///|
/// Lex one PDF delimiter token.
///
/// Recognizes `[`, `]`, `<<`, and `>>`. A single `<` or `>` is left unread and
/// returns `LexNone`.
pub fn pdf_cursor_lex_delimiter(cursor : @core.ByteCursor) -> PdfLexeme {
pdf_cursor_drop_whitespace(cursor)
let position = cursor.absolute_position()
match cursor.peek_byte() {
91 => {
cursor.nudge()
LexLeftSquare
}
93 => {
cursor.nudge()
LexRightSquare
}
value if value == 60 && cursor.byte_int_at_absolute(position + 1) == 60 => {
cursor.nudge()
cursor.nudge()
LexLeftDict
}
value if value == 62 && cursor.byte_int_at_absolute(position + 1) == 62 => {
cursor.nudge()
cursor.nudge()
LexRightDict
}
_ => LexNone
}
}
///|
/// Lex one token from the current cursor position.
///
/// This dispatcher skips white space, recognizes comments, primitives,
/// delimiters, names, literal and hex strings, and selected keywords. It returns
/// `StopLexing` at end of input and at markers that terminate object scanning,
/// such as `startxref` or inline image data boundaries.
pub fn pdf_cursor_lex_token(
cursor : @core.ByteCursor,
) -> PdfLexeme raise @core.PdfError {
pdf_cursor_drop_whitespace(cursor)
let position = cursor.absolute_position()
let value = cursor.peek_byte()
match value {
value if value == @core.pdf_no_more => StopLexing
37 => pdf_cursor_lex_comment(cursor)
116 | 102 => pdf_cursor_lex_bool(cursor)
47 => pdf_cursor_lex_name(cursor)
value if pdf_is_number_start_byte(value) => pdf_cursor_lex_number(cursor)
91 | 93 => pdf_cursor_lex_delimiter(cursor)
40 => pdf_cursor_lex_literal_string(cursor)
60 =>
if pdf_cursor_matches_ascii_at(cursor, position, [60, 60]) {
pdf_cursor_lex_delimiter(cursor)
} else {
pdf_cursor_lex_hex_string(cursor)
}
62 => pdf_cursor_lex_delimiter(cursor)
82 => {
cursor.nudge()
LexR
}
115 =>
// `startxref` terminates object lexing; any other `s...` marker is left
// for the stream reader so malformed spellings report stream errors.
StopLexing
73 => StopLexing
value if pdf_is_lowercase_keyword_start_byte(value) =>
pdf_cursor_lex_keyword(cursor)
_ => LexNone
}
}
///|
/// Lex tokens until a stop marker or unmatched input is reached.
///
/// `StopLexing` and `LexNone` terminate scanning and are not included in the
/// returned token array.
pub fn pdf_cursor_lex_tokens(
cursor : @core.ByteCursor,
) -> Array[PdfLexeme] raise @core.PdfError {
let tokens : Array[PdfLexeme] = []
let mut done = false
while !done {
match pdf_cursor_lex_token(cursor) {
StopLexing | LexNone => done = true
token => tokens.push(token)
}
}
tokens
}
///|
/// Lex a single token from a borrowed byte view.
pub fn pdf_lex_single_view(data : BytesView) -> PdfLexeme raise @core.PdfError {
pdf_cursor_lex_token(@core.byte_cursor_of_view(data))
}
///|
/// Lex all tokens from a borrowed byte view.
pub fn pdf_lex_view(data : BytesView) -> Array[PdfLexeme] raise @core.PdfError {
pdf_cursor_lex_tokens(@core.byte_cursor_of_view(data))
}
///|
/// Lex all tokens from owned PDF bytes.
pub fn pdf_lex_bytes(
data : @core.PdfBytes,
) -> Array[PdfLexeme] raise @core.PdfError {
pdf_lex_view(data)
}