#!/usr/bin/env python

#

# Copyright (c) 2009 Google Inc. All rights reserved.

#

# Redistribution and use in source and binary forms, with or without

# modification, are permitted provided that the following conditions are

# met:

#

# * Redistributions of source code must retain the above copyright

# notice, this list of conditions and the following disclaimer.

# * Redistributions in binary form must reproduce the above

# copyright notice, this list of conditions and the following disclaimer

# in the documentation and/or other materials provided with the

# distribution.

# * Neither the name of Google Inc. nor the names of its

# contributors may be used to endorse or promote products derived from

# this software without specific prior written permission.

#

# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS

# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT

# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR

# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT

# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,

# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT

# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,

# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY

# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT

# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE

# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

"""Does google-lint on c++ files.

The goal of this script is to identify places in the code that *may*

be in non-compliance with google style. It does not attempt to fix

up these problems -- the point is to educate. It does also not

attempt to find all problems, or to ensure that everything it does

find is legitimately a problem.

In particular, we can get very confused by /* and // inside strings!

We do a small hack, which is to ignore //'s with "'s after them on the

same line, but it is far from perfect (in either direction).

"""

import codecs

import copy

import getopt

import glob

import itertools

import math # for log

import os

import re

import sre_compile

import string

import sys

import sysconfig

import unicodedata

import xml.etree.ElementTree

# if empty, use defaults

_valid_extensions = set([])

__VERSION__ = '1.4.6'

try:

xrange # Python 2

except NameError:

# -- pylint: disable=redefined-builtin

xrange = range # Python 3

_USAGE = """

Syntax: cpplint.py [--verbose=#] [--output=emacs|eclipse|vs7|junit]

[--filter=-x,+y,...]

[--counting=total|toplevel|detailed] [--root=subdir]

[--repository=path]

[--linelength=digits] [--headers=x,y,...]

[--recursive]

[--exclude=path]

[--extensions=hpp,cpp,...]

[--quiet]

[--version]

<file> [file] ...

Style checker for C/C++ source files.

This is a fork of the Google style checker with minor extensions.

The style guidelines this tries to follow are those in

https://google.github.io/styleguide/cppguide.html

Every problem is given a confidence score from 1-5, with 5 meaning we are

certain of the problem, and 1 meaning it could be a legitimate construct.

This will miss some errors, and is not a substitute for a code review.

To suppress false-positive errors of a certain category, add a

'NOLINT(category)' comment to the line. NOLINT or NOLINT(*)

suppresses errors of all categories on that line.

The files passed in will be linted; at least one file must be provided.

Default linted extensions are %s.

Other file types will be ignored.

Change the extensions with the --extensions flag.

Flags:

output=emacs|eclipse|vs7|junit

By default, the output is formatted to ease emacs parsing. Visual Studio

compatible output (vs7) may also be used. Further support exists for

eclipse (eclipse), and JUnit (junit). XML parsers such as those used

in Jenkins and Bamboo may also be used. Other formats are unsupported.

verbose=#

Specify a number 0-5 to restrict errors to certain verbosity levels.

Errors with lower verbosity levels have lower confidence and are more

likely to be false positives.

quiet

Don't print anything if no errors are found.

filter=-x,+y,...

Specify a comma-separated list of category-filters to apply: only

error messages whose category names pass the filters will be printed.

(Category names are printed with the message and look like

"[whitespace/indent]".) Filters are evaluated left to right.

"-FOO" and "FOO" means "do not print categories that start with FOO".

"+FOO" means "do print categories that start with FOO".

Examples: --filter=-whitespace,+whitespace/braces

--filter=whitespace,runtime/printf,+runtime/printf_format

--filter=-,+build/include_what_you_use

To see a list of all the categories used in cpplint, pass no arg:

--filter=

counting=total|toplevel|detailed

The total number of errors found is always printed. If

'toplevel' is provided, then the count of errors in each of

the top-level categories like 'build' and 'whitespace' will

also be printed. If 'detailed' is provided, then a count

is provided for each category like 'build/class'.

repository=path

The top level directory of the repository, used to derive the header

guard CPP variable. By default, this is determined by searching for a

path that contains .git, .hg, or .svn. When this flag is specified, the

given path is used instead. This option allows the header guard CPP

variable to remain consistent even if members of a team have different

repository root directories (such as when checking out a subdirectory

with SVN). In addition, users of non-mainstream version control systems

can use this flag to ensure readable header guard CPP variables.

