eric: comparison Plugins/CheckerPlugins/CodeStyleChecker/pep8.py

-:47e29cc0f29c
+:081b69e00f3e
+#!/usr/bin/env python
 # -*- coding: utf-8 -*-
-#
 # pep8.py - Check Python source code formatting, according to PEP 8
 # Copyright (C) 2006-2009 Johann C. Rocholl <johann@rocholl.net>
-# Copyright (C) 2009-2013 Florent Xicluna <florent.xicluna@gmail.com>
+# Copyright (C) 2009-2014 Florent Xicluna <florent.xicluna@gmail.com>
 #
 # Permission is hereby granted, free of charge, to any person
 # obtaining a copy of this software and associated documentation files
 # (the "Software"), to deal in the Software without restriction,
 # including without limitation the rights to use, copy, modify, merge,
 # BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
 # ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
 # CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 # SOFTWARE.
-"""
+r"""
-Check Python source code formatting, according to PEP 8:
+Check Python source code formatting, according to PEP 8.
-http://www.python.org/dev/peps/pep-0008/
 For usage and a list of options, try this:
 $ python pep8.py -h
 This program and its regression test suite live here:
 500 line length
 600 deprecation
 700 statements
 900 syntax error
 """
+from __future__ import with_statement
 #
 # This is a modified version to make the original pep8.py better suitable
 # for being called from within the eric5 IDE. The modifications are as
 # follows:
 # - added code for eric5 integration
 #
 # Copyright (c) 2011 - 2014 Detlev Offenbach <detlev@die-offenbachs.de>
 #
-__version__ = '1.4.6'
+__version__ = '1.5.6'
 import os
 import sys
 import re
 import time
 WS_OPTIONAL_OPERATORS = ARITHMETIC_OP.union(['^', '&', '|', '<<', '>>', '%'])
 WS_NEEDED_OPERATORS = frozenset([
 '**=', '*=', '/=', '//=', '+=', '-=', '!=', '<>', '<', '>',
 '%=', '^=', '&=', '|=', '==', '<=', '>=', '<<=', '>>=', '='])
 WHITESPACE = frozenset(' \t')
-SKIP_TOKENS = frozenset([tokenize.COMMENT, tokenize.NL, tokenize.NEWLINE,
+NEWLINE = frozenset([tokenize.NL, tokenize.NEWLINE])
-tokenize.INDENT, tokenize.DEDENT])
+SKIP_TOKENS = NEWLINE.union([tokenize.INDENT, tokenize.DEDENT])
+# ERRORTOKEN is triggered by backticks in Python 3
+SKIP_COMMENTS = SKIP_TOKENS.union([tokenize.COMMENT, tokenize.ERRORTOKEN])
 BENCHMARK_KEYS = ['directories', 'files', 'logical lines', 'physical lines']
 INDENT_REGEX = re.compile(r'([ \t]*)')
 RAISE_COMMA_REGEX = re.compile(r'raise\s+\w+\s*,')
-RERAISE_COMMA_REGEX = re.compile(r'raise\s+\w+\s*,\s*\w+\s*,\s*\w+')
+RERAISE_COMMA_REGEX = re.compile(r'raise\s+\w+\s*,.*,\s*\w+\s*$')
 ERRORCODE_REGEX = re.compile(r'\b[A-Z]\d{3}\b')
 DOCSTRING_REGEX = re.compile(r'u?r?["\']')
 EXTRANEOUS_WHITESPACE_REGEX = re.compile(r'[[({] | []}),;:]')
 WHITESPACE_AFTER_COMMA_REGEX = re.compile(r'[,;:]\s*(?:  |\t)')
 COMPARE_SINGLETON_REGEX = re.compile(r'([=!]=)\s*(None|False|True)')
+COMPARE_NEGATIVE_REGEX = re.compile(r'\b(not)\s+[^[({ ]+\s+(in|is)\s')
 COMPARE_TYPE_REGEX = re.compile(r'(?:[=!]=|is(?:\s+not)?)\s*type(?:s.\w+Type'
 r'|\s*\(\s*([^)]*[^ )])\s*\))')
 KEYWORD_REGEX = re.compile(r'(\s*)\b(?:%s)\b(\s*)' % r'|'.join(KEYWORDS))
 OPERATOR_REGEX = re.compile(r'(?:[^,\s])(\s*)(?:[-+*/|!<=>%&^]+)(\s*)')
 LAMBDA_REGEX = re.compile(r'\blambda\b')
 # Plugins (check functions) for physical lines
 ##############################################################################
 def tabs_or_spaces(physical_line, indent_char):
-r"""
+r"""Never mix tabs and spaces.
-Never mix tabs and spaces.
 The most popular way of indenting Python is with spaces only.  The
 second-most popular way is with tabs only.  Code indented with a mixture
 of tabs and spaces should be converted to using spaces exclusively.  When
 invoking the Python command line interpreter with the -t option, it issues
 E101: if a == 0:\n        a = 1\n\tb = 1
 """
 indent = INDENT_REGEX.match(physical_line).group(1)
 for offset, char in enumerate(indent):
 if char != indent_char:
-return offset, "E101"
+return offset, "E101 indentation contains mixed spaces and tabs"
 def tabs_obsolete(physical_line):
-r"""
+r"""For new projects, spaces-only are strongly recommended over tabs.
-For new projects, spaces-only are strongly recommended over tabs.  Most
-editors have features that make this easy to do.
 Okay: if True:\n    return
 W191: if True:\n\treturn
 """
 indent = INDENT_REGEX.match(physical_line).group(1)
 if '\t' in indent:
-return indent.index('\t'), "W191"
+return indent.index('\t'), "W191 indentation contains tabs"
 def trailing_whitespace(physical_line):
-r"""
+r"""Trailing whitespace is superfluous.
-JCR: Trailing whitespace is superfluous.
-FBM: Except when it occurs as part of a blank line (i.e. the line is
-nothing but whitespace). According to Python docs[1] a line with only
-whitespace is considered a blank line, and is to be ignored. However,
-matching a blank line to its indentation level avoids mistakenly
-terminating a multi-line statement (e.g. class declaration) when
-pasting code into the standard Python interpreter.
-[1] http://docs.python.org/reference/lexical_analysis.html#blank-lines
 The warning returned varies on whether the line itself is blank, for easier
 filtering for those who want to indent their blank lines.
 Okay: spam(1)\n#
 physical_line = physical_line.rstrip('\r')    # chr(13), carriage return
 physical_line = physical_line.rstrip('\x0c')  # chr(12), form feed, ^L
 stripped = physical_line.rstrip(' \t\v')
 if physical_line != stripped:
 if stripped:
-return len(stripped), "W291"
+return len(stripped), "W291 trailing whitespace"
 else:
-return 0, "W293"
+return 0, "W293 blank line contains whitespace"
-def trailing_blank_lines(physical_line, lines, line_number):
+def trailing_blank_lines(physical_line, lines, line_number, total_lines):
-r"""
+r"""Trailing blank lines are superfluous.
-JCR: Trailing blank lines are superfluous.
 Okay: spam(1)
 W391: spam(1)\n
-"""
-if not physical_line.rstrip() and line_number == len(lines):
+However the last line should end with a new line (warning W292).
-return 0, "W391"
+"""
+if line_number == total_lines:
+stripped_last_line = physical_line.rstrip()
-def missing_newline(physical_line):
+if not stripped_last_line:
-"""
+return 0, "W391 blank line at end of file"
-JCR: The last line should have a newline.
+if stripped_last_line == physical_line:
+return len(physical_line), "W292 no newline at end of file"
-Reports warning W292.
