View Source Pdf.Reader.XRef.Classic (ExPDF v1.0.1)

Parses a classic PDF cross-reference table (keyword xref).

Per PDF spec § 7.5.4:

  • Starts with the xref keyword on its own line.
  • Followed by one or more subsections. Each subsection has a header line <first_obj_num> <count> and then exactly count 20-byte entries.
  • Each entry format: <10-digit-offset> <5-digit-gen> <n|f><EOL> where EOL is \r\n, \r, or \n (3 variants = 20 bytes total).
  • After all subsections, a trailer keyword + dictionary.

Summary

Types

entries()
xref_entry()

Functions

parse(binary, offset)

Parses a classic xref table starting at offset within binary.

Types

Link to this type

entries()

View Source 
@type entries() :: %{required({pos_integer(), non_neg_integer()}) => xref_entry()}
Link to this type

xref_entry()

View Source 
@type xref_entry() :: {:in_use, non_neg_integer(), non_neg_integer()} | :free

Functions

Link to this function

parse(binary, offset)

View Source 
@spec parse(binary(), non_neg_integer()) :: {:ok, entries()} | {:error, term()}

Parses a classic xref table starting at offset within binary.

Returns {:ok, entries_map} where keys are {obj_num, gen_num} and values are {:in_use, offset, gen} or :free.

Returns {:error, reason} if the binary at that offset is not a valid classic xref section.