kreuzberg 4.6.0 → 4.6.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 203e9719bcf3cf2cda1252dcd7a5c5782e7b73936a304b626a351894d4fcd909
-  data.tar.gz: 2c02a45c882ef6b6b6935896e9334c46f012aaf8bc6f6669fa3c0110b67398e5
+  metadata.gz: 7f2441d44e9083d36f9f0e60b2c155624ad251f97c6f2ca629bd3ef33aeeeb46
+  data.tar.gz: b443884a5e4bfffa1bc916928ae97997e1ca21cdfb80e76b845c454c0aa67256
 SHA512:
-  metadata.gz: d3dde81c8c38b1ee99bed3cae32e477e4c8941d401c6449fc9c3eec3608a5b771b47c20ab3a9679ccf75059fed5e6c09f9d91eefed83a7d9dc59eebf7acb5626
-  data.tar.gz: e590247800d9752175985ee3b8ad0c89c5926f1afa0669a881cf476455c8514332880d30ff35d46a2836cf9cdc18b752296fb06a545f42129b548b5675180a71
+  metadata.gz: e238336b3ceae6d2bed4bd530d393f32473a7a88f68f7e28a6300ffe2f7cadc5aee0ac834fa5712223810f9cf1f7a68a363b7e96079060565e1c9e10af9f8114
+  data.tar.gz: 48eae2f77c78ef1b9794ac3dde62c785b32ab99253b6ab7ce8da9d8bba8a57f209fb5fc6353a0bc8f5771bd825fc3b3b29652149133bd0ba1dc29e58101185bb

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=v4.6.1" 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/ext/kreuzberg_rb/native/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "kreuzberg-rb"
-version = "4.6.0"
+version = "4.6.1"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <naaman@kreuzberg.dev>"]

data/ext/kreuzberg_rb/native/src/config/types.rs

@@ -875,6 +875,17 @@ pub fn parse_extraction_config(ruby: &Ruby, opts: Option<RHash>) -> Result<Extra
             config.force_ocr = bool::try_convert(val)?;
         }
 
+        if let Some(val) = get_kw(ruby, hash, "force_ocr_pages")
+            && val.equal(ruby.qnil()).ok() != Some(true)
+        {
+            let pages_array = magnus::RArray::try_convert(val)?;
+            let pages: Vec<usize> = pages_array
+                .into_iter()
+                .map(|v| usize::try_convert(v))
+                .collect::<Result<Vec<_>, _>>()?;
+            config.force_ocr_pages = Some(pages);
+        }
+
         if let Some(val) = get_kw(ruby, hash, "include_document_structure") {
             config.include_document_structure = bool::try_convert(val)?;
         }
@@ -1013,6 +1024,13 @@ pub fn parse_extraction_config(ruby: &Ruby, opts: Option<RHash>) -> Result<Extra
                 }
             };
         }
+
+        if let Some(val) = get_kw(ruby, hash, "extraction_timeout_secs")
+            && val.equal(ruby.qnil()).ok() != Some(true)
+        {
+            let secs = u64::try_convert(val)?;
+            config.extraction_timeout_secs = Some(secs);
+        }
     }
 
     Ok(config)

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/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Kreuzberg
-  VERSION = '4.6.0'
+  VERSION = '4.6.1'
 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/vendor/Cargo.toml

@@ -2,7 +2,7 @@
 members = ["kreuzberg", "kreuzberg-ffi", "kreuzberg-tesseract", "kreuzberg-paddle-ocr", "kreuzberg-pdfium-render"]
 
 [workspace.package]
-version = "4.6.0"
+version = "4.6.1"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <naaman@kreuzberg.dev>"]
@@ -30,8 +30,8 @@ html-to-markdown-rs = { version = "2.29.0", default-features = false }
 image = { version = "0.25.10", default-features = false }
 itertools = "0.14"
 js-sys = "0.3"
-kreuzberg = { path = "./crates/kreuzberg", version = "4.6.0", default-features = false }
-kreuzberg-ffi = { path = "./crates/kreuzberg-ffi", version = "4.6.0" }
+kreuzberg = { path = "./crates/kreuzberg", version = "4.6.1", default-features = false }
+kreuzberg-ffi = { path = "./crates/kreuzberg-ffi", version = "4.6.1" }
 lazy_static = "1.5.0"
 libc = "0.2.183"
 log = "0.4"