Examples:

Assuming that Alice checks out ProjectName and Bob checks out

ProjectName/trunk and trunk contains src/chrome/ui/browser.h, then

with no --repository flag, the header guard CPP variable will be:

Alice => TRUNK_SRC_CHROME_BROWSER_UI_BROWSER_H_

Bob => SRC_CHROME_BROWSER_UI_BROWSER_H_

If Alice uses the --repository=trunk flag and Bob omits the flag or

uses --repository=. then the header guard CPP variable will be:

Alice => SRC_CHROME_BROWSER_UI_BROWSER_H_

Bob => SRC_CHROME_BROWSER_UI_BROWSER_H_

root=subdir

The root directory used for deriving header guard CPP variable.

This directory is relative to the top level directory of the repository

which by default is determined by searching for a directory that contains

.git, .hg, or .svn but can also be controlled with the --repository flag.

If the specified directory does not exist, this flag is ignored.

Examples:

Assuming that src is the top level directory of the repository (and

cwd=top/src), the header guard CPP variables for

src/chrome/browser/ui/browser.h are:

No flag => CHROME_BROWSER_UI_BROWSER_H_

--root=chrome => BROWSER_UI_BROWSER_H_

--root=chrome/browser => UI_BROWSER_H_

--root=.. => SRC_CHROME_BROWSER_UI_BROWSER_H_

linelength=digits

This is the allowed line length for the project. The default value is

80 characters.

Examples:

--linelength=120

recursive

Search for files to lint recursively. Each directory given in the list

of files to be linted is replaced by all files that descend from that

directory. Files with extensions not in the valid extensions list are

excluded.

exclude=path

Exclude the given path from the list of files to be linted. Relative

paths are evaluated relative to the current directory and shell globbing

is performed. This flag can be provided multiple times to exclude

multiple files.

Examples:

--exclude=one.cc

--exclude=src/*.cc

--exclude=src/*.cc --exclude=test/*.cc

extensions=extension,extension,...

The allowed file extensions that cpplint will check

Examples:

--extensions=%s

headers=x,y,...

The header extensions that cpplint will treat as .h in checks. Values are

automatically added to --extensions list.

(by default, only files with extensions %s will be assumed to be headers)

Examples:

--headers=%s

--headers=hpp,hxx

--headers=hpp

cpplint.py supports per-directory configurations specified in CPPLINT.cfg

files. CPPLINT.cfg file can contain a number of key=value pairs.

Currently the following options are supported:

set noparent

filter=+filter1,-filter2,...

exclude_files=regex

linelength=80

root=subdir

headers=x,y,...

"set noparent" option prevents cpplint from traversing directory tree

upwards looking for more .cfg files in parent directories. This option

is usually placed in the top-level project directory.

The "filter" option is similar in function to --filter flag. It specifies

message filters in addition to the |_DEFAULT_FILTERS| and those specified

through --filter command-line flag.

"exclude_files" allows to specify a regular expression to be matched against

a file name. If the expression matches, the file is skipped and not run

through the linter.

"linelength" allows to specify the allowed line length for the project.

The "root" option is similar in function to the --root flag (see example

above). Paths are relative to the directory of the CPPLINT.cfg.

The "headers" option is similar in function to the --headers flag

(see example above).

CPPLINT.cfg has an effect on files in the same directory and all

sub-directories, unless overridden by a nested configuration file.

Example file:

filter=-build/include_order,+build/include_alpha

exclude_files=.*\\.cc

The above example disables build/include_order warning and enables

build/include_alpha as well as excludes all .cc from being

processed by linter, in the current directory (where the .cfg

file is located) and all sub-directories.