-"""
-if physical_line.rstrip() == physical_line:
+def maximum_line_length(physical_line, max_line_length, multiline):
-return len(physical_line), "W292"
+r"""Limit all lines to a maximum of 79 characters.
-def maximum_line_length(physical_line, max_line_length):
-"""
-Limit all lines to a maximum of 79 characters.
 There are still many devices around that are limited to 80 character
 lines; plus, limiting windows to 80 characters makes it possible to have
 several windows side-by-side.  The default wrapping on such devices looks
 ugly.  Therefore, please limit all lines to a maximum of 79 characters.
 Reports error E501.
 """
 line = physical_line.rstrip()
 length = len(line)
 if length > max_line_length and not noqa(line):
+# Special case for long URLs in multi-line docstrings or comments,
+# but still report the error when the 72 first chars are whitespaces.
+chunks = line.split()
+if ((len(chunks) == 1 and multiline) or
+(len(chunks) == 2 and chunks[0] == '#')) and \
+len(line) - len(chunks[-1]) < max_line_length - 7:
+return
 if hasattr(line, 'decode'):   # Python 2
 # The line could contain multi-byte characters
 try:
 length = len(line.decode('utf-8'))
 except UnicodeError:
 pass
 if length > max_line_length:
-return max_line_length, "E501", length, max_line_length
+return (max_line_length, "E501 line too long ", length, max_line_length)
 ##############################################################################
 # Plugins (check functions) for logical lines
 ##############################################################################
 def blank_lines(logical_line, blank_lines, indent_level, line_number,
-previous_logical, previous_indent_level):
+blank_before, previous_logical, previous_indent_level):
-r"""
+r"""Separate top-level function and class definitions with two blank lines.
-Separate top-level function and class definitions with two blank lines.
 Method definitions inside a class are separated by a single blank line.
 Extra blank lines may be used (sparingly) to separate groups of related
 functions.  Blank lines may be omitted between a bunch of related
 """
 if line_number < 3 and not previous_logical:
 return  # Don't expect blank lines before the first line
 if previous_logical.startswith('@'):
 if blank_lines:
-yield 0, "E304"
+yield 0, "E304 blank lines found after function decorator"
 elif blank_lines > 2 or (indent_level and blank_lines == 2):
-yield 0, "E303", blank_lines
+yield 0, "E303 too many blank lines (%d)", blank_lines
 elif logical_line.startswith(('def ', 'class ', '@')):
 if indent_level:
-if not (blank_lines or previous_indent_level < indent_level or
+if not (blank_before or previous_indent_level < indent_level or
 DOCSTRING_REGEX.match(previous_logical)):
-yield 0, "E301"
+yield 0, "E301 expected 1 blank line, found 0"
-elif blank_lines != 2:
+elif blank_before != 2:
-yield 0, "E302", blank_lines
+yield 0, "E302 expected 2 blank lines, found %d", blank_before
 def extraneous_whitespace(logical_line):
-"""
+r"""Avoid extraneous whitespace.
-Avoid extraneous whitespace in the following situations:
+Avoid extraneous whitespace in these situations:
 - Immediately inside parentheses, brackets or braces.
 - Immediately before a comma, semicolon, or colon.
 Okay: spam(ham[1], {eggs: 2})
 E201: spam( ham[1], {eggs: 2})
 E201: spam(ham[ 1], {eggs: 2})
 text = match.group()
 char = text.strip()
 found = match.start()
 if text == char + ' ':
 # assert char in '([{'
-yield found + 1, "E201", char
+yield found + 1, "E201 whitespace after '%s'", char
 elif line[found - 1] != ',':
 code = ('E202' if char in '}])' else 'E203')  # if char in ',;:'
-yield found, code, char
+yield found, "%s whitespace before '%s'" % (code, char), char
 def whitespace_around_keywords(logical_line):
-r"""
+r"""Avoid extraneous whitespace around keywords.
-Avoid extraneous whitespace around keywords.
 Okay: True and False
 E271: True and  False
 E272: True  and False
 E273: True and\tFalse
 """
 for match in KEYWORD_REGEX.finditer(logical_line):
 before, after = match.groups()
 if '\t' in before:
-yield match.start(1), "E274"
+yield match.start(1), "E274 tab before keyword"
 elif len(before) > 1:
-yield match.start(1), "E272"
+yield match.start(1), "E272 multiple spaces before keyword"
 if '\t' in after:
-yield match.start(2), "E273"
+yield match.start(2), "E273 tab after keyword"
 elif len(after) > 1:
-yield match.start(2), "E271"
+yield match.start(2), "E271 multiple spaces after keyword"
 def missing_whitespace(logical_line):
-"""
+r"""Each comma, semicolon or colon should be followed by whitespace.
-JCR: Each comma, semicolon or colon should be followed by whitespace.
 Okay: [a, b]
 Okay: (3,)
 Okay: a[1:4]
 Okay: a[:4]
 if char == ':' and before.count('[') > before.count(']') and \
 before.rfind('{') < before.rfind('['):
 continue  # Slice syntax, no space required
 if char == ',' and line[index + 1] == ')':
 continue  # Allow tuple with only one element: (3,)
-yield index, "E231", char
+yield index, "E231 missing whitespace after '%s'", char
 def indentation(logical_line, previous_logical, indent_char,
 indent_level, previous_indent_level):
-r"""
+r"""Use 4 spaces per indentation level.
-Use 4 spaces per indentation level.
 For really old code that you don't want to mess up, you can continue to
 use 8-space tabs.
 Okay: a = 1
 Okay: a = 1\nb = 2
 E113: a = 1\n    b = 2
 """
 if indent_char == ' ' and indent_level % 4:
-yield 0, "E111"
+yield 0, "E111 indentation is not a multiple of four"
 indent_expect = previous_logical.endswith(':')
 if indent_expect and indent_level <= previous_indent_level:
-yield 0, "E112"
+yield 0, "E112 expected an indented block"
 if indent_level > previous_indent_level and not indent_expect:
-yield 0, "E113"
+yield 0, "E113 unexpected indentation"
 def continued_indentation(logical_line, tokens, indent_level, hang_closing,
-noqa, verbose):
+indent_char, noqa, verbose):
-r"""
+r"""Continuation lines indentation.
-Continuation lines should align wrapped elements either vertically using
-Python's implicit line joining inside parentheses, brackets and braces, or
+Continuation lines should align wrapped elements either vertically
-using a hanging indent.
+using Python's implicit line joining inside parentheses, brackets
+and braces, or using a hanging indent.
-When using a hanging indent the following considerations should be applied:
+When using a hanging indent these considerations should be applied:
 - there should be no arguments on the first line, and
 - further indentation should be used to clearly distinguish itself as a
 continuation line.
