kreuzberg 4.2.1 → 4.2.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: '010412940492f83bc170a4a4efc644ac7e3818502734035523796287837a3893'
-  data.tar.gz: af24cba007cc58283d678d4b15251ebae3a5740f60ade9a54cc19973a2022a82
+  metadata.gz: 2c6fc44b151014f7e56c82bd191f55244a4294a259b24b95fc494dba6f8eaba6
+  data.tar.gz: 6e40a732814ff3e2a164e718cdb1c7a6ae838b2b2210a66b232f8675c7f79a80
 SHA512:
-  metadata.gz: ad67348bec54a01ca3592ed72e9b2b8bc9e711a37e11b40ada31466c67f834132fc0de278c53a1c014fa6751da7abebae934cff2a9cc1835f7e056c895a273cb
-  data.tar.gz: ca2cdb076a5d1af67f0e807978a966d1a391cc286bcdf5499544e3403196140918a54674beab77ea09fc0e8bc7ab66f357da5d984326a511b1d21643a3d6cf41
+  metadata.gz: f9c3a45f31c3ad9e3857872d8705b397b40c4317844ef421f4da4c2918e57411f5a626df4f6706d7db4916f33b8644c736e7b41508b398fd0197f1a87170fa3c
+  data.tar.gz: 8b05a75be261dbe583c4873d9d21079efff97d6c9c0340bbd8a73a43c9d15955431f4de20cd8b4a8b7956872f52e4467c253f5da03177a1e7d3b6a10d202b59d

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    kreuzberg (4.2.1)
+    kreuzberg (4.2.3)
 
 GEM
   remote: https://rubygems.org/
@@ -61,7 +61,7 @@ GEM
     parser (3.3.10.1)
       ast (~> 2.4.1)
       racc
-    prism (1.8.0)
+    prism (1.9.0)
     pry (0.16.0)
       coderay (~> 1.1)
       method_source (~> 1.0)
@@ -98,7 +98,7 @@ GEM
       diff-lcs (>= 1.2.0, < 2.0)
       rspec-support (~> 3.13.0)
     rspec-support (3.13.6)
-    rubocop (1.82.1)
+    rubocop (1.84.0)
       json (~> 2.3)
       language_server-protocol (~> 3.17.0.2)
       lint_roller (~> 1.1.0)
@@ -106,7 +106,7 @@ GEM
       parser (>= 3.3.0.2)
       rainbow (>= 2.2.2, < 4.0)
       regexp_parser (>= 2.9.3, < 3.0)
-      rubocop-ast (>= 1.48.0, < 2.0)
+      rubocop-ast (>= 1.49.0, < 2.0)
       ruby-progressbar (~> 1.7)
       unicode-display_width (>= 2.4.0, < 4.0)
     rubocop-ast (1.49.0)
@@ -121,7 +121,7 @@ GEM
       rubocop (~> 1.81)
     ruby-progressbar (1.13.0)
     securerandom (0.4.1)
-    sorbet-runtime (0.6.12897)
+    sorbet-runtime (0.6.12903)
     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 (4.2.1)
+  kreuzberg (4.2.3)
   language_server-protocol (3.17.0.5) sha256=fd1e39a51a28bf3eec959379985a72e296e9f9acfce46f6a79d31ca8760803cc
   lint_roller (1.1.0) sha256=2c0c845b632a7d172cb849cc90c1bce937a28c5c8ccccb50dfd46a485003cc87
   listen (3.10.0) sha256=c6e182db62143aeccc2e1960033bebe7445309c7272061979bb098d03760c9d2
@@ -217,7 +217,7 @@ CHECKSUMS
   mutex_m (0.3.0) sha256=cfcb04ac16b69c4813777022fdceda24e9f798e48092a2b817eb4c0a782b0751
   parallel (1.27.0) sha256=4ac151e1806b755fb4e2dc2332cbf0e54f2e24ba821ff2d3dcf86bf6dc4ae130
   parser (3.3.10.1) sha256=06f6a725d2cd91e5e7f2b7c32ba143631e1f7c8ae2fb918fc4cebec187e6a688
-  prism (1.8.0) sha256=84453a16ef5530ea62c5f03ec16b52a459575ad4e7b9c2b360fd8ce2c39c1254
+  prism (1.9.0) sha256=7b530c6a9f92c24300014919c9dcbc055bf4cdf51ec30aed099b06cd6674ef85
   pry (0.16.0) sha256=d76c69065698ed1f85e717bd33d7942c38a50868f6b0673c636192b3d1b6054e
   pry-byebug (3.12.0) sha256=594e094ae8a8390a7ad4c7b36ae36e13304ed02664c67417d108dc5f7213d1b7
   racc (1.8.1) sha256=4a7f6929691dbec8b5209a0b373bc2614882b55fc5d2e447a21aaa691303d62f
@@ -236,13 +236,13 @@ CHECKSUMS
   rspec-expectations (3.13.5) sha256=33a4d3a1d95060aea4c94e9f237030a8f9eae5615e9bd85718fe3a09e4b58836
   rspec-mocks (3.13.7) sha256=0979034e64b1d7a838aaaddf12bf065ea4dc40ef3d4c39f01f93ae2c66c62b1c
   rspec-support (3.13.6) sha256=2e8de3702427eab064c9352fe74488cc12a1bfae887ad8b91cba480ec9f8afb2
-  rubocop (1.82.1) sha256=09f1a6a654a960eda767aebea33e47603080f8e9c9a3f019bf9b94c9cab5e273
+  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 (0.6.12897) sha256=0348ab8803c4c3646977fee298083ded9b7e74d5b34b50c567c63eb7e36eb286
+  sorbet-runtime (0.6.12903) sha256=c23968c0dcf5a5db57f32c003fe3db7fb588c168cdd57d92ea4dceaba063118a
   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=v4.2.1" alt="Go">
+    <img src="https://img.shields.io/github/v/tag/kreuzberg-dev/kreuzberg?label=Go&color=007ec6&filter=v4.2.3" 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/api_proxy.rb