"""

# We categorize each error message we print. Here are the categories.

# We want an explicit list so we can list them all in cpplint --filter=.

# If you add a new error message with a new category, add it to the list

# here! cpplint_unittest.py should tell you if you forget to do this.

_ERROR_CATEGORIES = [

'build/class',

'build/c++11',

'build/c++14',

'build/c++tr1',

'build/deprecated',

'build/endif_comment',

'build/explicit_make_pair',

'build/forward_decl',

'build/header_guard',

'build/include',

'build/include_subdir',

'build/include_alpha',

'build/include_order',

'build/include_what_you_use',

'build/namespaces_literals',

'build/namespaces',

'build/printf_format',

'build/storage_class',

'legal/copyright',

'readability/alt_tokens',

'readability/braces',

'readability/casting',

'readability/check',

'readability/constructors',

'readability/fn_size',

'readability/inheritance',

'readability/multiline_comment',

'readability/multiline_string',

'readability/namespace',

'readability/nolint',

'readability/nul',

'readability/strings',

'readability/todo',

'readability/utf8',

'runtime/arrays',

'runtime/casting',

'runtime/explicit',

'runtime/int',

'runtime/init',

'runtime/invalid_increment',

'runtime/member_string_references',

'runtime/memset',

'runtime/indentation_namespace',

'runtime/operator',

'runtime/printf',

'runtime/printf_format',

'runtime/references',

'runtime/string',

'runtime/threadsafe_fn',

'runtime/vlog',

'whitespace/blank_line',

'whitespace/braces',

'whitespace/comma',

'whitespace/comments',

'whitespace/empty_conditional_body',

'whitespace/empty_if_body',

'whitespace/empty_loop_body',

'whitespace/end_of_line',

'whitespace/ending_newline',

'whitespace/forcolon',

'whitespace/indent',

'whitespace/line_length',

'whitespace/newline',

'whitespace/operators',

'whitespace/parens',

'whitespace/semicolon',

'whitespace/tab',

'whitespace/todo',

]

# These error categories are no longer enforced by cpplint, but for backwards-

# compatibility they may still appear in NOLINT comments.

_LEGACY_ERROR_CATEGORIES = [

'readability/streams',

'readability/function',

]

# The default state of the category filter. This is overridden by the --filter=

# flag. By default all errors are on, so only add here categories that should be

# off by default (i.e., categories that must be enabled by the --filter= flags).

# All entries here should start with a '-' or '+', as in the --filter= flag.

_DEFAULT_FILTERS = ['-build/include_alpha']

# The default list of categories suppressed for C (not C++) files.

_DEFAULT_C_SUPPRESSED_CATEGORIES = [

'readability/casting',

]

# The default list of categories suppressed for Linux Kernel files.

_DEFAULT_KERNEL_SUPPRESSED_CATEGORIES = [

'whitespace/tab',

]

