kreuzberg 4.1.0 → 4.1.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: df75a873a0c3547b2e6e44fa1c04e939372f5c01839c1d81ea671414e5c98016
-  data.tar.gz: e4f60f88486c2807c6bf45959e6e6160d509ed096236a5d5bf0c70c26cf8f2f8
+  metadata.gz: 6550daabf58e5e396576e5a83c6a53f226e677f9c129920c9990bba309fbd7ba
+  data.tar.gz: 9595aa468666391d08a0962db589bbbc50d5bd1c8532e101efa234f6c523d7c5
 SHA512:
-  metadata.gz: fac3f1dc6ca132f71f8034536f3ca7b7df53542a760061932e447e5cde3906d61903bb9ade16e96b782de1b879be68c50ee59218c3b0ba908d26d515d90d4966
-  data.tar.gz: 79b7ab92c373b7a06d06fe324b3d06ad2a588c9e993a2c8cb23c474e3452d82914d2ac48eb9edc1b9475cbf337b4e09a160c242d3510b33364d7f511d264e36f
+  metadata.gz: 0dea911deebe061515dd4cbff2b76b3a7947c68f196fcc576001d42d80386f6c53f8ed63e0e4acb8e719ad6f95c21e689df7aef5f6cbbbc0d1c92ef96ddb673c
+  data.tar.gz: 0df091f80f7c73dda0c17d89d4aa0571cd01f0f2b697b187fd9bae28f8dbcf96cd2e3a269f9831a442b8cf46ce40608586d3d6a242d84bb394fe6056cba3b492

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    kreuzberg (4.1.0)
+    kreuzberg (4.1.2)
 
 GEM
   remote: https://rubygems.org/
@@ -121,7 +121,7 @@ GEM
       rubocop (~> 1.81)
     ruby-progressbar (1.13.0)
     securerandom (0.4.1)
-    sorbet-runtime (0.6.12894)
+    sorbet-runtime (0.6.12897)
     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 (4.1.0)
+  kreuzberg (4.1.2)
   language_server-protocol (3.17.0.5) sha256=fd1e39a51a28bf3eec959379985a72e296e9f9acfce46f6a79d31ca8760803cc
   lint_roller (1.1.0) sha256=2c0c845b632a7d172cb849cc90c1bce937a28c5c8ccccb50dfd46a485003cc87
   listen (3.10.0) sha256=c6e182db62143aeccc2e1960033bebe7445309c7272061979bb098d03760c9d2
@@ -242,7 +242,7 @@ CHECKSUMS
   rubocop-rspec (3.9.0) sha256=8fa70a3619408237d789aeecfb9beef40576acc855173e60939d63332fdb55e2
   ruby-progressbar (1.13.0) sha256=80fc9c47a9b640d6834e0dc7b3c94c9df37f08cb072b7761e4a71e22cff29b33
   securerandom (0.4.1) sha256=cc5193d414a4341b6e225f0cb4446aceca8e50d5e1888743fac16987638ea0b1
-  sorbet-runtime (0.6.12894) sha256=4f0cbe041d80dac973ec3a5a848679922074dd77cc19f46384b27a8b9ff4a90c
+  sorbet-runtime (0.6.12897) sha256=0348ab8803c4c3646977fee298083ded9b7e74d5b34b50c567c63eb7e36eb286
   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=v4.1.0" alt="Go">
+    <img src="https://img.shields.io/github/v/tag/kreuzberg-dev/kreuzberg?label=Go&color=007ec6&filter=v4.1.2" 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#">
@@ -86,10 +86,13 @@ gem 'kreuzberg'
 
 ### System Requirements
 
-- **Ruby 2.7+** required
+- **Ruby 3.2.0 or higher** required (including Ruby 4.x)
+- Ruby 4.0+ is fully supported with no code changes required
 - Optional: [ONNX Runtime](https://github.com/microsoft/onnxruntime/releases) version 1.22.x for embeddings support
 - Optional: [Tesseract OCR](https://github.com/tesseract-ocr/tesseract) for OCR functionality
 
+**Ruby 4.0 Compatibility:** Kreuzberg is fully compatible with Ruby 4.0 (released December 25, 2025) and all Ruby 4.x versions. All tests pass with 100% compatibility. The gem compiles without any breaking changes. Key Ruby 4.0 features like Ruby Box, ZJIT compiler, and Ractor improvements work seamlessly with Kreuzberg.
+
 
 
 ## Quick Start
@@ -202,9 +205,9 @@ puts "Processing time: #{result.metadata&.dig('processing_time')}ms"
 
 ## Features
 
-### Supported File Formats (56+)
+### Supported File Formats (57+)
 
-56 file formats across 8 major categories with intelligent format detection and comprehensive metadata extraction.
+57 file formats across 8 major categories with intelligent format detection and comprehensive metadata extraction.
 
