You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.

324 lines
11 KiB

# Copyright 2015 Google Inc. All Rights Reserved.
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# See the License for the specific language governing permissions and
# limitations under the License.
"""Enhanced token information for formatting."""
import keyword
import re
from lib2to3.pgen2 import token
from yapf.pytree import pytree_utils
from yapf.yapflib import py3compat
from yapf.yapflib import style
from yapf.yapflib import subtypes
_OPENING_BRACKETS = frozenset({'(', '[', '{'})
_CLOSING_BRACKETS = frozenset({')', ']', '}'})
def _TabbedContinuationAlignPadding(spaces, align_style, tab_width):
"""Build padding string for continuation alignment in tabbed indentation.
spaces: (int) The number of spaces to place before the token for alignment.
align_style: (str) The alignment style for continuation lines.
tab_width: (int) Number of columns of each tab character.
A padding string for alignment with style specified by align_style option.
if align_style in ('FIXED', 'VALIGN-RIGHT'):
if spaces > 0:
return '\t' * int((spaces + tab_width - 1) / tab_width)
return ''
return ' ' * spaces
class FormatToken(object):
"""Enhanced token information for formatting.
This represents the token plus additional information useful for reformatting
the code.
node: The original token node.
next_token: The token in the logical line after this token or None if this
is the last token in the logical line.
previous_token: The token in the logical line before this token or None if
this is the first token in the logical line.
matching_bracket: If a bracket token ('[', '{', or '(') the matching
parameters: If this and its following tokens make up a parameter list, then
this is a list of those parameters.
container_opening: If the object is in a container, this points to its
opening bracket.
container_elements: If this is the start of a container, a list of the
elements in the container.
whitespace_prefix: The prefix for the whitespace.
spaces_required_before: The number of spaces required before a token. This
is a lower-bound for the formatter and not a hard requirement. For
instance, a comment may have n required spaces before it. But the
formatter won't place n spaces before all comments. Only those that are
moved to the end of a line of code. The formatter may use different
spacing when appropriate.
total_length: The total length of the logical line up to and including
whitespace and this token. However, this doesn't include the initial
indentation amount.
split_penalty: The penalty for splitting the line before this token.
can_break_before: True if we're allowed to break before this token.
must_break_before: True if we're required to break before this token.
newlines: The number of newlines needed before this token.
def __init__(self, node, name):
node: (pytree.Leaf) The node that's being wrapped.
name: (string) The name of the node.
self.node = node = name
self.type = node.type
self.column = node.column
self.lineno = node.lineno
self.value = node.value
if self.is_continuation:
self.value = node.value.rstrip()
self.next_token = None
self.previous_token = None
self.matching_bracket = None
self.parameters = []
self.container_opening = None
self.container_elements = []
self.whitespace_prefix = ''
self.total_length = 0
self.split_penalty = 0
self.can_break_before = False
self.must_break_before = pytree_utils.GetNodeAnnotation(
node, pytree_utils.Annotation.MUST_SPLIT, default=False)
self.newlines = pytree_utils.GetNodeAnnotation(
node, pytree_utils.Annotation.NEWLINES)
self.spaces_required_before = 0
if self.is_comment:
self.spaces_required_before = style.Get('SPACES_BEFORE_COMMENT')
stypes = pytree_utils.GetNodeAnnotation(node,
self.subtypes = {subtypes.NONE} if not stypes else stypes
self.is_pseudo = hasattr(node, 'is_pseudo') and node.is_pseudo
def formatted_whitespace_prefix(self):
if style.Get('INDENT_BLANK_LINES'):
without_newlines = self.whitespace_prefix.lstrip('\n')
height = len(self.whitespace_prefix) - len(without_newlines)
if height:
return ('\n' + without_newlines) * height
return self.whitespace_prefix
def AddWhitespacePrefix(self, newlines_before, spaces=0, indent_level=0):
"""Register a token's whitespace prefix.
