kreuzberg 4.3.0

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
129
130
131
132
133
134
135
136
137

//! Metadata extraction from EPUB OPF files.
//!
//! Handles parsing of OPF (Open Packaging Format) files and extraction of
//! Dublin Core metadata following EPUB2 and EPUB3 standards.

use crate::Result;
use roxmltree;
use std::collections::BTreeMap;

/// Metadata extracted from OPF (Open Packaging Format) file
#[derive(Debug, Default, Clone)]
pub(super) struct OepbMetadata {
    pub(super) title: Option<String>,
    pub(super) creator: Option<String>,
    pub(super) date: Option<String>,
    pub(super) language: Option<String>,
    pub(super) identifier: Option<String>,
    pub(super) publisher: Option<String>,
    pub(super) subject: Option<String>,
    pub(super) description: Option<String>,
    pub(super) rights: Option<String>,
}

/// Extract metadata from EPUB OPF file
pub(super) fn extract_metadata(opf_xml: &str) -> Result<(OepbMetadata, BTreeMap<String, serde_json::Value>)> {
    let mut additional_metadata = BTreeMap::new();

    let (epub_metadata, _) = parse_opf(opf_xml)?;

    if let Some(ref identifier) = epub_metadata.identifier {
        additional_metadata.insert("identifier".to_string(), serde_json::json!(identifier.clone()));
    }

    if let Some(ref publisher) = epub_metadata.publisher {
        additional_metadata.insert("publisher".to_string(), serde_json::json!(publisher.clone()));
    }

    if let Some(ref subject) = epub_metadata.subject {
        additional_metadata.insert("subject".to_string(), serde_json::json!(subject.clone()));
    }

    if let Some(ref description) = epub_metadata.description {
        additional_metadata.insert("description".to_string(), serde_json::json!(description.clone()));
    }

    if let Some(ref rights) = epub_metadata.rights {
        additional_metadata.insert("rights".to_string(), serde_json::json!(rights.clone()));
    }

    Ok((epub_metadata, additional_metadata))
}

/// Parse OPF file and extract metadata and spine order
pub(super) fn parse_opf(xml: &str) -> Result<(OepbMetadata, Vec<String>)> {
    match roxmltree::Document::parse(xml) {
        Ok(doc) => {
            let root = doc.root();

            let mut metadata = OepbMetadata::default();
            let mut manifest: BTreeMap<String, String> = BTreeMap::new();
            let mut spine_order: Vec<String> = Vec::new();

            for node in root.descendants() {
                match node.tag_name().name() {
                    "title" => {
                        if let Some(text) = node.text() {
                            metadata.title = Some(text.trim().to_string());
                        }
                    }
                    "creator" => {
                        if let Some(text) = node.text() {
                            metadata.creator = Some(text.trim().to_string());
                        }
                    }
                    "date" => {
                        if let Some(text) = node.text() {
                            metadata.date = Some(text.trim().to_string());
                        }
                    }
                    "language" => {
                        if let Some(text) = node.text() {
                            metadata.language = Some(text.trim().to_string());
                        }
                    }
                    "identifier" => {
                        if let Some(text) = node.text() {
                            metadata.identifier = Some(text.trim().to_string());
                        }
                    }
                    "publisher" => {
                        if let Some(text) = node.text() {
                            metadata.publisher = Some(text.trim().to_string());
                        }
                    }
                    "subject" => {
                        if let Some(text) = node.text() {
                            metadata.subject = Some(text.trim().to_string());
                        }
                    }
                    "description" => {
                        if let Some(text) = node.text() {
                            metadata.description = Some(text.trim().to_string());
                        }
                    }
                    "rights" => {
                        if let Some(text) = node.text() {
                            metadata.rights = Some(text.trim().to_string());
                        }
                    }
                    "item" => {
                        if let Some(id) = node.attribute("id")
                            && let Some(href) = node.attribute("href")
                        {
                            manifest.insert(id.to_string(), href.to_string());
                        }
                    }
                    _ => {}
                }
            }

            for node in root.descendants() {
                if node.tag_name().name() == "itemref"
                    && let Some(idref) = node.attribute("idref")
                    && let Some(href) = manifest.get(idref)
                {
                    spine_order.push(href.clone());
                }
            }

            Ok((metadata, spine_order))
        }
        Err(e) => Err(crate::KreuzbergError::Parsing {
            message: format!("Failed to parse OPF file: {}", e),
            source: None,
        }),
    }
}