pdfni 0.1.0

Extract tables and Markdown from text-embedded PDFs, with a built-in pure-Rust PDF reader adapted from Mozilla pdf.js.
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
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153

//! wasm から JS へ公開する入口

use crate::extract::{ExtractOptions, PagePart, extract_from_parts};
use serde::Deserialize;
use wasm_bindgen::prelude::*;

/// JS から受け取る入力
#[derive(Deserialize)]
struct Input {
    pages: Vec<PagePart>,
    #[serde(default)]
    options: ExtractOptions,
}

/// 文字と線分の JSON から表抽出し Document の JSON を返す
#[wasm_bindgen]
pub fn extract_tables(input_json: &str) -> Result<String, JsError> {
    let input: Input =
        serde_json::from_str(input_json).map_err(|e| JsError::new(&e.to_string()))?;
    let doc = extract_from_parts(input.pages, &input.options);
    serde_json::to_string(&doc).map_err(|e| JsError::new(&e.to_string()))
}

/// PDF バイト列から純Rustリーダーで表抽出し Document の JSON を返す
#[cfg(feature = "reader")]
#[wasm_bindgen]
pub fn extract_tables_from_pdf(
    bytes: &[u8],
    options_json: Option<String>,
    password: Option<String>,
) -> Result<String, JsError> {
    let options: ExtractOptions = match options_json {
        Some(s) if !s.is_empty() => {
            serde_json::from_str(&s).map_err(|e| JsError::new(&e.to_string()))?
        }
        _ => ExtractOptions::default(),
    };
    let doc = crate::reader::extract_from_bytes(bytes, password.as_deref(), &options)
        .map_err(|e| JsError::new(&e.to_string()))?;
    serde_json::to_string(&doc).map_err(|e| JsError::new(&e.to_string()))
}

/// options_json から読み取り層の設定を取り出す
///
/// 検出系の項目は無視される。None・空文字は既定値
#[cfg(feature = "reader")]
fn reader_settings_from_json(
    options_json: Option<String>,
) -> Result<crate::extract::ReaderSettings, JsError> {
    match options_json {
        Some(s) if !s.is_empty() => {
            let options: ExtractOptions =
                serde_json::from_str(&s).map_err(|e| JsError::new(&e.to_string()))?;
            Ok(options.reader)
        }
        _ => Ok(crate::extract::ReaderSettings::default()),
    }
}

/// PDF バイト列から座標付き文字を抽出し TextDoc の JSON を返す
#[cfg(feature = "reader")]
#[wasm_bindgen]
pub fn extract_text_from_pdf(
    bytes: &[u8],
    password: Option<String>,
    options_json: Option<String>,
) -> Result<String, JsError> {
    let reader = reader_settings_from_json(options_json)?;
    let doc = crate::reader::extract_text_from_bytes_with(bytes, password.as_deref(), &reader)
        .map_err(|e| JsError::new(&e.to_string()))?;
    serde_json::to_string(&doc).map_err(|e| JsError::new(&e.to_string()))
}

/// PDF バイト列から TextItem 互換ビューを抽出しページごとの items の JSON を返す
///
/// bidi は既定オフ。有効化は Rust の `build_text_items_with` か文書入口の options を使う
#[cfg(feature = "reader")]
#[wasm_bindgen]
pub fn extract_text_items_from_pdf(
    bytes: &[u8],
    password: Option<String>,
    options_json: Option<String>,
) -> Result<String, JsError> {
    let reader = reader_settings_from_json(options_json)?;
    let doc = crate::reader::extract_text_from_bytes_with(bytes, password.as_deref(), &reader)
        .map_err(|e| JsError::new(&e.to_string()))?;
    let pages: Vec<Vec<crate::text::TextItem>> = doc
        .pages
        .iter()
        .map(crate::text::build_text_items)
        .collect();
    serde_json::to_string(&pages).map_err(|e| JsError::new(&e.to_string()))
}

/// PDF バイト列から行・単語ビューを抽出しページごとの lines の JSON を返す
///
/// bidi は既定オフ。有効化は Rust の `build_text_lines_with` か文書入口の options を使う
#[cfg(feature = "reader")]
#[wasm_bindgen]
pub fn extract_text_lines_from_pdf(
    bytes: &[u8],
    password: Option<String>,
    options_json: Option<String>,
) -> Result<String, JsError> {
    let reader = reader_settings_from_json(options_json)?;
    let doc = crate::reader::extract_text_from_bytes_with(bytes, password.as_deref(), &reader)
        .map_err(|e| JsError::new(&e.to_string()))?;
    let pages: Vec<Vec<crate::text::TextLine>> = doc
        .pages
        .into_iter()
        .map(|pg| crate::text::build_text_lines(&pg))
        .collect();
    serde_json::to_string(&pages).map_err(|e| JsError::new(&e.to_string()))
}

/// PDF バイト列から文書モデルを抽出し DocDoc の JSON を返す
#[cfg(feature = "document")]
#[wasm_bindgen]
pub fn extract_document_from_pdf(
    bytes: &[u8],
    options_json: Option<String>,
    password: Option<String>,
) -> Result<String, JsError> {
    let options: ExtractOptions = match options_json {
        Some(s) if !s.is_empty() => {
            serde_json::from_str(&s).map_err(|e| JsError::new(&e.to_string()))?
        }
        _ => ExtractOptions::default(),
    };
    let doc =
        crate::document::extract_document_from_bytes(bytes, password.as_deref(), &options)
            .map_err(|e| JsError::new(&e.to_string()))?;
    serde_json::to_string(&doc).map_err(|e| JsError::new(&e.to_string()))
}

/// PDF バイト列から Markdown 文字列を返す
#[cfg(feature = "document")]
#[wasm_bindgen]
pub fn extract_markdown_from_pdf(
    bytes: &[u8],
    options_json: Option<String>,
    password: Option<String>,
) -> Result<String, JsError> {
    let options: ExtractOptions = match options_json {
        Some(s) if !s.is_empty() => {
            serde_json::from_str(&s).map_err(|e| JsError::new(&e.to_string()))?
        }
        _ => ExtractOptions::default(),
    };
    crate::document::extract_markdown_from_bytes(bytes, password.as_deref(), &options)
        .map_err(|e| JsError::new(&e.to_string()))
}