 Okay: a = (\n)
 E123: a = (\n    )
 Okay: a = (\n    42)
 E121: a = (\n   42)
 E122: a = (\n42)
 E123: a = (\n    42\n    )
 E124: a = (24,\n     42\n)
-E125: if (a or\n    b):\n    pass
+E125: if (\n    b):\n    pass
 E126: a = (\n        42)
 E127: a = (24,\n      42)
 E128: a = (24,\n    42)
+E129: if (a or\n    b):\n    pass
+E131: a = (\n    42\n 24)
 """
 first_row = tokens[0][2][0]
 nrows = 1 + tokens[-1][2][0] - first_row
 if noqa or nrows == 1:
 return
 # indents on the final continuation line; in turn, some other
 # indents are allowed to have an extra 4 spaces.
 indent_next = logical_line.endswith(':')
 row = depth = 0
+valid_hangs = (4,) if indent_char != '\t' else (4, 8)
 # remember how many brackets were opened on each line
 parens = [0] * nrows
 # relative indents of physical lines
 rel_indent = [0] * nrows
+# for each depth, collect a list of opening rows
+open_rows = [[0]]
+# for each depth, memorize the hanging indentation
+hangs = [None]
 # visual indents
 indent_chances = {}
 last_indent = tokens[0][2]
+visual_indent = None
+# for each depth, memorize the visual indent column
 indent = [last_indent[1]]
 if verbose >= 3:
 print(">>> " + tokens[0][4].rstrip())
 for token_type, text, start, end, line in tokens:
 last_token_multiline = (start[0] != end[0])
 newline = row < start[0] - first_row
 if newline:
 row = start[0] - first_row
-newline = (not last_token_multiline and
+newline = not last_token_multiline and token_type not in NEWLINE
-token_type not in (tokenize.NL, tokenize.NEWLINE))
 if newline:
 # this is the beginning of a continuation line.
 last_indent = start
 if verbose >= 3:
 print("... " + line.rstrip())
 # record the initial indent.
 rel_indent[row] = expand_indent(line) - indent_level
-if depth:
+# identify closing bracket
-# a bracket expression in a continuation line.
-# find the line that it was opened on
-for open_row in range(row - 1, -1, -1):
-if parens[open_row]:
-break
-else:
-# an unbracketed continuation line (ie, backslash)
-open_row = 0
-hang = rel_indent[row] - rel_indent[open_row]
 close_bracket = (token_type == tokenize.OP and text in ']})')
+# is the indent relative to an opening bracket line?
+for open_row in reversed(open_rows[depth]):
+hang = rel_indent[row] - rel_indent[open_row]
+hanging_indent = hang in valid_hangs
+if hanging_indent:
+break
+if hangs[depth]:
+hanging_indent = (hang == hangs[depth])
+# is there any chance of visual indent?
 visual_indent = (not close_bracket and hang > 0 and
 indent_chances.get(start[1]))
 if close_bracket and indent[depth]:
 # closing bracket for visual indent
 if start[1] != indent[depth]:
-yield start, "E124"
+yield (start, "E124 closing bracket does not match "
+"visual indentation")
 elif close_bracket and not hang:
 # closing bracket matches indentation of opening bracket's line
 if hang_closing:
-yield start, "E133"
+yield start, "E133 closing bracket is missing indentation"
+elif indent[depth] and start[1] < indent[depth]:
+if visual_indent is not True:
+# visual indent is broken
+yield (start, "E128 continuation line "
+"under-indented for visual indent")
+elif hanging_indent or (indent_next and rel_indent[row] == 8):
+# hanging indent is verified
+if close_bracket and not hang_closing:
+yield (start, "E123 closing bracket does not match "
+"indentation of opening bracket's line")
+hangs[depth] = hang
 elif visual_indent is True:
 # visual indent is verified
-if not indent[depth]:
+indent[depth] = start[1]
-indent[depth] = start[1]
 elif visual_indent in (text, str):
 # ignore token lined up with matching one from a previous line
 pass
-elif indent[depth] and start[1] < indent[depth]:
-# visual indent is broken
-yield start, "E128"
-elif hang == 4 or (indent_next and rel_indent[row] == 8):
-# hanging indent is verified
-if close_bracket and not hang_closing:
-yield (start, "E123")
 else:
 # indent is broken
 if hang <= 0:
-error = "E122"
+error = "E122", "missing indentation or outdented"
 elif indent[depth]:
-error = "E127"
+error = "E127", "over-indented for visual indent"
-elif hang % 4:
+elif not close_bracket and hangs[depth]:
-error = "E121"
+error = "E131", "unaligned for hanging indent"
 else:
-error = "E126"
+hangs[depth] = hang
-yield start, error
+if hang > 4:
+error = "E126", "over-indented for hanging indent"
+else:
+error = "E121", "under-indented for hanging indent"
+yield start, "%s continuation line %s" % error
 # look for visual indenting
 if (parens[row] and token_type not in (tokenize.NL, tokenize.COMMENT)
 and not indent[depth]):
 indent[depth] = start[1]
 text in ('u', 'ur', 'b', 'br')):
 indent_chances[start[1]] = str
 # special case for the "if" statement because len("if (") == 4
 elif not indent_chances and not row and not depth and text == 'if':
 indent_chances[end[1] + 1] = True
+elif text == ':' and line[end[1]:].isspace():
+open_rows[depth].append(row)
 # keep track of bracket depth
 if token_type == tokenize.OP:
 if text in '([{':
 depth += 1
 indent.append(0)
+hangs.append(None)
+if len(open_rows) == depth:
+open_rows.append([])
+open_rows[depth].append(row)
 parens[row] += 1
 if verbose >= 4:
 print("bracket depth %s seen, col %s, visual min = %s" %
 (depth, start[1], indent[depth]))
 elif text in ')]}' and depth > 0:
 # parent indents should not be more than this one
 prev_indent = indent.pop() or last_indent[1]
+hangs.pop()
 for d in range(depth):
 if indent[d] > prev_indent:
 indent[d] = 0
 for ind in list(indent_chances):
 if ind >= prev_indent:
 del indent_chances[ind]
+del open_rows[depth + 1:]
 depth -= 1
 if depth:
 indent_chances[indent[depth]] = True
 for idx in range(row, -1, -1):
 if parens[idx]:
 parens[idx] -= 1
-rel_indent[row] = rel_indent[idx]
 break
 assert len(indent) == depth + 1
 if start[1] not in indent_chances:
 # allow to line up tokens
 indent_chances[start[1]] = text
+if last_token_multiline:
+rel_indent[end[0] - first_row] = rel_indent[row]
 if indent_next and expand_indent(line) == indent_level + 4:
-yield last_indent, "E125"
+pos = (start[0], indent[0] + 4)
+if visual_indent:
+code = "E129 visually indented line"
+else:
+code = "E125 continuation line"
+yield pos, "%s with same indent as next logical line" % code
 def whitespace_before_parameters(logical_line, tokens):
-"""
+r"""Avoid extraneous whitespace.
 Avoid extraneous whitespace in the following situations:
+- before the open parenthesis that starts the argument list of a
-- Immediately before the open parenthesis that starts the argument
+function call.
-list of a function call.
+- before the open parenthesis that starts an indexing or slicing.
-- Immediately before the open parenthesis that starts an indexing or
-slicing.
 Okay: spam(1)
 E211: spam (1)
 Okay: dict['key'] = list[index]
 (prev_type == tokenize.NAME or prev_text in '}])') and
 # Syntax "class A (B):" is allowed, but avoid it
 (index < 2 or tokens[index - 2][1] != 'class') and
 # Allow "return (a.foo for a in range(5))"
 not keyword.iskeyword(prev_text)):
-yield prev_end, "E211", text
+yield prev_end, "E211 whitespace before '%s'", text
 prev_type = token_type
 prev_text = text
 prev_end = end
 def whitespace_around_operator(logical_line):
-"""
+r"""Avoid extraneous whitespace around an operator.
-Avoid extraneous whitespace in the following situations:
-- More than one space around an assignment (or other) operator to
-align it with another.
