pdfluent_extract/
error.rs

1//! Error types for `pdf-extract` operations.
2
3// Copyright (c) 2026 Innovation Trigger B.V.
4//
5// PDFluent is available under two licences, at your option: the GNU AGPLv3, or
6// the PDFluent Commercial Licence. See the LICENSE file in this repository --
7// that file travels with the copy you received, which a URL does not.
8
9use thiserror::Error;
10
11/// Errors returned by the `pdf-extract` crate while reading text or images
12/// from a PDF.
13///
14/// Customers usually see these wrapped inside `pdfluent::Error` after they
15/// bubble up through the facade. The inner variant tells the caller what
16/// kind of input prevented extraction.
17#[derive(Debug, Error)]
18pub enum ExtractError {
19    /// The underlying PDF byte stream could not be parsed.
20    ///
21    /// Typically a malformed cross-reference table, a truncated stream, or
22    /// an unsupported PDF construct. The wrapped `lopdf::Error` has the
23    /// detail; surface it to the user when reporting the problem.
24    #[error("PDF error: {0}")]
25    Pdf(#[from] lopdf::Error),
26
27    /// An I/O error occurred while reading the document or writing extracted
28    /// output (for image extraction targets).
29    #[error("IO error: {0}")]
30    Io(#[from] std::io::Error),
31
32    /// The caller asked for a page index that does not exist in the
33    /// document.
34    ///
35    /// Fields: `(requested_page, total_pages)`. Page numbers are 1-based at
36    /// the public API; this variant carries the same convention.
37    #[error("page {0} out of range (document has {1} pages)")]
38    PageOutOfRange(u32, u32),
39
40    /// The image embedded on the page uses a PDF stream filter that this
41    /// crate cannot decode (for example uncommon JPX profiles or unknown
42    /// custom filters). The wrapped string identifies the filter by its
43    /// PDF name.
44    #[error("unsupported image filter: {0}")]
45    UnsupportedFilter(String),
46
47    /// Image bytes were extracted but could not be decoded by the image
48    /// backend (corrupt JPEG/PNG/JPX payload, invalid colorspace, etc.).
49    #[error("image decode error: {0}")]
50    ImageDecode(String),
51
52    /// A non-categorised extraction failure. Reserved for cases the more
53    /// specific variants do not cover; the message describes the situation.
54    #[error("{0}")]
55    Other(String),
56}
57
58/// Convenience `Result` alias for fallible `pdf-extract` operations.
59pub type Result<T> = std::result::Result<T, ExtractError>;
pdfluent_extract/error.rs

pdfluent_extract/
error.rs