 #### Office Documents
 
@@ -230,7 +233,7 @@ puts "Processing time: #{result.metadata&.dig('processing_time')}ms"
 |----------|---------|----------|
 | **Markup** | `.html`, `.htm`, `.xhtml`, `.xml`, `.svg` | DOM parsing, metadata (Open Graph, Twitter Card), link extraction |
 | **Structured Data** | `.json`, `.yaml`, `.yml`, `.toml`, `.csv`, `.tsv` | Schema detection, nested structures, validation |
-| **Text & Markdown** | `.txt`, `.md`, `.markdown`, `.rst`, `.org`, `.rtf` | CommonMark, GFM, reStructuredText, Org Mode |
+| **Text & Markdown** | `.txt`, `.md`, `.markdown`, `.djot`, `.rst`, `.org`, `.rtf` | CommonMark, GFM, Djot, reStructuredText, Org Mode |
 
 #### Email & Archives
 

data/ext/kreuzberg_rb/native/Cargo.toml

@@ -31,7 +31,7 @@ embeddings = ["kreuzberg/embeddings"]
 
 [dependencies]
 async-trait = "0.1.89"
-kreuzberg = { path = "../../../../../crates/kreuzberg", default-features = false, features = [
+kreuzberg = { path = "../../../vendor/kreuzberg", default-features = false, features = [
     "pdf",
     "excel",
     "office",
@@ -51,7 +51,7 @@ kreuzberg = { path = "../../../../../crates/kreuzberg", default-features = false
     "bundled-pdfium",
     "tokio-runtime",
 ] }
-kreuzberg-ffi = { path = "../../../../../crates/kreuzberg-ffi" }
+kreuzberg-ffi = { path = "../../../vendor/kreuzberg-ffi" }
 magnus = { git = "https://github.com/matsadler/magnus", rev = "f6db11769efb517427bf7f121f9c32e18b059b38", features = [
     "rb-sys",
 ] }

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

@@ -1025,8 +1025,10 @@ pub fn config_from_file(path: String) -> Result<RHash, Error> {
         .and_then(|v| magnus::RHash::try_convert(v).map_err(|_| validation_error("Config must be a Hash")))
 }
 
-/// Discover extraction config from current directory
+/// Discover extraction config from current directory or parent directories
 pub fn config_discover() -> Result<Value, Error> {
+    use std::path::PathBuf;
+
     let ruby = Ruby::get().expect("Ruby not initialized");
 
     // Search for config files in order of precedence
@@ -1038,19 +1040,27 @@ pub fn config_discover() -> Result<Value, Error> {
         (".kreuzbergrc", "json"),
     ];
 
-    for (name, format) in config_files {
-        if let Ok(content) = fs::read_to_string(name) {
-            let json_value: serde_json::Value = match format {
-                "toml" => toml::from_str(&content)
-                    .map_err(|e| validation_error(format!("Invalid TOML in {}: {}", name, e)))?,
-                "yaml" => serde_yaml_ng::from_str(&content)
-                    .map_err(|e| validation_error(format!("Invalid YAML in {}: {}", name, e)))?,
-                "json" => serde_json::from_str(&content)
-                    .map_err(|e| validation_error(format!("Invalid JSON in {}: {}", name, e)))?,
-                _ => unreachable!(),
-            };
-            return json_value_to_ruby(&ruby, &json_value);
+    // Start from current directory and search up to parent directories
+    let mut current_dir: Option<PathBuf> = std::env::current_dir().ok();
+
+    while let Some(dir) = current_dir {
+        for (name, format) in &config_files {
+            let config_path = dir.join(name);
+            if let Ok(content) = fs::read_to_string(&config_path) {
+                let json_value: serde_json::Value = match *format {
+                    "toml" => toml::from_str(&content)
+                        .map_err(|e| validation_error(format!("Invalid TOML in {}: {}", config_path.display(), e)))?,
+                    "yaml" => serde_yaml_ng::from_str(&content)
+                        .map_err(|e| validation_error(format!("Invalid YAML in {}: {}", config_path.display(), e)))?,
+                    "json" => serde_json::from_str(&content)
+                        .map_err(|e| validation_error(format!("Invalid JSON in {}: {}", config_path.display(), e)))?,
+                    _ => unreachable!(),
+                };
+                return json_value_to_ruby(&ruby, &json_value);
+            }
         }
+        // Move to parent directory
+        current_dir = dir.parent().map(|p| p.to_path_buf());
     }
 
     // Return nil if no config found

data/kreuzberg.gemspec

@@ -165,7 +165,7 @@ Gem::Specification.new do |spec|
   DESC
   spec.homepage = 'https://github.com/kreuzberg-dev/kreuzberg'
   spec.license = 'MIT'
-  spec.required_ruby_version = '>= 3.2.0'
+  spec.required_ruby_version = '>= 3.2.0', '< 5.0'
 
   spec.metadata = {
     'homepage_uri' => spec.homepage,

data/lib/kreuzberg/api_proxy.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require 'open3'
-require 'pathname'
 
 module Kreuzberg
   # @example Start the server

data/lib/kreuzberg/cli_proxy.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require 'open3'
-require 'pathname'
 
 module Kreuzberg
   # @example

data/lib/kreuzberg/config.rb

@@ -733,11 +733,42 @@ module Kreuzberg
       # @example Load from YAML
       #   config = Kreuzberg::Config::Extraction.from_file("config.yaml")
       #
+      # 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
+        postprocessor token_reduction keywords html_options pages
+        max_concurrent_extractions
+      ].freeze
+
+      # Aliases for backward compatibility
+      KEY_ALIASES = {
+        images: :image_extraction
+      }.freeze
+
       def self.from_file(path)
         hash = Kreuzberg._config_from_file_native(path)
-        new(**hash.transform_keys(&:to_sym))
+        new(**normalize_hash_keys(hash))
       end
 
+      # Normalize hash keys from native function
+      # - Converts string keys to symbols
+      # - Maps aliased keys to their canonical names
+      # - Filters out unknown keys
+      def self.normalize_hash_keys(hash)
+        symbolized = hash.transform_keys(&:to_sym)
+
+        # Apply key aliases
+        KEY_ALIASES.each do |from, to|
+          symbolized[to] = symbolized.delete(from) if symbolized.key?(from) && !symbolized.key?(to)
+        end
+
+        # Filter to only allowed keys
+        symbolized.slice(*ALLOWED_KEYS)
+      end
+
+      private_class_method :normalize_hash_keys
+
       # Discover configuration file in current or parent directories.
       #
       # Searches for kreuzberg.toml, kreuzberg.yaml, or kreuzberg.json in the current
@@ -755,7 +786,7 @@ module Kreuzberg
         hash = Kreuzberg._config_discover_native
         return nil if hash.nil?
 
