mirror of
https://github.com/python/cpython
synced 2024-09-06 07:13:55 +00:00
6715f91edc
This commit replaces the Python implementation of the tokenize module with an implementation that reuses the real C tokenizer via a private extension module. The tokenize module now implements a compatibility layer that transforms tokens from the C tokenizer into Python tokenize tokens for backward compatibility. As the C tokenizer does not emit some tokens that the Python tokenizer provides (such as comments and non-semantic newlines), a new special mode has been added to the C tokenizer mode that currently is only used via the extension module that exposes it to the Python layer. This new mode forces the C tokenizer to emit these new extra tokens and add the appropriate metadata that is needed to match the old Python implementation. Co-authored-by: Pablo Galindo <pablogsal@gmail.com>
458 lines
16 KiB
Python
458 lines
16 KiB
Python
"""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
|
|
members:
|
|
|
|
the token type (see token.py)
|
|
the token (a string)
|
|
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, '
|
|
'Michael Foord')
|
|
from builtins import open as _builtin_open
|
|
from codecs import lookup, BOM_UTF8
|
|
import collections
|
|
import functools
|
|
from io import TextIOWrapper
|
|
import itertools as _itertools
|
|
import re
|
|
import sys
|
|
from token import *
|
|
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)
|
|
|
|
import token
|
|
__all__ = token.__all__ + ["tokenize", "generate_tokens", "detect_encoding",
|
|
"untokenize", "TokenInfo"]
|
|
del token
|
|
|
|
class TokenInfo(collections.namedtuple('TokenInfo', 'type string start end line')):
|
|
def __repr__(self):
|
|
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))
|
|
|
|
@property
|
|
def exact_type(self):
|
|
if self.type == OP and self.string in EXACT_TOKEN_TYPES:
|
|
return EXACT_TOKEN_TYPES[self.string]
|
|
else:
|
|
return self.type
|
|
|
|
|
|
class TokenError(Exception): pass
|
|
|
|
|
|
class StopTokenizing(Exception): pass
|
|
|
|
class Untokenizer:
|
|
|
|
def __init__(self):
|
|
self.tokens = []
|
|
self.prev_row = 1
|
|
self.prev_col = 0
|
|
self.encoding = None
|
|
|
|
def add_whitespace(self, start):
|
|
row, col = 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
|
|
if row_offset:
|
|
self.tokens.append("\\\n" * row_offset)
|
|
self.prev_col = 0
|
|
col_offset = col - self.prev_col
|
|
if col_offset:
|
|
self.tokens.append(" " * col_offset)
|
|
|
|
def untokenize(self, iterable):
|
|
it = iter(iterable)
|
|
indents = []
|
|
startline = False
|
|
for t in it:
|
|
if len(t) == 2:
|
|
self.compat(t, it)
|
|
break
|
|
tok_type, token, start, end, line = t
|
|
if tok_type == ENCODING:
|
|
self.encoding = token
|
|
continue
|
|
if tok_type == ENDMARKER:
|
|
break
|
|
if tok_type == INDENT:
|
|
indents.append(token)
|
|
continue
|
|
elif tok_type == DEDENT:
|
|
indents.pop()
|
|
self.prev_row, self.prev_col = end
|
|
continue
|
|
elif tok_type in (NEWLINE, NL):
|
|
startline = True
|
|
elif startline and indents:
|
|
indent = indents[-1]
|
|
if start[1] >= len(indent):
|
|
self.tokens.append(indent)
|
|
self.prev_col = len(indent)
|
|
startline = False
|
|
elif tok_type == FSTRING_MIDDLE:
|
|
if '{' in token or '}' in token:
|
|
end_line, end_col = end
|
|
end = (end_line, end_col + token.count('{') + token.count('}'))
|
|
token = re.sub('{', '{{', token)
|
|
token = re.sub('}', '}}', token)
|
|
|
|
|
|
self.add_whitespace(start)
|
|
self.tokens.append(token)
|
|
self.prev_row, self.prev_col = end
|
|
if tok_type in (NEWLINE, NL):
|
|
self.prev_row += 1
|
|
self.prev_col = 0
|
|
return "".join(self.tokens)
|
|
|
|
def compat(self, token, iterable):
|
|
indents = []
|
|
toks_append = self.tokens.append
|
|
startline = token[0] in (NEWLINE, NL)
|
|
prevstring = False
|
|
|
|
for tok in _itertools.chain([token], iterable):
|
|
toknum, tokval = tok[:2]