 Okay: a = 12 + 3
 E221: a = 4  + 5
 E222: a = 4 +  5
 E223: a = 4\t+ 5
 """
 for match in OPERATOR_REGEX.finditer(logical_line):
 before, after = match.groups()
 if '\t' in before:
-yield match.start(1), "E223"
+yield match.start(1), "E223 tab before operator"
 elif len(before) > 1:
-yield match.start(1), "E221"
+yield match.start(1), "E221 multiple spaces before operator"
 if '\t' in after:
-yield match.start(2), "E224"
+yield match.start(2), "E224 tab after operator"
 elif len(after) > 1:
-yield match.start(2), "E222"
+yield match.start(2), "E222 multiple spaces after operator"
 def missing_whitespace_around_operator(logical_line, tokens):
-r"""
+r"""Surround operators with a single space on either side.
 - Always surround these binary operators with a single space on
 either side: assignment (=), augmented assignment (+=, -= etc.),
-comparisons (==, <, >, !=, <>, <=, >=, in, not in, is, is not),
+comparisons (==, <, >, !=, <=, >=, in, not in, is, is not),
 Booleans (and, or, not).
-- Use spaces around arithmetic operators.
+- If operators with different priorities are used, consider adding
+whitespace around the operators with the lowest priorities.
 Okay: i = i + 1
 Okay: submitted += 1
 Okay: x = x * 2 - 1
 Okay: hypot2 = x * x + y * y
 parens = 0
 need_space = False
 prev_type = tokenize.OP
 prev_text = prev_end = None
 for token_type, text, start, end, line in tokens:
-if token_type in (tokenize.NL, tokenize.NEWLINE, tokenize.ERRORTOKEN):
+if token_type in SKIP_COMMENTS:
-# ERRORTOKEN is triggered by backticks in Python 3
 continue
 if text in ('(', 'lambda'):
 parens += 1
 elif text == ')':
 parens -= 1
 if need_space:
 if start != prev_end:
 # Found a (probably) needed space
 if need_space is not True and not need_space[1]:
-yield need_space[0], "E225"
+yield (need_space[0],
+"E225 missing whitespace around operator")
 need_space = False
 elif text == '>' and prev_text in ('<', '-'):
 # Tolerate the "<>" operator, even if running Python 3
 # Deal with Python 3's annotated return value "->"
 pass
 else:
 if need_space is True or need_space[1]:
 # A needed trailing space was not found
-yield prev_end, "E225"
+yield prev_end, "E225 missing whitespace around operator"
 else:
-code = 'E226'
+code, optype = 'E226', 'arithmetic'
 if prev_text == '%':
-code = 'E228'
+code, optype = 'E228', 'modulo'
 elif prev_text not in ARITHMETIC_OP:
-code = 'E227'
+code, optype = 'E227', 'bitwise or shift'
-yield need_space[0], code
+yield (need_space[0], "%s missing whitespace "
+"around %s operator" % (code, optype))
 need_space = False
 elif token_type == tokenize.OP and prev_end is not None:
 if text == '=' and parens:
 # Allow keyword args or defaults: foo(bar=None).
 pass
 need_space = True
 elif text in UNARY_OPERATORS:
 # Check if the operator is being used as a binary operator
 # Allow unary operators: -123, -x, +1.
 # Allow argument unpacking: foo(*args, **kwargs).
-if prev_type == tokenize.OP:
+if (prev_text in '}])' if prev_type == tokenize.OP
-binary_usage = (prev_text in '}])')
+else prev_text not in KEYWORDS):
-elif prev_type == tokenize.NAME:
-binary_usage = (prev_text not in KEYWORDS)
-else:
-binary_usage = (prev_type not in SKIP_TOKENS)
-if binary_usage:
 need_space = None
 elif text in WS_OPTIONAL_OPERATORS:
 need_space = None
 if need_space is None:
 # Surrounding space is optional, but ensure that
 # trailing space matches opening space
 need_space = (prev_end, start != prev_end)
 elif need_space and start == prev_end:
 # A needed opening space was not found
-yield prev_end, "E225"
+yield prev_end, "E225 missing whitespace around operator"
 need_space = False
 prev_type = token_type
 prev_text = text
 prev_end = end
 def whitespace_around_comma(logical_line):
-"""
+r"""Avoid extraneous whitespace after a comma or a colon.
-Avoid extraneous whitespace in the following situations:
-- More than one space around an assignment (or other) operator to
-align it with another.
-JCR: This should also be applied around comma etc.
 Note: these checks are disabled by default
 Okay: a = (1, 2)
 E241: a = (1,  2)
 E242: a = (1,\t2)
 """
 line = logical_line
 for m in WHITESPACE_AFTER_COMMA_REGEX.finditer(line):
 found = m.start() + 1
 if '\t' in m.group():
-yield found, "E242", m.group()[0]
+yield found, "E242 tab after '%s'", m.group()[0]
 else:
-yield found, "E241", m.group()[0]
+yield found, "E241 multiple spaces after '%s'", m.group()[0]
 def whitespace_around_named_parameter_equals(logical_line, tokens):
-"""
+r"""Don't use spaces around the '=' sign in function arguments.
 Don't use spaces around the '=' sign when used to indicate a
 keyword argument or a default parameter value.
 Okay: def complex(real, imag=0.0):
 Okay: return magic(r=real, i=imag)
 E251: return magic(r = real, i = imag)
 """
 parens = 0
 no_space = False
 prev_end = None
-message = "E251"
+message = "E251 unexpected spaces around keyword / parameter equals"
 for token_type, text, start, end, line in tokens:
+if token_type == tokenize.NL:
+continue
 if no_space:
 no_space = False
 if start != prev_end:
-yield prev_end, message
+yield (prev_end, message)
 elif token_type == tokenize.OP:
 if text == '(':
 parens += 1
 elif text == ')':
 parens -= 1
 elif parens and text == '=':
 no_space = True
 if start != prev_end:
-yield prev_end, message
+yield (prev_end, message)
 prev_end = end
-def whitespace_before_inline_comment(logical_line, tokens):
+def whitespace_before_comment(logical_line, tokens):
-"""
+r"""Separate inline comments by at least two spaces.
-Separate inline comments by at least two spaces.
 An inline comment is a comment on the same line as a statement.  Inline
 comments should be separated by at least two spaces from the statement.
 They should start with a # and a single space.
+Each line of a block comment starts with a # and a single space
+(unless it is indented text inside the comment).
 Okay: x = x + 1  # Increment x
 Okay: x = x + 1    # Increment x
+Okay: # Block comment
 E261: x = x + 1 # Increment x
 E262: x = x + 1  #Increment x
 E262: x = x + 1  #  Increment x
+E265: #Block comment
 """
 prev_end = (0, 0)
 for token_type, text, start, end, line in tokens:
 if token_type == tokenize.COMMENT:
-if not line[:start[1]].strip():
+inline_comment = line[:start[1]].strip()
-continue
+if inline_comment:
 if prev_end[0] == start[0] and start[1] < prev_end[1] + 2:
-yield prev_end, "E261"
+yield (prev_end,
+"E261 at least two spaces before inline comment")
 symbol, sp, comment = text.partition(' ')
-if symbol not in ('#', '#:') or comment[:1].isspace():
+bad_prefix = symbol not in ('#', '#:')
-yield start, "E262"
+if inline_comment:
+if bad_prefix or comment[:1].isspace():
+yield start, "E262 inline comment should start with '# '"
+elif bad_prefix:
+if text.rstrip('#') and (start[0] > 1 or symbol[1] != '!'):
+yield start, "E265 block comment should start with '# '"
 elif token_type != tokenize.NL:
 prev_end = end
 def imports_on_separate_lines(logical_line):
-r"""
+r"""Imports should usually be on separate lines.
-Imports should usually be on separate lines.