-        new(**hash.transform_keys(&:to_sym))
+        new(**normalize_hash_keys(hash))
       end
 
       def initialize(

data/lib/kreuzberg/mcp_proxy.rb

@@ -1,7 +1,6 @@
 # frozen_string_literal: true
 
 require 'open3'
-require 'pathname'
 require 'json'
 
 module Kreuzberg

data/lib/kreuzberg/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Kreuzberg
-  VERSION = '4.1.0'
+  VERSION = '4.1.2'
 end

data/spec/fixtures/config.toml

@@ -21,7 +21,7 @@ extract_images = true
 passwords = ["secret", "backup"]
 extract_metadata = true
 
-[images]
+[image_extraction]
 extract_images = true
 target_dpi = 600
 max_image_dimension = 2000

data/spec/fixtures/config.yaml

@@ -23,7 +23,7 @@ pdf_options:
     - password2
   extract_metadata: true
 
-images:
+image_extraction:
   extract_images: true
   target_dpi: 300
   max_image_dimension: 4096

data/vendor/Cargo.toml

@@ -3,7 +3,7 @@ members = ["kreuzberg", "kreuzberg-tesseract", "kreuzberg-ffi"]
 resolver = "2"
 
 [workspace.package]
-version = "4.1.0"
+version = "4.1.2"
 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 = "4.1.0"
+version = "4.1.2"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <nhirschfeld@gmail.com>"]
@@ -142,7 +142,7 @@ mime_guess = "2.0"
 rmp-serde = "1.3"
 thiserror = { workspace = true }
 tokio = { workspace = true, optional = true }
-uuid = { version = "1.19.0", features = ["v4", "js"] }
+uuid = { version = "1.20.0", features = ["v4", "js"] }
 indexmap = "2.13.0"
 tracing = { workspace = true }
 pdfium-render = { package = "kreuzberg-pdfium-render", version = "0.9.0", features = [
@@ -174,7 +174,7 @@ rst_parser = { version = "0.4", optional = true }
 fb2 = { version = "0.4", optional = true }
 typst-syntax = { version = "0.14", optional = true }
 
-kreuzberg-tesseract = { path = "../kreuzberg-tesseract", optional = true }
+kreuzberg-tesseract = { path = "../kreuzberg-tesseract", version = "4.1", optional = true }
 image = { workspace = true, default-features = false, features = [
     "png",
     "jpeg",
@@ -198,7 +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 }
-rmcp = { version = "0.13.0", features = [
+rmcp = { version = "0.14.0", features = [
     "server",
     "macros",
     "base64",

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 4.1.0 Release**
+> **🚀 Version 4.1.2 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/core/mime.rs

@@ -57,6 +57,11 @@ static EXT_TO_MIME: Lazy<HashMap<&'static str, &'static str>> = Lazy::new(|| {
     m.insert("ods", OPENDOC_SPREADSHEET_MIME_TYPE);
 
     m.insert("pptx", POWER_POINT_MIME_TYPE);
+    m.insert(
+        "ppsx",
+        "application/vnd.openxmlformats-officedocument.presentationml.slideshow",
+    );
+    m.insert("pptm", "application/vnd.ms-powerpoint.presentation.macroEnabled.12");
     m.insert("ppt", LEGACY_POWERPOINT_MIME_TYPE);
 
     m.insert("docx", DOCX_MIME_TYPE);
@@ -180,6 +185,8 @@ static SUPPORTED_MIME_TYPES: Lazy<HashSet<&'static str>> = Lazy::new(|| {
 
     set.insert(PDF_MIME_TYPE);
     set.insert(POWER_POINT_MIME_TYPE);
+    set.insert("application/vnd.openxmlformats-officedocument.presentationml.slideshow"); // PPSX
+    set.insert("application/vnd.ms-powerpoint.presentation.macroEnabled.12"); // PPTM
     set.insert(LEGACY_WORD_MIME_TYPE);
     set.insert(LEGACY_POWERPOINT_MIME_TYPE);
     set.insert(HTML_MIME_TYPE);
@@ -459,6 +466,14 @@ mod tests {
             ("test.xlsx", EXCEL_MIME_TYPE),
             ("test.xls", EXCEL_BINARY_MIME_TYPE),
             ("test.pptx", POWER_POINT_MIME_TYPE),
+            (
+                "test.ppsx",
+                "application/vnd.openxmlformats-officedocument.presentationml.slideshow",
+            ),
+            (
+                "test.pptm",
+                "application/vnd.ms-powerpoint.presentation.macroEnabled.12",
+            ),
             ("test.ppt", LEGACY_POWERPOINT_MIME_TYPE),
             ("test.docx", DOCX_MIME_TYPE),
             ("test.doc", LEGACY_WORD_MIME_TYPE),

data/vendor/kreuzberg/src/extraction/pptx/parser.rs

@@ -60,9 +60,12 @@ fn parse_group(node: &Node) -> Result<Vec<SlideElement>> {
     match tag_name {
         "sp" => {
             let position = extract_position(node);
-            match parse_sp(node)? {
-                ParsedContent::Text(text) => elements.push(SlideElement::Text(text, position)),
-                ParsedContent::List(list) => elements.push(SlideElement::List(list, position)),
+            // parse_sp returns None for shapes without txBody (e.g., image placeholders)
+            if let Some(content) = parse_sp(node)? {
+                match content {
+                    ParsedContent::Text(text) => elements.push(SlideElement::Text(text, position)),
+                    ParsedContent::List(list) => elements.push(SlideElement::List(list, position)),
+                }
             }
         }
         "graphicFrame" => {
@@ -85,11 +88,17 @@ fn parse_group(node: &Node) -> Result<Vec<SlideElement>> {
     Ok(elements)
 }
 