# We used to check for high-bit characters, but after much discussion we

# decided those were OK, as long as they were in UTF-8 and didn't represent

# hard-coded international strings, which belong in a separate i18n file.

# C++ headers

_CPP_HEADERS = frozenset([

# Legacy

'algobase.h',

'algo.h',

'alloc.h',

'builtinbuf.h',

'bvector.h',

'complex.h',

'defalloc.h',

'deque.h',

'editbuf.h',

'fstream.h',

'function.h',

'hash_map',

'hash_map.h',

'hash_set',

'hash_set.h',

'hashtable.h',

'heap.h',

'indstream.h',

'iomanip.h',

'iostream.h',

'istream.h',

'iterator.h',

'list.h',

'map.h',

'multimap.h',

'multiset.h',

'ostream.h',

'pair.h',

'parsestream.h',

'pfstream.h',

'procbuf.h',

'pthread_alloc',

'pthread_alloc.h',

'rope',

'rope.h',

'ropeimpl.h',

'set.h',

'slist',

'slist.h',

'stack.h',

'stdiostream.h',

'stl_alloc.h',

'stl_relops.h',

'streambuf.h',

'stream.h',

'strfile.h',

'strstream.h',

'tempbuf.h',

'tree.h',

'type_traits.h',

'vector.h',

# 17.6.1.2 C++ library headers

'algorithm',

'array',

'atomic',

'bitset',

'chrono',

'codecvt',

'complex',

'condition_variable',

'deque',

'exception',

'forward_list',

'fstream',

'functional',

'future',

'initializer_list',

'iomanip',

'ios',

'iosfwd',

'iostream',

'istream',

'iterator',

'limits',

'list',

'locale',

'map',

'memory',

'mutex',

'new',

'numeric',

'ostream',

'queue',

'random',

'ratio',

'regex',

'scoped_allocator',

'set',

'sstream',

'stack',

'stdexcept',

'streambuf',

'string',

'strstream',

'system_error',

'thread',

'tuple',

'typeindex',

'typeinfo',

'type_traits',

'unordered_map',

'unordered_set',

'utility',

'valarray',

'vector',

# 17.6.1.2 C++14 headers

'shared_mutex',

# 17.6.1.2 C++17 headers

'any',

'charconv',

'codecvt',

'execution',

'filesystem',

'memory_resource',

'optional',

'string_view',

'variant',

# 17.6.1.2 C++ headers for C library facilities

'cassert',

'ccomplex',

'cctype',

'cerrno',

'cfenv',

'cfloat',

'cinttypes',

'ciso646',

'climits',

'clocale',

'cmath',

'csetjmp',

'csignal',

'cstdalign',

'cstdarg',

'cstdbool',

'cstddef',

'cstdint',

'cstdio',

'cstdlib',

'cstring',

'ctgmath',

'ctime',

'cuchar',

'cwchar',

'cwctype',

])

# Type names

_TYPES = re.compile(

r'^(?:'

# [dcl.type.simple]

r'(char(16_t|32_t)?)|wchar_t|'

r'bool|short|int|long|signed|unsigned|float|double|'

# [support.types]

r'(ptrdiff_t|size_t|max_align_t|nullptr_t)|'

# [cstdint.syn]

r'(u?int(_fast|_least)?(8|16|32|64)_t)|'

r'(u?int(max|ptr)_t)|'

r')$')

# These headers are excluded from [build/include] and [build/include_order]

# checks:

# - Anything not following google file name conventions (containing an

# uppercase character, such as Python.h or nsStringAPI.h, for example).

# - Lua headers.

_THIRD_PARTY_HEADERS_PATTERN = re.compile(

r'^(?:[^/]*[A-Z][^/]*\.h|lua\.h|lauxlib\.h|lualib\.h)$')

# Pattern for matching FileInfo.BaseName() against test file name

_test_suffixes = ['_test', '_regtest', '_unittest']

_TEST_FILE_SUFFIX = '(' + '|'.join(_test_suffixes) + r')$'

# Pattern that matches only complete whitespace, possibly across multiple lines.

_EMPTY_CONDITIONAL_BODY_PATTERN = re.compile(r'^\s*$', re.DOTALL)

# Assertion macros. These are defined in base/logging.h and

# testing/base/public/gunit.h.

_CHECK_MACROS = [

'DCHECK', 'CHECK',

'EXPECT_TRUE', 'ASSERT_TRUE',

'EXPECT_FALSE', 'ASSERT_FALSE',

]

# Replacement macros for CHECK/DCHECK/EXPECT_TRUE/EXPECT_FALSE

_CHECK_REPLACEMENT = dict([(macro_var, {}) for macro_var in _CHECK_MACROS])

for op, replacement in [('==', 'EQ'), ('!=', 'NE'),

('>=', 'GE'), ('>', 'GT'),

('<=', 'LE'), ('<', 'LT')]:

_CHECK_REPLACEMENT['DCHECK'][op] = 'DCHECK_%s' % replacement

_CHECK_REPLACEMENT['CHECK'][op] = 'CHECK_%s' % replacement

_CHECK_REPLACEMENT['EXPECT_TRUE'][op] = 'EXPECT_%s' % replacement

_CHECK_REPLACEMENT['ASSERT_TRUE'][op] = 'ASSERT_%s' % replacement

for op, inv_replacement in [('==', 'NE'), ('!=', 'EQ'),

('>=', 'LT'), ('>', 'LE'),

('<=', 'GT'), ('<', 'GE')]:

_CHECK_REPLACEMENT['EXPECT_FALSE'][op] = 'EXPECT_%s' % inv_replacement

_CHECK_REPLACEMENT['ASSERT_FALSE'][op] = 'ASSERT_%s' % inv_replacement

# Alternative tokens and their replacements. For full list, see section 2.5

# Alternative tokens [lex.digraph] in the C++ standard.

#

# Digraphs (such as '%:') are not included here since it's a mess to

# match those on a word boundary.

_ALT_TOKEN_REPLACEMENT = {

'and': '&&',

'bitor': '|',

'or': '||',

'xor': '^',

'compl': '~',

'bitand': '&',

'and_eq': '&=',

'or_eq': '|=',

'xor_eq': '^=',

'not': '!',

'not_eq': '!='