 Okay: import os\nimport sys
 E401: import sys, os
 Okay: from subprocess import Popen, PIPE
 """
 line = logical_line
 if line.startswith('import '):
 found = line.find(',')
 if -1 < found and ';' not in line[:found]:
-yield found, "E401"
+yield found, "E401 multiple imports on one line"
 def compound_statements(logical_line):
-r"""
+r"""Compound statements (on the same line) are generally discouraged.
-Compound statements (multiple statements on the same line) are
-generally discouraged.
 While sometimes it's okay to put an if/for/while with a small body
-on the same line, never do this for multi-clause statements. Also
+on the same line, never do this for multi-clause statements.
-avoid folding such long lines!
+Also avoid folding such long lines!
 Okay: if foo == 'blah':\n    do_blah_thing()
 Okay: do_one()
 Okay: do_two()
 Okay: do_three()
 before = line[:found]
 if (before.count('{') <= before.count('}') and  # {'a': 1} (dict)
 before.count('[') <= before.count(']') and  # [1:2] (slice)
 before.count('(') <= before.count(')') and  # (Python 3 annotation)
 not LAMBDA_REGEX.search(before)):       # lambda x: x
-yield found, "E701"
+yield found, "E701 multiple statements on one line (colon)"
 found = line.find(':', found + 1)
 found = line.find(';')
 while -1 < found:
 if found < last_char:
-yield found, "E702"
+yield found, "E702 multiple statements on one line (semicolon)"
 else:
-yield found, "E703"
+yield found, "E703 statement ends with a semicolon"
 found = line.find(';', found + 1)
 def explicit_line_join(logical_line, tokens):
-r"""
+r"""Avoid explicit line join between brackets.
-Avoid explicit line join between brackets.
 The preferred way of wrapping long lines is by using Python's implied line
 continuation inside parentheses, brackets and braces.  Long lines can be
 broken over multiple lines by wrapping expressions in parentheses.  These
 should be used in preference to using a backslash for line continuation.
 """
 prev_start = prev_end = parens = 0
 backslash = None
 for token_type, text, start, end, line in tokens:
 if start[0] != prev_start and parens and backslash:
-yield backslash, "E502"
+yield backslash, "E502 the backslash is redundant between brackets"
 if end[0] != prev_end:
 if line.rstrip('\r\n').endswith('\\'):
 backslash = (end[0], len(line.splitlines()[-1]) - 1)
 else:
 backslash = None
 elif text in ')]}':
 parens -= 1
 def comparison_to_singleton(logical_line, noqa):
-"""
+r"""Comparison to singletons should use "is" or "is not".
 Comparisons to singletons like None should always be done
 with "is" or "is not", never the equality operators.
 Okay: if arg is not None:
 E711: if arg != None:
 else:
 code = 'E712'
 nonzero = ((singleton == 'True' and same) or
 (singleton == 'False' and not same))
 msg += " or 'if %scond:'" % ('' if nonzero else 'not ')
-yield match.start(1), code, singleton, msg
+yield (match.start(1), "%s comparison to %s should be %s" %
+(code, singleton, msg), singleton, msg)
+def comparison_negative(logical_line):
+r"""Negative comparison should be done using "not in" and "is not".
+Okay: if x not in y:\n    pass
+Okay: assert (X in Y or X is Z)
+Okay: if not (X in Y):\n    pass
+Okay: zz = x is not y
+E713: Z = not X in Y
+E713: if not X.B in Y:\n    pass
+E714: if not X is Y:\n    pass
+E714: Z = not X.B is Y
+"""
+match = COMPARE_NEGATIVE_REGEX.search(logical_line)
+if match:
+pos = match.start(1)
+if match.group(2) == 'in':
+yield pos, "E713 test for membership should be 'not in'"
+else:
+yield pos, "E714 test for object identity should be 'is not'"
 def comparison_type(logical_line):
-"""
+r"""Object type comparisons should always use isinstance().
-Object type comparisons should always use isinstance() instead of
-comparing types directly.
+Do not compare types directly.
 Okay: if isinstance(obj, int):
 E721: if type(obj) is type(1):
 When checking if an object is a string, keep in mind that it might be a
 match = COMPARE_TYPE_REGEX.search(logical_line)
 if match:
 inst = match.group(1)
 if inst and isidentifier(inst) and inst not in SINGLETONS:
 return  # Allow comparison for types which are not obvious
-yield match.start(), "E721"
+yield match.start(), "E721 do not compare types, use 'isinstance()'"
-def python_3000_has_key(logical_line):
+def python_3000_has_key(logical_line, noqa):
-r"""
+r"""The {}.has_key() method is removed in Python 3: use the 'in' operator.
-The {}.has_key() method is removed in the Python 3.
-Use the 'in' operation instead.
 Okay: if "alph" in d:\n    print d["alph"]
 W601: assert d.has_key('alph')
 """
 pos = logical_line.find('.has_key(')
-if pos > -1:
+if pos > -1 and not noqa:
-yield pos, "W601"
+yield pos, "W601 .has_key() is deprecated, use 'in'"
 def python_3000_raise_comma(logical_line):
-"""
+r"""When raising an exception, use "raise ValueError('message')".
-When raising an exception, use "raise ValueError('message')"
-instead of the older form "raise ValueError, 'message'".
+The older form is removed in Python 3.
-The paren-using form is preferred because when the exception arguments
-are long or include string formatting, you don't need to use line
-continuation characters thanks to the containing parentheses.  The older
-form is removed in Python 3.
 Okay: raise DummyError("Message")
 W602: raise DummyError, "Message"
 """
 match = RAISE_COMMA_REGEX.match(logical_line)
 if match and not RERAISE_COMMA_REGEX.match(logical_line):
-yield match.end() - 1, "W602"
+yield match.end() - 1, "W602 deprecated form of raising exception"
 def python_3000_not_equal(logical_line):
-"""
+r"""New code should always use != instead of <>.
-!= can also be written <>, but this is an obsolete usage kept for
-backwards compatibility only. New code should always use !=.
 The older syntax is removed in Python 3.
 Okay: if a != 'no':
 W603: if a <> 'no':
 """
 pos = logical_line.find('<>')
 if pos > -1:
-yield pos, "W603"
+yield pos, "W603 '<>' is deprecated, use '!='"
 def python_3000_backticks(logical_line):
-"""
+r"""Backticks are removed in Python 3: use repr() instead.
-Backticks are removed in Python 3.
-Use repr() instead.
 Okay: val = repr(1 + 2)
 W604: val = `1 + 2`
 """
 pos = logical_line.find('`')
 if pos > -1:
-yield pos, "W604"
+yield pos, "W604 backticks are deprecated, use 'repr()'"
 ##############################################################################
 # Helper functions
 ##############################################################################
 if '' == ''.encode():
 # Python 2: implicit encoding.
 def readlines(filename):
-f = open(filename)
+"""Read the source code."""
-try:
+with open(filename) as f:
 return f.readlines()
-finally:
-f.close()
 isidentifier = re.compile(r'[a-zA-Z_]\w*').match
 stdin_get_value = sys.stdin.read
 else:
 # Python 3
-def readlines(filename):                            # __IGNORE_WARNING__
+def readlines(filename):
-f = open(filename, 'rb')
+"""Read the source code."""
 try:
-coding, lines = tokenize.detect_encoding(f.readline)
+with open(filename, 'rb') as f:
-f = TextIOWrapper(f, coding, line_buffering=True)
+(coding, lines) = tokenize.detect_encoding(f.readline)
-return [l.decode(coding) for l in lines] + f.readlines()
+f = TextIOWrapper(f, coding, line_buffering=True)
+return [l.decode(coding) for l in lines] + f.readlines()
 except (LookupError, SyntaxError, UnicodeError):
-f.close()
+# Fall back if file encoding is improperly declared
-# Fall back if files are improperly declared
+with open(filename, encoding='latin-1') as f:
-f = open(filename, encoding='latin-1')
+return f.readlines()
-return f.readlines()
-finally:
-f.close()
 isidentifier = str.isidentifier
 def stdin_get_value():
 return TextIOWrapper(sys.stdin.buffer, errors='ignore').read()
-readlines.__doc__ = "    Read the source code."
 noqa = re.compile(r'# no(?:qa|pep8)\b', re.I).search
 def expand_indent(line):
-r"""
+r"""Return the amount of indentation.