-fn parse_sp(sp_node: &Node) -> Result<ParsedContent> {
-    let tx_body_node = sp_node
+fn parse_sp(sp_node: &Node) -> Result<Option<ParsedContent>> {
+    // Some shapes like image placeholders (<p:ph type="pic"/>) don't have txBody.
+    // These should be skipped gracefully - they contain no text to extract.
+    // GitHub Issue #321 Bug 1
+    let tx_body_node = match sp_node
         .children()
         .find(|n| n.tag_name().name() == "txBody" && n.tag_name().namespace() == Some(P_NAMESPACE))
-        .ok_or_else(|| KreuzbergError::parsing("No txBody found".to_string()))?;
+    {
+        Some(node) => node,
+        None => return Ok(None), // Skip shapes without txBody
+    };
 
     let is_list = tx_body_node.descendants().any(|n| {
         n.is_element()
@@ -103,9 +112,9 @@ fn parse_sp(sp_node: &Node) -> Result<ParsedContent> {
     });
 
     if is_list {
-        Ok(ParsedContent::List(parse_list(&tx_body_node)?))
+        Ok(Some(ParsedContent::List(parse_list(&tx_body_node)?)))
     } else {
-        Ok(ParsedContent::Text(parse_text(&tx_body_node)?))
+        Ok(Some(ParsedContent::Text(parse_text(&tx_body_node)?)))
     }
 }
 