}

# Compile regular expression that matches all the above keywords. The "[ =()]"

# bit is meant to avoid matching these keywords outside of boolean expressions.

#

# False positives include C-style multi-line comments and multi-line strings

# but those have always been troublesome for cpplint.

_ALT_TOKEN_REPLACEMENT_PATTERN = re.compile(

r'[ =()](' + ('|'.join(_ALT_TOKEN_REPLACEMENT.keys())) + r')(?=[ (]|$)')

# These constants define types of headers for use with

# _IncludeState.CheckNextIncludeOrder().

_C_SYS_HEADER = 1

_CPP_SYS_HEADER = 2

_LIKELY_MY_HEADER = 3

_POSSIBLE_MY_HEADER = 4

_OTHER_HEADER = 5

# These constants define the current inline assembly state

_NO_ASM = 0 # Outside of inline assembly block

_INSIDE_ASM = 1 # Inside inline assembly block

_END_ASM = 2 # Last line of inline assembly block

_BLOCK_ASM = 3 # The whole block is an inline assembly block

# Match start of assembly blocks

_MATCH_ASM = re.compile(r'^\s*(?:asm|_asm|__asm|__asm__)'

r'(?:\s+(volatile|__volatile__))?'

r'\s*[{(]')

# Match strings that indicate we're working on a C (not C++) file.

_SEARCH_C_FILE = re.compile(r'\b(?:LINT_C_FILE|'

r'vim?:\s*.*(\s*|:)filetype=c(\s*|:|$))')

# Match string that indicates we're working on a Linux Kernel file.

_SEARCH_KERNEL_FILE = re.compile(r'\b(?:LINT_KERNEL_FILE)')

_regexp_compile_cache = {}

# {str, set(int)}: a map from error categories to sets of linenumbers

# on which those errors are expected and should be suppressed.

_error_suppressions = {}

# The root directory used for deriving header guard CPP variable.

# This is set by --root flag.

_root = None

_root_debug = False

# The top level repository directory. If set, _root is calculated relative to

# this directory instead of the directory containing version control artifacts.

# This is set by the --repository flag.

_repository = None

# Files to exclude from linting. This is set by the --exclude flag.

_excludes = None

# Whether to supress PrintInfo messages

_quiet = False

# The allowed line length of files.

# This is set by --linelength flag.

_line_length = 80

try:

unicode

except NameError:

# -- pylint: disable=redefined-builtin

basestring = unicode = str

try:

long

except NameError:

# -- pylint: disable=redefined-builtin

long = int

if sys.version_info < (3,):

# -- pylint: disable=no-member

# BINARY_TYPE = str

itervalues = dict.itervalues

iteritems = dict.iteritems

else:

# BINARY_TYPE = bytes

itervalues = dict.values

iteritems = dict.items

def unicode_escape_decode(x):

if sys.version_info < (3,):

return codecs.unicode_escape_decode(x)[0]

else:

return x

# Treat all headers starting with 'h' equally: .h, .hpp, .hxx etc.

# This is set by --headers flag.

_hpp_headers = set([])

# {str, bool}: a map from error categories to booleans which indicate if the

# category should be suppressed for every line.

_global_error_suppressions = {}

def ProcessHppHeadersOption(val):

global _hpp_headers

try:

_hpp_headers = {ext.strip() for ext in val.split(',')}

except ValueError:

PrintUsage('Header extensions must be comma separated list.')

def IsHeaderExtension(file_extension):

return file_extension in GetHeaderExtensions()

def GetHeaderExtensions():

if _hpp_headers:

return _hpp_headers

if _valid_extensions:

return {h for h in _valid_extensions if 'h' in h}

return set(['h', 'hh', 'hpp', 'hxx', 'h++', 'cuh'])

# The allowed extensions for file names

# This is set by --extensions flag

def GetAllExtensions():

return GetHeaderExtensions().union(_valid_extensions or set(

['c', 'cc', 'cpp', 'cxx', 'c++', 'cu']))

def ProcessExtensionsOption(val):

global _valid_extensions

try:

extensions = [ext.strip() for ext in val.split(',')]

_valid_extensions = set(extensions)

except ValueError:

PrintUsage('Extensions should be a comma-separated list of values;'

'for example: extensions=hpp,cpp\n'

'This could not be parsed: "%s"' % (val,))

def GetNonHeaderExtensions():

return GetAllExtensions().difference(GetHeaderExtensions())

def ParseNolintSuppressions(filename, raw_line, linenum, error):

"""Updates the global list of line error-suppressions.

