kreuzberg 4.7.4-aarch64-linux → 4.8.0-aarch64-linux

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 00cdc7fc0ab8f3b23ba177771f8fd4782e0b241a1dbce1cbde3daab2713f3932
-  data.tar.gz: f4aec1401be34830b59426141d31df2e40a1a26f22af5a0b4100e8bf153ee21f
+  metadata.gz: 7b8c670c1b98342788190baf143a5ae2ae605fcfec4974a162f931d36052a87f
+  data.tar.gz: e6a35b64afbb09e7c228d6cf8e148be71cd70215c8be46ba9d291f265597a942
 SHA512:
-  metadata.gz: 1be6b7d0ed3aee0ad35639e285fb35fd23a2df6df870aae88ad7a038ab0a9ee078b94f5fa4e3ee035ce93bda5e7c053c365fbdfe836a65dac527484a7b04f03d
-  data.tar.gz: 284e3fa2c4d038f8345fe5b73ca2497890c5c6bdf23106c2f714e19e1573cd720ef830af4d06458a3a8658cb36fc616b1a8f61fefec1da4acd85d4db801ac066
+  metadata.gz: 4981b0051cc6a8db5de244957c455eaced589b0404da51dd09278d742a9e0d79cb459b79788950d2608c6c3728abfe5dc38d294eb324178cfa31c284080224d9
+  data.tar.gz: 4e3848edfb9d39632727ae5018ccb745ae16d80acc638d43ecc2e7119489c1a02f986819879f5bb7762d25c21692f56da2520c282516b6b43d91640bb0423558

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=v4.7.4" alt="Go">
+    <img src="https://img.shields.io/github/v/tag/kreuzberg-dev/kreuzberg?label=Go&color=007ec6&filter=v4.8.0" 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#">
@@ -42,7 +42,7 @@
 
   <!-- Project Info -->
   <a href="https://github.com/kreuzberg-dev/kreuzberg/blob/main/LICENSE">
-    <img src="https://img.shields.io/badge/License-MIT-007ec6" alt="License">
+    <img src="https://img.shields.io/badge/License-Elastic--2.0-blue.svg" alt="License">
   </a>
   <a href="https://docs.kreuzberg.dev">
     <img src="https://img.shields.io/badge/docs-kreuzberg.dev-007ec6" alt="Documentation">
@@ -419,7 +419,7 @@ Contributions are welcome! See [Contributing Guide](https://github.com/kreuzberg
 
 ## License
 
-MIT License - see LICENSE file for details.
+Elastic License 2.0 (ELv2) - see [LICENSE](../../LICENSE) for details.
 
 ## Support
 

data/lib/kreuzberg/config.rb

@@ -856,6 +856,40 @@ module Kreuzberg
       end
     end
 
+    # Content filter configuration for controlling extraction of headers, footers,
+    # watermarks, and repeating text across document formats.
+    #
+    # @example Include headers and footers
+    #   filter = ContentFilter.new(include_headers: true, include_footers: true)
+    #
+    # @example Disable repeating text removal
+    #   filter = ContentFilter.new(strip_repeating_text: false)
+    #
+    class ContentFilter
+      attr_reader :include_headers, :include_footers, :strip_repeating_text, :include_watermarks
+
+      def initialize(
+        include_headers: false,
+        include_footers: false,
+        strip_repeating_text: true,
+        include_watermarks: false
+      )
+        @include_headers = include_headers ? true : false
+        @include_footers = include_footers ? true : false
+        @strip_repeating_text = strip_repeating_text ? true : false
+        @include_watermarks = include_watermarks ? true : false
+      end
+
+      def to_h
+        {
+          include_headers: @include_headers,
+          include_footers: @include_footers,
+          strip_repeating_text: @strip_repeating_text,
+          include_watermarks: @include_watermarks
+        }
+      end
+    end
+
     # Layout detection configuration
     #
     # @example Basic usage
@@ -951,7 +985,7 @@ module Kreuzberg
                   :max_concurrent_extractions, :output_format, :result_format,
                   :security_limits, :layout, :concurrency,
                   :cache_namespace, :cache_ttl_secs, :extraction_timeout_secs,
-                  :max_archive_depth, :acceleration, :email
+                  :max_archive_depth, :acceleration, :email, :content_filter
 