data/vendor/kreuzberg/tests/pptx_regression_tests.rs

@@ -0,0 +1,504 @@
+//! Regression tests for PPTX/PPSX extraction bugs
+//!
+//! GitHub Issue #321: PPTX extraction fails on shapes without txBody (image placeholders) + PPSX not supported
+//!
+//! Bug 1: "No txBody found" - PPTX extraction fails when any shape lacks a text body
+//! Bug 2: PPSX not supported - PowerPoint Show files rejected entirely
+
+#![cfg(feature = "office")]
+
+use kreuzberg::{ExtractionConfig, extract_file};
+use std::io::Write;
+use tempfile::NamedTempFile;
+use zip::CompressionMethod;
+use zip::write::{FileOptions, ZipWriter};
+
+/// Test that PPSX (PowerPoint Show) files are extracted correctly.
+///
+/// PPSX files use MIME type `application/vnd.openxmlformats-officedocument.presentationml.slideshow`
+/// instead of PPTX's `application/vnd.openxmlformats-officedocument.presentationml.presentation`.
+///
+/// The internal structure is identical to PPTX - same slide XML format.
+///
+/// GitHub Issue #321 Bug 2
+#[tokio::test]
+async fn test_ppsx_slideshow_extraction() {
+    let workspace_root = std::path::Path::new(env!("CARGO_MANIFEST_DIR"))
+        .parent()
+        .unwrap()
+        .parent()
+        .unwrap();
+    let test_file = workspace_root.join("test_documents/presentations/sample.ppsx");
+
+    if !test_file.exists() {
+        println!("Skipping test: PPSX test file not found at {:?}", test_file);
+        return;
+    }
+
+    let result = extract_file(&test_file, None, &ExtractionConfig::default()).await;
+
+    match result {
+        Ok(extraction) => {
+            assert!(!extraction.content.is_empty(), "PPSX content should not be empty");
+            println!("✅ PPSX extraction succeeded!");
+            println!("   Content length: {} chars", extraction.content.len());
+            println!(
+                "   Content preview: {}",
+                &extraction.content[..extraction.content.len().min(200)]
+            );
+        }
+        Err(e) => {
+            panic!(
+                "PPSX extraction failed with error: {:?}\n\
+                 This is GitHub Issue #321 Bug 2: PPSX files should be supported.\n\
+                 PPSX MIME type (application/vnd.openxmlformats-officedocument.presentationml.slideshow) \
+                 needs to be added to extension-to-MIME mapping.",
+                e
+            );
+        }
+    }
+}
+
+/// Test that PPSX files can be extracted when MIME type is explicitly provided.
+///
+/// This validates that the PPTX extractor can handle PPSX content correctly
+/// (the XML structure is identical), even if MIME detection fails.
+///
+/// GitHub Issue #321 Bug 2
+#[tokio::test]
+async fn test_ppsx_with_explicit_mime_type() {
+    let workspace_root = std::path::Path::new(env!("CARGO_MANIFEST_DIR"))
+        .parent()
+        .unwrap()
+        .parent()
+        .unwrap();
+    let test_file = workspace_root.join("test_documents/presentations/sample.ppsx");
+
+    if !test_file.exists() {
+        println!("Skipping test: PPSX test file not found at {:?}", test_file);
+        return;
+    }
+
+    // Explicitly provide the PPSX MIME type
+    let result = extract_file(
+        &test_file,
+        Some("application/vnd.openxmlformats-officedocument.presentationml.slideshow"),
+        &ExtractionConfig::default(),
+    )
+    .await;
+
+    match result {
+        Ok(extraction) => {
+            assert!(!extraction.content.is_empty(), "PPSX content should not be empty");
+            println!("✅ PPSX extraction with explicit MIME type succeeded!");
+        }
+        Err(e) => {
+            panic!(
+                "PPSX extraction with explicit MIME type failed: {:?}\n\
+                 The PPTX extractor should handle PPSX content (identical XML structure).",
+                e
+            );
+        }
+    }
+}
+
+/// Test that PPTX files with image placeholder shapes (no txBody) are extracted correctly.
+///
+/// Some shapes in PPTX files, like image placeholders (`<p:ph type="pic"/>`), don't have
+/// `<p:txBody>` children because they're designed to hold images, not text.
+///
+/// The parser should skip shapes without txBody gracefully instead of failing.
+///
+/// GitHub Issue #321 Bug 1
+#[tokio::test]
+async fn test_pptx_with_image_placeholder_no_txbody() {
+    // Create a minimal PPTX with a shape that has no txBody (image placeholder)
+    let mut temp_file = NamedTempFile::with_suffix(".pptx").expect("Failed to create temp file");
+
+    {
+        let mut zip = ZipWriter::new(&mut temp_file);
+        let options: FileOptions<()> = FileOptions::default().compression_method(CompressionMethod::Stored);
+
+        // Add [Content_Types].xml
+        zip.start_file("[Content_Types].xml", options).unwrap();
+        zip.write_all(br#"<?xml version="1.0" encoding="UTF-8"?>
+<Types xmlns="http://schemas.openxmlformats.org/package/2006/content-types">
+  <Default Extension="rels" ContentType="application/vnd.openxmlformats-package.relationships+xml"/>
+  <Default Extension="xml" ContentType="application/xml"/>
+  <Override PartName="/ppt/presentation.xml" ContentType="application/vnd.openxmlformats-officedocument.presentationml.presentation.main+xml"/>
+  <Override PartName="/ppt/slides/slide1.xml" ContentType="application/vnd.openxmlformats-officedocument.presentationml.slide+xml"/>
+</Types>"#).unwrap();
+
+        // Add _rels/.rels
+        zip.start_file("_rels/.rels", options).unwrap();
+        zip.write_all(br#"<?xml version="1.0" encoding="UTF-8"?>
+<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">
+  <Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument" Target="ppt/presentation.xml"/>
+</Relationships>"#).unwrap();
+
+        // Add ppt/presentation.xml
+        zip.start_file("ppt/presentation.xml", options).unwrap();
+        zip.write_all(
+            br#"<?xml version="1.0" encoding="UTF-8"?>
+<p:presentation xmlns:p="http://schemas.openxmlformats.org/presentationml/2006/main"
+                xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships"
+                xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main">
+  <p:sldIdLst>
+    <p:sldId id="256" r:id="rId2"/>
+  </p:sldIdLst>
+</p:presentation>"#,
+        )
+        .unwrap();
+
