"""Tokenization help for Python programs.
tokenize(readline) is a generator that breaks a stream of bytes into
Python tokens. It decodes the bytes according to PEP-0263 for
determining source file encoding.
It accepts a readline-like method which is called repeatedly to get the
next line of input (or b"" for EOF). It generates 5-tuples with these
the token type (see token.py)
the starting (row, column) indices of the token (a 2-tuple of ints)
the ending (row, column) indices of the token (a 2-tuple of ints)
the original line (string)
It is designed to match the working of the Python tokenizer exactly, except
that it produces COMMENT tokens for comments and gives type OP for all
operators. Additionally, all token lists start with an ENCODING token
which tells you which encoding was used to decode the bytes stream.
__author__ = 'Ka-Ping Yee <ping@lfw.org>'
__credits__ = ('GvR, ESR, Tim Peters, Thomas Wouters, Fred Drake, '
'Skip Montanaro, Raymond Hettinger, Trent Nelson, '
from builtins import open as _builtin_open
from codecs import lookup, BOM_UTF8
from io import TextIOWrapper
import itertools as _itertools
from token import EXACT_TOKEN_TYPES
cookie_re = re.compile(r'^[ \t\f]*#.*?coding[:=][ \t]*([-\w.]+)', re.ASCII)
blank_re = re.compile(br'^[ \t\f]*(?:[#\r\n]|$)', re.ASCII)
__all__ = token.__all__ + ["tokenize", "generate_tokens", "detect_encoding",
"untokenize", "TokenInfo"]
class TokenInfo(collections.namedtuple('TokenInfo', 'type string start end line')):
annotated_type = '%d (%s)' % (self.type, tok_name[self.type])
return ('TokenInfo(type=%s, string=%r, start=%r, end=%r, line=%r)' %
self._replace(type=annotated_type))
if self.type == OP and self.string in EXACT_TOKEN_TYPES:
return EXACT_TOKEN_TYPES[self.string]
def group(*choices): return '(' + '|'.join(choices) + ')'
def any(*choices): return group(*choices) + '*'
def maybe(*choices): return group(*choices) + '?'
# Note: we use unicode matching for names ("\w") but ascii matching for
Ignore = Whitespace + any(r'\\\r?\n' + Whitespace) + maybe(Comment)
Hexnumber = r'0[xX](?:_?[0-9a-fA-F])+'
Binnumber = r'0[bB](?:_?[01])+'
Octnumber = r'0[oO](?:_?[0-7])+'
Decnumber = r'(?:0(?:_?0)*|[1-9](?:_?[0-9])*)'
Intnumber = group(Hexnumber, Binnumber, Octnumber, Decnumber)
Exponent = r'[eE][-+]?[0-9](?:_?[0-9])*'
Pointfloat = group(r'[0-9](?:_?[0-9])*\.(?:[0-9](?:_?[0-9])*)?',
r'\.[0-9](?:_?[0-9])*') + maybe(Exponent)
Expfloat = r'[0-9](?:_?[0-9])*' + Exponent
Floatnumber = group(Pointfloat, Expfloat)
Imagnumber = group(r'[0-9](?:_?[0-9])*[jJ]', Floatnumber + r'[jJ]')
Number = group(Imagnumber, Floatnumber, Intnumber)
# Return the empty string, plus all of the valid string prefixes.
def _all_string_prefixes():
# The valid string prefixes. Only contain the lower case versions,
# and don't contain any permutations (include 'fr', but not
# 'rf'). The various permutations will be generated.
_valid_string_prefixes = ['b', 'r', 'u', 'f', 'br', 'fr']
# if we add binary f-strings, add: ['fb', 'fbr']
for prefix in _valid_string_prefixes:
for t in _itertools.permutations(prefix):
# create a list with upper and lower versions of each
for u in _itertools.product(*[(c, c.upper()) for c in t]):
return re.compile(expr, re.UNICODE)
# Note that since _all_string_prefixes includes the empty string,
# StringPrefix can be the empty string (making it optional).
StringPrefix = group(*_all_string_prefixes())
Single = r"[^'\\]*(?:\\.[^'\\]*)*'"
Double = r'[^"\\]*(?:\\.[^"\\]*)*"'