This is the whitespace that will be output before a token's string.
newlines_before: (int) The number of newlines to place before the token.
spaces: (int) The number of spaces to place before the token.
indent_level: (int) The indentation level.
if style.Get('USE_TABS'):
if newlines_before > 0:
indent_before = '\t' * indent_level + _TabbedContinuationAlignPadding(
spaces, style.Get('CONTINUATION_ALIGN_STYLE'),
indent_before = '\t' * indent_level + ' ' * spaces
indent_before = (' ' * indent_level * style.Get('INDENT_WIDTH') +
' ' * spaces)
if self.is_comment:
comment_lines = [s.lstrip() for s in self.value.splitlines()]
self.value = ('\n' + indent_before).join(comment_lines)
# Update our own value since we are changing node value
self.value = self.value
if not self.whitespace_prefix:
self.whitespace_prefix = ('\n' * (self.newlines or newlines_before) +
self.whitespace_prefix += indent_before
def AdjustNewlinesBefore(self, newlines_before):
"""Change the number of newlines before this token."""
self.whitespace_prefix = ('\n' * newlines_before +
def RetainHorizontalSpacing(self, first_column, depth):
"""Retains a token's horizontal spacing."""
previous = self.previous_token
if not previous:
if previous.is_pseudo:
previous = previous.previous_token
if not previous:
cur_lineno = self.lineno
prev_lineno = previous.lineno
if previous.is_multiline_string:
prev_lineno += previous.value.count('\n')
if (cur_lineno != prev_lineno or
(previous.is_pseudo and previous.value != ')' and
cur_lineno != previous.previous_token.lineno)):
self.spaces_required_before = (
self.column - first_column + depth * style.Get('INDENT_WIDTH'))
cur_column = self.column
prev_column = previous.node.column
prev_len = len(previous.value)
if previous.is_pseudo and previous.value == ')':
prev_column -= 1
prev_len = 0
if previous.is_multiline_string:
prev_len = len(previous.value.split('\n')[-1])
if '\n' in previous.value:
prev_column = 0 # Last line starts in column 0.
self.spaces_required_before = cur_column - (prev_column + prev_len)
def OpensScope(self):
return self.value in _OPENING_BRACKETS
def ClosesScope(self):
return self.value in _CLOSING_BRACKETS
def AddSubtype(self, subtype):
def __repr__(self):
msg = ('FormatToken(name={0}, value={1}, column={2}, lineno={3}, '
'DOCSTRING' if self.is_docstring else, self.value,
self.column, self.lineno, self.split_penalty))
msg += ', pseudo)' if self.is_pseudo else ')'
return msg
def node_split_penalty(self):
"""Split penalty attached to the pytree node of this token."""
return pytree_utils.GetNodeAnnotation(
self.node, pytree_utils.Annotation.SPLIT_PENALTY, default=0)
def is_binary_op(self):
"""Token is a binary operator."""
return subtypes.BINARY_OPERATOR in self.subtypes
def is_arithmetic_op(self):
"""Token is an arithmetic operator."""
return self.value in frozenset({
'+', # Add
'-', # Subtract
'*', # Multiply
'@', # Matrix Multiply
'/', # Divide
'//', # Floor Divide
'%', # Modulo
'<<', # Left Shift
'>>', # Right Shift
'|', # Bitwise Or
'&', # Bitwise Add
'^', # Bitwise Xor
'**', # Power
def is_simple_expr(self):
"""Token is an operator in a simple expression."""
return subtypes.SIMPLE_EXPRESSION in self.subtypes
def is_subscript_colon(self):
"""Token is a subscript colon."""
return subtypes.SUBSCRIPT_COLON in self.subtypes
def is_comment(self):
return self.type == token.COMMENT
def is_continuation(self):
return self.type == CONTINUATION
def is_keyword(self):
return keyword.iskeyword(self.value)
def is_name(self):
return self.type == token.NAME and not self.is_keyword
def is_number(self):
return self.type == token.NUMBER
def is_string(self):
return self.type == token.STRING
def is_multiline_string(self):
"""Test if this string is a multiline string.
A multiline string always ends with triple quotes, so if it is a string
token, inspect the last 3 characters and return True if it is a triple
double or triple single quote mark.
return self.is_string and self.value.endswith(('"""', "'''"))
def is_docstring(self):
return self.is_string and self.previous_token is None
def is_pylint_comment(self):
return self.is_comment and re.match(r'#.*\bpylint:\s*(disable|enable)=',
def is_pytype_comment(self):
return self.is_comment and re.match(r'#.*\bpytype:\s*(disable|enable)=',
def is_copybara_comment(self):
return self.is_comment and re.match(
r'#.*\bcopybara:\s*(strip|insert|replace)', self.value)