cpplint.py

#!/usr/bin/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 math  # for log
import os
import re
import sre_compile
import string
import sys
import unicodedata


_USAGE = """
Syntax: cpplint.py [--verbose=#] [--output=vs7] [--filter=-x,+y,...]
                   [--counting=total|toplevel|detailed] [--root=subdir]
                   [--linelength=digits]
        <file> [file] ...

  The style guidelines this tries to follow are those in
    http://google-styleguide.googlecode.com/svn/trunk/cppguide.xml

  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 .cc, .cpp, .cu, .cuh and .h.  Change the
  extensions with the --extensions flag.

  Flags:

    output=vs7
      By default, the output is formatted to ease emacs parsing.  Visual Studio
      compatible output (vs7) may also be used.  Other formats are unsupported.

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

    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'.

    root=subdir
      The root directory used for deriving header guard CPP variable.
      By default, the header guard CPP variable is calculated as the relative
      path to the directory that contains .git, .hg, or .svn.  When this flag
      is specified, the relative path is calculated from the specified
      directory. If the specified directory does not exist, this flag is
      ignored.

      Examples:
        Assuing that src/.git exists, 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_

    linelength=digits
      This is the allowed line length for the project. The default value is
      80 characters.

      Examples:
        --linelength=120

    extensions=extension,extension,...
      The allowed file extensions that cpplint will check

      Examples:
        --extensions=hpp,cpp
"""

# 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/deprecated',
  'build/endif_comment',
  'build/explicit_make_pair',
  'build/forward_decl',
  'build/header_guard',
  'build/include',
  'build/include_alpha',
  'build/include_order',
  'build/include_what_you_use',
  '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/function',
  'readability/multiline_comment',
  'readability/multiline_string',
  'readability/namespace',
  'readability/nolint',
  'readability/nul',
  'readability/streams',
  'readability/todo',
  'readability/utf8',
  'runtime/arrays',
  'runtime/casting',
  'runtime/explicit',
  'runtime/int',
  'runtime/init',
  'runtime/invalid_increment',
  'runtime/member_string_references',
  'runtime/memset',
  'runtime/operator',
  'runtime/printf',
  'runtime/printf_format',
  'runtime/references',
  'runtime/sizeof',
  'runtime/string',
  'runtime/threadsafe_fn',
  'runtime/vlog',
  'whitespace/blank_line',
  'whitespace/braces',
  'whitespace/comma',
  'whitespace/comments',
  'whitespace/empty_conditional_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'
  ]

# The default state of the category filter. This is overrided 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']

# 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',
    '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++ 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',
    ])

# Assertion macros.  These are defined in base/logging.h and
# testing/base/gunit.h.  Note that the _M versions need to come first
# for substring matching to work.
_CHECK_MACROS = [
    'DCHECK', 'CHECK',
    'EXPECT_TRUE_M', 'EXPECT_TRUE',
    'ASSERT_TRUE_M', 'ASSERT_TRUE',
    'EXPECT_FALSE_M', 'EXPECT_FALSE',
    'ASSERT_FALSE_M', 'ASSERT_FALSE',
    ]

# Replacement macros for CHECK/DCHECK/EXPECT_TRUE/EXPECT_FALSE
_CHECK_REPLACEMENT = dict([(m, {}) for m 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
  _CHECK_REPLACEMENT['EXPECT_TRUE_M'][op] = 'EXPECT_%s_M' % replacement
  _CHECK_REPLACEMENT['ASSERT_TRUE_M'][op] = 'ASSERT_%s_M' % 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
  _CHECK_REPLACEMENT['EXPECT_FALSE_M'][op] = 'EXPECT_%s_M' % inv_replacement
  _CHECK_REPLACEMENT['ASSERT_FALSE_M'][op] = 'ASSERT_%s_M' % 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*[{(]')