@@ -6,9 +6,9 @@ module Kreuzberg
   # @example Start the server
   # @example With block
   module APIProxy
-    Error = Class.new(Kreuzberg::Errors::Error)
-    MissingBinaryError = Class.new(Error)
-    ServerError = Class.new(Error)
+    class Error < Kreuzberg::Errors::Error; end
+    class MissingBinaryError < Error; end
+    class ServerError < Error; end
 
     # API server instance
     class Server

data/lib/kreuzberg/cli_proxy.rb

@@ -5,8 +5,8 @@ require 'open3'
 module Kreuzberg
   # @example
   module CLIProxy
-    Error = Class.new(Kreuzberg::Errors::Error)
-    MissingBinaryError = Class.new(Error)
+    class Error < Kreuzberg::Errors::Error; end
+    class MissingBinaryError < Error; end
 
     # CLI execution error with stderr and exit status
     class CLIExecutionError < Error

data/lib/kreuzberg/config.rb

@@ -684,13 +684,6 @@ module Kreuzberg
     #   image = Config::ImageExtraction.new(extract_images: true, target_dpi: 600)
     #   config = Extraction.new(image_extraction: image)
     #
-    # @example With preprocessing
-    #   preprocessing = Config::ImagePreprocessing.new(
-    #     binarization_method: "sauvola",
-    #     denoise: true
-    #   )
-    #   config = Extraction.new(image_preprocessing: preprocessing)
-    #
     # @example With post-processing
     #   postprocessor = Config::PostProcessor.new(
     #     enabled: true,
@@ -708,14 +701,13 @@ module Kreuzberg
     #     language_detection: Config::LanguageDetection.new(enabled: true),
     #     pdf_options: Config::PDF.new(extract_images: true, passwords: ["secret"]),
     #     image_extraction: Config::ImageExtraction.new(target_dpi: 600),
-    #     image_preprocessing: Config::ImagePreprocessing.new(denoise: true),
     #     postprocessor: Config::PostProcessor.new(enabled: true)
     #   )
     #
     class Extraction
       attr_reader :use_cache, :enable_quality_processing, :force_ocr,
                   :ocr, :chunking, :language_detection, :pdf_options,
-                  :images, :image_preprocessing, :postprocessor,
+                  :images, :postprocessor,
                   :token_reduction, :keywords, :html_options, :pages,
                   :max_concurrent_extractions, :output_format, :result_format
 