# Tail end of ''' string.
Single3 = r"[^'\\]*(?:(?:\\.|'(?!''))[^'\\]*)*'''"
# Tail end of """ string.
Double3 = r'[^"\\]*(?:(?:\\.|"(?!""))[^"\\]*)*"""'
Triple = group(StringPrefix + "'''", StringPrefix + '"""')
# Single-line ' or " string.
String = group(StringPrefix + r"'[^\n'\\]*(?:\\.[^\n'\\]*)*'",
StringPrefix + r'"[^\n"\\]*(?:\\.[^\n"\\]*)*"')
# Sorting in reverse order puts the long operators before their prefixes.
# Otherwise if = came before ==, == would get recognized as two instances
Special = group(*map(re.escape, sorted(EXACT_TOKEN_TYPES, reverse=True)))
Funny = group(r'\r?\n', Special)
PlainToken = group(Number, Funny, String, Name)
Token = Ignore + PlainToken
# First (or only) line of ' or " string.
ContStr = group(StringPrefix + r"'[^\n'\\]*(?:\\.[^\n'\\]*)*" +
StringPrefix + r'"[^\n"\\]*(?:\\.[^\n"\\]*)*' +
PseudoExtras = group(r'\\\r?\n|\Z', Comment, Triple)
PseudoToken = Whitespace + group(PseudoExtras, Number, Funny, ContStr, Name)
# For a given string prefix plus quotes, endpats maps it to a regex
# to match the remainder of that string. _prefix can be empty, for
# a normal single or triple quoted string (with no prefix).
for _prefix in _all_string_prefixes():
endpats[_prefix + "'"] = Single
endpats[_prefix + '"'] = Double
endpats[_prefix + "'''"] = Single3
endpats[_prefix + '"""'] = Double3
# A set of all of the single and triple quoted string prefixes,
# including the opening quotes.
for t in _all_string_prefixes():
for u in (t + '"', t + "'"):
for u in (t + '"""', t + "'''"):
class TokenError(Exception): pass
class StopTokenizing(Exception): pass
def add_whitespace(self, start):
if row < self.prev_row or row == self.prev_row and col < self.prev_col:
raise ValueError("start ({},{}) precedes previous end ({},{})"
.format(row, col, self.prev_row, self.prev_col))
row_offset = row - self.prev_row
self.tokens.append("\\\n" * row_offset)
col_offset = col - self.prev_col
self.tokens.append(" " * col_offset)
def untokenize(self, iterable):
tok_type, token, start, end, line = t
if tok_type == ENDMARKER:
self.prev_row, self.prev_col = end
elif tok_type in (NEWLINE, NL):
elif startline and indents:
if start[1] >= len(indent):
self.tokens.append(indent)
self.prev_col = len(indent)
self.add_whitespace(start)
self.tokens.append(token)
self.prev_row, self.prev_col = end
if tok_type in (NEWLINE, NL):
return "".join(self.tokens)
def compat(self, token, iterable):
toks_append = self.tokens.append
startline = token[0] in (NEWLINE, NL)
for tok in _itertools.chain([token], iterable):
if toknum in (NAME, NUMBER):
# Insert a space between two consecutive strings
elif toknum in (NEWLINE, NL):
elif startline and indents:
def untokenize(iterable):
"""Transform tokens back into Python source code.
It returns a bytes object, encoded using the ENCODING
token, which is the first token sequence output by tokenize.
Each element returned by the iterable must be a token sequence
with at least two elements, a token number and token value. If
only two tokens are passed, the resulting output is poor.
Round-trip invariant for full input:
Untokenized source will match input source exactly
Round-trip invariant for limited input:
# Output bytes will tokenize back to the input
t1 = [tok[:2] for tok in tokenize(f.readline)]
readline = BytesIO(newcode).readline
t2 = [tok[:2] for tok in tokenize(readline)]
out = ut.untokenize(iterable)
if ut.encoding is not None:
out = out.encode(ut.encoding)
def _get_normal_name(orig_enc):
"""Imitates get_normal_name in tokenizer.c."""