data/vendor/kreuzberg/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "kreuzberg"
-version = "4.6.0"
+version = "4.6.1"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <naaman@kreuzberg.dev>"]
@@ -53,7 +53,7 @@ office = [
 ]
 hwp = ["dep:cfb", "dep:flate2"]
 iwork = ["dep:zip", "dep:snap"]
-email = ["dep:mail-parser", "dep:cfb"]
+email = ["dep:mail-parser", "dep:cfb", "dep:outlook-pst", "dep:tempfile", "dep:chrono"]
 html = ["dep:html-to-markdown-rs"]
 xml = ["dep:quick-xml", "dep:roxmltree"]
 archives = ["dep:zip", "dep:tar", "dep:sevenz-rust2", "dep:flate2"]
@@ -238,6 +238,7 @@ bytes = { version = "1", features = ["serde"] }
 calamine = { version = "0.34.0", features = ["dates"], optional = true }
 cfb = { version = "0.14", optional = true }
 chardetng = { version = "0.1.17", optional = true }
+chrono = { version = "0.4", optional = true }
 dashmap = "6.1"
 dbase = { version = "0.7", optional = true }
 encoding_rs = { version = "0.8.35" }
@@ -287,6 +288,7 @@ ort = { version = "2.0.0-rc.12", default-features = false, features = [
     "ndarray",
     "api-18",
 ], optional = true }
+outlook-pst = { version = "1.2.0", optional = true }
 parking_lot = "0.12.5"
 pastey = "0.2"
 pdf_oxide = { version = "0.3.17", default-features = false, optional = true }
@@ -315,6 +317,7 @@ sha2 = { version = "0.10", optional = true }
 simdutf8 = { version = "0.1", optional = true }
 snap = { version = "1.1", optional = true }
 tar = { version = "0.4.45", optional = true }
+tempfile = { version = "3.27.0", optional = true }
 text-splitter = { version = "0.29.3", features = ["markdown"], optional = true }
 thiserror = "2.0.18"
 tiff = { version = "0.11", optional = true }

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 4.6.0 Release**
+> **🚀 Version 4.6.1 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

@@ -115,6 +115,7 @@ impl ApiError {
             KreuzbergError::Plugin { .. } => "PluginError",
             KreuzbergError::LockPoisoned(_) => "LockPoisonedError",
             KreuzbergError::UnsupportedFormat(_) => "UnsupportedFormatError",
+            KreuzbergError::Timeout { .. } => "TimeoutError",
             KreuzbergError::Other(_) => "Error",
         };
 

data/vendor/kreuzberg/src/api/openapi.rs

@@ -74,6 +74,12 @@ use utoipa::OpenApi;
             crate::types::extraction::ElementId,
             crate::types::extraction::ElementType,
             crate::types::extraction::BoundingBox,
+            crate::types::ocr_elements::OcrElement,
+            crate::types::ocr_elements::OcrBoundingGeometry,
+            crate::types::ocr_elements::OcrConfidence,
+            crate::types::ocr_elements::OcrRotation,
+            crate::types::ocr_elements::OcrElementLevel,
+            crate::types::ocr_elements::OcrElementConfig,
             crate::types::metadata::Metadata,
             crate::types::tables::Table,
             crate::types::page::PageContent,

data/vendor/kreuzberg/src/core/config/extraction/core.rs

@@ -47,6 +47,16 @@ pub struct ExtractionConfig {
     #[serde(default)]
     pub force_ocr: bool,
 
+    /// Force OCR on specific pages only (1-indexed page numbers, must be >= 1).
+    ///
+    /// When set, only the listed pages are OCR'd regardless of text layer quality.
+    /// Unlisted pages use native text extraction. Ignored when `force_ocr` is `true`.
+    /// Only applies to PDF documents. Duplicates are automatically deduplicated.
+    /// An `ocr` config is recommended for backend/language selection; defaults are used if absent.
+    #[serde(default)]
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub force_ocr_pages: Option<Vec<usize>>,
+
     /// Text chunking configuration (None = chunking disabled)
     #[serde(default)]
     pub chunking: Option<ChunkingConfig>,
@@ -89,6 +99,14 @@ pub struct ExtractionConfig {
     #[serde(default)]
     pub html_options: Option<html_to_markdown_rs::ConversionOptions>,
 