@@ -739,7 +731,7 @@ module Kreuzberg
       # Keys that are allowed in the Extraction config
       ALLOWED_KEYS = %i[
         use_cache enable_quality_processing force_ocr ocr chunking
-        language_detection pdf_options image_extraction image_preprocessing
+        language_detection pdf_options image_extraction
         postprocessor token_reduction keywords html_options pages
         max_concurrent_extractions output_format result_format
       ].freeze
@@ -800,14 +792,13 @@ module Kreuzberg
 
       def initialize(hash = nil,
                      use_cache: true,
-                     enable_quality_processing: false,
+                     enable_quality_processing: true,
                      force_ocr: false,
                      ocr: nil,
                      chunking: nil,
                      language_detection: nil,
                      pdf_options: nil,
                      image_extraction: nil,
-                     image_preprocessing: nil,
                      postprocessor: nil,
                      token_reduction: nil,
                      keywords: nil,
@@ -820,7 +811,7 @@ module Kreuzberg
           use_cache: use_cache, enable_quality_processing: enable_quality_processing,
           force_ocr: force_ocr, ocr: ocr, chunking: chunking, language_detection: language_detection,
           pdf_options: pdf_options, image_extraction: image_extraction,
-          image_preprocessing: image_preprocessing, postprocessor: postprocessor,
+          postprocessor: postprocessor,
           token_reduction: token_reduction, keywords: keywords, html_options: html_options,
           pages: pages, max_concurrent_extractions: max_concurrent_extractions,
           output_format: output_format, result_format: result_format
@@ -846,7 +837,6 @@ module Kreuzberg
         @language_detection = normalize_config(params[:language_detection], LanguageDetection)
         @pdf_options = normalize_config(params[:pdf_options], PDF)
         @images = normalize_config(params[:image_extraction], ImageExtraction)
-        @image_preprocessing = normalize_config(params[:image_preprocessing], ImagePreprocessing)
         @postprocessor = normalize_config(params[:postprocessor], PostProcessor)
         @token_reduction = normalize_config(params[:token_reduction], TokenReduction)
         @keywords = normalize_config(params[:keywords], Keywords)
@@ -878,7 +868,6 @@ module Kreuzberg
       end
 
       # rubocop:disable Metrics/CyclomaticComplexity
-      # rubocop:disable Metrics/MethodLength
       def to_h
         {
           use_cache: @use_cache,
@@ -889,7 +878,6 @@ module Kreuzberg
           language_detection: @language_detection&.to_h,
           pdf_options: @pdf_options&.to_h,
           images: @images&.to_h,
-          image_preprocessing: @image_preprocessing&.to_h,
           postprocessor: @postprocessor&.to_h,
           token_reduction: @token_reduction&.to_h,
           keywords: @keywords&.to_h,
@@ -900,7 +888,6 @@ module Kreuzberg
           result_format: @result_format
         }.compact
       end
-      # rubocop:enable Metrics/MethodLength
       # rubocop:enable Metrics/CyclomaticComplexity
 