-Return the amount of indentation.
 Tabs are expanded to the next multiple of 8.
 >>> expand_indent('    ')
 4
 >>> expand_indent('\t')
-8
->>> expand_indent('    \t')
 8
 >>> expand_indent('       \t')
 8
 >>> expand_indent('        \t')
 16
 break
 return result
 def mute_string(text):
-"""
+"""Replace contents with 'xxx' to prevent syntax matching.
-Replace contents with 'xxx' to prevent syntax matching.
 >>> mute_string('"abc"')
 '"xxx"'
 >>> mute_string("'''abc'''")
 "'''xxx'''"
 if line[:1] != '-':
 nrows -= 1
 continue
 if line[:3] == '@@ ':
 hunk_match = HUNK_REGEX.match(line)
-row, nrows = [int(g or '1') for g in hunk_match.groups()]
+(row, nrows) = [int(g or '1') for g in hunk_match.groups()]
 rv[path].update(range(row, row + nrows))
 elif line[:3] == '+++':
 path = line[4:].split('\t', 1)[0]
 if path[:2] == 'b/':
 path = path[2:]
 return dict([(os.path.join(parent, path), rows)
 for (path, rows) in rv.items()
 if rows and filename_match(path, patterns)])
+def normalize_paths(value, parent=os.curdir):
+"""Parse a comma-separated list of paths.
+Return a list of absolute paths.
+"""
+if not value or isinstance(value, list):
+return value
+paths = []
+for path in value.split(','):
+if '/' in path:
+path = os.path.abspath(os.path.join(parent, path))
+paths.append(path.rstrip('/'))
+return paths
 def filename_match(filename, patterns, default=True):
-"""
+"""Check if patterns contains a pattern that matches filename.
-Check if patterns contains a pattern that matches filename.
 If patterns is unspecified, this always returns True.
 """
 if not patterns:
 return default
 return any(fnmatch(filename, pattern) for pattern in patterns)
+if COMMENT_WITH_NL:
+def _is_eol_token(token):
+return (token[0] in NEWLINE or
+(token[0] == tokenize.COMMENT and token[1] == token[4]))
+else:
+def _is_eol_token(token):
+return token[0] in NEWLINE
 ##############################################################################
 # Framework to run all checks
 ##############################################################################
 _checks = {'physical_line': {}, 'logical_line': {}, 'tree': {}}
 def register_check(check, codes=None):
-"""
+"""Register a new check object."""
-Register a new check object.
-"""
 def _add_check(check, kind, codes, args):
 if check in _checks[kind]:
 _checks[kind][check][0].extend(codes or [])
 else:
 _checks[kind][check] = (codes or [''], args)
 if inspect.getargspec(check.__init__)[0][:2] == ['self', 'tree']:
 _add_check(check, 'tree', codes, None)
 def init_checks_registry():
-"""
+"""Register all globally visible functions.
-Register all globally visible functions where the first argument name
-is 'physical_line' or 'logical_line'.
+The first argument name is either 'physical_line' or 'logical_line'.
 """
 mod = inspect.getmodule(register_check)
 for (name, function) in inspect.getmembers(mod, inspect.isfunction):
 register_check(function)
 init_checks_registry()
 class Checker(object):
-"""
+"""Load a Python source file, tokenize it, check coding style."""
-Load a Python source file, tokenize it, check coding style.
-"""
 def __init__(self, filename=None, lines=None,
 options=None, report=None, **kwargs):
 if options is None:
 options = StyleGuide(kwargs).options
 self._io_error = None
 self._physical_checks = options.physical_checks
 self._logical_checks = options.logical_checks
 self._ast_checks = options.ast_checks
 self.max_line_length = options.max_line_length
+self.multiline = False  # in a multiline string?
 self.hang_closing = options.hang_closing
 self.verbose = options.verbose
 self.filename = filename
 if filename is None:
 self.filename = 'stdin'
 self.lines = stdin_get_value().splitlines(True)
 elif lines is None:
 try:
 self.lines = readlines(filename)
 except IOError:
-exc_type, exc = sys.exc_info()[:2]
+(exc_type, exc) = sys.exc_info()[:2]
 self._io_error = '%s: %s' % (exc_type.__name__, exc)
 self.lines = []
 else:
 self.lines = lines
 if self.lines:
 # added for eric5 integration
 self.options = options
 def report_invalid_syntax(self):
-exc_type, exc = sys.exc_info()[:2]
+"""Check if the syntax is valid."""
+(exc_type, exc) = sys.exc_info()[:2]
 if len(exc.args) > 1:
 offset = exc.args[1]
 if len(offset) > 2:
 offset = offset[1:3]
 else:
 offset = (1, 0)
 self.report_error_args(offset[0], offset[1] or 0,
 'E901', self.report_invalid_syntax,
 exc_type.__name__, exc.args[0])
-report_invalid_syntax.__doc__ = "    Check if the syntax is valid."
 def readline(self):
-"""
+"""Get the next line from the input buffer."""
-Get the next line from the input buffer.
+if self.line_number >= self.total_lines:
-"""
+return ''
+line = self.lines[self.line_number]
 self.line_number += 1
-if self.line_number > len(self.lines):
+if self.indent_char is None and line[:1] in WHITESPACE:
-return ''
+self.indent_char = line[0]
-return self.lines[self.line_number - 1]
-def readline_check_physical(self):
-"""
-Check and return the next physical line. This method can be
-used to feed tokenize.generate_tokens.
-"""
-line = self.readline()
-if line:
-self.check_physical(line)
 return line
 def run_check(self, check, argument_names):
-"""
+"""Run a check plugin."""
-Run a check plugin.
-"""
 arguments = []
 for name in argument_names:
 arguments.append(getattr(self, name))
 return check(*arguments)
 def check_physical(self, line):
-"""
+"""Run all physical checks on a raw input line."""
-Run all physical checks on a raw input line.
-"""
 self.physical_line = line
-if self.indent_char is None and line[:1] in WHITESPACE:
-self.indent_char = line[0]
 for name, check, argument_names in self._physical_checks:
 result = self.run_check(check, argument_names)
 if result is not None:
-offset, code = result[:2]
+(offset, text) = result[:2]
 args = result[2:]
 self.report_error_args(
-self.line_number, offset, code, check, *args)
+self.line_number, offset, text, check, *args)
+if text[:4] == 'E101':
+self.indent_char = line[0]
 def build_tokens_line(self):
-"""
+"""Build a logical line from tokens."""
-Build a logical line from tokens.