+    /// Default per-file timeout in seconds for batch extraction.
+    ///
+    /// When set, each file in a batch will be canceled after this duration
+    /// unless overridden by [`FileExtractionConfig::timeout_secs`].
+    /// `None` means no timeout (unbounded extraction time).
+    #[serde(default)]
+    pub extraction_timeout_secs: Option<u64>,
+
     /// Maximum concurrent extractions in batch operations (None = (num_cpus × 1.5).ceil()).
     ///
     /// Limits parallelism to prevent resource exhaustion when processing
@@ -201,6 +219,7 @@ impl Default for ExtractionConfig {
             enable_quality_processing: true,
             ocr: None,
             force_ocr: false,
+            force_ocr_pages: None,
             chunking: None,
             images: None,
             #[cfg(feature = "pdf")]
@@ -213,6 +232,7 @@ impl Default for ExtractionConfig {
             postprocessor: None,
             #[cfg(feature = "html")]
             html_options: None,
+            extraction_timeout_secs: None,
             max_concurrent_extractions: None,
             #[cfg(feature = "archives")]
             security_limits: None,
@@ -259,6 +279,7 @@ impl ExtractionConfig {
             ref enable_quality_processing,
             ref ocr,
             ref force_ocr,
+            ref force_ocr_pages,
             ref chunking,
             ref images,
             #[cfg(feature = "pdf")]
@@ -276,6 +297,7 @@ impl ExtractionConfig {
             ref include_document_structure,
             #[cfg(feature = "layout-detection")]
             ref layout,
+            ref timeout_secs,
         } = *overrides;
 
         let mut config = self.clone();
@@ -289,6 +311,9 @@ impl ExtractionConfig {
         if let Some(v) = force_ocr {
             config.force_ocr = *v;
         }
+        if let Some(v) = force_ocr_pages {
+            config.force_ocr_pages = Some(v.clone());
+        }
         if let Some(v) = chunking {
             config.chunking = Some(v.clone());
         }
@@ -332,6 +357,9 @@ impl ExtractionConfig {
         if let Some(v) = layout {
             config.layout = Some(v.clone());
         }
+        if let Some(v) = timeout_secs {
+            config.extraction_timeout_secs = Some(*v);
+        }
 
         config
     }

data/vendor/kreuzberg/src/core/config/extraction/file_config.rs

@@ -57,6 +57,10 @@ pub struct FileExtractionConfig {
     #[serde(skip_serializing_if = "Option::is_none")]
     pub force_ocr: Option<bool>,
 
+    /// Override force OCR pages for this file (1-indexed page numbers).
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub force_ocr_pages: Option<Vec<usize>>,
+
     /// Override chunking configuration for this file.
     #[serde(skip_serializing_if = "Option::is_none")]
     pub chunking: Option<ChunkingConfig>,
@@ -112,4 +116,12 @@ pub struct FileExtractionConfig {
     #[cfg(feature = "layout-detection")]
     #[serde(skip_serializing_if = "Option::is_none")]
     pub layout: Option<super::super::layout::LayoutDetectionConfig>,
+
+    /// Override per-file extraction timeout in seconds.
+    ///
+    /// When set, the extraction for this file will be canceled after the
+    /// specified duration. A timed-out file produces an error result without
+    /// affecting other files in the batch.
+    #[serde(skip_serializing_if = "Option::is_none")]
+    pub timeout_secs: Option<u64>,
 }

data/vendor/kreuzberg/src/core/extractor/batch.rs

@@ -63,11 +63,12 @@ where
     Ok(results.into_iter().map(|r| r.unwrap()).collect())
 }
 
-/// Run a single extraction task with semaphore gating, timing, and batch mode.
+/// Run a single extraction task with semaphore gating, timing, optional timeout, and batch mode.
 #[cfg(feature = "tokio-runtime")]
 async fn run_timed_extraction<F, Fut>(
     index: usize,
     semaphore: Arc<tokio::sync::Semaphore>,
+    timeout_secs: Option<u64>,
     extract_fn: F,
 ) -> (usize, Result<ExtractionResult>, u64)
 where
@@ -76,7 +77,23 @@ where
 {
     let _permit = semaphore.acquire().await.unwrap();
     let start = Instant::now();
-    let mut result = crate::core::batch_mode::with_batch_mode(extract_fn()).await;
+
+    let extraction_future = crate::core::batch_mode::with_batch_mode(extract_fn());
+
+    let mut result = match timeout_secs {
+        Some(secs) => match tokio::time::timeout(std::time::Duration::from_secs(secs), extraction_future).await {
+            Ok(inner) => inner,
+            Err(_elapsed) => {
+                let elapsed_ms = start.elapsed().as_millis() as u64;
+                Err(KreuzbergError::Timeout {
+                    elapsed_ms,
+                    limit_ms: secs * 1000,
+                })
+            }
+        },
+        None => extraction_future.await,
+    };
+
     let elapsed_ms = start.elapsed().as_millis() as u64;
 
