kreuzberg 4.8.4 → 4.8.5

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 16deeaa47cb35ded0b844af72d43b74da5539084f21a79d17513be2da9ac2f0b
-  data.tar.gz: 64715b14cffac78a796853e9f5d9a2d0969427de9d59a243c87a5d20699dcce3
+  metadata.gz: 198e3a74bbf86420f51a7d5b8f80e5e110cc93f2e144fadfe129f8f90155fe40
+  data.tar.gz: e54cfaf29ae9b910f5cc365a8095298e312ef93098c185dafa497b80a5f5859b
 SHA512:
-  metadata.gz: e362717e5db0fad6a9494737e53c2444a84cb76fd274c70283a6650eef0891e9ced2af424b2ed9501eb749f21fcfb2ca3b4f8c7b336d1a248bb99f4a7e69131e
-  data.tar.gz: 5d05d862a170f0efe0f6f6a9867846bb3b000136f638b1efe6ddee5e94310dc92495a40a7dff204b4098f92989175e1691ba10897be884ca960477d48dcbc6ca
+  metadata.gz: e5d6e2d8a21b12d6984bd4fc3d4cdc7858b46d1d21e09e238c832da830f8329af93f33ac6997590a342b52f68c7448aa68d68b8a323397116682e1373be28d38
+  data.tar.gz: d5a7c6df3f4a90f76e8554fffcb3cba45e274521f2efc96ecf4c882228b8398ed4ec0cb3322ccf1090b615b3ecae5b19b8f947e779fb3bf1a72c6e101d4184f1

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=v4.8.4" alt="Go">
+    <img src="https://img.shields.io/github/v/tag/kreuzberg-dev/kreuzberg?label=Go&color=007ec6&filter=v4.8.5" 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#">
@@ -39,10 +39,13 @@
   <a href="https://github.com/kreuzberg-dev/kreuzberg/pkgs/container/kreuzberg">
     <img src="https://img.shields.io/badge/Docker-007ec6?logo=docker&logoColor=white" alt="Docker">
   </a>
+  <a href="https://artifacthub.io/packages/search?repo=kreuzberg">
+    <img src="https://img.shields.io/endpoint?url=https://artifacthub.io/badge/repository/kreuzberg" alt="Artifact Hub">
+  </a>
 
   <!-- Project Info -->
   <a href="https://github.com/kreuzberg-dev/kreuzberg/blob/main/LICENSE">
-    <img src="https://img.shields.io/badge/License-Elastic--2.0-blue.svg" alt="License">
+    <img src="https://img.shields.io/badge/License-MIT-007ec6" alt="License">
   </a>
   <a href="https://docs.kreuzberg.dev">
     <img src="https://img.shields.io/badge/docs-kreuzberg.dev-007ec6" alt="Documentation">
@@ -419,7 +422,7 @@ Contributions are welcome! See [Contributing Guide](https://github.com/kreuzberg
 
 ## License
 
-Elastic License 2.0 (ELv2) - see [LICENSE](../../LICENSE) for details.
+MIT License - see LICENSE file for details.
 
 ## Support
 

data/ext/kreuzberg_rb/native/Cargo.lock

@@ -326,9 +326,9 @@ dependencies = [
 
 [[package]]
 name = "axum"
-version = "0.8.8"
+version = "0.8.9"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "8b52af3cb4058c895d37317bb27508dccc8e5f2d39454016b297bf4a400597b8"
+checksum = "31b698c5f9a010f6573133b09e0de5408834d0c82f8d7475a89fc1867a71cd90"
 dependencies = [
  "axum-core",
  "axum-macros",
@@ -380,9 +380,9 @@ dependencies = [
 
 [[package]]
 name = "axum-macros"
-version = "0.5.0"
+version = "0.5.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "604fde5e028fea851ce1d8570bbdc034bec850d157f7569d10f347d06808c05c"
+checksum = "7aa268c23bfbbd2c4363b9cd302a4f504fb2a9dfe7e3451d66f35dd392e20aca"
 dependencies = [
  "proc-macro2",
  "quote",
@@ -766,7 +766,7 @@ checksum = "6f8d983286843e49675a4b7a2d174efe136dc93a18d69130dd18198a6c167601"
 dependencies = [
  "cfg-if",
  "cpufeatures 0.3.0",
- "rand_core 0.10.0",
+ "rand_core 0.10.1",
 ]
 
