kreuzberg 4.6.0-aarch64-linux → 4.6.2-aarch64-linux

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: e120139c5153d954f686166a3c23f0a3c64a24dcbbcf2ff1a9d17abeafbdc22d
-  data.tar.gz: a17475f255eceefe8a6c0702866cf00c401cdbc294c8b5eace4cae78b9401161
+  metadata.gz: 406c791db1a8cb29e3ff2e89a60e0a29ed73c0ff0548338b60c3574bd5944f6f
+  data.tar.gz: 7565bbe0708afceadc13f43b57103a03529992f705caf21f1ddaa00e68ad27d6
 SHA512:
-  metadata.gz: '01525768eda28284eca9d077377d052d6fb90834282a26a73d8f8b2c3dbfcbad573bb71c51bbf0679329e59f205cd097f23080bdb10a7ad0432d700092aa7501'
-  data.tar.gz: ef96711a4c445cb7487d253731cfd697a0dd179f675efe8d3c9d2a6f6033b2257ada11f4a343cf0cf41a2160e814b3565e4160a0cf3f0a24f08d1765d59e6107
+  metadata.gz: a680fd2406f8dac338a53ab303ef29d117462cbf376d48be4af48846f7d4f82d4e7999e2aeac17a404e5da7feb79f1cccfb8f240153c0c36a4274e16eeb50f6f
+  data.tar.gz: dc1c1dc215020a65560490e159e38d6780b703bae597279cb0e3d2c369afce7a7bee06984972987d1c3a69fe2040da1cb1e2812df5c542590300dfcb09e5e7d7

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=v4.6.0" alt="Go">
+    <img src="https://img.shields.io/github/v/tag/kreuzberg-dev/kreuzberg?label=Go&color=007ec6&filter=v4.6.2" 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

@@ -926,14 +926,14 @@ module Kreuzberg
     #   )
     #
     class Extraction
-      attr_reader :use_cache, :enable_quality_processing, :force_ocr,
+      attr_reader :use_cache, :enable_quality_processing, :force_ocr, :force_ocr_pages,
                   :include_document_structure,
                   :ocr, :chunking, :language_detection, :pdf_options,
                   :images, :postprocessor,
                   :token_reduction, :keywords, :html_options, :pages,
                   :max_concurrent_extractions, :output_format, :result_format,
                   :security_limits, :layout, :concurrency,
-                  :cache_namespace, :cache_ttl_secs
+                  :cache_namespace, :cache_ttl_secs, :extraction_timeout_secs
 
       # Alias for backward compatibility - image_extraction is the canonical name
       alias image_extraction images
@@ -954,11 +954,11 @@ module Kreuzberg
       #
       # Keys that are allowed in the Extraction config
       ALLOWED_KEYS = %i[
-        use_cache enable_quality_processing force_ocr include_document_structure ocr chunking
+        use_cache enable_quality_processing force_ocr force_ocr_pages include_document_structure ocr chunking
         language_detection pdf_options image_extraction
         postprocessor token_reduction keywords html_options pages
         max_concurrent_extractions output_format result_format
-        security_limits layout concurrency cache_namespace cache_ttl_secs
+        security_limits layout concurrency cache_namespace cache_ttl_secs extraction_timeout_secs
       ].freeze
 
       # Aliases for backward compatibility
@@ -1019,6 +1019,7 @@ module Kreuzberg
                      use_cache: true,
                      enable_quality_processing: true,
                      force_ocr: false,
+                     force_ocr_pages: nil,
                      include_document_structure: false,
                      ocr: nil,
                      chunking: nil,
@@ -1037,10 +1038,12 @@ module Kreuzberg
                      layout: nil,
                      concurrency: nil,
                      cache_namespace: nil,
-                     cache_ttl_secs: nil)
+                     cache_ttl_secs: nil,
+                     extraction_timeout_secs: nil)
         kwargs = {
           use_cache: use_cache, enable_quality_processing: enable_quality_processing,
-          force_ocr: force_ocr, include_document_structure: include_document_structure,
+          force_ocr: force_ocr, force_ocr_pages: force_ocr_pages,
+          include_document_structure: include_document_structure,
           ocr: ocr, chunking: chunking, language_detection: language_detection,
           pdf_options: pdf_options, image_extraction: image_extraction,
           postprocessor: postprocessor,
@@ -1050,7 +1053,8 @@ module Kreuzberg
           security_limits: security_limits, layout: layout,
           concurrency: concurrency,
           cache_namespace: cache_namespace,
-          cache_ttl_secs: cache_ttl_secs
+          cache_ttl_secs: cache_ttl_secs,
+          extraction_timeout_secs: extraction_timeout_secs
         }
         extracted = extract_from_hash(hash, kwargs)
 