_regexp_compile_cache = {}

# Finds occurrences of NOLINT or NOLINT(...).
_RE_SUPPRESSION = re.compile(r'\bNOLINT\b(\([^)]*\))?')

# {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

# The allowed line length of files.
# This is set by --linelength flag.
_line_length = 80

# The allowed extensions for file names
# This is set by --extensions flag.
_valid_extensions = set(['cc', 'h', 'cpp', 'cu', 'cuh'])

def ParseNolintSuppressions(filename, raw_line, linenum, error):
  """Updates the global list of 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.
  """
  # FIXME(adonovan): "NOLINT(" is misparsed as NOLINT(*).
  matched = _RE_SUPPRESSION.search(raw_line)
  if matched:
    category = matched.group(1)
    if category in (None, '(*)'):  # => "suppress all"
      _error_suppressions.setdefault(None, set()).add(linenum)
    else:
      if category.startswith('(') and category.endswith(')'):
        category = category[1:-1]
        if category in _ERROR_CATEGORIES:
          _error_suppressions.setdefault(category, set()).add(linenum)
        else:
          error(filename, linenum, 'readability/nolint', 5,
                'Unknown NOLINT error category: %s' % category)


def ResetNolintSuppressions():
  "Resets the set of NOLINT suppressions to empty."
  _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/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.
  """
  return (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)


class _IncludeState(dict):
  """Tracks line numbers for includes, and the order in which includes appear.

  As a dict, an _IncludeState object serves as a mapping between include
  filename and line number on which that file was included.

  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):
    dict.__init__(self)
    self.ResetSection()

  def ResetSection(self):
    # The name of the current section.
    self._section = self._INITIAL_SECTION
    # The path of last found header.
    self._last_header = ''

  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
        not Match(r'^\s*$', 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 = ''

    return ''


class _CppLintState(object):
  """Maintains module-wide state.."""

  def __init__(self):
    self.verbose_level = 1  # global setting.
    self.error_count = 0    # global count of reported errors
    # filters to apply when emitting error messages
    self.filters = _DEFAULT_FILTERS[:]
    self.counting = 'total'  # In what way are we counting errors?
    self.errors_by_category = {}  # string to int dict storing error counts

    # output format:
    # "emacs" - format that emacs can parse (default)
    # "vs7" - format that Microsoft Visual Studio 7 can parse
    self.output_format = 'emacs'

  def SetOutputFormat(self, output_format):
    """Sets the output format for errors."""
    self.output_format = output_format

  def SetVerboseLevel(self, level):
    """Sets the module's verbosity, and returns the previous setting."""
    last_verbose_level = self.verbose_level
    self.verbose_level = level
    return last_verbose_level

  def SetCountingStyle(self, counting_style):
    """Sets the module's counting options."""
    self.counting = counting_style

  def SetFilters(self, filters):
    """Sets the error-message filters.

    These filters are applied when deciding whether to emit a given
    error message.

    Args:
      filters: A string of comma-separated filters (eg "+whitespace/indent").
               Each filter should start with + or -; else we die.

    Raises:
      ValueError: The comma-separated filters did not all start with '+' or '-'.
                  E.g. "-,+whitespace,-whitespace/indent,whitespace/badfilter"
    """
    # Default filters always have less priority than the flag ones.
    self.filters = _DEFAULT_FILTERS[:]
    for filt in filters.split(','):
      clean_filt = filt.strip()
      if clean_filt:
        self.filters.append(clean_filt)
    for filt in self.filters:
      if not (filt.startswith('+') or filt.startswith('-')):
        raise ValueError('Every filter in --filters must start with + or -'
                         ' (%s does not)' % filt)

  def ResetErrorCounts(self):
    """Sets the module's error statistic back to zero."""
    self.error_count = 0
    self.errors_by_category = {}

  def IncrementErrorCount(self, category):
    """Bumps the module's error statistic."""
    self.error_count += 1
    if self.counting in ('toplevel', 'detailed'):
      if self.counting != 'detailed':
        category = category.split('/')[0]
      if category not in self.errors_by_category:
        self.errors_by_category[category] = 0
      self.errors_by_category[category] += 1

  def PrintErrorCounts(self):
    """Print a summary of errors by category, and the total."""
    for category, count in self.errors_by_category.iteritems():
      sys.stderr.write('Category \'%s\' errors found: %d\n' %
                       (category, count))
    sys.stderr.write('Total errors found: %d\n' % self.error_count)

