kreuzberg 4.0.0.pre.rc.6 → 4.0.0.pre.rc.7

data/vendor/kreuzberg/Cargo.toml

@@ -19,12 +19,14 @@ crate-type = ["rlib"]
 [features]
 default = ["tokio-runtime"]
 
-tokio-runtime = []
 profiling = ["dep:pprof"]
 
+# Runtime features
+tokio-runtime = ["dep:tokio"]
+
 # Format extractors
 pdf = ["dep:pdfium-render", "dep:lopdf", "dep:image"]
-excel = ["dep:calamine", "dep:polars"]
+excel = ["dep:calamine", "dep:polars", "tokio-runtime"]
 office = [
     "dep:roxmltree",
     "dep:zip",
@@ -37,7 +39,8 @@ office = [
     "dep:rst_parser",
     "dep:fb2",
     "dep:typst-syntax",
-    "html",  # EPUB needs HTML parsing (zip + roxmltree + html-to-markdown-rs)
+    "html",               # EPUB needs HTML parsing (zip + roxmltree + html-to-markdown-rs)
+    "tokio-runtime",
 ]
 email = ["dep:mail-parser", "dep:msg_parser"]
 html = ["dep:html-to-markdown-rs"]
@@ -48,6 +51,7 @@ archives = ["dep:zip", "dep:tar", "dep:sevenz-rust"]
 ocr = [
     "dep:kreuzberg-tesseract",
     "dep:image",
+    "dep:tiff",
     "dep:fast_image_resize",
     "dep:ndarray",
     "dep:kamadak-exif",
@@ -55,7 +59,7 @@ ocr = [
 ]
 language-detection = ["dep:whatlang"]
 chunking = ["dep:text-splitter"]
-embeddings = ["dep:fastembed", "chunking"]
+embeddings = ["dep:fastembed", "dep:reqwest", "chunking", "tokio-runtime"]
 stopwords = [] # Stopwords for keyword extraction and token reduction
 quality = ["dep:unicode-normalization", "dep:chardetng", "dep:encoding_rs", "stopwords"]
 
@@ -65,12 +69,15 @@ keywords-rake = ["dep:rake", "stopwords"]
 keywords = ["keywords-yake", "keywords-rake"]
 
 # Server features
-api = ["dep:axum", "dep:tower", "dep:tower-http"]
-mcp = ["dep:rmcp"]
+api = ["dep:axum", "dep:tower", "dep:tower-http", "tokio-runtime"]
+mcp = ["dep:rmcp", "tokio-runtime"]
 
 # Observability features
 otel = ["dep:opentelemetry", "dep:opentelemetry_sdk", "dep:tracing-opentelemetry"]
 
+# WASM-compatible feature bundle
+wasm-target = ["html", "xml", "email", "language-detection", "chunking", "quality"]
+
 # Convenience bundles
 full = [
     "pdf",
@@ -114,11 +121,14 @@ toml = "0.9.8"
 mime_guess = "2.0"
 rmp-serde = "1.3"
 thiserror = { workspace = true }
-tokio = { workspace = true }
-uuid = { version = "1.19.0", features = ["v4"] }
+tokio = { workspace = true, optional = true }
+uuid = { version = "1.19.0", features = ["v4", "js"] }
 indexmap = "2.12.1"
 tracing = { workspace = true }
-reqwest = { workspace = true, default-features = false, features = ["json", "rustls-tls"] }
+reqwest = { workspace = true, default-features = false, features = [
+    "json",
+    "rustls-tls",
+], optional = true }
 # Format extractors (optional)
 pdfium-render = { version = "0.8.37", features = ["thread_safe", "image"], optional = true }
 lopdf = { version = "0.38.0", optional = true }
@@ -128,7 +138,7 @@ roxmltree = { version = "0.21.1", optional = true }
 zip = { version = "6.0.0", optional = true }
 mail-parser = { version = "0.11.1", optional = true }
 msg_parser = { version = "0.1.1", optional = true }
-html-to-markdown-rs = { version = "2.12.0", features = ["inline-images"], optional = true }
+html-to-markdown-rs = { version = "2.14.1", features = ["inline-images"], optional = true }
 quick-xml = { version = "0.38.4", features = ["serialize"], optional = true }
 tar = { version = "0.4.44", optional = true }
 sevenz-rust = { version = "0.6.1", optional = true }
@@ -143,7 +153,7 @@ fb2 = { version = "0.4", optional = true }
 typst-syntax = { version = "0.14", optional = true }
 
 # Processing features (optional)
-kreuzberg-tesseract = { version = "4.0.0-rc.6", optional = true }
+kreuzberg-tesseract = { version = "4.0.0-rc.7", optional = true }
 image = { workspace = true, default-features = false, features = [
     "png",
     "jpeg",
@@ -153,6 +163,7 @@ image = { workspace = true, default-features = false, features = [
     "gif",
     "rayon",
 ], optional = true }
+tiff = { version = "0.10", optional = true }
 fast_image_resize = { version = "5.4.0", optional = true }
 ndarray = { version = "0.17.1", optional = true }
 kamadak-exif = { version = "0.6.1", optional = true }
@@ -202,3 +213,6 @@ harness = false
 # Only build profiling tooling on non-Windows targets (pprof depends on Unix APIs)
 [target.'cfg(not(target_os = "windows"))'.dependencies]
 pprof = { version = "0.15.0", features = ["flamegraph"], optional = true }
+
+[target.'cfg(target_arch = "wasm32")'.dependencies]
+wasm-bindgen-rayon = "1.2"

