Crate rpdfium_parser 

PDF file structure parser for rpdfium — a faithful Rust port of PDFium.

This crate implements the PDF file format parser, including:

• Object model: PDF objects (null, boolean, integer, real, string, name, array, dictionary, stream, reference) with lazy resolution.
• Tokenizer: Low-level byte-to-token conversion.
• Header parsing: %PDF-X.Y version detection.
• Cross-reference tables: Traditional xref and PDF 1.5+ xref streams.
• Trailer parsing: startxref location, /Prev chain following.
• Object streams: ObjStm (PDF 1.5+) decompression and extraction.
• ObjectStore: Central thread-safe lazy-parsing object repository.
• Linearization detection: Checks for linearized PDF markers.
• Content stream tokenization: Parses PostScript-like operator sequences.

Design Principles

• #![forbid(unsafe_code)]
• All deep operations are iterative (explicit Vec stacks), never recursive.
• OnceLock-based lazy parsing: each object is parsed at most once.
• Stream /Length uses direct-object-only + endstream scan fallback.
• Security limits enforced: MAX_OBJECT_NUMBER, MAX_RECURSION_DEPTH, etc.

Re-exports

pub use content_stream::Operand;

pub use content_stream::Operator;

pub use content_stream::TextArrayElement;

pub use content_stream::tokenize_content_stream;

pub use crypto::CryptoError;

pub use filter::resolve_filter_chain;

pub use header::PdfVersion;

pub use hint_tables::HintTables;

pub use hint_tables::PageOffsetHintTable;

pub use linearized_header::LinearizedInfo;

pub use linearized_header::detect_linearized;

pub use object::Object;

pub use object::StreamData;

pub use object_walker::ObjectStats;

pub use object_walker::ObjectVisitor;

pub use object_walker::ObjectWalker;

pub use security::Permissions;

pub use security::SecurityError;

pub use security::SecurityHandler;

pub use store::ObjectStore;

pub use trailer::TrailerInfo;

pub use xref::XrefEntry;

pub use xref::XrefEntryType;

pub use xref::XrefSection;

pub use xref::XrefTable;

Modules

content_stream

Content stream operator tokenization (Stage 1).

crypto

Low-level cryptographic primitives for PDF encryption.

filter

Filter chain resolution — maps stream dictionary /Filter and /DecodeParms entries to codec types.

header

PDF header parsing — detects %PDF-X.Y and returns the version.

hint_tables

Linearization hint tables – page offset and shared object hint table parsing.

linearized_header

Linearization detection.

object

PDF object model — ObjectId, StreamData, and the Object enum.

object_parser

PDF object parsing — builds Object values from token streams.

object_stream

Object stream (ObjStm) parsing.

object_walker

Object graph walker – iterative BFS traversal with cycle detection.

security

PDF Standard Security Handler (R2–R6).

store

ObjectStore — the central data structure for PDF object access.

tokenizer

Low-level PDF tokenizer.

trailer

Trailer parsing — locates startxref, parses trailer dictionary, and follows the /Prev chain to build the full cross-reference table.

xref

Cross-reference table parsing (traditional xref format).

xref_stream

Cross-reference stream parsing (PDF 1.5+).

Structs

ObjectId

Unique identifier for a PDF indirect object.