       # Alias for backward compatibility - image_extraction is the canonical name
       alias image_extraction images
@@ -977,7 +1011,7 @@ module Kreuzberg
         postprocessor token_reduction keywords html_options pages
         max_concurrent_extractions output_format result_format
         security_limits layout concurrency cache_namespace cache_ttl_secs extraction_timeout_secs
-        max_archive_depth acceleration email
+        max_archive_depth acceleration email content_filter
       ].freeze
 
       # Aliases for backward compatibility
@@ -1062,7 +1096,8 @@ module Kreuzberg
                      extraction_timeout_secs: nil,
                      max_archive_depth: 3,
                      acceleration: nil,
-                     email: nil)
+                     email: nil,
+                     content_filter: nil)
         kwargs = {
           use_cache: use_cache, enable_quality_processing: enable_quality_processing,
           force_ocr: force_ocr, disable_ocr: disable_ocr, force_ocr_pages: force_ocr_pages,
@@ -1080,7 +1115,8 @@ module Kreuzberg
           extraction_timeout_secs: extraction_timeout_secs,
           max_archive_depth: max_archive_depth,
           acceleration: acceleration,
-          email: email
+          email: email,
+          content_filter: content_filter
         }
         extracted = extract_from_hash(hash, kwargs)
 
@@ -1115,6 +1151,7 @@ module Kreuzberg
         @concurrency = normalize_config(params[:concurrency], Concurrency)
         @acceleration = normalize_config(params[:acceleration], Acceleration)
         @email = normalize_config(params[:email], Email)
+        @content_filter = normalize_config(params[:content_filter], ContentFilter)
         @max_concurrent_extractions = params[:max_concurrent_extractions]&.to_i
         @max_archive_depth = params[:max_archive_depth]&.to_i || 3
         @output_format = validate_output_format(params[:output_format])
@@ -1175,7 +1212,8 @@ module Kreuzberg
           token_reduction: @token_reduction&.to_h, keywords: @keywords&.to_h,
           html_options: @html_options&.to_h, pages: @pages&.to_h,
           layout: @layout&.to_h, concurrency: @concurrency&.to_h,
-          acceleration: @acceleration&.to_h, email: @email&.to_h
+          acceleration: @acceleration&.to_h, email: @email&.to_h,
+          content_filter: @content_filter&.to_h
         }
       end
 

data/lib/kreuzberg/errors.rb

@@ -11,6 +11,7 @@ module Kreuzberg
   ERROR_CODE_PARSING = 5
   ERROR_CODE_OCR = 6
   ERROR_CODE_MISSING_DEPENDENCY = 7
+  ERROR_CODE_EMBEDDING = 8
 
   module Errors
     class PanicContext
@@ -112,5 +113,8 @@ module Kreuzberg
 
     # Raised when an unsupported file format or MIME type is encountered
     class UnsupportedFormatError < Error; end
+
+    # Raised when embedding fails
+    class EmbeddingError < Error; end
   end
 end

data/lib/kreuzberg/extraction_api.rb

@@ -236,6 +236,41 @@ module Kreuzberg
       results
     end
 
+    # Asynchronously generate embeddings for multiple texts.
+    #
+    # Non-blocking embedding generation from a list of strings.
+    #
+    # @param texts [Array<String>] List of strings to embed.
+    # @param config [Config::Embedding, Hash, nil] Embedding configuration.
+    #
+    # @return [Array<Array<Float>>] Array of embedding vectors.
+    #
+    # @raise [Errors::EmbeddingError] If embedding generation fails.
+    #
+    # @example Generate embeddings asynchronously
+    #   texts = ["Hello, world!", "Kreuzberg is awesome."]
+    #   embeddings = Kreuzberg.embed(texts: texts)
+    #   puts embeddings.first.length # 384
+    def embed(texts:, config: nil)
+      opts = normalize_config(config)
+      native_embed(texts: texts.map(&:to_s), config: opts)
+    end
+
+    # Synchronously generate embeddings for multiple texts.
+    #
+    # Blocking embedding generation from a list of strings.
+    #
+    # @param texts [Array<String>] List of strings to embed.
+    # @param config [Config::Embedding, Hash, nil] Embedding configuration.
+    #
+    # @return [Array<Array<Float>>] Array of embedding vectors.
+    #
+    # @raise [Errors::EmbeddingError] If embedding generation fails.
+    def embed_sync(texts:, config: nil)
+      opts = normalize_config(config)
+      native_embed_sync(texts: texts.map(&:to_s), config: opts)
+    end
+
     # Synchronously extract content from multiple byte data sources.
     #
     # Processes multiple in-memory binary documents in a single batch operation. Results

