 600, denoise: true)
-      extraction = Kreuzberg::Config::Extraction.new(image_preprocessing: preprocessing)
-
-      expect(extraction.image_preprocessing).to be_a described_class
-      expect(extraction.image_preprocessing.target_dpi).to eq 600
-      expect(extraction.image_preprocessing.denoise).to be true
-    end
-
-    it 'accepts hash in Extraction config' do
-      extraction = Kreuzberg::Config::Extraction.new(
-        image_preprocessing: { target_dpi: 600, binarization_method: 'sauvola' }
-      )
-
-      expect(extraction.image_preprocessing).to be_a described_class
-      expect(extraction.image_preprocessing.target_dpi).to eq 600
-      expect(extraction.image_preprocessing.binarization_method).to eq 'sauvola'
-    end
-
     it 'can be nested in Tesseract config' do
       preprocessing = described_class.new(denoise: true)
       tesseract = Kreuzberg::Config::Tesseract.new(preprocessing: preprocessing)

data/vendor/Cargo.toml

@@ -3,7 +3,7 @@ members = ["kreuzberg", "kreuzberg-tesseract", "kreuzberg-ffi"]
 resolver = "2"
 
 [workspace.package]
-version = "4.2.3"
+version = "4.2.5"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <nhirschfeld@gmail.com>"]
@@ -47,7 +47,7 @@ hex = "0.4.3"
 toml = "0.9.11"
 num_cpus = "1.17.0"
 once_cell = "1.21.3"
-html-to-markdown-rs = { version = "2.23.4", default-features = false }
+html-to-markdown-rs = { version = "2.24.1", default-features = false }
 reqwest = { version = "0.13.1", default-features = false, features = ["json", "rustls"] }
 image = { version = "0.25.9", default-features = false }
 lzma-rust2 = { version = "0.15.7" }

data/vendor/kreuzberg/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "kreuzberg"
-version = "4.2.3"
+version = "4.2.5"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <nhirschfeld@gmail.com>"]

data/vendor/kreuzberg/README.md

@@ -17,7 +17,7 @@ High-performance document intelligence library for Rust. Extract text, metadata,
 
 This is the core Rust library that powers the Python, TypeScript, and Ruby bindings.
 
-> **🚀 Version 4.2.3 Release**
+> **🚀 Version 4.2.5 Release**
 > This is a pre-release version. We invite you to test the library and [report any issues](https://github.com/kreuzberg-dev/kreuzberg/issues) you encounter.
 >
 > **Note**: The Rust crate is not currently published to crates.io for this RC. Use git dependencies or language bindings (Python, TypeScript, Ruby) instead.

data/vendor/kreuzberg/src/keywords/config.rs

@@ -3,6 +3,14 @@
 use super::types::KeywordAlgorithm;
 use serde::{Deserialize, Serialize};
 
+fn default_max_keywords() -> usize {
+    10
+}
+
+fn default_ngram_range() -> (usize, usize) {
+    (1, 3)
+}
+
 /// YAKE-specific parameters.
 #[cfg(feature = "keywords-yake")]
 #[derive(Debug, Clone, Serialize, Deserialize)]
@@ -45,15 +53,18 @@ impl Default for RakeParams {
 #[derive(Debug, Clone, Serialize, Deserialize)]
 pub struct KeywordConfig {
     /// Algorithm to use for extraction.
+    #[serde(default)]
     pub algorithm: KeywordAlgorithm,
 
     /// Maximum number of keywords to extract (default: 10).
+    #[serde(default = "default_max_keywords")]
     pub max_keywords: usize,
 
     /// Minimum score threshold (0.0-1.0, default: 0.0).
     ///
     /// Keywords with scores below this threshold are filtered out.
     /// Note: Score ranges differ between algorithms.
+    #[serde(default)]
     pub min_score: f32,
 
     /// N-gram range for keyword extraction (min, max).
@@ -61,6 +72,7 @@ pub struct KeywordConfig {
     /// (1, 1) = unigrams only
     /// (1, 2) = unigrams and bigrams
     /// (1, 3) = unigrams, bigrams, and trigrams (default)
+    #[serde(default = "default_ngram_range")]
     pub ngram_range: (usize, usize),
 
     /// Language code for stopword filtering (e.g., "en", "de", "fr").

data/vendor/kreuzberg-ffi/Cargo.toml

@@ -28,7 +28,7 @@ serde_json = { workspace = true }
 serde = { workspace = true }
 async-trait = { workspace = true }
 tokio = { workspace = true }
-html-to-markdown-rs = { version = "2.23.4", default-features = false }
+html-to-markdown-rs = { version = "2.24.1", default-features = false }
 rayon = { version = "1.11", optional = true }
 log = "0.4"
 

data/vendor/kreuzberg-tesseract/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "kreuzberg-tesseract"
-version = "4.2.3"
+version = "4.2.5"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <nhirschfeld@gmail.com>"]
@@ -24,7 +24,7 @@ thiserror = { workspace = true }
 image = { workspace = true, features = ["png"] }
 
 [build-dependencies]
-cc = { version = "^1.2.54", optional = true }
+cc = { version = "^1.2.55", optional = true }
 cmake = { version = "0.1.57", optional = true }
 zip = { version = "7.2.0", optional = true }
 

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: kreuzberg
 version: !ruby/object:Gem::Version
-  version: 4.2.3
+  version: 4.2.5
 platform: ruby
 authors:
 - Na'aman Hirschfeld
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-01-28 00:00:00.000000000 Z
+date: 2026-01-30 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler