252 lines
10 KiB
Python
252 lines
10 KiB
Python
# -*- coding: utf-8 -*-
|
|
"""
|
|
pygments.lexers.c_cpp
|
|
~~~~~~~~~~~~~~~~~~~~~
|
|
|
|
Lexers for C/C++ languages.
|
|
|
|
:copyright: Copyright 2006-2017 by the Pygments team, see AUTHORS.
|
|
:license: BSD, see LICENSE for details.
|
|
"""
|
|
|
|
import re
|
|
|
|
from pygments.lexer import RegexLexer, include, bygroups, using, \
|
|
this, inherit, default, words
|
|
from pygments.util import get_bool_opt
|
|
from pygments.token import Text, Comment, Operator, Keyword, Name, String, \
|
|
Number, Punctuation, Error
|
|
|
|
__all__ = ['CLexer', 'CppLexer']
|
|
|
|
|
|
class CFamilyLexer(RegexLexer):
|
|
"""
|
|
For C family source code. This is used as a base class to avoid repetitious
|
|
definitions.
|
|
"""
|
|
|
|
#: optional Comment or Whitespace
|
|
_ws = r'(?:\s|//.*?\n|/[*].*?[*]/)+'
|
|
|
|
# The trailing ?, rather than *, avoids a geometric performance drop here.
|
|
#: only one /* */ style comment
|
|
_ws1 = r'\s*(?:/[*].*?[*]/\s*)?'
|
|
|
|
tokens = {
|
|
'whitespace': [
|
|
# preprocessor directives: without whitespace
|
|
('^#if\s+0', Comment.Preproc, 'if0'),
|
|
('^#', Comment.Preproc, 'macro'),
|
|
# or with whitespace
|
|
('^(' + _ws1 + r')(#if\s+0)',
|
|
bygroups(using(this), Comment.Preproc), 'if0'),
|
|
('^(' + _ws1 + ')(#)',
|
|
bygroups(using(this), Comment.Preproc), 'macro'),
|
|
(r'\n', Text),
|
|
(r'\s+', Text),
|
|
(r'\\\n', Text), # line continuation
|
|
(r'//(\n|[\w\W]*?[^\\]\n)', Comment.Single),
|
|
(r'/(\\\n)?[*][\w\W]*?[*](\\\n)?/', Comment.Multiline),
|
|
# Open until EOF, so no ending delimeter
|
|
(r'/(\\\n)?[*][\w\W]*', Comment.Multiline),
|
|
],
|
|
'statements': [
|
|
(r'(L?)(")', bygroups(String.Affix, String), 'string'),
|
|
(r"(L?)(')(\\.|\\[0-7]{1,3}|\\x[a-fA-F0-9]{1,2}|[^\\\'\n])(')",
|
|
bygroups(String.Affix, String.Char, String.Char, String.Char)),
|
|
(r'(\d+\.\d*|\.\d+|\d+)[eE][+-]?\d+[LlUu]*', Number.Float),
|
|
(r'(\d+\.\d*|\.\d+|\d+[fF])[fF]?', Number.Float),
|
|
(r'0x[0-9a-fA-F]+[LlUu]*', Number.Hex),
|
|
(r'0[0-7]+[LlUu]*', Number.Oct),
|
|
(r'\d+[LlUu]*', Number.Integer),
|
|
(r'\*/', Error),
|
|
(r'[~!%^&*+=|?:<>/-]', Operator),
|
|
(r'[()\[\],.]', Punctuation),
|
|
(words(('asm', 'auto', 'break', 'case', 'const', 'continue',
|
|
'default', 'do', 'else', 'enum', 'extern', 'for', 'goto',
|
|
'if', 'register', 'restricted', 'return', 'sizeof',
|
|
'static', 'struct', 'switch', 'typedef', 'union',
|
|
'volatile', 'while'),
|
|
suffix=r'\b'), Keyword),
|
|
(r'(bool|int|long|float|short|double|char|unsigned|signed|void)\b',
|
|
Keyword.Type),
|
|
(words(('inline', '_inline', '__inline', 'naked', 'restrict',
|
|
'thread', 'typename'), suffix=r'\b'), Keyword.Reserved),
|
|
# Vector intrinsics
|
|
(r'(__m(128i|128d|128|64))\b', Keyword.Reserved),
|
|
# Microsoft-isms
|
|
(words((
|
|
'asm', 'int8', 'based', 'except', 'int16', 'stdcall', 'cdecl',
|
|
'fastcall', 'int32', 'declspec', 'finally', 'int64', 'try',
|
|
'leave', 'wchar_t', 'w64', 'unaligned', 'raise', 'noop',
|
|
'identifier', 'forceinline', 'assume'),
|
|
prefix=r'__', suffix=r'\b'), Keyword.Reserved),
|
|
(r'(true|false|NULL)\b', Name.Builtin),
|
|
(r'([a-zA-Z_]\w*)(\s*)(:)(?!:)', bygroups(Name.Label, Text, Punctuation)),
|
|
('[a-zA-Z_]\w*', Name),
|
|
],
|
|
'root': [
|
|
include('whitespace'),
|
|
# functions
|
|
(r'((?:[\w*\s])+?(?:\s|[*]))' # return arguments
|
|
r'([a-zA-Z_]\w*)' # method name
|
|
r'(\s*\([^;]*?\))' # signature