Parses any NOLINT comments on the current line, updating the global

error_suppressions store. Reports an error if the NOLINT comment

was malformed.

Args:

filename: str, the name of the input file.

raw_line: str, the line of input text, with comments.

linenum: int, the number of the current line.

error: function, an error handler.

"""

matched = Search(r'\bNOLINT(NEXTLINE)?\b(\([^)]+\))?', raw_line)

if matched:

if matched.group(1):

suppressed_line = linenum + 1

else:

suppressed_line = linenum

category = matched.group(2)

if category in (None, '(*)'): # => "suppress all"

_error_suppressions.setdefault(None, set()).add(suppressed_line)

else:

if category.startswith('(') and category.endswith(')'):

category = category[1:-1]

if category in _ERROR_CATEGORIES:

_error_suppressions.setdefault(category, set()).add(suppressed_line)

elif category not in _LEGACY_ERROR_CATEGORIES:

error(filename, linenum, 'readability/nolint', 5,

'Unknown NOLINT error category: %s' % category)

def ProcessGlobalSuppresions(lines):

"""Updates the list of global error suppressions.

Parses any lint directives in the file that have global effect.

Args:

lines: An array of strings, each representing a line of the file, with the

last element being empty if the file is terminated with a newline.

"""

for line in lines:

if _SEARCH_C_FILE.search(line):

for category in _DEFAULT_C_SUPPRESSED_CATEGORIES:

_global_error_suppressions[category] = True

if _SEARCH_KERNEL_FILE.search(line):

for category in _DEFAULT_KERNEL_SUPPRESSED_CATEGORIES:

_global_error_suppressions[category] = True

def ResetNolintSuppressions():

"""Resets the set of NOLINT suppressions to empty."""

_error_suppressions.clear()

_global_error_suppressions.clear()

def IsErrorSuppressedByNolint(category, linenum):

"""Returns true if the specified error category is suppressed on this line.

Consults the global error_suppressions map populated by

ParseNolintSuppressions/ProcessGlobalSuppresions/ResetNolintSuppressions.

Args:

category: str, the category of the error.

linenum: int, the current line number.

Returns:

bool, True iff the error should be suppressed due to a NOLINT comment or

global suppression.

"""

return (_global_error_suppressions.get(category, False) or

linenum in _error_suppressions.get(category, set()) or

linenum in _error_suppressions.get(None, set()))

def Match(pattern, s):

"""Matches the string with the pattern, caching the compiled regexp."""

# The regexp compilation caching is inlined in both Match and Search for

# performance reasons; factoring it out into a separate function turns out

# to be noticeably expensive.

if pattern not in _regexp_compile_cache:

_regexp_compile_cache[pattern] = sre_compile.compile(pattern)

return _regexp_compile_cache[pattern].match(s)

def ReplaceAll(pattern, rep, s):

"""Replaces instances of pattern in a string with a replacement.

The compiled regex is kept in a cache shared by Match and Search.

Args:

pattern: regex pattern

rep: replacement text

s: search string

Returns:

string with replacements made (or original string if no replacements)

"""

if pattern not in _regexp_compile_cache:

_regexp_compile_cache[pattern] = sre_compile.compile(pattern)

return _regexp_compile_cache[pattern].sub(rep, s)

def Search(pattern, s):

"""Searches the string for the pattern, caching the compiled regexp."""

if pattern not in _regexp_compile_cache:

_regexp_compile_cache[pattern] = sre_compile.compile(pattern)

return _regexp_compile_cache[pattern].search(s)

def _IsSourceExtension(s):

"""File extension (excluding dot) matches a source file extension."""

return s in GetNonHeaderExtensions()

class _IncludeState(object):

"""Tracks line numbers for includes, and the order in which includes appear.

include_list contains list of lists of (header, line number) pairs.

It's a lists of lists rather than just one flat list to make it

easier to update across preprocessor boundaries.

Call CheckNextIncludeOrder() once for each header in the file, passing

in the type constants defined above. Calls in an illegal order will

raise an _IncludeError with an appropriate error message.