 [[package]]
@@ -1858,7 +1858,7 @@ dependencies = [
  "js-sys",
  "libc",
  "r-efi 6.0.0",
- "rand_core 0.10.0",
+ "rand_core 0.10.1",
  "wasip2",
  "wasip3",
  "wasm-bindgen",
@@ -2098,7 +2098,7 @@ dependencies = [
  "indicatif 0.17.11",
  "libc",
  "log",
- "rand 0.9.3",
+ "rand 0.9.4",
  "serde",
  "serde_json",
  "thiserror 2.0.18",
@@ -2117,7 +2117,7 @@ dependencies = [
  "indicatif 0.18.4",
  "libc",
  "log",
- "rand 0.9.3",
+ "rand 0.9.4",
  "serde",
  "serde_json",
  "thiserror 2.0.18",
@@ -2529,7 +2529,7 @@ dependencies = [
  "itertools 0.14.0",
  "nalgebra",
  "num",
- "rand 0.9.3",
+ "rand 0.9.4",
  "rand_distr",
 ]
 
@@ -2931,7 +2931,7 @@ dependencies = [
 
 [[package]]
 name = "kreuzberg-rb"
-version = "4.8.4"
+version = "4.8.5"
 dependencies = [
  "async-trait",
  "html-to-markdown-rs",
@@ -3138,9 +3138,9 @@ dependencies = [
 
 [[package]]
 name = "lru"
-version = "0.16.3"
+version = "0.16.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "a1dc47f592c06f33f8e3aea9591776ec7c9f9e4124778ff8a3c3b87159f7e593"
+checksum = "7f66e8d5d03f609abc3a39e6f08e4164ebf1447a732906d39eb9b99b7919ef39"
 dependencies = [
  "hashbrown 0.16.1",
 ]
@@ -3707,7 +3707,7 @@ dependencies = [
  "futures-util",
  "opentelemetry",
  "percent-encoding",
- "rand 0.9.3",
+ "rand 0.9.4",
  "thiserror 2.0.18",
  "tokio",
  "tokio-stream",
@@ -4074,7 +4074,7 @@ dependencies = [
  "bytes",
  "getrandom 0.3.4",
  "lru-slab",
- "rand 0.9.3",
+ "rand 0.9.4",
  "ring",
  "rustc-hash",
  "rustls",
@@ -4140,9 +4140,9 @@ dependencies = [
 
 [[package]]
 name = "rand"
-version = "0.9.3"
+version = "0.9.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "7ec095654a25171c2124e9e3393a930bddbffdc939556c914957a4c3e0a87166"
+checksum = "44c5af06bb1b7d3216d91932aed5265164bf384dc89cd6ba05cf59a35f5f76ea"
 dependencies = [
  "rand_chacha",
  "rand_core 0.9.5",
@@ -4156,7 +4156,7 @@ checksum = "d2e8e8bcc7961af1fdac401278c6a831614941f6164ee3bf4ce61b7edb162207"
 dependencies = [
  "chacha20",
  "getrandom 0.4.2",
- "rand_core 0.10.0",
+ "rand_core 0.10.1",
 ]
 
 [[package]]
@@ -4180,9 +4180,9 @@ dependencies = [
 
 [[package]]
 name = "rand_core"
-version = "0.10.0"
+version = "0.10.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "0c8d0fd677905edcbeedbf2edb6494d676f0e98d54d5cf9bda0b061cb8fb8aba"
+checksum = "63b8176103e19a2643978565ca18b50549f6101881c443590420e4dc998a3c69"
 
 [[package]]
 name = "rand_distr"
@@ -4191,7 +4191,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "6a8615d50dcf34fa31f7ab52692afec947c4dd0ab803cc87cb3b0b4570ff7463"
 dependencies = [
  "num-traits",
- "rand 0.9.3",
+ "rand 0.9.4",
 ]
 
 [[package]]
@@ -4227,7 +4227,7 @@ dependencies = [
  "num-traits",
  "paste",
  "profiling",
- "rand 0.9.3",
+ "rand 0.9.4",
  "rand_chacha",
  "simd_helpers",
  "thiserror 2.0.18",
@@ -4258,9 +4258,9 @@ checksum = "60a357793950651c4ed0f3f52338f53b2f809f32d83a07f72909fa13e4c6c1e3"
 