data/vendor/kreuzberg/README.md

@@ -1,14 +1,19 @@
-# Kreuzberg (Rust Core)
+# Kreuzberg
+
+[![Rust](https://img.shields.io/crates/v/kreuzberg?label=Rust)](https://crates.io/crates/kreuzberg)
+[![Python](https://img.shields.io/pypi/v/kreuzberg?label=Python)](https://pypi.org/project/kreuzberg/)
+[![TypeScript](https://img.shields.io/npm/v/@kreuzberg/node?label=TypeScript)](https://www.npmjs.com/package/@kreuzberg/node)
+[![WASM](https://img.shields.io/npm/v/@kreuzberg/wasm?label=WASM)](https://www.npmjs.com/package/@kreuzberg/wasm)
+[![Ruby](https://img.shields.io/gem/v/kreuzberg?label=Ruby)](https://rubygems.org/gems/kreuzberg)
+[![Java](https://img.shields.io/maven-central/v/dev.kreuzberg/kreuzberg?label=Java)](https://central.sonatype.com/artifact/dev.kreuzberg/kreuzberg)
+[![Go](https://img.shields.io/github/v/tag/kreuzberg-dev/kreuzberg?label=Go)](https://pkg.go.dev/github.com/kreuzberg-dev/kreuzberg)
+[![C#](https://img.shields.io/nuget/v/Goldziher.Kreuzberg?label=C%23)](https://www.nuget.org/packages/Goldziher.Kreuzberg/)
 
-[![Crates.io](https://img.shields.io/crates/v/kreuzberg)](https://crates.io/crates/kreuzberg)
-[![PyPI](https://img.shields.io/pypi/v/kreuzberg)](https://pypi.org/project/kreuzberg/)
-[![npm](https://img.shields.io/npm/v/kreuzberg)](https://www.npmjs.com/package/kreuzberg)
-[![RubyGems](https://img.shields.io/gem/v/kreuzberg)](https://rubygems.org/gems/kreuzberg)
-[![docs.rs](https://docs.rs/kreuzberg/badge.svg)](https://docs.rs/kreuzberg)
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
-[![Documentation](https://img.shields.io/badge/docs-kreuzberg.dev-blue)](https://kreuzberg.dev)
+[![Documentation](https://img.shields.io/badge/docs-kreuzberg.dev-blue)](https://kreuzberg.dev/)
+[![Discord](https://img.shields.io/badge/Discord-Join%20our%20community-7289da)](https://discord.gg/pXxagNK2zN)
 
-High-performance document intelligence library for Rust. Extract text, metadata, and structured information from PDFs, Office documents, images, and 50+ formats.
+High-performance document intelligence library for Rust. Extract text, metadata, and structured information from PDFs, Office documents, images, and 56 formats.
 
 This is the core Rust library that powers the Python, TypeScript, and Ruby bindings.
 

data/vendor/kreuzberg/build.rs

@@ -12,6 +12,12 @@ fn main() {
 
     println!("cargo::rustc-check-cfg=cfg(coverage)");
 
+    // Skip pdfium linking if the pdf feature is not enabled
+    if !cfg!(feature = "pdf") {
+        tracing::debug!("PDF feature not enabled, skipping pdfium linking");
+        return;
+    }
+
     let (download_url, lib_name) = get_pdfium_url_and_lib(&target);
 
     let pdfium_dir = out_dir.join("pdfium");
@@ -59,7 +65,10 @@ fn main() {
 
     let lib_dir = pdfium_dir.join("lib");
     println!("cargo:rustc-link-search=native={}", lib_dir.display());
-    println!("cargo:rustc-link-lib=dylib={}", lib_name);
+
+    // WASM requires static linking
+    let link_type = if target.contains("wasm") { "static" } else { "dylib" };
+    println!("cargo:rustc-link-lib={}={}", link_type, lib_name);
 
     if target.contains("darwin") {
         println!("cargo:rustc-link-arg=-Wl,-rpath,@loader_path");
@@ -121,11 +130,12 @@ fn get_pdfium_url_and_lib(target: &str) -> (String, String) {
             .unwrap_or_else(|| get_latest_version("paulocoutinhox/pdfium-lib"));
         tracing::debug!("Using pdfium-lib version: {}", version);
 
-        let wasm_arch = if target.contains("wasm32") { "wasm32" } else { "wasm64" };
+        // WASM builds use a single 'wasm.tgz' asset regardless of architecture
+        // The archive contains both wasm32 and wasm64 if available
         return (
             format!(
-                "https://github.com/paulocoutinhox/pdfium-lib/releases/download/{}/pdfium-{}.tar.gz",
-                version, wasm_arch
+                "https://github.com/paulocoutinhox/pdfium-lib/releases/download/{}/wasm.tgz",
+                version
             ),
             "pdfium".to_string(),
         );
@@ -314,7 +324,6 @@ fn copy_lib_to_package(pdfium_dir: &Path, target: &str) {
             );
         }
 
-        // Also copy to target/{profile} for Java FFI (Maven expects it here)
         let simple_target_dir = workspace_root.join("target").join(&profile);
         if simple_target_dir != target_dir {
             fs::create_dir_all(&simple_target_dir).ok();
@@ -416,7 +425,9 @@ fn codesign_if_needed(target: &str, binary: &Path) {
 }
 
 fn runtime_library_info(target: &str) -> (String, &'static str) {
-    if target.contains("windows") {
+    if target.contains("wasm") {
+        ("libpdfium.a".to_string(), "lib")
+    } else if target.contains("windows") {
         ("pdfium.dll".to_string(), "bin")
     } else if target.contains("darwin") {
         ("libpdfium.dylib".to_string(), "lib")

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

@@ -1,3 +1,5 @@
+#![cfg(feature = "api")]
+
 //! REST API server for Kreuzberg document extraction.
 //!
 //! This module provides an Axum-based HTTP server for document extraction