@@ -1068,6 +1072,7 @@ module Kreuzberg
         @use_cache = params[:use_cache] ? true : false
         @enable_quality_processing = params[:enable_quality_processing] ? true : false
         @force_ocr = params[:force_ocr] ? true : false
+        @force_ocr_pages = params[:force_ocr_pages]
         @include_document_structure = params[:include_document_structure] ? true : false
         @ocr = normalize_config(params[:ocr], OCR)
         @chunking = normalize_config(params[:chunking], Chunking)
@@ -1086,6 +1091,7 @@ module Kreuzberg
         @result_format = validate_result_format(params[:result_format])
         @cache_namespace = params[:cache_namespace]
         @cache_ttl_secs = params[:cache_ttl_secs]&.to_i
+        @extraction_timeout_secs = params[:extraction_timeout_secs]&.to_i
         @security_limits = params[:security_limits]
       end
 
@@ -1118,12 +1124,14 @@ module Kreuzberg
           use_cache: @use_cache,
           enable_quality_processing: @enable_quality_processing,
           force_ocr: @force_ocr,
+          force_ocr_pages: @force_ocr_pages,
           include_document_structure: @include_document_structure,
           max_concurrent_extractions: @max_concurrent_extractions,
           output_format: @output_format,
           result_format: @result_format,
           cache_namespace: @cache_namespace,
-          cache_ttl_secs: @cache_ttl_secs
+          cache_ttl_secs: @cache_ttl_secs,
+          extraction_timeout_secs: @extraction_timeout_secs
         }
       end
 
@@ -1250,6 +1258,8 @@ module Kreuzberg
           @enable_quality_processing = value ? true : false
         when :force_ocr
           @force_ocr = value ? true : false
+        when :force_ocr_pages
+          @force_ocr_pages = value
         when :include_document_structure
           @include_document_structure = value ? true : false
         when :ocr
@@ -1286,6 +1296,8 @@ module Kreuzberg
           @cache_namespace = value
         when :cache_ttl_secs
           @cache_ttl_secs = value&.to_i
+        when :extraction_timeout_secs
+          @extraction_timeout_secs = value&.to_i
         else
           raise ArgumentError, "Unknown configuration key: #{key}"
         end
@@ -1345,6 +1357,7 @@ module Kreuzberg
         @use_cache = merged.use_cache
         @enable_quality_processing = merged.enable_quality_processing
         @force_ocr = merged.force_ocr
+        @force_ocr_pages = merged.force_ocr_pages
         @include_document_structure = merged.include_document_structure
         @ocr = merged.ocr
         @chunking = merged.chunking
@@ -1369,6 +1382,7 @@ module Kreuzberg
         @result_format = merged.result_format
         @cache_namespace = merged.cache_namespace
         @cache_ttl_secs = merged.cache_ttl_secs
+        @extraction_timeout_secs = merged.extraction_timeout_secs
       end
     end
   end

data/lib/kreuzberg/extraction_api.rb

@@ -319,6 +319,43 @@ module Kreuzberg
       results
     end
 
+    # Render a single PDF page as a PNG image.
+    #
+    # @param path [String, Pathname] Path to the PDF file
+    # @param page_index [Integer] Zero-based page index
+    # @param dpi [Integer] Rendering resolution (default 150)
+    # @return [String] PNG-encoded binary string
+    # @raise [Errors::IOError] If the file cannot be read
+    # @raise [Errors::ParsingError] If rendering fails
+    def render_pdf_page(path, page_index, dpi: 150)
+      path_str = path.to_s
+      raise ArgumentError, 'page_index must be non-negative' if page_index.negative?
+      raise Errors::IOError, "File not found: #{path_str}" unless File.exist?(path_str)
+
+      native_render_pdf_page(path_str, page_index, dpi)
+    end
+
+    # Iterate over pages of a PDF lazily, yielding each page as it is rendered.
+    #
+    # Each page is rendered via the native FFI iterator, so only one page is in
+    # memory at a time.
+    #
+    # @param path [String, Pathname] Path to the PDF file
+    # @param dpi [Integer] Rendering resolution (default 150)
+    # @yieldparam page_index [Integer] Zero-based page index
+    # @yieldparam png_bytes [String] PNG-encoded binary string for the page
+    # @return [Enumerator] if no block is given
+    # @raise [Errors::IOError] If the file cannot be read
+    # @raise [Errors::ParsingError] If rendering fails
+    def render_pdf_pages_iter(path, dpi: 150, &block)
+      path_str = path.to_s
+      raise Errors::IOError, "File not found: #{path_str}" unless File.exist?(path_str)
+
+      return enum_for(:render_pdf_pages_iter, path, dpi: dpi) unless block
+
+      native_render_pdf_pages_iter(path_str, dpi, &block)
+    end
+
     def normalize_config(config)
       return {} if config.nil?
       return config if config.is_a?(Hash)

data/lib/kreuzberg/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Kreuzberg
-  VERSION = '4.6.0'
+  VERSION = '4.6.2'
 end

