tokenize-rt

The stdlib tokenize module does not properly roundtrip. This wrapper around the stdlib provides two additional tokens ESCAPED_NL and UNIMPORTANT_WS, and a Token data type. Use src_to_tokens and tokens_to_src to roundtrip.

This library is useful if you're writing a refactoring tool based on the python tokenization.

Installation

pip install tokenize-rt

Usage

datastructures

tokenize_rt.Offset(line=None, utf8_byte_offset=None)

A token offset, useful as a key when cross referencing the ast and the tokenized source.

tokenize_rt.Token(name, src, line=None, utf8_byte_offset=None)

Construct a token

• name: one of the token names listed in token.tok_name or ESCAPED_NL or UNIMPORTANT_WS
• src: token's source as text
• line: the line number that this token appears on.
• utf8_byte_offset: the utf8 byte offset that this token appears on in the line.

tokenize_rt.Token.offset

Retrieves an Offset for this token.

converting to and from Token representations

tokenize_rt.src_to_tokens(text: str) -> List[Token]

tokenize_rt.tokens_to_src(Iterable[Token]) -> str

additional tokens added by tokenize-rt

tokenize_rt.ESCAPED_NL

tokenize_rt.UNIMPORTANT_WS

helpers

tokenize_rt.NON_CODING_TOKENS

A frozenset containing tokens which may appear between others while not affecting control flow or code:

• COMMENT
• ESCAPED_NL
• NL
• UNIMPORTANT_WS

tokenize_rt.parse_string_literal(text: str) -> Tuple[str, str]

parse a string literal into its prefix and string content

>>> parse_string_literal('f"foo"')
('f', '"foo"')

tokenize_rt.reversed_enumerate(Sequence[Token]) -> Iterator[Tuple[int, Token]]

yields (index, token) pairs. Useful for rewriting source.

tokenize_rt.rfind_string_parts(Sequence[Token], i) -> Tuple[int, ...]

find the indices of the string parts of a (joined) string literal

• i should start at the end of the string literal
• returns () (an empty tuple) for things which are not string literals

>>> tokens = src_to_tokens('"foo" "bar".capitalize()')
>>> rfind_string_parts(tokens, 2)
(0, 2)
>>> tokens = src_to_tokens('("foo" "bar").capitalize()')
>>> rfind_string_parts(tokens, 4)
(1, 3)

Differences from tokenize

• tokenize-rt adds ESCAPED_NL for a backslash-escaped newline "token"
• tokenize-rt adds UNIMPORTANT_WS for whitespace (discarded in tokenize)
• tokenize-rt normalizes string prefixes, even if they are not parsed -- for instance, this means you'll see Token('STRING', "f'foo'", ...) even in python 2.
• tokenize-rt normalizes python 2 long literals (4l / 4L) and octal literals (0755) in python 3 (for easier rewriting of python 2 code while running python 3).

Sample usage

• https://github.com/asottile/add-trailing-comma
• https://github.com/asottile/future-annotations
• https://github.com/asottile/future-fstrings
• https://github.com/asottile/pyupgrade
• https://github.com/asottile/yesqa