"""

# self._section will move monotonically through this set. If it ever

# needs to move backwards, CheckNextIncludeOrder will raise an error.

_INITIAL_SECTION = 0

_MY_H_SECTION = 1

_C_SECTION = 2

_CPP_SECTION = 3

_OTHER_H_SECTION = 4

_TYPE_NAMES = {

_C_SYS_HEADER: 'C system header',

_CPP_SYS_HEADER: 'C++ system header',

_LIKELY_MY_HEADER: 'header this file implements',

_POSSIBLE_MY_HEADER: 'header this file may implement',

_OTHER_HEADER: 'other header',

}

_SECTION_NAMES = {

_INITIAL_SECTION: "... nothing. (This can't be an error.)",

_MY_H_SECTION: 'a header this file implements',

_C_SECTION: 'C system header',

_CPP_SECTION: 'C++ system header',

_OTHER_H_SECTION: 'other header',

}

def __init__(self):

self.include_list = [[]]

self._section = None

self._last_header = None

self.ResetSection('')

def FindHeader(self, header):

"""Check if a header has already been included.

Args:

header: header to check.

Returns:

Line number of previous occurrence, or -1 if the header has not

been seen before.

"""

for section_list in self.include_list:

for f in section_list:

if f[0] == header:

return f[1]

return -1

def ResetSection(self, directive):

"""Reset section checking for preprocessor directive.

Args:

directive: preprocessor directive (e.g. "if", "else").

"""

# The name of the current section.

self._section = self._INITIAL_SECTION

# The path of last found header.

self._last_header = ''

# Update list of includes. Note that we never pop from the

# include list.

if directive in ('if', 'ifdef', 'ifndef'):

self.include_list.append([])

elif directive in ('else', 'elif'):

self.include_list[-1] = []

def SetLastHeader(self, header_path):

self._last_header = header_path

def CanonicalizeAlphabeticalOrder(self, header_path):

"""Returns a path canonicalized for alphabetical comparison.

- replaces "-" with "_" so they both cmp the same.

- removes '-inl' since we don't require them to be after the main header.

- lowercase everything, just in case.

Args:

header_path: Path to be canonicalized.

Returns:

Canonicalized path.

"""

return header_path.replace('-inl.h', '.h').replace('-', '_').lower()

def IsInAlphabeticalOrder(self, clean_lines, linenum, header_path):

"""Check if a header is in alphabetical order with the previous header.

Args:

clean_lines: A CleansedLines instance containing the file.

linenum: The number of the line to check.

header_path: Canonicalized header to be checked.

Returns:

Returns true if the header is in alphabetical order.

"""

# If previous section is different from current section, _last_header will

# be reset to empty string, so it's always less than current header.

#

# If previous line was a blank line, assume that the headers are

# intentionally sorted the way they are.

if (self._last_header > header_path and

Match(r'^\s*#\s*include\b', clean_lines.elided[linenum - 1])):

return False

return True

def CheckNextIncludeOrder(self, header_type):

"""Returns a non-empty error message if the next header is out of order.

This function also updates the internal state to be ready to check

the next include.

Args:

header_type: One of the _XXX_HEADER constants defined above.

Returns:

The empty string if the header is in the right order, or an

error message describing what's wrong.

"""

error_message = ('Found %s after %s' %

(self._TYPE_NAMES[header_type],

self._SECTION_NAMES[self._section]))

last_section = self._section

if header_type == _C_SYS_HEADER:

if self._section <= self._C_SECTION:

self._section = self._C_SECTION

else:

self._last_header = ''

return error_message

elif header_type == _CPP_SYS_HEADER:

if self._section <= self._CPP_SECTION:

self._section = self._CPP_SECTION

else:

self._last_header = ''

return error_message

elif header_type == _LIKELY_MY_HEADER:

if self._section <= self._MY_H_SECTION:

self._section = self._MY_H_SECTION

else:

self._section = self._OTHER_H_SECTION

elif header_type == _POSSIBLE_MY_HEADER:

if self._section <= self._MY_H_SECTION:

self._section = self._MY_H_SECTION

else:

# This will always be the fallback because we're not sure

# enough that the header is associated with this file.

self._section = self._OTHER_H_SECTION

else:

assert header_type == _OTHER_HEADER

self._section = self._OTHER_H_SECTION

if last_section != self._section:

self._last_header = ''