data/sig/kreuzberg.rbs

@@ -481,7 +481,9 @@ module Kreuzberg
       attr_reader enable_quality_processing: bool
       attr_reader cache_namespace: String?
       attr_reader cache_ttl_secs: Integer?
+      attr_reader extraction_timeout_secs: Integer?
       attr_reader force_ocr: bool
+      attr_reader force_ocr_pages: Array[Integer]?
       attr_reader include_document_structure: bool
       attr_reader ocr: OCR?
       attr_reader chunking: Chunking?
@@ -508,6 +510,7 @@ module Kreuzberg
         ?use_cache: bool,
         ?enable_quality_processing: bool,
         ?force_ocr: bool,
+        ?force_ocr_pages: Array[Integer]?,
         ?include_document_structure: bool,
         ?ocr: (OCR | Hash[Symbol, untyped])?,
         ?chunking: (Chunking | Hash[Symbol, untyped])?,
@@ -525,7 +528,8 @@ module Kreuzberg
         ?output_format: String?,
         ?result_format: String?,
         ?cache_namespace: String?,
-        ?cache_ttl_secs: Integer?
+        ?cache_ttl_secs: Integer?,
+        ?extraction_timeout_secs: Integer?
       ) -> void
       def to_h: () -> Hash[Symbol, untyped]
       def to_json: (*untyped) -> String

data/spec/binding/render_spec.rb

@@ -0,0 +1,91 @@
+# Hand-written binding-specific edge case tests for PDF rendering.
+# Happy-path render tests are auto-generated from fixtures in e2e/.
+# These tests cover error handling, validation, and lifecycle patterns
+# that vary per language and can't be generated uniformly.
+
+# frozen_string_literal: true
+
+require 'spec_helper'
+
+RSpec.describe 'PDF Rendering' do
+  it 'exposes rendering methods' do
+    expect(Kreuzberg).to respond_to(:render_pdf_page)
+    expect(Kreuzberg).to respond_to(:render_pdf_pages_iter)
+  end
+
+  describe '.render_pdf_page' do
+    it 'raises an error for a nonexistent file' do
+      expect do
+        Kreuzberg.render_pdf_page('/nonexistent/path/to/document.pdf', 0)
+      end.to raise_error(Kreuzberg::Errors::IOError)
+    end
+
+    it 'raises an error for an out-of-bounds page index' do
+      pdf_path = test_document_path('pdf/tiny.pdf')
+      skip 'Test PDF not available' unless File.exist?(pdf_path)
+
+      expect do
+        Kreuzberg.render_pdf_page(pdf_path, 9999)
+      end.to raise_error(StandardError)
+    end
+  end
+
+  describe '.render_pdf_page with negative index' do
+    it 'raises ArgumentError for a negative page index' do
+      pdf_path = test_document_path('pdf/tiny.pdf')
+      skip 'Test PDF not available' unless File.exist?(pdf_path)
+
+      expect do
+        Kreuzberg.render_pdf_page(pdf_path, -1)
+      end.to raise_error(ArgumentError)
+    end
+  end
+
+  describe '.render_pdf_pages_iter' do
+    it 'raises an error for a nonexistent file' do
+      expect do
+        Kreuzberg.render_pdf_pages_iter('/nonexistent/path/to/document.pdf') { |_, _| nil }
+      end.to raise_error(Kreuzberg::Errors::IOError)
+    end
+  end
+
+  describe '.render_pdf_page with empty path' do
+    it 'raises an error for an empty path' do
+      expect do
+        Kreuzberg.render_pdf_page('', 0)
+      end.to raise_error(StandardError)
+    end
+  end
+
+  describe '.render_pdf_pages_iter cleanup' do
+    it 'handles iterator cleanup without fully consuming' do
+      pdf_path = test_document_path('pdf/tiny.pdf')
+      skip 'Test PDF not available' unless File.exist?(pdf_path)
+
+      # Iterate but stop immediately — no crash
+      Kreuzberg.render_pdf_pages_iter(pdf_path) do |_page_index, _png_data|
+        break
+      end
+    end
+  end
+
+  describe '.render_pdf_pages_iter early termination' do
+    it 'returns valid PNG for the first page then stops' do
+      pdf_path = test_document_path('pdf/tiny.pdf')
+      skip 'Test PDF not available' unless File.exist?(pdf_path)
+
+      first_png = nil
+      Kreuzberg.render_pdf_pages_iter(pdf_path) do |page_index, png_data|
+        expect(page_index).to eq(0)
+        expect(png_data).to be_a(String)
+        expect(png_data.bytesize).to be > 8
+        # PNG magic bytes
+        expect(png_data.bytes[0..3]).to eq([0x89, 0x50, 0x4E, 0x47])
+        first_png = png_data
+        break
+      end
+
+      expect(first_png).not_to be_nil
+    end
+  end
+end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: kreuzberg
 version: !ruby/object:Gem::Version
-  version: 4.6.0
+  version: 4.6.2
 platform: aarch64-linux
 authors:
 - Na'aman Hirschfeld
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-03-25 00:00:00.000000000 Z
+date: 2026-03-26 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -220,6 +220,7 @@ files:
 - spec/binding/plugins/ocr_backend_spec.rb
 - spec/binding/plugins/postprocessor_spec.rb
 - spec/binding/plugins/validator_spec.rb
+- spec/binding/render_spec.rb
 - spec/binding/tables_spec.rb
 - spec/serialization_spec.rb
 - spec/smoke/package_spec.rb