kreuzberg 4.4.2

High-performance document intelligence library for Rust. Extract text, metadata, and structured data from PDFs, Office documents, images, and 75+ formats with async/sync APIs.
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128

//! Kreuzberg - High-Performance Document Intelligence Library
//!
//! Kreuzberg is a Rust-first document extraction library with language-agnostic plugin support.
//! It provides fast, accurate extraction from PDFs, images, Office documents, emails, and more.
//!
//! # Quick Start
//!
//! ```rust,no_run
//! use kreuzberg::{extract_file_sync, ExtractionConfig};
//!
//! # fn main() -> kreuzberg::Result<()> {
//! // Extract content from a file
//! let config = ExtractionConfig::default();
//! let result = extract_file_sync("document.pdf", None, &config)?;
//! println!("Extracted: {}", result.content);
//! # Ok(())
//! # }
//! ```
//!
//! # Architecture
//!
//! - **Core Module** (`core`): Main extraction orchestration, MIME detection, config loading
//! - **Plugin System**: Language-agnostic plugin architecture
//! - **Extractors**: Format-specific extraction (PDF, images, Office docs, email, etc.)
//! - **OCR**: Multiple OCR backend support (Tesseract, EasyOCR, PaddleOCR)
//!
//! # Features
//!
//! - Fast parallel processing with async/await
//! - Priority-based extractor selection
//! - Comprehensive MIME type detection (118+ file extensions)
//! - Configurable caching and quality processing
//! - Cross-language plugin support (Python, Node.js planned)

#![deny(unsafe_code)]

pub mod cache;
pub mod core;
pub mod error;
pub mod extraction;
pub mod extractors;
pub mod panic_context;
pub mod plugins;
pub mod text;
pub mod types;
pub mod utils;

#[cfg(feature = "api")]
pub mod api;

#[cfg(feature = "mcp")]
pub mod mcp;

#[cfg(feature = "chunking")]
pub mod chunking;

#[cfg(feature = "embeddings")]
pub mod embeddings;

#[cfg(feature = "ocr")]
pub mod image;

#[cfg(feature = "language-detection")]
pub mod language_detection;

// Note: `image` module (DPI, resize, preprocessing) requires full `ocr` feature
// due to fast_image_resize dependency. The `ocr` module requires tokio and native
// deps (JP2, JBIG2), so it stays `ocr`-only. WASM OCR uses the JS bridge instead.

#[cfg(feature = "stopwords")]
pub mod stopwords;

#[cfg(any(feature = "keywords-yake", feature = "keywords-rake"))]
pub mod keywords;

#[cfg(feature = "ocr")]
pub mod ocr;

#[cfg(any(feature = "paddle-ocr", feature = "embeddings"))]
pub mod ort_discovery;

#[cfg(feature = "paddle-ocr")]
pub mod paddle_ocr;

#[cfg(feature = "pdf")]
pub mod pdf;

pub use error::{KreuzbergError, Result};
pub use types::*;

#[cfg(feature = "tokio-runtime")]
pub use core::extractor::{batch_extract_bytes, batch_extract_file};
pub use core::extractor::{extract_bytes, extract_file};

pub use core::extractor::{batch_extract_bytes_sync, extract_bytes_sync};

#[cfg(feature = "tokio-runtime")]
pub use core::extractor::{batch_extract_file_sync, extract_file_sync};

pub use core::config::{
    ChunkerType, ChunkingConfig, EmbeddingConfig, EmbeddingModelType, ExtractionConfig, ImageExtractionConfig,
    LanguageDetectionConfig, OcrConfig, OutputFormat, PageConfig, PostProcessorConfig, TokenReductionConfig,
};

#[cfg(feature = "api")]
pub use core::server_config::ServerConfig;

#[cfg(feature = "pdf")]
pub use core::config::{HierarchyConfig, PdfConfig};

#[cfg(feature = "paddle-ocr")]
pub use paddle_ocr::{CacheStats, ModelManager, ModelPaths, PaddleLanguage, PaddleOcrBackend, PaddleOcrConfig};

pub use core::mime::{
    DOCX_MIME_TYPE, EXCEL_MIME_TYPE, HTML_MIME_TYPE, JSON_MIME_TYPE, MARKDOWN_MIME_TYPE, PDF_MIME_TYPE,
    PLAIN_TEXT_MIME_TYPE, POWER_POINT_MIME_TYPE, SupportedFormat, XML_MIME_TYPE, detect_mime_type,
    detect_mime_type_from_bytes, detect_or_validate, get_extensions_for_mime, list_supported_formats,
    validate_mime_type,
};

pub use core::formats::{KNOWN_FORMATS, is_valid_format_field};

pub use plugins::registry::{
    get_document_extractor_registry, get_ocr_backend_registry, get_post_processor_registry, get_validator_registry,
};

#[cfg(feature = "embeddings")]
pub use embeddings::{EMBEDDING_PRESETS, EmbeddingPreset, get_preset, list_presets};