# Only care about the first 12 characters.
enc = orig_enc[:12].lower().replace("_", "-")
if enc == "utf-8" or enc.startswith("utf-8-"):
if enc in ("latin-1", "iso-8859-1", "iso-latin-1") or \
enc.startswith(("latin-1-", "iso-8859-1-", "iso-latin-1-")):
def detect_encoding(readline):
The detect_encoding() function is used to detect the encoding that should
be used to decode a Python source file. It requires one argument, readline,
in the same way as the tokenize() generator.
It will call readline a maximum of twice, and return the encoding used
(as a string) and a list of any lines (left as bytes) it has read in.
It detects the encoding from the presence of a utf-8 bom or an encoding
cookie as specified in pep-0263. If both a bom and a cookie are present,
but disagree, a SyntaxError will be raised. If the encoding cookie is an
invalid charset, raise a SyntaxError. Note that if a utf-8 bom is found,
If no encoding is specified, then the default of 'utf-8' will be returned.
filename = readline.__self__.name
# Decode as UTF-8. Either the line is an encoding declaration,
# in which case it should be pure ASCII, or it must be UTF-8
line_string = line.decode('utf-8')
except UnicodeDecodeError:
msg = "invalid or missing encoding declaration"
msg = '{} for {!r}'.format(msg, filename)
match = cookie_re.match(line_string)
encoding = _get_normal_name(match.group(1))
# This behaviour mimics the Python interpreter
msg = "unknown encoding: " + encoding
msg = "unknown encoding for {!r}: {}".format(filename,
# This behaviour mimics the Python interpreter
msg = 'encoding problem: utf-8'
msg = 'encoding problem for {!r}: utf-8'.format(filename)
if first.startswith(BOM_UTF8):
encoding = find_cookie(first)
if not blank_re.match(first):
encoding = find_cookie(second)
return encoding, [first, second]
return default, [first, second]
"""Open a file in read only mode using the encoding detected by
buffer = _builtin_open(filename, 'rb')
encoding, lines = detect_encoding(buffer.readline)
text = TextIOWrapper(buffer, encoding, line_buffering=True)
The tokenize() generator requires one argument, readline, which
must be a callable object which provides the same interface as the
readline() method of built-in file objects. Each call to the function
should return one line of input as bytes. Alternatively, readline
can be a callable function terminating with StopIteration:
readline = open(myfile, 'rb').__next__ # Example of alternate readline
The generator produces 5-tuples with these members: the token type; the
token string; a 2-tuple (srow, scol) of ints specifying the row and
column where the token begins in the source; a 2-tuple (erow, ecol) of
ints specifying the row and column where the token ends in the source;
and the line on which the token was found. The line passed is the
The first token sequence will always be an ENCODING token
which tells you which encoding was used to decode the bytes stream.
encoding, consumed = detect_encoding(readline)
empty = _itertools.repeat(b"")
rl_gen = _itertools.chain(consumed, iter(readline, b""), empty)
return _tokenize(rl_gen.__next__, encoding)
def _tokenize(readline, encoding):
lnum = parenlev = continued = 0
contstr, needcont = '', 0
if encoding == "utf-8-sig":
# BOM will already have been stripped.
yield TokenInfo(ENCODING, encoding, (0, 0), (0, 0), '')
while True: # loop over lines in stream
# We capture the value of the line variable here because
# readline uses the empty string '' to signal end of input,
# hence `line` itself will always be overwritten at the end
line = line.decode(encoding)
if contstr: # continued string
raise TokenError("EOF in multi-line string", strstart)
endmatch = endprog.match(line)
pos = end = endmatch.end(0)
yield TokenInfo(STRING, contstr + line[:end],
strstart, (lnum, end), contline + line)
contstr, needcont = '', 0
elif needcont and line[-2:] != '\\\n' and line[-3:] != '\\\r\n':
yield TokenInfo(ERRORTOKEN, contstr + line,
strstart, (lnum, len(line)), contline)
contline = contline + line
elif parenlev == 0 and not continued: # new statement
while pos < max: # measure leading whitespace
column = (column//tabsize + 1)*tabsize
if line[pos] in '#\r\n': # skip comments or blank lines
comment_token = line[pos:].rstrip('\r\n')
yield TokenInfo(COMMENT, comment_token,
(lnum, pos), (lnum, pos + len(comment_token)), line)
It is recommended that you Edit text format, this type of Fix handles quite a lot in one request