 [[package]]
 name = "rayon"
-version = "1.11.0"
+version = "1.12.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "368f01d005bf8fd9b1206fb6fa653e6c4a81ceb1466406b81792d87c5677a58f"
+checksum = "fb39b166781f92d482534ef4b4b1b2568f42613b53e5b6c160e24cfbfa30926d"
 dependencies = [
  "either",
  "rayon-core",
@@ -4625,9 +4625,9 @@ checksum = "f87165f0995f63a9fbeea62b64d10b4d9d8e78ec6d7d51fb2125fda7bb36788f"
 
 [[package]]
 name = "rustls-webpki"
-version = "0.103.11"
+version = "0.103.12"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "20a6af516fea4b20eccceaf166e8aa666ac996208e8a644ce3ef5aa783bc7cd4"
+checksum = "8279bb85272c9f10811ae6a6c547ff594d6a7f3c6c6b02ee9726d1d0dcfcdd06"
 dependencies = [
  "aws-lc-rs",
  "ring",
@@ -5378,7 +5378,7 @@ dependencies = [
  "monostate",
  "onig",
  "paste",
- "rand 0.9.3",
+ "rand 0.9.4",
  "rayon",
  "rayon-cond",
  "regex",

data/ext/kreuzberg_rb/native/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "kreuzberg-rb"
-version = "4.8.4"
+version = "4.8.5"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <naaman@kreuzberg.dev>"]

data/ext/kreuzberg_rb/native/src/plugins/ocr_backend.rs

@@ -111,6 +111,7 @@ impl OcrBackend for RubyOcrBackend {
                 uris: None,
                 ocr_internal_document: None,
                 structured_output: None,
+                llm_usage: None,
             })
         })
     }

data/ext/kreuzberg_rb/native/src/result.rs

@@ -712,6 +712,45 @@ pub fn extraction_result_to_ruby(ruby: &Ruby, result: RustExtractionResult) -> R
     }
     set_hash_entry(ruby, &hash, "processing_warnings", warnings_array.into_value_with(ruby))?;
 
+    // Convert LLM usage
+    if let Some(usages) = result.llm_usage {
+        let usage_array = ruby.ary_new();
+        for usage in usages {
+            let usage_hash = ruby.hash_new();
+            usage_hash.aset("model", usage.model.as_str())?;
+            usage_hash.aset("source", usage.source.as_str())?;
+            if let Some(input_tokens) = usage.input_tokens {
+                usage_hash.aset("input_tokens", input_tokens as i64)?;
+            } else {
+                usage_hash.aset("input_tokens", ruby.qnil().as_value())?;
+            }
+            if let Some(output_tokens) = usage.output_tokens {
+                usage_hash.aset("output_tokens", output_tokens as i64)?;
+            } else {
+                usage_hash.aset("output_tokens", ruby.qnil().as_value())?;
+            }
+            if let Some(total_tokens) = usage.total_tokens {
+                usage_hash.aset("total_tokens", total_tokens as i64)?;
+            } else {
+                usage_hash.aset("total_tokens", ruby.qnil().as_value())?;
+            }
+            if let Some(cost) = usage.estimated_cost {
+                usage_hash.aset("estimated_cost", ruby.float_from_f64(cost).into_value_with(ruby))?;
+            } else {
+                usage_hash.aset("estimated_cost", ruby.qnil().as_value())?;
+            }
+            if let Some(reason) = usage.finish_reason {
+                usage_hash.aset("finish_reason", reason.as_str())?;
+            } else {
+                usage_hash.aset("finish_reason", ruby.qnil().as_value())?;
+            }
+            usage_array.push(usage_hash)?;
+        }
+        set_hash_entry(ruby, &hash, "llm_usage", usage_array.into_value_with(ruby))?;
+    } else {
+        set_hash_entry(ruby, &hash, "llm_usage", ruby.qnil().as_value())?;
+    }
+
     // Convert annotations
     if let Some(annotations) = result.annotations {
         let annotations_array = ruby.ary_new();

data/lib/kreuzberg/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Kreuzberg
-  VERSION = '4.8.4'
+  VERSION = '4.8.5'
 end

data/sig/kreuzberg.rbs

@@ -18,7 +18,8 @@ module Kreuzberg
   type element_type = 'title' | 'narrative_text' | 'heading' | 'list_item' | 'table' | 'image' | 'page_break' | 'code_block' | 'block_quote' | 'footer' | 'header'
 