       # Serialize configuration to JSON string
@@ -1025,8 +1012,6 @@ module Kreuzberg
           @pdf_options = normalize_config(value, PDF)
         when :image_extraction
           @images = normalize_config(value, ImageExtraction)
-        when :image_preprocessing
-          @image_preprocessing = normalize_config(value, ImagePreprocessing)
         when :postprocessor
           @postprocessor = normalize_config(value, PostProcessor)
         when :token_reduction
@@ -1101,7 +1086,6 @@ module Kreuzberg
         @language_detection = merged.language_detection
         @pdf_options = merged.pdf_options
         @images = merged.image_extraction
-        @image_preprocessing = merged.image_preprocessing
         @postprocessor = merged.postprocessor
         @token_reduction = merged.token_reduction
         @keywords = merged.keywords

data/lib/kreuzberg/mcp_proxy.rb

@@ -6,9 +6,9 @@ require 'json'
 module Kreuzberg
   # @example Start MCP server
   module MCPProxy
-    Error = Class.new(Kreuzberg::Errors::Error)
-    MissingBinaryError = Class.new(Error)
-    ServerError = Class.new(Error)
+    class Error < Kreuzberg::Errors::Error; end
+    class MissingBinaryError < Error; end
+    class ServerError < Error; end
 
     # MCP server instance
     class Server

data/lib/kreuzberg/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Kreuzberg
-  VERSION = '4.2.1'
+  VERSION = '4.2.3'
 end

data/spec/binding/config_spec.rb

@@ -309,7 +309,7 @@ RSpec.describe Kreuzberg::Config do
       config = described_class.new
 
       expect(config.use_cache).to be true
-      expect(config.enable_quality_processing).to be false
+      expect(config.enable_quality_processing).to be true
       expect(config.force_ocr).to be false
       expect(config.ocr).to be_nil
       expect(config.chunking).to be_nil

data/spec/unit/config/extraction_config_spec.rb

@@ -6,7 +6,7 @@ RSpec.describe Kreuzberg::Config::Extraction do
       config = described_class.new
 
       expect(config.use_cache).to be true
-      expect(config.enable_quality_processing).to be false
+      expect(config.enable_quality_processing).to be true
       expect(config.force_ocr).to be false
       expect(config.ocr).to be_nil
       expect(config.chunking).to be_nil
@@ -103,7 +103,7 @@ RSpec.describe Kreuzberg::Config::Extraction do
       hash = config.to_h
 
       expect(hash[:use_cache]).to be true
-      expect(hash[:enable_quality_processing]).to be false
+      expect(hash[:enable_quality_processing]).to be true
       expect(hash[:force_ocr]).to be false
     end
   end

data/vendor/Cargo.toml

@@ -3,7 +3,7 @@ members = ["kreuzberg", "kreuzberg-tesseract", "kreuzberg-ffi"]
 resolver = "2"
 
 [workspace.package]
-version = "4.2.1"
+version = "4.2.3"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <nhirschfeld@gmail.com>"]

data/vendor/kreuzberg/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "kreuzberg"
-version = "4.2.1"
+version = "4.2.3"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <nhirschfeld@gmail.com>"]
@@ -71,7 +71,7 @@ keywords-yake = ["dep:yake-rust", "stopwords"]
 keywords-rake = ["dep:rake", "stopwords"]
 keywords = ["keywords-yake", "keywords-rake"]
 
-api = ["dep:axum", "dep:tower", "dep:tower-http", "tokio-runtime"]
+api = ["dep:axum", "dep:tower", "dep:tower-http", "dep:utoipa", "tokio-runtime"]
 mcp = ["dep:rmcp", "tokio-runtime"]
 mcp-http = ["mcp", "api"]
 
@@ -198,6 +198,7 @@ rake = { version = "0.3.6", optional = true }
 axum = { version = "0.8", features = ["macros", "json", "multipart"], optional = true }
 tower = { version = "0.5", optional = true }
 tower-http = { version = "0.6", features = ["cors", "trace", "limit"], optional = true }
+utoipa = { version = "5.4", features = ["axum_extras"], optional = true }
 rmcp = { version = "0.14.0", features = [
     "server",
     "macros",

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 4.2.1 Release**
+> **🚀 Version 4.2.3 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/api/error.rs

