parse

type ObjectLocation struct {
	IsDirect      bool  // True if object is at a direct byte offset
	ByteOffset    int64 // For direct objects: byte offset in PDF
	StreamObjNum  int   // For object stream objects: containing stream's object number
	IndexInStream int   // For object stream objects: index within the stream
}

type ObjectRef struct {
	Number       int   // Object number
	Generation   int   // Generation number (usually 0)
	Offset       int64 // Byte offset in PDF (0 if in object stream)
	InStream     bool  // True if object is stored in an object stream
	StreamObjNum int   // Object stream number (if InStream is true)
	StreamIndex  int   // Index within object stream (if InStream is true)
}

type ObjectStreamEntry struct {
	StreamObjNum  int // The object stream that contains this object
	IndexInStream int // Index within the object stream
}

type PDF struct {
	// contains filtered or unexported fields
}

type PDFDocument struct {
	RawBytes  []byte         // Original complete PDF bytes
	Header    *PDFHeader     // PDF header info
	Revisions []*PDFRevision // All revisions, oldest first
}

type PDFHeader struct {
	Version      string // PDF version (e.g., "1.7")
	MajorVersion int    // Major version number (e.g., 1)
	MinorVersion int    // Minor version number (e.g., 7)
	RawBytes     []byte // Exact header bytes including binary marker (up to first object)
}

type PDFRawObject struct {
	Number     int    // Object number
	Generation int    // Generation number
	Offset     int64  // Byte offset in file where object starts
	EndOffset  int64  // Byte offset where object ends (after "endobj")
	RawBytes   []byte // Complete raw bytes from "N G obj" through "endobj" (inclusive)

	// Parsed stream components (populated for stream objects)
	IsStream    bool   // True if this is a stream object
	DictRaw     []byte // Raw dictionary bytes (including << >>)
	StreamRaw   []byte // Raw stream data (between "stream\n" and "\nendstream", excluding keywords)
	DictStart   int    // Offset within RawBytes where dictionary starts
	DictEnd     int    // Offset within RawBytes where dictionary ends
	StreamStart int    // Offset within RawBytes where stream data starts
	StreamEnd   int    // Offset within RawBytes where stream data ends
}

type PDFRevision struct {
	Number    int                   // Revision number (1-indexed)
	Objects   map[int]*PDFRawObject // Objects added/modified in this revision (keyed by object number)
	XRef      *XRefData             // Cross-reference data for this revision
	Trailer   *TrailerData          // Trailer dictionary for this revision
	StartXRef int64                 // startxref value for this revision
	EOFOffset int64                 // Byte offset where %%EOF starts
	EndOffset int64                 // Byte offset after %%EOF (and any trailing newlines)
}

type PDFTrailer struct {
	RootRef    string // Root reference (e.g., "/Root 204 0 R")
	EncryptRef string // Encrypt reference if present
	InfoRef    string // Info reference if present
	StartXRef  int64  // Byte offset from startxref
}

type ParseOptions struct {
	Password    []byte // Password for encrypted PDFs (empty for unencrypted)
	Verbose     bool   // Enable verbose logging
	BytePerfect bool   // Preserve exact bytes for reconstruction
}

type TrailerData struct {
	Offset   int64  // Byte offset where "trailer" keyword starts (0 for xref stream)
	RawBytes []byte // Raw bytes of trailer dictionary (including "trailer\n<<...>>")

	// Parsed values for convenience
	Size    int      // /Size value
	Root    string   // /Root reference (e.g., "1 0 R")
	Encrypt string   // /Encrypt reference (if encrypted)
	Info    string   // /Info reference (if present)
	Prev    int64    // /Prev value (offset of previous xref, 0 if none)
	ID      [][]byte // /ID array (two byte strings)
}

type TrailerInfo struct {
	Size       int    // Number of objects in the file
	RootRef    string // Reference to document catalog (e.g., "1 0 R")
	InfoRef    string // Reference to document info dictionary
	EncryptRef string // Reference to encryption dictionary
	IDArray    []byte // File identifier array
}

type XRef struct {
	Objects map[int]*ObjectRef // Object number -> reference info
	Size    int                // Total number of objects
}

type XRefData struct {
	Type     XRefType // Traditional table or stream
	Offset   int64    // Byte offset where xref section starts
	RawBytes []byte   // Complete raw bytes of xref section (for traditional: "xref" through entries)

	// Parsed entries for convenience
	Entries []XRefEntry

	// For xref streams: the object containing the stream
	StreamObject *PDFRawObject
}

type XRefEntry struct {
	ObjectNum  int   // Object number
	Generation int   // Generation number
	Offset     int64 // For type 1: byte offset in file
	InUse      bool  // true = 'n' (in use), false = 'f' (free)

	// For type 2 entries (object in object stream)
	InObjectStream bool // True if this object is in an object stream
	StreamObjNum   int  // Object stream number (if InObjectStream)
	IndexInStream  int  // Index within object stream (if InObjectStream)

	// Raw entry bytes for byte-perfect reconstruction (20 bytes for traditional)
	RawBytes []byte
}

type XRefResult struct {
	// Regular objects (Type 1): objNum -> byte offset
	Objects map[int]int64
	// Objects in object streams (Type 2): objNum -> ObjectStreamEntry
	ObjectStreams map[int]ObjectStreamEntry
}

type XRefType int