   # Bounding box coordinates for element positioning (T::Struct from types.rb)
-  class BoundingBox     attr_reader x0: Float
+  class BoundingBox
+    attr_reader x0: Float
     attr_reader y0: Float
     attr_reader x1: Float
     attr_reader y1: Float
@@ -28,7 +29,8 @@ module Kreuzberg
   end
 
   # Metadata for a semantic element (T::Struct from types.rb)
-  class ElementMetadata     attr_reader page_number: Integer?
+  class ElementMetadata
+    attr_reader page_number: Integer?
     attr_reader filename: String?
     attr_reader coordinates: BoundingBox?
     attr_reader element_index: Integer?
@@ -39,7 +41,8 @@ module Kreuzberg
   end
 
   # Semantic element extracted from document (T::Struct from types.rb)
-  class Element     attr_reader element_id: String
+  class Element
+    attr_reader element_id: String
     attr_reader element_type: String
     attr_reader text: String
     attr_reader metadata: ElementMetadata
@@ -49,7 +52,8 @@ module Kreuzberg
   end
 
   # Header/Heading metadata (T::Struct from types.rb)
-  class HeaderMetadata     attr_reader level: Integer
+  class HeaderMetadata
+    attr_reader level: Integer
     attr_reader text: String
     attr_reader id: String?
     attr_reader depth: Integer
@@ -60,7 +64,8 @@ module Kreuzberg
   end
 
   # Link metadata (T::Struct from types.rb)
-  class LinkMetadata     attr_reader href: String
+  class LinkMetadata
+    attr_reader href: String
     attr_reader text: String
     attr_reader title: String?
     attr_reader link_type: String
@@ -72,7 +77,8 @@ module Kreuzberg
   end
 
   # Image metadata (T::Struct from types.rb)
-  class ImageMetadata     attr_reader src: String
+  class ImageMetadata
+    attr_reader src: String
     attr_reader alt: String?
     attr_reader title: String?
     attr_reader dimensions: Array[Integer]?
@@ -84,7 +90,8 @@ module Kreuzberg
   end
 
   # Structured data metadata (T::Struct from types.rb)
-  class StructuredData     attr_reader data_type: String
+  class StructuredData
+    attr_reader data_type: String
     attr_reader raw_json: String
     attr_reader schema_type: String?
 
@@ -210,7 +217,8 @@ module Kreuzberg
   end
 
   # HTML metadata (T::Struct from types.rb)
-  class HtmlMetadata     attr_reader title: String?
+  class HtmlMetadata
+    attr_reader title: String?
     attr_reader description: String?
     attr_reader author: String?
     attr_reader copyright: String?
@@ -261,7 +269,8 @@ module Kreuzberg
   end
 
   # Extracted keyword with relevance metadata (T::Struct from types.rb)
-  class ExtractedKeyword     attr_reader text: String
+  class ExtractedKeyword
+    attr_reader text: String
     attr_reader score: Float
     attr_reader algorithm: String
     attr_reader positions: Array[Integer]?
@@ -271,15 +280,39 @@ module Kreuzberg
   end
 
   # Processing warning from a pipeline stage (T::Struct from types.rb)
-  class ProcessingWarning     attr_reader source: String
+  class ProcessingWarning
+    attr_reader source: String
     attr_reader message: String
 
     def initialize: (source: String, message: String) -> void
     def serialize: () -> Hash[Symbol, untyped]
   end
 
+  # LLM token usage from an LLM-assisted extraction step (T::Struct from types.rb)
+  class LlmUsage
+    attr_reader model: String
+    attr_reader source: String
+    attr_reader input_tokens: Integer?
+    attr_reader output_tokens: Integer?
+    attr_reader total_tokens: Integer?
+    attr_reader estimated_cost: Float?
+    attr_reader finish_reason: String?
+
+    def initialize: (
+      model: String,
+      source: String,
+      ?input_tokens: Integer?,
+      ?output_tokens: Integer?,
+      ?total_tokens: Integer?,
+      ?estimated_cost: Float?,
+      ?finish_reason: String?
+    ) -> void
+    def serialize: () -> Hash[Symbol, untyped]
+  end
+
   # Bounding box for document node positioning (T::Struct from types.rb)
-  class DocumentBoundingBox     attr_reader x0: Float
+  class DocumentBoundingBox
+    attr_reader x0: Float
     attr_reader y0: Float
     attr_reader x1: Float
     attr_reader y1: Float
@@ -289,7 +322,8 @@ module Kreuzberg
   end
 