|
|
if toknum == ENCODING:
|
|
self.encoding = tokval
|
|
continue
|
|
|
|
if toknum in (NAME, NUMBER):
|
|
tokval += ' '
|
|
|
|
# Insert a space between two consecutive strings
|
|
if toknum == STRING:
|
|
if prevstring:
|
|
tokval = ' ' + tokval
|
|
prevstring = True
|
|
else:
|
|
prevstring = False
|
|
|
|
if toknum == INDENT:
|
|
indents.append(tokval)
|
|
continue
|
|
elif toknum == DEDENT:
|
|
indents.pop()
|
|
continue
|
|
elif toknum in (NEWLINE, NL):
|
|
startline = True
|
|
elif startline and indents:
|
|
toks_append(indents[-1])
|
|
startline = False
|
|
elif toknum == FSTRING_MIDDLE:
|
|
if '{' in tokval or '}' in tokval:
|
|
tokval = re.sub('{', '{{', tokval)
|
|
tokval = re.sub('}', '}}', tokval)
|
|
|
|
toks_append(tokval)
|
|
|
|
|
|
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)]
|
|
newcode = untokenize(t1)
|
|
readline = BytesIO(newcode).readline
|
|
t2 = [tok[:2] for tok in tokenize(readline)]
|
|
assert t1 == t2
|
|
"""
|
|
ut = Untokenizer()
|
|
out = ut.untokenize(iterable)
|
|
if ut.encoding is not None:
|
|
out = out.encode(ut.encoding)
|
|
return out
|
|
|
|
|
|
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-"):
|
|
return "utf-8"
|
|
if enc in ("latin-1", "iso-8859-1", "iso-latin-1") or \
|
|
enc.startswith(("latin-1-", "iso-8859-1-", "iso-latin-1-")):
|
|
return "iso-8859-1"
|
|
return orig_enc
|
|
|
|
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,
|
|
'utf-8-sig' is returned.
|
|
|
|
If no encoding is specified, then the default of 'utf-8' will be returned.
|
|
"""
|
|
try:
|
|
filename = readline.__self__.name
|
|
except AttributeError:
|
|
filename = None
|
|
bom_found = False
|
|
encoding = None
|
|
default = 'utf-8'
|
|
def read_or_stop():
|
|
try:
|
|
return readline()
|
|
except StopIteration:
|
|
return b''
|
|
|
|
def find_cookie(line):
|
|
try:
|
|
# 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
|
|
# per default encoding.
|
|
line_string = line.decode('utf-8')
|
|
except UnicodeDecodeError:
|
|
msg = "invalid or missing encoding declaration"
|
|
if filename is not None:
|
|
msg = '{} for {!r}'.format(msg, filename)
|
|
raise SyntaxError(msg)
|
|
|
|
match = cookie_re.match(line_string)
|
|
if not match:
|
|
return None
|
|
encoding = _get_normal_name(match.group(1))
|
|
try:
|
|
codec = lookup(encoding)
|
|
except LookupError:
|
|
# This behaviour mimics the Python interpreter
|
|
if filename is None:
|
|
msg = "unknown encoding: " + encoding
|
|
else:
|
|
msg = "unknown encoding for {!r}: {}".format(filename,
|
|
encoding)
|
|
raise SyntaxError(msg)
|
|
|
|
if bom_found:
|
|
if encoding != 'utf-8':
|
|
# This behaviour mimics the Python interpreter
|
|
if filename is None:
|
|
msg = 'encoding problem: utf-8'
|
|
else:
|
|
msg = 'encoding problem for {!r}: utf-8'.format(filename)
|
|
raise SyntaxError(msg)
|
|
encoding += '-sig'
|
|
return encoding
|
|
|
|
first = read_or_stop()
|
|
if first.startswith(BOM_UTF8):
|
|
bom_found = True
|
|
first = first[3:]
|
|
default = 'utf-8-sig'
|
|
if not first:
|
|
return default, []
|
|
|
|
encoding = find_cookie(first)
|
|
if encoding:
|
|
return encoding, [first]
|
|
if not blank_re.match(first):
|
|
return default, [first]
|
|
|
|
second = read_or_stop()
|
|
if not second:
|
|
return default, [first]
|
|
|
|
encoding = find_cookie(second)
|
|
if encoding:
|
|
return encoding, [first, second]
|
|
|
|
return default, [first, second]
|
|
|
|
|
|
def open(filename):
|
|
"""Open a file in read only mode using the encoding detected by
|
|
detect_encoding().
|
|
"""
|
|
buffer = _builtin_open(filename, 'rb')
|
|
try:
|
|
encoding, lines = detect_encoding(buffer.readline)
|
|
buffer.seek(0)
|
|
text = TextIOWrapper(buffer, encoding, line_buffering=True)
|
|
text.mode = 'r'
|
|
return text
|
|
except:
|
|
buffer.close()
|
|
raise
|
|
|
|
def tokenize(readline):
|
|
"""
|
|
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
|
|
physical line.