-"""
-self.mapping = []
 logical = []
 comments = []
 length = 0
-previous = None
+prev_row = prev_col = mapping = None
-for token in self.tokens:
+for token_type, text, start, end, line in self.tokens:
-token_type, text = token[0:2]
+if token_type in SKIP_TOKENS:
+continue
+if not mapping:
+mapping = [(0, start)]
 if token_type == tokenize.COMMENT:
 comments.append(text)
 continue
-if token_type in SKIP_TOKENS:
-continue
 if token_type == tokenize.STRING:
 text = mute_string(text)
-if previous:
+if prev_row:
-end_row, end = previous[3]
+(start_row, start_col) = start
-start_row, start = token[2]
+if prev_row != start_row:    # different row
-if end_row != start_row:    # different row
+prev_text = self.lines[prev_row - 1][prev_col - 1]
-prev_text = self.lines[end_row - 1][end - 1]
 if prev_text == ',' or (prev_text not in '{[('
 and text not in '}])'):
-logical.append(' ')
+text = ' ' + text
-length += 1
+elif prev_col != start_col:  # different column
-elif end != start:  # different column
+text = line[prev_col:start_col] + text
-fill = self.lines[end_row - 1][end:start]
-logical.append(fill)
-length += len(fill)
-self.mapping.append((length, token))
 logical.append(text)
 length += len(text)
-previous = token
+mapping.append((length, end))
+(prev_row, prev_col) = end
 self.logical_line = ''.join(logical)
 self.noqa = comments and noqa(''.join(comments))
-# With Python 2, if the line ends with '\r\r\n' the assertion fails
+return mapping
-# assert self.logical_line.strip() == self.logical_line
 def check_logical(self):
-"""
+"""Build a line from tokens and run all logical checks on it."""
-Build a line from tokens and run all logical checks on it.
-"""
-self.build_tokens_line()
 self.report.increment_logical_line()
-first_line = self.lines[self.mapping[0][1][2][0] - 1]
+mapping = self.build_tokens_line()
-indent = first_line[:self.mapping[0][1][2][1]]
+(start_row, start_col) = mapping[0][1]
-self.previous_indent_level = self.indent_level
+start_line = self.lines[start_row - 1]
-self.indent_level = expand_indent(indent)
+self.indent_level = expand_indent(start_line[:start_col])
+if self.blank_before < self.blank_lines:
+self.blank_before = self.blank_lines
 if self.verbose >= 2:
 print(self.logical_line[:80].rstrip())
 for name, check, argument_names in self._logical_checks:
 if self.verbose >= 4:
 print('   ' + name)
 for result in self.run_check(check, argument_names):
-offset, code = result[:2]
+offset, text = result[:2]
 args = result[2:]
-if isinstance(offset, tuple):
+if not isinstance(offset, tuple):
-orig_number, orig_offset = offset
+for token_offset, pos in mapping:
-else:
+if offset <= token_offset:
-for token_offset, token in self.mapping:
+break
-if offset >= token_offset:
+offset = (pos[0], pos[1] + offset - token_offset)
-orig_number = token[2][0]
-orig_offset = (token[2][1] + offset - token_offset)
 self.report_error_args(
-orig_number, orig_offset, code, check, *args)
+offset[0], offset[1], text, check, *args)
-self.previous_logical = self.logical_line
+if self.logical_line:
+self.previous_indent_level = self.indent_level
+self.previous_logical = self.logical_line
+self.blank_lines = 0
+self.tokens = []
 def check_ast(self):
+"""Build the file's AST and run all AST checks."""
 try:
 tree = compile(''.join(self.lines), '', 'exec', ast.PyCF_ONLY_AST)
 except (SyntaxError, TypeError):
 return self.report_invalid_syntax()
-for name, cls, _ in self._ast_checks:
+for name, cls, __ in self._ast_checks:
 # extended API for eric5 integration
 checker = cls(tree, self.filename, self.options)
 for args in checker.run():
 lineno = args[0]
-if not noqa(self.lines[lineno - 1]):
+if not self.lines or not noqa(self.lines[lineno - 1]):
 self.report_error_args(lineno, *args[1:])
 def generate_tokens(self):
+"""Tokenize the file, run physical line checks and yield tokens."""
 if self._io_error:
-self.report_error(1, 0, 'E902 %s' % self._io_error, readlines)
+self.report_error_args(1, 0, 'E902', self._io_error, readlines)
-tokengen = tokenize.generate_tokens(self.readline_check_physical)
+tokengen = tokenize.generate_tokens(self.readline)
 try:
 for token in tokengen:
+self.maybe_check_physical(token)
 yield token
 except (SyntaxError, tokenize.TokenError):
 self.report_invalid_syntax()
+def maybe_check_physical(self, token):
+"""If appropriate (based on token), check current physical line(s)."""
+# Called after every token, but act only on end of line.
+if _is_eol_token(token):
+# Obviously, a newline token ends a single physical line.
+self.check_physical(token[4])
+elif token[0] == tokenize.STRING and '\n' in token[1]:
+# Less obviously, a string that contains newlines is a
+# multiline string, either triple-quoted or with internal
+# newlines backslash-escaped. Check every physical line in the
+# string *except* for the last one: its newline is outside of
+# the multiline string, so we consider it a regular physical
+# line, and will check it like any other physical line.
+#
+# Subtleties:
+# - we don't *completely* ignore the last line; if it contains
+#   the magical "# noqa" comment, we disable all physical
+#   checks for the entire multiline string
+# - have to wind self.line_number back because initially it
+#   points to the last line of the string, and we want
+#   check_physical() to give accurate feedback
+if noqa(token[4]):
+return
+self.multiline = True
+self.line_number = token[2][0]
+for line in token[1].split('\n')[:-1]:
+self.check_physical(line + '\n')
+self.line_number += 1
+self.multiline = False
 def check_all(self, expected=None, line_offset=0):
-"""
+"""Run all checks on the input file."""
-Run all checks on the input file.
-"""
 self.report.init_file(self.filename, self.lines, expected, line_offset)
+self.total_lines = len(self.lines)
 if self._ast_checks:
 self.check_ast()
 self.line_number = 0
 self.indent_char = None
-self.indent_level = 0
+self.indent_level = self.previous_indent_level = 0
 self.previous_logical = ''
 self.tokens = []
-self.blank_lines = blank_lines_before_comment = 0
+self.blank_lines = self.blank_before = 0
 parens = 0
 for token in self.generate_tokens():
 self.tokens.append(token)
 token_type, text = token[0:2]
 if self.verbose >= 3:
 if text in '([{':
 parens += 1
 elif text in '}])':
 parens -= 1
 elif not parens:
-if token_type == tokenize.NEWLINE:
+if token_type in NEWLINE:
-if self.blank_lines < blank_lines_before_comment:
+if token_type == tokenize.NEWLINE:
-self.blank_lines = blank_lines_before_comment
+self.check_logical()
-self.check_logical()
+self.blank_before = 0
-self.tokens = []
+elif len(self.tokens) == 1:
-self.blank_lines = blank_lines_before_comment = 0
-elif token_type == tokenize.NL:
-if len(self.tokens) == 1:
 # The physical line contains only this token.
 self.blank_lines += 1
-self.tokens = []
+del self.tokens[0]
-elif token_type == tokenize.COMMENT and len(self.tokens) == 1:
+else:
-if blank_lines_before_comment < self.blank_lines:
+self.check_logical()
-blank_lines_before_comment = self.blank_lines
+elif COMMENT_WITH_NL and token_type == tokenize.COMMENT:
-self.blank_lines = 0
+if len(self.tokens) == 1:
-if COMMENT_WITH_NL:
 # The comment also ends a physical line
-self.tokens = []
+token = list(token)
+token[1] = text.rstrip('\r\n')
+token[3] = (token[2][0], token[2][1] + len(token[1]))
+self.tokens = [tuple(token)]
+self.check_logical()
+if len(self.tokens) > 1 and (token_type == tokenize.ENDMARKER and
+self.tokens[-2][0] not in SKIP_TOKENS):
+self.tokens.pop()
+self.check_physical(self.tokens[-1][4])
+self.check_logical()
 return self.report.get_file_results()
 class BaseReport(object):
 """Collect the results of the checks."""
 print_filename = False
 def __init__(self, options):
 self._benchmark_keys = options.benchmark_keys
 self._ignore_code = options.ignore_code
 print(self.filename)
 self.file_errors += 1
 self.total_errors += 1
 return code
-def error_args(self, line_number, offset, code, check, *args):
+def error_args(self, line_number, offset, text, check, *args):
 """Report an error, according to options."""