   # Annotation for a document node (T::Struct from types.rb)
-  class DocumentAnnotation     attr_reader key: String
+  class DocumentAnnotation
+    attr_reader key: String
     attr_reader value: String
 
     def initialize: (key: String, value: String) -> void
@@ -297,7 +331,8 @@ module Kreuzberg
   end
 
   # Single node in the document structure tree (T::Struct from types.rb)
-  class DocumentNode     attr_reader id: String
+  class DocumentNode
+    attr_reader id: String
     attr_reader content: String
     attr_reader parent: Integer?
     attr_reader children: Array[Integer]
@@ -322,7 +357,8 @@ module Kreuzberg
   end
 
   # Structured document representation (T::Struct from types.rb)
-  class DocumentStructure     attr_reader nodes: Array[DocumentNode]
+  class DocumentStructure
+    attr_reader nodes: Array[DocumentNode]
 
     def initialize: (nodes: Array[DocumentNode]) -> void
     def serialize: () -> Hash[Symbol, untyped]
@@ -927,7 +963,10 @@ module Kreuzberg
     extracted_keywords: Array[extracted_keyword_hash]?,
     quality_score: Float?,
     processing_warnings: Array[processing_warning_hash]?,
-    annotations: Array[pdf_annotation_hash]?
+    annotations: Array[pdf_annotation_hash]?,
+    uris: Array[uri_hash]?,
+    children: Array[archive_entry_hash]?,
+    llm_usage: Array[llm_usage_hash]?
   }
 
   type extracted_keyword_hash = {
@@ -942,6 +981,29 @@ module Kreuzberg
     message: String
   }
 
+  type llm_usage_hash = {
+    model: String,
+    source: String,
+    input_tokens: Integer?,
+    output_tokens: Integer?,
+    total_tokens: Integer?,
+    estimated_cost: Float?,
+    finish_reason: String?
+  }
+
+  type uri_hash = {
+    url: String,
+    label: String?,
+    page: Integer?,
+    kind: String
+  }
+
+  type archive_entry_hash = {
+    path: String,
+    mime_type: String,
+    result: extraction_result_hash?
+  }
+
   type page_content_hash = {
     page_number: Integer,
     content: String,
@@ -1483,6 +1545,9 @@ module Kreuzberg
     attr_reader quality_score: Float?
     attr_reader processing_warnings: Array[ProcessingWarning]?
     attr_reader annotations: Array[PdfAnnotation]?
+    attr_reader uris: Array[uri_hash]?
+    attr_reader children: Array[archive_entry_hash]?
+    attr_reader llm_usage: Array[LlmUsage]?
 
     # PDF annotation extracted from a document page (Struct from result.rb)
     class PdfAnnotation
@@ -1521,6 +1586,11 @@ module Kreuzberg
     def parse_document_structure: (Hash[String, untyped]? document_data) -> DocumentStructure?
     def parse_extracted_keywords: (Array[extracted_keyword_hash]? keywords_data) -> Array[ExtractedKeyword]?
     def parse_processing_warnings: (Array[processing_warning_hash]? warnings_data) -> Array[ProcessingWarning]
+    def parse_uris: (Array[uri_hash]? uris_data) -> Array[uri_hash]?
+    def build_uri: (Hash[String, untyped] u_hash) -> uri_hash
+    def parse_children: (Array[untyped]? children_data) -> Array[archive_entry_hash]?
+    def build_archive_entry: (Hash[String, untyped] c_hash) -> archive_entry_hash
+    def parse_llm_usage: (Array[llm_usage_hash]? usage_data) -> Array[LlmUsage]?
     def get_value: (Hash[String | Symbol, untyped] hash, String key, ?untyped default) -> untyped
     def serialize_tables: () -> Array[table_hash]
     def serialize_chunks: () -> Array[chunk_hash]?