|
|
r'([^;{]*)(\{)',
|
|
bygroups(using(this), Name.Function, using(this), using(this),
|
|
Punctuation),
|
|
'function'),
|
|
# function declarations
|
|
(r'((?:[\w*\s])+?(?:\s|[*]))' # return arguments
|
|
r'([a-zA-Z_]\w*)' # method name
|
|
r'(\s*\([^;]*?\))' # signature
|
|
r'([^;]*)(;)',
|
|
bygroups(using(this), Name.Function, using(this), using(this),
|
|
Punctuation)),
|
|
default('statement'),
|
|
],
|
|
'statement': [
|
|
include('whitespace'),
|
|
include('statements'),
|
|
('[{}]', Punctuation),
|
|
(';', Punctuation, '#pop'),
|
|
],
|
|
'function': [
|
|
include('whitespace'),
|
|
include('statements'),
|
|
(';', Punctuation),
|
|
(r'\{', Punctuation, '#push'),
|
|
(r'\}', Punctuation, '#pop'),
|
|
],
|
|
'string': [
|
|
(r'"', String, '#pop'),
|
|
(r'\\([\\abfnrtv"\']|x[a-fA-F0-9]{2,4}|'
|
|
r'u[a-fA-F0-9]{4}|U[a-fA-F0-9]{8}|[0-7]{1,3})', String.Escape),
|
|
(r'[^\\"\n]+', String), # all other characters
|
|
(r'\\\n', String), # line continuation
|
|
(r'\\', String), # stray backslash
|
|
],
|
|
'macro': [
|
|
(r'(include)(' + _ws1 + r')([^\n]+)',
|
|
bygroups(Comment.Preproc, Text, Comment.PreprocFile)),
|
|
(r'[^/\n]+', Comment.Preproc),
|
|
(r'/[*](.|\n)*?[*]/', Comment.Multiline),
|
|
(r'//.*?\n', Comment.Single, '#pop'),
|
|
(r'/', Comment.Preproc),
|
|
(r'(?<=\\)\n', Comment.Preproc),
|
|
(r'\n', Comment.Preproc, '#pop'),
|
|
],
|
|
'if0': [
|
|
(r'^\s*#if.*?(?<!\\)\n', Comment.Preproc, '#push'),
|
|
(r'^\s*#el(?:se|if).*\n', Comment.Preproc, '#pop'),
|
|
(r'^\s*#endif.*?(?<!\\)\n', Comment.Preproc, '#pop'),
|
|
(r'.*?\n', Comment),
|
|
]
|
|
}
|
|
|
|
stdlib_types = set((
|
|
'size_t', 'ssize_t', 'off_t', 'wchar_t', 'ptrdiff_t', 'sig_atomic_t', 'fpos_t',
|
|
'clock_t', 'time_t', 'va_list', 'jmp_buf', 'FILE', 'DIR', 'div_t', 'ldiv_t',
|
|
'mbstate_t', 'wctrans_t', 'wint_t', 'wctype_t'))
|
|
c99_types = set((
|
|
'_Bool', '_Complex', 'int8_t', 'int16_t', 'int32_t', 'int64_t', 'uint8_t',
|
|
'uint16_t', 'uint32_t', 'uint64_t', 'int_least8_t', 'int_least16_t',
|
|
'int_least32_t', 'int_least64_t', 'uint_least8_t', 'uint_least16_t',
|
|
'uint_least32_t', 'uint_least64_t', 'int_fast8_t', 'int_fast16_t', 'int_fast32_t',
|
|
'int_fast64_t', 'uint_fast8_t', 'uint_fast16_t', 'uint_fast32_t', 'uint_fast64_t',
|
|
'intptr_t', 'uintptr_t', 'intmax_t', 'uintmax_t'))
|
|
linux_types = set((
|
|
'clockid_t', 'cpu_set_t', 'cpumask_t', 'dev_t', 'gid_t', 'id_t', 'ino_t', 'key_t',
|
|
'mode_t', 'nfds_t', 'pid_t', 'rlim_t', 'sig_t', 'sighandler_t', 'siginfo_t',
|
|
'sigset_t', 'sigval_t', 'socklen_t', 'timer_t', 'uid_t'))
|
|
|
|
def __init__(self, **options):
|
|
self.stdlibhighlighting = get_bool_opt(options, 'stdlibhighlighting', True)
|
|
self.c99highlighting = get_bool_opt(options, 'c99highlighting', True)
|
|
self.platformhighlighting = get_bool_opt(options, 'platformhighlighting', True)
|
|
RegexLexer.__init__(self, **options)
|
|
|
|
def get_tokens_unprocessed(self, text):
|
|
for index, token, value in \
|
|
RegexLexer.get_tokens_unprocessed(self, text):
|
|
if token is Name:
|
|
if self.stdlibhighlighting and value in self.stdlib_types:
|
|
token = Keyword.Type
|
|
elif self.c99highlighting and value in self.c99_types:
|
|
token = Keyword.Type
|
|
elif self.platformhighlighting and value in self.linux_types:
|
|
token = Keyword.Type
|
|
yield index, token, value
|
|
|
|
|
|
class CLexer(CFamilyLexer):
|
|
"""
|
|
For C source code with preprocessor directives.