@@ -2,14 +2,67 @@
 
 use axum::{
     Json,
+    body::to_bytes,
+    extract::{FromRequest, Request, rejection::JsonRejection},
     http::StatusCode,
     response::{IntoResponse, Response},
 };
+use serde::de::DeserializeOwned;
 
 use crate::error::KreuzbergError;
 
 use super::types::ErrorResponse;
 
+/// Custom JSON extractor that returns JSON error responses instead of plain text.
+///
+/// This wraps axum's `Json` extractor but uses `ApiError` as the rejection type,
+/// ensuring that all JSON parsing errors are returned as JSON with proper content type.
+///
+/// Additionally, this extractor validates that the root JSON value is an object (not an array),
+/// which prevents serde from incorrectly deserializing JSON arrays into struct fields.
+#[derive(Debug, Clone, Copy, Default)]
+pub struct JsonApi<T>(pub T);
+
+impl<T, S> FromRequest<S> for JsonApi<T>
+where
+    T: DeserializeOwned,
+    S: Send + Sync,
+{
+    type Rejection = ApiError;
+
+    async fn from_request(req: Request, state: &S) -> Result<Self, Self::Rejection> {
+        // First, extract the body to check if it's a valid JSON object (not array)
+        let (parts, body) = req.into_parts();
+        let bytes = to_bytes(body, usize::MAX).await.map_err(|_| {
+            ApiError::new(
+                StatusCode::BAD_REQUEST,
+                KreuzbergError::Other("Failed to read request body".to_string()),
+            )
+        })?;
+
+        // Validate that the root JSON is an object, not an array
+        if !bytes.is_empty() {
+            let trimmed = std::str::from_utf8(&bytes).unwrap_or("").trim_start();
+            if trimmed.starts_with('[') {
+                return Err(ApiError::new(
+                    StatusCode::BAD_REQUEST,
+                    KreuzbergError::validation(
+                        "Expected JSON object, but received JSON array. \
+                         Please wrap your data in an object with appropriate fields.",
+                    ),
+                ));
+            }
+        }
+
+        // Reconstruct the request and use the standard Json extractor
+        let req = Request::from_parts(parts, axum::body::Body::from(bytes));
+        match Json::<T>::from_request(req, state).await {
+            Ok(Json(value)) => Ok(JsonApi(value)),
+            Err(rejection) => Err(ApiError::from(rejection)),
+        }
+    }
+}
+
 /// API-specific error wrapper.
 #[derive(Debug)]
 pub struct ApiError {
@@ -79,3 +132,39 @@ impl From<KreuzbergError> for ApiError {
         }
     }
 }
+
+impl From<JsonRejection> for ApiError {
+    fn from(rejection: JsonRejection) -> Self {
+        let (status, message) = match rejection {
+            JsonRejection::JsonDataError(err) => (
+                StatusCode::UNPROCESSABLE_ENTITY,
+                format!(
+                    "Failed to deserialize the JSON body into the target type: {}",
+                    err.body_text()
+                ),
+            ),
+            JsonRejection::JsonSyntaxError(err) => (
+                StatusCode::BAD_REQUEST,
+                format!("Failed to parse the request body as JSON: {}", err.body_text()),
+            ),
+            JsonRejection::MissingJsonContentType(_) => (
+                StatusCode::UNSUPPORTED_MEDIA_TYPE,
+                "Expected request with `Content-Type: application/json`".to_string(),
+            ),
+            JsonRejection::BytesRejection(err) => {
+                (StatusCode::BAD_REQUEST, format!("Failed to read request body: {}", err))
+            }
+            _ => (StatusCode::BAD_REQUEST, "Unknown JSON parsing error".to_string()),
+        };
+
+        Self {
+            status,
+            body: ErrorResponse {
+                error_type: "JsonParsingError".to_string(),
+                message,
+                traceback: None,
+                status_code: status.as_u16(),
+            },
+        }
+    }
+}