+code = text[:4]
 if self._ignore_code(code):
 return
 if code in self.counters:
 self.counters[code] += 1
 else:
 """Return the total count of errors and warnings."""
 return sum([self.counters[key]
 for key in self.messages if key.startswith(prefix)])
 def get_statistics(self, prefix=''):
-"""
+"""Get statistics for message codes that start with the prefix.
-Get statistics for message codes that start with the prefix.
 prefix='' matches all errors and warnings
 prefix='E' matches all errors
 prefix='W' matches all warnings
 prefix='E4' matches all errors that have to do with imports
 if line_number > len(self.lines):
 line = ''
 else:
 line = self.lines[line_number - 1]
 print(line.rstrip())
-print(' ' * offset + '^')
+print(re.sub(r'\S', ' ', line[:offset]) + '^')
 if self._show_pep8 and doc:
-print(doc.lstrip('\n').rstrip())
+print('    ' + doc.strip())
 return self.file_errors
 class DiffReport(StandardReport):
 """Collect and print the results for the changed lines only."""
 # build options from the command line
 self.checker_class = kwargs.pop('checker_class', Checker)
 parse_argv = kwargs.pop('parse_argv', False)
 config_file = kwargs.pop('config_file', None)
 parser = kwargs.pop('parser', None)
+# build options from dict
+options_dict = dict(*args, **kwargs)
+arglist = None if parse_argv else options_dict.get('paths', None)
 options, self.paths = process_options(
-parse_argv=parse_argv, config_file=config_file, parser=parser)
+arglist, parse_argv, config_file, parser)
-if args or kwargs:
+if options_dict:
-# build options from dict
-options_dict = dict(*args, **kwargs)
 options.__dict__.update(options_dict)
 if 'paths' in options_dict:
 self.paths = options_dict['paths']
 self.runner = self.input_file
 self.options = options
 if not options.reporter:
 options.reporter = BaseReport if options.quiet else StandardReport
-for index, value in enumerate(options.exclude):
-options.exclude[index] = value.rstrip('/')
 options.select = tuple(options.select or ())
 if not (options.select or options.ignore or
 options.testsuite or options.doctest) and DEFAULT_IGNORE:
 # The default choice: ignore controversial checks
 options.ignore = tuple(DEFAULT_IGNORE.split(','))
 if ((filename_match(filename, filepatterns) and
 not self.excluded(filename, root))):
 runner(os.path.join(root, filename))
 def excluded(self, filename, parent=None):
-"""
+"""Check if the file should be excluded.
-Check if options.exclude contains a pattern that matches filename.
+Check if 'options.exclude' contains a pattern that matches filename.
 """
 if not self.options.exclude:
 return False
 basename = os.path.basename(filename)
 if filename_match(basename, self.options.exclude):
 return True
 if parent:
 filename = os.path.join(parent, filename)
+filename = os.path.abspath(filename)
 return filename_match(filename, self.options.exclude)
 def ignore_code(self, code):
-"""
+"""Check if the error code should be ignored.
-Check if the error code should be ignored.
 If 'options.select' contains a prefix of the error code,
 return False.  Else, if 'options.ignore' contains a prefix of
 the error code, return True.
 """
+if len(code) < 4 and any(s.startswith(code)
+for s in self.options.select):
+return False
 return (code.startswith(self.options.ignore) and
 not code.startswith(self.options.select))
 def get_checks(self, argument_name):
-"""
+"""Get all the checks for this category.
 Find all globally visible functions where the first argument name
 starts with argument_name and which contain selected tests.
 """
 checks = []
 for check, attrs in _checks[argument_name].items():
 if user_conf and os.path.isfile(user_conf):
 if options.verbose:
 print('user configuration: %s' % user_conf)
 config.read(user_conf)
+local_dir = os.curdir
 parent = tail = args and os.path.abspath(os.path.commonprefix(args))
 while tail:
 if config.read([os.path.join(parent, fn) for fn in PROJECT_CONFIG]):
+local_dir = parent
 if options.verbose:
 print('local configuration: in %s' % parent)
 break
-parent, tail = os.path.split(parent)
+(parent, tail) = os.path.split(parent)
 pep8_section = parser.prog
 if config.has_section(pep8_section):
 option_list = dict([(o.dest, o.type or o.action)
 for o in parser.option_list])
 # First, read the default values
-new_options, _ = parser.parse_args([])
+(new_options, __) = parser.parse_args([])
 # Second, parse the configuration
 for opt in config.options(pep8_section):
 if options.verbose > 1:
 print("  %s = %s" % (opt, config.get(pep8_section, opt)))
 opt_type = option_list[normalized_opt]
 if opt_type in ('int', 'count'):
 value = config.getint(pep8_section, opt)
 elif opt_type == 'string':
 value = config.get(pep8_section, opt)
+if normalized_opt == 'exclude':
+value = normalize_paths(value, local_dir)
 else:
 assert opt_type in ('store_true', 'store_false')
 value = config.getboolean(pep8_section, opt)
 setattr(new_options, normalized_opt, value)
 # Third, overwrite with the command-line options
-options, _ = parser.parse_args(arglist, values=new_options)
+(options, __) = parser.parse_args(arglist, values=new_options)
 options.doctest = options.testsuite = False
 return options
 def process_options(arglist=None, parse_argv=False, config_file=None,
 parser=None):
 """Process options passed either via arglist or via command line args."""
-if not arglist and not parse_argv:
-# Don't read the command line if the module is used as a library.
-arglist = []
 if not parser:
 parser = get_parser()
 if not parser.has_option('--config'):
 if config_file is True:
 config_file = DEFAULT_CONFIG
 "tox.ini file or the setup.cfg file located in any parent folder "
 "of the path(s) being processed.  Allowed options are: %s." %
 (parser.prog, ', '.join(parser.config_options))))
 group.add_option('--config', metavar='path', default=config_file,
 help="user config file location (default: %default)")
-options, args = parser.parse_args(arglist)
+# Don't read the command line if the module is used as a library.
+if not arglist and not parse_argv:
+arglist = []
+# If parse_argv is True and arglist is None, arguments are
+# parsed from the command line (sys.argv)
+(options, args) = parser.parse_args(arglist)
 options.reporter = None
 if options.ensure_value('testsuite', False):
 args.append(options.testsuite)
 elif not options.ensure_value('doctest', False):
 parser.error('input not specified')
 options = read_config(options, args, arglist, parser)
 options.reporter = parse_argv and options.quiet == 1 and FileReport
 options.filename = options.filename and options.filename.split(',')
-options.exclude = options.exclude.split(',')
+options.exclude = normalize_paths(options.exclude)
 options.select = options.select and options.select.split(',')
 options.ignore = options.ignore and options.ignore.split(',')
 if options.diff:
 options.reporter = DiffReport

Mercurial Repositories > eric / file comparison

comparison: Plugins/CheckerPlugins/CodeStyleChecker/pep8.py

Plugins/CheckerPlugins/CodeStyleChecker/pep8.py