_cpplint_state = _CppLintState()


def _OutputFormat():
  """Gets the module's output format."""
  return _cpplint_state.output_format


def _SetOutputFormat(output_format):
  """Sets the module's output format."""
  _cpplint_state.SetOutputFormat(output_format)


def _VerboseLevel():
  """Returns the module's verbosity setting."""
  return _cpplint_state.verbose_level


def _SetVerboseLevel(level):
  """Sets the module's verbosity, and returns the previous setting."""
  return _cpplint_state.SetVerboseLevel(level)


def _SetCountingStyle(level):
  """Sets the module's counting options."""
  _cpplint_state.SetCountingStyle(level)


def _Filters():
  """Returns the module's list of output filters, as a list."""
  return _cpplint_state.filters


def _SetFilters(filters):
  """Sets the module's error-message filters.

  These filters are applied when deciding whether to emit a given
  error message.

  Args:
    filters: A string of comma-separated filters (eg "whitespace/indent").
             Each filter should start with + or -; else we die.
  """
  _cpplint_state.SetFilters(filters)


class _FunctionState(object):
  """Tracks current function name and the number of lines in its body."""

  _NORMAL_TRIGGER = 250  # for --v=0, 500 for --v=1, etc.
  _TEST_TRIGGER = 400    # about 50% more than _NORMAL_TRIGGER.

  def __init__(self):
    self.in_a_function = False
    self.lines_in_function = 0
    self.current_function = ''

  def Begin(self, function_name):
    """Start analyzing function body.

    Args:
      function_name: The name of the function being tracked.
    """
    self.in_a_function = True
    self.lines_in_function = 0
    self.current_function = function_name

  def Count(self):
    """Count line in current function body."""
    if self.in_a_function:
      self.lines_in_function += 1

  def Check(self, error, filename, linenum):
    """Report if too many lines in function body.

    Args:
      error: The function to call with any errors found.
      filename: The name of the current file.
      linenum: The number of the line to check.
    """
    if Match(r'T(EST|est)', self.current_function):
      base_trigger = self._TEST_TRIGGER
    else:
      base_trigger = self._NORMAL_TRIGGER
    trigger = base_trigger * 2**_VerboseLevel()

    if self.lines_in_function > trigger:
      error_level = int(math.log(self.lines_in_function / base_trigger, 2))
      # 50 => 0, 100 => 1, 200 => 2, 400 => 3, 800 => 4, 1600 => 5, ...
      if error_level > 5:
        error_level = 5
      error(filename, linenum, 'readability/fn_size', error_level,
            'Small and focused functions are preferred:'
            ' %s has %d non-comment lines'
            ' (error triggered by exceeding %d lines).'  % (
                self.current_function, self.lines_in_function, trigger))

  def End(self):
    """Stop analyzing function body."""
    self.in_a_function = False


class _IncludeError(Exception):
  """Indicates a problem with the include order in a file."""
  pass


class FileInfo:
  """Provides utility functions for filenames.

  FileInfo provides easy access to the components of a file's path
  relative to the project root.
  """

  def __init__(self, filename):
    self._filename = filename

  def FullName(self):
    """Make Windows paths like Unix."""
    return os.path.abspath(self._filename).replace('\\', '/')

  def RepositoryName(self):
    """FullName after removing the local path to the repository.