|
|
"""
|
|
name = 'C'
|
|
aliases = ['c']
|
|
filenames = ['*.c', '*.h', '*.idc']
|
|
mimetypes = ['text/x-chdr', 'text/x-csrc']
|
|
priority = 0.1
|
|
|
|
def analyse_text(text):
|
|
if re.search('^\s*#include [<"]', text, re.MULTILINE):
|
|
return 0.1
|
|
if re.search('^\s*#ifn?def ', text, re.MULTILINE):
|
|
return 0.1
|
|
|
|
|
|
class CppLexer(CFamilyLexer):
|
|
"""
|
|
For C++ source code with preprocessor directives.
|
|
"""
|
|
name = 'C++'
|
|
aliases = ['cpp', 'c++']
|
|
filenames = ['*.cpp', '*.hpp', '*.c++', '*.h++',
|
|
'*.cc', '*.hh', '*.cxx', '*.hxx',
|
|
'*.C', '*.H', '*.cp', '*.CPP']
|
|
mimetypes = ['text/x-c++hdr', 'text/x-c++src']
|
|
priority = 0.1
|
|
|
|
tokens = {
|
|
'statements': [
|
|
(words((
|
|
'catch', 'const_cast', 'delete', 'dynamic_cast', 'explicit',
|
|
'export', 'friend', 'mutable', 'namespace', 'new', 'operator',
|
|
'private', 'protected', 'public', 'reinterpret_cast',
|
|
'restrict', 'static_cast', 'template', 'this', 'throw', 'throws',
|
|
'try', 'typeid', 'typename', 'using', 'virtual',
|
|
'constexpr', 'nullptr', 'decltype', 'thread_local',
|
|
'alignas', 'alignof', 'static_assert', 'noexcept', 'override',
|
|
'final'), suffix=r'\b'), Keyword),
|
|
(r'char(16_t|32_t)\b', Keyword.Type),
|
|
(r'(class)(\s+)', bygroups(Keyword, Text), 'classname'),
|
|
# C++11 raw strings
|
|
(r'(R)(")([^\\()\s]{,16})(\()((?:.|\n)*?)(\)\3)(")',
|
|
bygroups(String.Affix, String, String.Delimiter, String.Delimiter,
|
|
String, String.Delimiter, String)),
|
|
# C++11 UTF-8/16/32 strings
|
|
(r'(u8|u|U)(")', bygroups(String.Affix, String), 'string'),
|
|
inherit,
|
|
],
|
|
'root': [
|
|
inherit,
|
|
# C++ Microsoft-isms
|
|
(words(('virtual_inheritance', 'uuidof', 'super', 'single_inheritance',
|
|
'multiple_inheritance', 'interface', 'event'),
|
|
prefix=r'__', suffix=r'\b'), Keyword.Reserved),
|
|
# Offload C++ extensions, http://offload.codeplay.com/
|
|
(r'__(offload|blockingoffload|outer)\b', Keyword.Pseudo),
|
|
],
|
|
'classname': [
|
|
(r'[a-zA-Z_]\w*', Name.Class, '#pop'),
|
|
# template specification
|
|
(r'\s*(?=>)', Text, '#pop'),
|
|
],
|
|
}
|
|
|
|
def analyse_text(text):
|
|
if re.search('#include <[a-z_]+>', text):
|
|
return 0.2
|
|
if re.search('using namespace ', text):
|
|
return 0.4
|