+        // Add ppt/_rels/presentation.xml.rels
+        zip.start_file("ppt/_rels/presentation.xml.rels", options).unwrap();
+        zip.write_all(br#"<?xml version="1.0" encoding="UTF-8"?>
+<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">
+  <Relationship Id="rId2" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/slide" Target="slides/slide1.xml"/>
+</Relationships>"#).unwrap();
+
+        // Add ppt/slides/slide1.xml with a shape WITHOUT txBody (image placeholder)
+        // This is the critical test case - a <p:sp> element with no <p:txBody>
+        zip.start_file("ppt/slides/slide1.xml", options).unwrap();
+        zip.write_all(
+            br#"<?xml version="1.0" encoding="UTF-8"?>
+<p:sld xmlns:p="http://schemas.openxmlformats.org/presentationml/2006/main"
+       xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main"
+       xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships">
+  <p:cSld>
+    <p:spTree>
+      <p:nvGrpSpPr>
+        <p:cNvPr id="1" name=""/>
+        <p:cNvGrpSpPr/>
+        <p:nvPr/>
+      </p:nvGrpSpPr>
+      <p:grpSpPr>
+        <a:xfrm>
+          <a:off x="0" y="0"/>
+          <a:ext cx="0" cy="0"/>
+          <a:chOff x="0" y="0"/>
+          <a:chExt cx="0" cy="0"/>
+        </a:xfrm>
+      </p:grpSpPr>
+
+      <!-- Normal text shape WITH txBody - this should be extracted -->
+      <p:sp>
+        <p:nvSpPr>
+          <p:cNvPr id="2" name="Title"/>
+          <p:cNvSpPr/>
+          <p:nvPr/>
+        </p:nvSpPr>
+        <p:spPr>
+          <a:xfrm>
+            <a:off x="0" y="0"/>
+            <a:ext cx="100000" cy="100000"/>
+          </a:xfrm>
+          <a:prstGeom prst="rect"><a:avLst/></a:prstGeom>
+        </p:spPr>
+        <p:txBody>
+          <a:bodyPr/>
+          <a:lstStyle/>
+          <a:p>
+            <a:r>
+              <a:rPr lang="en-US"/>
+              <a:t>This is the title text</a:t>
+            </a:r>
+          </a:p>
+        </p:txBody>
+      </p:sp>
+
+      <!-- IMAGE PLACEHOLDER shape WITHOUT txBody - this caused the "No txBody found" error -->
+      <!-- This is a valid PPTX structure - image placeholders don't contain text -->
+      <p:sp>
+        <p:nvSpPr>
+          <p:cNvPr id="99" name="Image Placeholder"/>
+          <p:cNvSpPr>
+            <a:spLocks noGrp="1"/>
+          </p:cNvSpPr>
+          <p:nvPr>
+            <p:ph type="pic" idx="1"/>
+          </p:nvPr>
+        </p:nvSpPr>
+        <p:spPr>
+          <a:xfrm>
+            <a:off x="0" y="0"/>
+            <a:ext cx="100000" cy="100000"/>
+          </a:xfrm>
+          <a:prstGeom prst="rect"><a:avLst/></a:prstGeom>
+        </p:spPr>
+        <!-- NOTE: No <p:txBody> here - this is valid for image placeholders -->
+      </p:sp>
+
+      <!-- Another normal text shape - should also be extracted -->
+      <p:sp>
+        <p:nvSpPr>
+          <p:cNvPr id="3" name="Content"/>
+          <p:cNvSpPr/>
+          <p:nvPr/>
+        </p:nvSpPr>
+        <p:spPr>
+          <a:xfrm>
+            <a:off x="0" y="200000"/>
+            <a:ext cx="100000" cy="100000"/>
+          </a:xfrm>
+          <a:prstGeom prst="rect"><a:avLst/></a:prstGeom>
+        </p:spPr>
+        <p:txBody>
+          <a:bodyPr/>
+          <a:lstStyle/>
+          <a:p>
+            <a:r>
+              <a:rPr lang="en-US"/>
+              <a:t>Content after image placeholder</a:t>
+            </a:r>
+          </a:p>
+        </p:txBody>
+      </p:sp>
+
+    </p:spTree>
+  </p:cSld>
+</p:sld>"#,
+        )
+        .unwrap();
+
+        // Add ppt/slides/_rels/slide1.xml.rels (empty)
+        zip.start_file("ppt/slides/_rels/slide1.xml.rels", options).unwrap();
+        zip.write_all(
+            br#"<?xml version="1.0" encoding="UTF-8"?>
+<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">
+</Relationships>"#,
+        )
+        .unwrap();
+
+        zip.finish().unwrap();
+    }
+
+    // Extract the PPTX file
+    let result = extract_file(
+        temp_file.path(),
+        Some("application/vnd.openxmlformats-officedocument.presentationml.presentation"),
+        &ExtractionConfig::default(),
+    )
+    .await;
+
+    match result {
+        Ok(extraction) => {
+            assert!(!extraction.content.is_empty(), "Content should not be empty");
+
+            // Verify we extracted text from shapes that DO have txBody
+            assert!(
+                extraction.content.contains("title text"),
+                "Should extract text from first shape with txBody. Got: {}",
+                extraction.content
+            );
+            assert!(
+                extraction.content.contains("Content after"),
+                "Should extract text from shape after image placeholder. Got: {}",
+                extraction.content
+            );
+
+            println!("✅ PPTX with image placeholder (no txBody) extraction succeeded!");
+            println!("   Content: {}", extraction.content);
+        }
+        Err(e) => {
+            let error_msg = format!("{:?}", e);
+            if error_msg.contains("No txBody found") {
+                panic!(
+                    "PPTX extraction failed with 'No txBody found' error!\n\
+                     This is GitHub Issue #321 Bug 1.\n\
+                     The parser should skip shapes without txBody (image placeholders) \
+                     instead of failing.\n\
+                     Error: {:?}",
+                    e
+                );
+            } else {
+                panic!("PPTX extraction failed with unexpected error: {:?}", e);
+            }
+        }
+    }
+}
+
+/// Test extraction of PPTX with multiple shapes, some with txBody, some without.
+///
+/// This test verifies that:
+/// 1. Shapes WITH txBody are extracted
+/// 2. Shapes WITHOUT txBody (image placeholders, etc.) are skipped gracefully
+/// 3. The extraction continues and doesn't fail on the first shape without txBody
+///
+/// GitHub Issue #321 Bug 1
+#[tokio::test]
+async fn test_pptx_mixed_shapes_extraction() {
+    // Create a PPTX with multiple slides, each containing mixed shapes
+    let mut temp_file = NamedTempFile::with_suffix(".pptx").expect("Failed to create temp file");
+
+    {
+        let mut zip = ZipWriter::new(&mut temp_file);