data/vendor/Cargo.toml

@@ -2,7 +2,7 @@
 members = ["kreuzberg", "kreuzberg-ffi", "kreuzberg-tesseract", "kreuzberg-paddle-ocr", "kreuzberg-pdfium-render"]
 
 [workspace.package]
-version = "4.8.4"
+version = "4.8.5"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <naaman@kreuzberg.dev>"]
@@ -32,8 +32,8 @@ html-to-markdown-rs = { version = "3.1.0", default-features = false }
 image = { version = "0.25.10", default-features = false }
 itertools = "0.14"
 js-sys = "0.3"
-kreuzberg = { path = "./crates/kreuzberg", version = "4.8.4", default-features = false }
-kreuzberg-ffi = { path = "./crates/kreuzberg-ffi", version = "4.8.4" }
+kreuzberg = { path = "./crates/kreuzberg", version = "4.8.5", default-features = false }
+kreuzberg-ffi = { path = "./crates/kreuzberg-ffi", version = "4.8.5" }
 lazy_static = "1.5.0"
 libc = "0.2.185"
 liter-llm = { version = "1.2", features = ["native-http", "tracing"], default-features = false }
@@ -47,7 +47,7 @@ ort = { version = "2.0.0-rc.12", features = ["std", "api-18"], default-features
 parking_lot = "0.12.5"
 pdf_oxide = { version = "0.3.30", default-features = false }
 pdfium-render = { package = "kreuzberg-pdfium-render", path = "crates/kreuzberg-pdfium-render", version = "4.3" }
-rayon = "1.11.0"
+rayon = "1.12.0"
 reqwest = { version = "0.13.2", default-features = false }
 serde = { version = "1.0.228", features = ["derive"] }
 serde_json = { version = "1.0.149" }
@@ -57,7 +57,7 @@ thiserror = "2.0.18"
 tokio = { version = "1.51.1", features = ["rt", "rt-multi-thread", "macros", "sync", "process", "fs", "time", "io-util"] }
 toml = "1.1.2"
 tracing = "0.1"
-tree-sitter-language-pack = { version = "1.4.2", features = ["serde"], default-features = false }
+tree-sitter-language-pack = { version = "1.6.0", features = ["serde"], default-features = false }
 wasm-bindgen = { version = "0.2", features = ["enable-interning"] }
 wasm-bindgen-futures = "0.4"
 web-sys = { version = "0.3", features = ["Blob", "File", "FileReader", "console", "TextDecoder", "ImageData", "Window", "Response"] }

data/vendor/kreuzberg/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "kreuzberg"
-version = "4.8.4"
+version = "4.8.5"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <naaman@kreuzberg.dev>"]
@@ -319,7 +319,7 @@ pdfium-render = { package = "kreuzberg-pdfium-render", path = "../kreuzberg-pdfi
 pulldown-cmark = { version = "0.13" }
 quick-xml = { version = "0.39.2", features = ["serialize"], optional = true }
 rake = { version = "0.3.6", optional = true }
-rayon = "1.11.0"
+rayon = "1.12.0"
 regex = "1.12.3"
 rmcp = { version = "1.4.0", features = [
     "server",
@@ -392,7 +392,7 @@ optional = true
 # Override getrandom to enable js feature for WASM targets
 # This is needed because ring/rustls (via ureq) depend on getrandom without js feature
 getrandom = { version = "0.4.2", features = ["wasm_js"] }
-tree-sitter-language-pack = { version = "1.4.2", features = ["wasm", "serde"], default-features = false, optional = true }
+tree-sitter-language-pack = { version = "1.6.0", features = ["serde"], default-features = false, optional = true }
 wasm-bindgen-rayon = { version = "1.3", optional = true }
 
 [build-dependencies]

data/vendor/kreuzberg/README.md

@@ -18,7 +18,7 @@ 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 4.8.4 Release**
+> **🚀 Version 4.8.5 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/api/handlers.rs

@@ -680,7 +680,7 @@ pub async fn extract_structured_handler(
     };
 
     // Run structured extraction on the extracted content
-    let structured_output = crate::llm::structured::extract_structured(&result.content, &structured_config)
+    let (structured_output, _usage) = crate::llm::structured::extract_structured(&result.content, &structured_config)
         .await
         .map_err(ApiError::internal)?;