    If we have a real absolute path name here we can try to do something smart:
    detecting the root of the checkout and truncating /path/to/checkout from
    the name so that we get header guards that don't include things like
    "C:\Documents and Settings\..." or "/home/username/..." in them and thus
    people on different computers who have checked the source out to different
    locations won't see bogus errors.
    """
    fullname = self.FullName()

    if os.path.exists(fullname):
      project_dir = os.path.dirname(fullname)

      if os.path.exists(os.path.join(project_dir, ".svn")):
        # If there's a .svn file in the current directory, we recursively look
        # up the directory tree for the top of the SVN checkout
        root_dir = project_dir
        one_up_dir = os.path.dirname(root_dir)
        while os.path.exists(os.path.join(one_up_dir, ".svn")):
          root_dir = os.path.dirname(root_dir)
          one_up_dir = os.path.dirname(one_up_dir)

        prefix = os.path.commonprefix([root_dir, project_dir])
        return fullname[len(prefix) + 1:]

      # Not SVN <= 1.6? Try to find a git, hg, or svn top level directory by
      # searching up from the current path.
      root_dir = os.path.dirname(fullname)
      while (root_dir != os.path.dirname(root_dir) and
             not os.path.exists(os.path.join(root_dir, ".git")) and
             not os.path.exists(os.path.join(root_dir, ".hg")) and
             not os.path.exists(os.path.join(root_dir, ".svn"))):
        root_dir = os.path.dirname(root_dir)

      if (os.path.exists(os.path.join(root_dir, ".git")) or
          os.path.exists(os.path.join(root_dir, ".hg")) or
          os.path.exists(os.path.join(root_dir, ".svn"))):
        prefix = os.path.commonprefix([root_dir, project_dir])
        return fullname[len(prefix) + 1:]

    # Don't know what to do; header guard warnings may be wrong...
    return fullname

  def Split(self):
    """Splits the file into the directory, basename, and extension.

    For 'chrome/browser/browser.cc', Split() would
    return ('chrome/browser', 'browser', '.cc')

    Returns:
      A tuple of (directory, basename, extension).
    """

    googlename = self.RepositoryName()
    project, rest = os.path.split(googlename)
    return (project,) + os.path.splitext(rest)

  def BaseName(self):
    """File base name - text after the final slash, before the final period."""
    return self.Split()[1]

  def Extension(self):
    """File extension - text following the final period."""
    return self.Split()[2]

  def NoExtension(self):
    """File has no source file extension."""
    return '/'.join(self.Split()[0:2])

  def IsSource(self):
    """File has a source file extension."""
    return self.Extension()[1:] in ('c', 'cc', 'cpp', 'cxx')


def _ShouldPrintError(category, confidence, linenum):
  """If confidence >= verbose, category passes filter and is not suppressed."""

  # There are three ways we might decide not to print an error message:
  # a "NOLINT(category)" comment appears in the source,
  # the verbosity level isn't high enough, or the filters filter it out.
  if IsErrorSuppressedByNolint(category, linenum):
    return False
  if confidence < _cpplint_state.verbose_level:
    return False

  is_filtered = False
  for one_filter in _Filters():
    if one_filter.startswith('-'):
      if category.startswith(one_filter[1:]):
        is_filtered = True
    elif one_filter.startswith('+'):
      if category.startswith(one_filter[1:]):
        is_filtered = False
    else:
      assert False  # should have been checked for in SetFilter.
  if is_filtered:
    return False

  return True


def Error(filename, linenum, category, confidence, message):
  """Logs the fact we've found a lint error.

  We log where the error was found, and also our confidence in the error,
  that is, how certain we are this is a legitimate style regression, and
  not a misidentification or a use that's sometimes justified.

  False positives can be suppressed by the use of
  "cpplint(category)"  comments on the offending line.  These are
  parsed into _error_suppressions.

  Args:
    filename: The name of the file containing the error.
    linenum: The number of the line containing the error.
    category: A string used to describe the "category" this bug
      falls under: "whitespace", say, or "runtime".  Categories
      may have a hierarchy separated by slashes: "whitespace/indent".
    confidence: A number from 1-5 representing a confidence score for
      the error, with 5 meaning that we are certain of the problem,
      and 1 meaning that it could be a legitimate construct.
    message: The error message.
  """
  if _ShouldPrintError(category, confidence, linenum):
    _cpplint_state.IncrementErrorCount(category)