Struct PdfProcessResult 

pub struct PdfProcessResult {

    pub pdf_type: PdfType,
    pub markdown: Option<String>,
    pub page_count: u32,
    pub processing_time_ms: u64,
    pub pages_needing_ocr: Vec<u32>,
    pub ocr_reasons_by_page: Vec<PageOcrReasons>,
    pub title: Option<String>,
    pub author: Option<String>,
    pub subject: Option<String>,
    pub keywords: Option<String>,
    pub creator: Option<String>,
    pub producer: Option<String>,
    pub creation_date: Option<String>,
    pub mod_date: Option<String>,
    pub confidence: f32,
    pub layout: LayoutComplexity,
    pub has_encoding_issues: bool,
    pub cmap_gaps: Vec<FontCMapGaps>,
}

High-level PDF processing result.

Fields

pdf_type: PdfType

The detected PDF type.

markdown: Option<String>

Markdown output (populated in ProcessMode::Full, None otherwise).

page_count: u32

Page count.

processing_time_ms: u64

Processing time in milliseconds.

pages_needing_ocr: Vec<u32>

1-indexed page numbers that need OCR.

ocr_reasons_by_page: Vec<PageOcrReasons>

Machine-readable OCR reasons by 1-indexed page.

title: Option<String>

The /Title of the document information dictionary, decoded as a PDF text string (UTF-16 or UTF-8 after a byte order mark, PDFDocEncoding otherwise; see PdfTypeResult::title). None when the entry is missing or not a string. The entries below follow the same decoding and missing-value rule.

author: Option<String>

The document information dictionary’s /Author.

subject: Option<String>

The document information dictionary’s /Subject.

keywords: Option<String>

The document information dictionary’s /Keywords.

creator: Option<String>

The document information dictionary’s /Creator: the application the document was authored in.

producer: Option<String>

The document information dictionary’s /Producer: the application that wrote the PDF.

creation_date: Option<String>

The document information dictionary’s /CreationDate as written, a PDF date string such as D:20240115103000+01'00', neither validated nor converted.

mod_date: Option<String>

The document information dictionary’s /ModDate as written, like creation_date.

confidence: f32

Detection confidence score (0.0–1.0).

layout: LayoutComplexity

Layout complexity analysis (tables, multi-column detection).

has_encoding_issues: bool

true when broken font encodings are detected (garbled text, replacement characters). Clients should fall back to OCR.

cmap_gaps: Vec<FontCMapGaps>

The fonts whose ToUnicode CMap — or, for a font without one, the embedded program’s cmap table — lacked an entry for a code the document shows through it, with the counts of codes shown, read from their neighbours and left as U+FFFD (see FontCMapGaps). Always empty in ProcessMode::DetectOnly, which decodes no text; otherwise empty when every such code had an entry.

Trait Implementations

impl Debug for PdfProcessResult

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.