+        let options: FileOptions<()> = FileOptions::default().compression_method(CompressionMethod::Stored);
+
+        // Add [Content_Types].xml
+        zip.start_file("[Content_Types].xml", options).unwrap();
+        zip.write_all(br#"<?xml version="1.0" encoding="UTF-8"?>
+<Types xmlns="http://schemas.openxmlformats.org/package/2006/content-types">
+  <Default Extension="rels" ContentType="application/vnd.openxmlformats-package.relationships+xml"/>
+  <Default Extension="xml" ContentType="application/xml"/>
+  <Override PartName="/ppt/presentation.xml" ContentType="application/vnd.openxmlformats-officedocument.presentationml.presentation.main+xml"/>
+  <Override PartName="/ppt/slides/slide1.xml" ContentType="application/vnd.openxmlformats-officedocument.presentationml.slide+xml"/>
+</Types>"#).unwrap();
+
+        // Add _rels/.rels
+        zip.start_file("_rels/.rels", options).unwrap();
+        zip.write_all(br#"<?xml version="1.0" encoding="UTF-8"?>
+<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">
+  <Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/officeDocument" Target="ppt/presentation.xml"/>
+</Relationships>"#).unwrap();
+
+        // Add ppt/presentation.xml
+        zip.start_file("ppt/presentation.xml", options).unwrap();
+        zip.write_all(
+            br#"<?xml version="1.0" encoding="UTF-8"?>
+<p:presentation xmlns:p="http://schemas.openxmlformats.org/presentationml/2006/main"
+                xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships"
+                xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main">
+  <p:sldIdLst>
+    <p:sldId id="256" r:id="rId2"/>
+  </p:sldIdLst>
+</p:presentation>"#,
+        )
+        .unwrap();
+
+        // Add ppt/_rels/presentation.xml.rels
+        zip.start_file("ppt/_rels/presentation.xml.rels", options).unwrap();
+        zip.write_all(br#"<?xml version="1.0" encoding="UTF-8"?>
+<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">
+  <Relationship Id="rId2" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/slide" Target="slides/slide1.xml"/>
+</Relationships>"#).unwrap();
+
+        // Add slide with various shapes - some with txBody, some without
+        zip.start_file("ppt/slides/slide1.xml", options).unwrap();
+        zip.write_all(
+            br#"<?xml version="1.0" encoding="UTF-8"?>
+<p:sld xmlns:p="http://schemas.openxmlformats.org/presentationml/2006/main"
+       xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main"
+       xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships">
+  <p:cSld>
+    <p:spTree>
+      <p:nvGrpSpPr>
+        <p:cNvPr id="1" name=""/>
+        <p:cNvGrpSpPr/>
+        <p:nvPr/>
+      </p:nvGrpSpPr>
+      <p:grpSpPr/>
+
+      <!-- Shape 1: Normal text -->
+      <p:sp>
+        <p:nvSpPr><p:cNvPr id="2" name="Title"/><p:cNvSpPr/><p:nvPr/></p:nvSpPr>
+        <p:spPr/>
+        <p:txBody>
+          <a:bodyPr/><a:lstStyle/>
+          <a:p><a:r><a:t>First Text Shape</a:t></a:r></a:p>
+        </p:txBody>
+      </p:sp>
+
+      <!-- Shape 2: Image placeholder (NO txBody) -->
+      <p:sp>
+        <p:nvSpPr>
+          <p:cNvPr id="10" name="Picture Placeholder"/>
+          <p:cNvSpPr><a:spLocks noGrp="1"/></p:cNvSpPr>
+          <p:nvPr><p:ph type="pic"/></p:nvPr>
+        </p:nvSpPr>
+        <p:spPr/>
+      </p:sp>
+
+      <!-- Shape 3: Another text shape -->
+      <p:sp>
+        <p:nvSpPr><p:cNvPr id="3" name="Body"/><p:cNvSpPr/><p:nvPr/></p:nvSpPr>
+        <p:spPr/>
+        <p:txBody>
+          <a:bodyPr/><a:lstStyle/>
+          <a:p><a:r><a:t>Second Text Shape</a:t></a:r></a:p>
+        </p:txBody>
+      </p:sp>
+
+      <!-- Shape 4: Chart placeholder (NO txBody) -->
+      <p:sp>
+        <p:nvSpPr>
+          <p:cNvPr id="11" name="Chart Placeholder"/>
+          <p:cNvSpPr><a:spLocks noGrp="1"/></p:cNvSpPr>
+          <p:nvPr><p:ph type="chart"/></p:nvPr>
+        </p:nvSpPr>
+        <p:spPr/>
+      </p:sp>
+
+      <!-- Shape 5: Content placeholder (NO txBody - empty) -->
+      <p:sp>
+        <p:nvSpPr>
+          <p:cNvPr id="12" name="Content Placeholder"/>
+          <p:cNvSpPr><a:spLocks noGrp="1"/></p:cNvSpPr>
+          <p:nvPr><p:ph type="body"/></p:nvPr>
+        </p:nvSpPr>
+        <p:spPr/>
+      </p:sp>
+
+      <!-- Shape 6: Final text shape -->
+      <p:sp>
+        <p:nvSpPr><p:cNvPr id="4" name="Footer"/><p:cNvSpPr/><p:nvPr/></p:nvSpPr>
+        <p:spPr/>
+        <p:txBody>
+          <a:bodyPr/><a:lstStyle/>
+          <a:p><a:r><a:t>Third Text Shape</a:t></a:r></a:p>
+        </p:txBody>
+      </p:sp>
+
+    </p:spTree>
+  </p:cSld>
+</p:sld>"#,
+        )
+        .unwrap();
+
+        // Add empty rels
+        zip.start_file("ppt/slides/_rels/slide1.xml.rels", options).unwrap();
+        zip.write_all(
+            br#"<?xml version="1.0" encoding="UTF-8"?>
+<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">
+</Relationships>"#,
+        )
+        .unwrap();
+
+        zip.finish().unwrap();
+    }
+
+    let result = extract_file(
+        temp_file.path(),
+        Some("application/vnd.openxmlformats-officedocument.presentationml.presentation"),
+        &ExtractionConfig::default(),
+    )
+    .await;
+
+    match result {
+        Ok(extraction) => {
+            // All three text shapes should be extracted
+            assert!(
+                extraction.content.contains("First Text Shape"),
+                "Should extract first text shape"
+            );
+            assert!(
+                extraction.content.contains("Second Text Shape"),
+                "Should extract second text shape (after image placeholder)"
+            );
+            assert!(
+                extraction.content.contains("Third Text Shape"),
+                "Should extract third text shape (after multiple placeholders)"
+            );
+
+            println!("✅ PPTX mixed shapes extraction succeeded!");
+            println!("   All text shapes extracted despite image/chart/content placeholders without txBody");
+        }
+        Err(e) => {
+            panic!(
+                "PPTX extraction failed: {:?}\n\
+                 Shapes without txBody should be skipped gracefully.",
+                e
+            );
+        }
+    }
+}