     if let Ok(ref mut r) = result {
@@ -182,7 +199,8 @@ pub async fn batch_extract_file(
         async move {
             let (ref path, ref file_config) = items[index];
             let resolved = resolve_config(&cfg, file_config);
-            run_timed_extraction(index, sem, || {
+            let timeout = resolved.extraction_timeout_secs;
+            run_timed_extraction(index, sem, timeout, || {
                 let path = path.clone();
                 async move { extract_file(&path, None, &resolved).await }
             })
@@ -282,7 +300,8 @@ pub async fn batch_extract_bytes(
         async move {
             let (bytes, mime_type, file_config) = slots[index].lock().take().expect("batch item already consumed");
             let resolved = resolve_config(&cfg, &file_config);
-            run_timed_extraction(index, sem, || async move {
+            let timeout = resolved.extraction_timeout_secs;
+            run_timed_extraction(index, sem, timeout, || async move {
                 extract_bytes(&bytes, &mime_type, &resolved).await
             })
             .await

data/vendor/kreuzberg/src/core/mime.rs

@@ -35,7 +35,9 @@ pub const LEGACY_POWERPOINT_MIME_TYPE: &str = "application/vnd.ms-powerpoint";
 
 pub const EML_MIME_TYPE: &str = "message/rfc822";
 pub const MSG_MIME_TYPE: &str = "application/vnd.ms-outlook";
+pub const PST_MIME_TYPE: &str = "application/vnd.ms-outlook-pst";
 pub const JSON_MIME_TYPE: &str = "application/json";
+pub const JSONL_MIME_TYPE: &str = "application/x-ndjson";
 pub const YAML_MIME_TYPE: &str = "application/x-yaml";
 pub const TOML_MIME_TYPE: &str = "application/toml";
 pub const XML_MIME_TYPE: &str = "application/xml";
@@ -368,6 +370,11 @@ static FORMATS: &[FormatEntry] = &[
         mime_type: "application/csl+json",
         aliases: &[],
     },
+    FormatEntry {
+        extensions: &["jsonl", "ndjson"],
+        mime_type: "application/x-ndjson",
+        aliases: &["application/jsonl", "application/x-jsonlines"],
+    },
     FormatEntry {
         extensions: &["yaml", "yml"],
         mime_type: "application/x-yaml",
@@ -399,6 +406,11 @@ static FORMATS: &[FormatEntry] = &[
         mime_type: "application/vnd.ms-outlook",
         aliases: &[],
     },
+    FormatEntry {
+        extensions: &["pst"],
+        mime_type: "application/vnd.ms-outlook-pst",
+        aliases: &[],
+    },
     // ── Archives ────────────────────────────────────────────────────────
     FormatEntry {
         extensions: &["zip"],

data/vendor/kreuzberg/src/error.rs

@@ -124,6 +124,9 @@ pub enum KreuzbergError {
     #[error("Unsupported format: {0}")]
     UnsupportedFormat(String),
 
+    #[error("Extraction timed out after {elapsed_ms}ms (limit: {limit_ms}ms)")]
+    Timeout { elapsed_ms: u64, limit_ms: u64 },
+
     #[error("{0}")]
     Other(String),
 }

data/vendor/kreuzberg/src/extraction/mod.rs

@@ -22,6 +22,9 @@ pub mod archive;
 #[cfg(feature = "email")]
 pub mod email;
 
+#[cfg(feature = "email")]
+pub mod pst;
+
 #[cfg(any(feature = "excel", feature = "excel-wasm"))]
 pub mod excel;
 
@@ -77,6 +80,9 @@ pub use archive::{
 #[cfg(feature = "email")]
 pub use email::{build_email_text_output, extract_email_content, parse_eml_content, parse_msg_content};
 
+#[cfg(feature = "email")]
+pub use pst::extract_pst_messages;
+
 #[cfg(any(feature = "excel", feature = "excel-wasm"))]
 pub use excel::{excel_to_markdown, read_excel_bytes, read_excel_file};