kreuzberg 4.1.0 → 4.1.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: df75a873a0c3547b2e6e44fa1c04e939372f5c01839c1d81ea671414e5c98016
-  data.tar.gz: e4f60f88486c2807c6bf45959e6e6160d509ed096236a5d5bf0c70c26cf8f2f8
+  metadata.gz: 0c1c0519fb3a58c45ec553994bd982b4f284835bd35ea0758461f6f381accfd6
+  data.tar.gz: 161c18cfabdd20bdaa520abda521cb16072dcc00f5fd2e41152d9da4acdb9d08
 SHA512:
-  metadata.gz: fac3f1dc6ca132f71f8034536f3ca7b7df53542a760061932e447e5cde3906d61903bb9ade16e96b782de1b879be68c50ee59218c3b0ba908d26d515d90d4966
-  data.tar.gz: 79b7ab92c373b7a06d06fe324b3d06ad2a588c9e993a2c8cb23c474e3452d82914d2ac48eb9edc1b9475cbf337b4e09a160c242d3510b33364d7f511d264e36f
+  metadata.gz: 7c6e1768022dcfdef5eaaaa3557a8388e8ad45158a69ed022d852b07f7658cbb885ca7860fa32eda8f29b5f7ea4216f93033aad77614afb82578b9157ed92710
+  data.tar.gz: 96ca3f1f3c6d6f9ea6dc826f7704c1d741ffaab24f524f099c4eb294652211ac6b9f324874974bb24d417193e76b1bce951315330d6f1ccd8eb5b014fa7fc71f

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    kreuzberg (4.1.0)
+    kreuzberg (4.1.1)
 
 GEM
   remote: https://rubygems.org/
@@ -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 (4.1.1)
   language_server-protocol (3.17.0.5) sha256=fd1e39a51a28bf3eec959379985a72e296e9f9acfce46f6a79d31ca8760803cc
   lint_roller (1.1.0) sha256=2c0c845b632a7d172cb849cc90c1bce937a28c5c8ccccb50dfd46a485003cc87
   listen (3.10.0) sha256=c6e182db62143aeccc2e1960033bebe7445309c7272061979bb098d03760c9d2

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=v4.1.1" 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/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/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Kreuzberg
-  VERSION = '4.1.0'
+  VERSION = '4.1.1'
 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 = "4.1.1"
 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 = "4.1.1"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <nhirschfeld@gmail.com>"]
@@ -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",

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 4.1.1 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/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 = "4.1.1"
 edition = "2024"
 rust-version = "1.91"
 authors = ["Na'aman Hirschfeld <nhirschfeld@gmail.com>"]

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: kreuzberg
 version: !ruby/object:Gem::Version
-  version: 4.1.0
+  version: 4.1.1
 platform: ruby
 authors:
 - Na'aman Hirschfeld
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2026-01-22 00:00:00.000000000 Z
+date: 2026-01-23 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