|
|
|
|
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)
|
|
rl_gen = _itertools.chain(consumed, iter(readline, b""))
|
|
if encoding is not None:
|
|
if encoding == "utf-8-sig":
|
|
# BOM will already have been stripped.
|
|
encoding = "utf-8"
|
|
yield TokenInfo(ENCODING, encoding, (0, 0), (0, 0), '')
|
|
yield from _tokenize(rl_gen, encoding)
|
|
|
|
def _tokenize(rl_gen, encoding):
|
|
source = b"".join(rl_gen).decode(encoding)
|
|
token = None
|
|
for token in _generate_tokens_from_c_tokenizer(source, extra_tokens=True):
|
|
# TODO: Marta -> limpiar esto
|
|
if 6 < token.type <= 54:
|
|
token = token._replace(type=OP)
|
|
if token.type in {ASYNC, AWAIT}:
|
|
token = token._replace(type=NAME)
|
|
if token.type == NEWLINE:
|
|
l_start, c_start = token.start
|
|
l_end, c_end = token.end
|
|
token = token._replace(string='\n', start=(l_start, c_start), end=(l_end, c_end+1))
|
|
|
|
yield token
|
|
if token is not None:
|
|
last_line, _ = token.start
|
|
yield TokenInfo(ENDMARKER, '', (last_line + 1, 0), (last_line + 1, 0), '')
|
|
|
|
|
|
def generate_tokens(readline):
|
|
"""Tokenize a source reading Python code as unicode strings.
|
|
|
|
This has the same API as tokenize(), except that it expects the *readline*
|
|
callable to return str objects instead of bytes.
|
|
"""
|
|
def _gen():
|
|
while True:
|
|
try:
|
|
line = readline()
|
|
except StopIteration:
|
|
return
|
|
if not line:
|
|
return
|
|
yield line.encode()
|
|
return _tokenize(_gen(), 'utf-8')
|
|
|
|
def main():
|
|
import argparse
|
|
|
|
# Helper error handling routines
|
|
def perror(message):
|
|
sys.stderr.write(message)
|
|
sys.stderr.write('\n')
|
|
|
|
def error(message, filename=None, location=None):
|
|
if location:
|
|
args = (filename,) + location + (message,)
|
|
perror("%s:%d:%d: error: %s" % args)
|
|
elif filename:
|
|
perror("%s: error: %s" % (filename, message))
|
|
else:
|
|
perror("error: %s" % message)
|
|
sys.exit(1)
|
|
|
|
# Parse the arguments and options
|
|
parser = argparse.ArgumentParser(prog='python -m tokenize')
|
|
parser.add_argument(dest='filename', nargs='?',
|
|
metavar='filename.py',
|
|
help='the file to tokenize; defaults to stdin')
|
|
parser.add_argument('-e', '--exact', dest='exact', action='store_true',
|
|
help='display token names using the exact type')
|
|
args = parser.parse_args()
|
|
|
|
try:
|
|
# Tokenize the input
|
|
if args.filename:
|
|
filename = args.filename
|
|
with _builtin_open(filename, 'rb') as f:
|
|
tokens = list(tokenize(f.readline))
|
|
else:
|
|
filename = "<stdin>"
|
|
tokens = _tokenize(
|
|
(x.encode('utf-8') for x in iter(sys.stdin.readline, "")
|
|
), "utf-8")
|
|
|
|
|
|
# Output the tokenization
|
|
for token in tokens:
|
|
token_type = token.type
|
|
if args.exact:
|
|
token_type = token.exact_type
|
|
token_range = "%d,%d-%d,%d:" % (token.start + token.end)
|
|
print("%-20s%-15s%-15r" %
|
|
(token_range, tok_name[token_type], token.string))
|
|
except IndentationError as err:
|
|
line, column = err.args[1][1:3]
|
|
error(err.args[0], filename, (line, column))
|
|
except TokenError as err:
|
|
line, column = err.args[1]
|
|
error(err.args[0], filename, (line, column))
|
|
except SyntaxError as err:
|
|
error(err, filename)
|
|
except OSError as err:
|
|
error(err)
|
|
except KeyboardInterrupt:
|
|
print("interrupted\n")
|
|
except Exception as err:
|
|
perror("unexpected error: %s" % err)
|
|
raise
|
|
|
|
def _generate_tokens_from_c_tokenizer(source, extra_tokens=False):
|
|
"""Tokenize a source reading Python code as unicode strings using the internal C tokenizer"""
|
|
import _tokenize as c_tokenizer
|
|
for info in c_tokenizer.TokenizerIter(source, extra_tokens=extra_tokens):
|
|
tok, type, lineno, end_lineno, col_off, end_col_off, line = info
|
|
yield TokenInfo(type, tok, (lineno, col_off), (end_lineno, end_col_off), line)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
main()
|