data/vendor/kreuzberg-tesseract/Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "kreuzberg-tesseract"
-version = "4.1.0"
+version = "4.1.2"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <nhirschfeld@gmail.com>"]
@@ -21,10 +21,10 @@ libc = { workspace = true }
 thiserror = { workspace = true }
 
 [dev-dependencies]
-image = { workspace = true }
+image = { workspace = true, features = ["png"] }
 
 [build-dependencies]
-cc = { version = "^1.2.53", optional = true }
+cc = { version = "^1.2.54", optional = true }
 cmake = { version = "0.1.57", optional = true }
 zip = { version = "7.2.0", optional = true }
 

data/vendor/kreuzberg-tesseract/build.rs

@@ -38,7 +38,7 @@ mod build_tesseract {
                 return None;
             }
         }
-        Some(path.join("tesseract-rs-cache"))
+        Some(path.join("kreuzberg-tesseract-cache"))
     }
 
     fn get_preferred_out_dir() -> PathBuf {
@@ -63,14 +63,14 @@ mod build_tesseract {
             PathBuf::from(home_dir)
                 .join("Library")
                 .join("Application Support")
-                .join("tesseract-rs")
+                .join("kreuzberg-tesseract")
         } else if cfg!(target_os = "linux") {
             let home_dir = env::var("HOME").unwrap_or_else(|_| {
                 env::var("USER")
                     .map(|user| format!("/home/{}", user))
                     .expect("Neither HOME nor USER environment variable set")
             });
-            PathBuf::from(home_dir).join(".tesseract-rs")
+            PathBuf::from(home_dir).join(".kreuzberg-tesseract")
         } else {
             panic!("Unsupported operating system");
         }
@@ -117,7 +117,7 @@ mod build_tesseract {
                     "cargo:warning=Failed to create cache dir {:?}: {}. Falling back to temp dir.",
                     preferred, err
                 );
-                let fallback = env::temp_dir().join("tesseract-rs-cache");
+                let fallback = env::temp_dir().join("kreuzberg-tesseract-cache");
                 fs::create_dir_all(&fallback).expect("Failed to create fallback cache directory in temp dir");
                 fallback
             }

data/vendor/kreuzberg-tesseract/src/lib.rs

@@ -6,16 +6,16 @@
 #![allow(clippy::not_unsafe_ptr_arg_deref)]
 #![allow(clippy::cmp_null)]
 
-//! # tesseract-rs
+//! # kreuzberg-tesseract
 //!
-//! `tesseract-rs` provides safe Rust bindings for Tesseract OCR with built-in compilation
+//! `kreuzberg-tesseract` provides safe Rust bindings for Tesseract OCR with built-in compilation
 //! of Tesseract and Leptonica libraries. This crate aims to make OCR functionality
 //! easily accessible in Rust projects while handling the complexity of interfacing
 //! with the underlying C++ libraries.
 //!
 //! ## Usage
 //!
-//! Here's a basic example of how to use `tesseract-rs`:
+//! Here's a basic example of how to use `kreuzberg-tesseract`:
 //!
 //! ```rust
 //! use std::path::PathBuf;
@@ -28,16 +28,16 @@
 //!         PathBuf::from(home_dir)
 //!             .join("Library")
 //!             .join("Application Support")
-//!             .join("tesseract-rs")
+//!             .join("kreuzberg-tesseract")
 //!             .join("tessdata")
 //!     } else if cfg!(target_os = "linux") {
 //!         let home_dir = std::env::var("HOME").expect("HOME environment variable not set");
 //!         PathBuf::from(home_dir)
-//!             .join(".tesseract-rs")
+//!             .join(".kreuzberg-tesseract")
 //!             .join("tessdata")
 //!     } else if cfg!(target_os = "windows") {
 //!         PathBuf::from(std::env::var("APPDATA").expect("APPDATA environment variable not set"))
-//!             .join("tesseract-rs")
+//!             .join("kreuzberg-tesseract")
 //!             .join("tessdata")
 //!     } else {
 //!         panic!("Unsupported operating system");

data/vendor/kreuzberg-tesseract/tests/integration_test.rs

@@ -7,7 +7,7 @@ fn get_default_tessdata_dir() -> PathBuf {
         PathBuf::from(home_dir)
             .join("Library")
             .join("Application Support")
-            .join("tesseract-rs")
+            .join("kreuzberg-tesseract")
             .join("tessdata")
     } else if cfg!(target_os = "linux") {
         let system_paths = [
@@ -20,10 +20,10 @@ fn get_default_tessdata_dir() -> PathBuf {
             }
         }
         let home_dir = std::env::var("HOME").expect("HOME environment variable not set");
-        PathBuf::from(home_dir).join(".tesseract-rs").join("tessdata")
+        PathBuf::from(home_dir).join(".kreuzberg-tesseract").join("tessdata")
     } else if cfg!(target_os = "windows") {
         PathBuf::from(std::env::var("APPDATA").expect("APPDATA environment variable not set"))
-            .join("tesseract-rs")
+            .join("kreuzberg-tesseract")
             .join("tessdata")
     } else {
         panic!("Unsupported operating system");

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: kreuzberg
 version: !ruby/object:Gem::Version
-  version: 4.1.0
+  version: 4.1.2
 platform: ruby
 authors:
 - Na'aman Hirschfeld
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-01-22 00:00:00.000000000 Z
+date: 2026-01-25 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -761,6 +761,7 @@ files:
 - vendor/kreuzberg/tests/plugin_postprocessor_test.rs
 - vendor/kreuzberg/tests/plugin_system.rs
 - vendor/kreuzberg/tests/plugin_validator_test.rs
+- vendor/kreuzberg/tests/pptx_regression_tests.rs
 - vendor/kreuzberg/tests/registry_integration_tests.rs
 - vendor/kreuzberg/tests/rst_extractor_tests.rs
 - vendor/kreuzberg/tests/rtf_extractor_tests.rs
@@ -790,6 +791,9 @@ required_ruby_version: !ruby/object:Gem::Requirement
   - - ">="
     - !ruby/object:Gem::Version
       version: 3.2.0
+  - - "<"
+    - !ruby/object:Gem::Version
+      version: '5.0'
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="