data/lib/kreuzberg/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Kreuzberg
-  VERSION = '4.7.4'
+  VERSION = '4.8.0'
 end

data/lib/kreuzberg.rb

@@ -59,10 +59,13 @@ module Kreuzberg
     alias native_batch_extract_bytes batch_extract_bytes
     alias native_clear_cache clear_cache
     alias native_cache_stats cache_stats
+    alias native_embed_sync embed_sync
+    alias native_embed embed
 
     private :native_extract_file_sync, :native_extract_bytes_sync, :native_batch_extract_files_sync
     private :native_extract_file, :native_extract_bytes, :native_batch_extract_files
     private :native_batch_extract_bytes_sync, :native_batch_extract_bytes
+    private :native_embed_sync, :native_embed
   end
 
   module_function :register_post_processor
@@ -94,6 +97,10 @@ module Kreuzberg
   module_function :validate_mime_type
 
   module_function :get_extensions_for_mime
+
+  module_function :embed_sync
+
+  module_function :embed
 end
 
 require_relative 'kreuzberg/cache_api'

data/sig/kreuzberg.rbs

@@ -12,6 +12,7 @@ module Kreuzberg
   ERROR_CODE_PARSING: Integer
   ERROR_CODE_OCR: Integer
   ERROR_CODE_MISSING_DEPENDENCY: Integer
+  ERROR_CODE_EMBEDDING: Integer
 
   # Semantic element type classification (T.type_alias)
   type element_type = 'title' | 'narrative_text' | 'heading' | 'list_item' | 'table' | 'image' | 'page_break' | 'code_block' | 'block_quote' | 'footer' | 'header'
@@ -747,6 +748,16 @@ module Kreuzberg
       def to_h: () -> Hash[Symbol, untyped]
     end
 
+    class ContentFilter
+      attr_reader include_headers: bool
+      attr_reader include_footers: bool
+      attr_reader strip_repeating_text: bool
+      attr_reader include_watermarks: bool
+
+      def initialize: (?include_headers: bool, ?include_footers: bool, ?strip_repeating_text: bool, ?include_watermarks: bool) -> void
+      def to_h: () -> Hash[Symbol, untyped]
+    end
+
     class LayoutDetection
       attr_reader preset: String
       attr_reader confidence_threshold: Float?
@@ -813,6 +824,7 @@ module Kreuzberg
       attr_reader concurrency: Concurrency?
       attr_reader acceleration: Acceleration?
       attr_reader email: Email?
+      attr_reader content_filter: ContentFilter?
       attr_reader tree_sitter: TreeSitterConfig?
       attr_reader max_concurrent_extractions: Integer?
       attr_reader max_archive_depth: Integer
@@ -844,6 +856,7 @@ module Kreuzberg
         ?concurrency: (Concurrency | Hash[Symbol, untyped])?,
         ?acceleration: (Acceleration | Hash[Symbol, untyped])?,
         ?email: (Email | Hash[Symbol, untyped])?,
+        ?content_filter: (ContentFilter | Hash[Symbol, untyped])?,
         ?tree_sitter: (TreeSitterConfig | Hash[Symbol, untyped])?,
         ?max_concurrent_extractions: Integer?,
         ?max_archive_depth: Integer,
@@ -1572,6 +1585,10 @@ module Kreuzberg
     ?file_configs: Array[file_config_input]?
   ) -> Array[Result]
 
+  # Standalone embedding
+  def self.embed_sync: (texts: Array[String], ?config: Hash[Symbol, untyped]?) -> Array[Array[Float]]
+  def self.embed: (texts: Array[String], ?config: Hash[Symbol, untyped]?) -> Array[Array[Float]]
+
   # Cache API
   def self.clear_cache: () -> void
   def self.cache_stats: () -> Hash[Symbol | String, Integer]
@@ -1722,6 +1739,9 @@ module Kreuzberg
 
     class UnsupportedFormatError < Error
     end
+
+    class EmbeddingError < 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: 4.7.4
+  version: 4.8.0
 platform: aarch64-linux
 authors:
 - Na'aman Hirschfeld
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-04-06 00:00:00.000000000 Z
+date: 2026-04-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -223,7 +223,7 @@ files:
 - spec/spec_helper.rb
 homepage: https://github.com/kreuzberg-dev/kreuzberg
 licenses:
-- MIT
+- Elastic-2.0
 metadata:
   homepage_uri: https://github.com/kreuzberg-dev/kreuzberg
   source_code_uri: https://github.com/kreuzberg-dev/kreuzberg