kreuzberg 4.9.7

High-performance document intelligence library for Rust. Extract text, metadata, and structured data from PDFs, Office documents, images, and 91+ formats and 248 programming languages via tree-sitter code intelligence 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

//! pdf_oxide backend for PDF extraction.
//!
//! Pure-Rust alternative to the pdfium-render backend. Provides text extraction,
//! metadata parsing, annotation extraction, image extraction, table detection,
//! and font metrics for heading hierarchy detection.

pub(crate) mod annotations;
pub(crate) mod hierarchy;
pub(crate) mod images;
pub(crate) mod metadata;
pub(crate) mod table;
pub(crate) mod text;

use crate::Result;
use crate::error::KreuzbergError;
use std::path::Path;

/// Wraps a [`pdf_oxide::PdfDocument`] with convenient constructors that map
/// pdf_oxide errors into [`KreuzbergError::Parsing`].
pub(crate) struct OxideDocument {
    pub doc: pdf_oxide::PdfDocument,
}

impl OxideDocument {
    /// Open a PDF from a file path.
    #[allow(dead_code)]
    pub fn open_file(path: &Path) -> Result<Self> {
        let doc = pdf_oxide::PdfDocument::open(path).map_err(|e| KreuzbergError::Parsing {
            message: format!("pdf_oxide: failed to open file: {e}"),
            source: None,
        })?;
        Ok(Self { doc })
    }

    /// Open a PDF from in-memory bytes.
    pub fn open_bytes(bytes: &[u8]) -> Result<Self> {
        let doc = pdf_oxide::PdfDocument::from_bytes(bytes.to_vec()).map_err(|e| KreuzbergError::Parsing {
            message: format!("pdf_oxide: failed to load bytes: {e}"),
            source: None,
        })?;
        Ok(Self { doc })
    }
}