[209] | 1 | #!/usr/bin/env python
|
---|
| 2 | #
|
---|
| 3 | # Copyright (c) 2009 Google Inc. All rights reserved.
|
---|
| 4 | #
|
---|
| 5 | # Redistribution and use in source and binary forms, with or without
|
---|
| 6 | # modification, are permitted provided that the following conditions are
|
---|
| 7 | # met:
|
---|
| 8 | #
|
---|
| 9 | # * Redistributions of source code must retain the above copyright
|
---|
| 10 | # notice, this list of conditions and the following disclaimer.
|
---|
| 11 | # * Redistributions in binary form must reproduce the above
|
---|
| 12 | # copyright notice, this list of conditions and the following disclaimer
|
---|
| 13 | # in the documentation and/or other materials provided with the
|
---|
| 14 | # distribution.
|
---|
| 15 | # * Neither the name of Google Inc. nor the names of its
|
---|
| 16 | # contributors may be used to endorse or promote products derived from
|
---|
| 17 | # this software without specific prior written permission.
|
---|
| 18 | #
|
---|
| 19 | # THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
|
---|
| 20 | # "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
|
---|
| 21 | # LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
|
---|
| 22 | # A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
|
---|
| 23 | # OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
|
---|
| 24 | # SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
|
---|
| 25 | # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
---|
| 26 | # DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
---|
| 27 | # THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
---|
| 28 | # (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
|
---|
| 29 | # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
---|
| 30 |
|
---|
| 31 | # Here are some issues that I've had people identify in my code during reviews,
|
---|
| 32 | # that I think are possible to flag automatically in a lint tool. If these were
|
---|
| 33 | # caught by lint, it would save time both for myself and that of my reviewers.
|
---|
| 34 | # Most likely, some of these are beyond the scope of the current lint framework,
|
---|
| 35 | # but I think it is valuable to retain these wish-list items even if they cannot
|
---|
| 36 | # be immediately implemented.
|
---|
| 37 | #
|
---|
| 38 | # Suggestions
|
---|
| 39 | # -----------
|
---|
| 40 | # - Check for no 'explicit' for multi-arg ctor
|
---|
| 41 | # - Check for boolean assign RHS in parens
|
---|
| 42 | # - Check for ctor initializer-list colon position and spacing
|
---|
| 43 | # - Check that if there's a ctor, there should be a dtor
|
---|
| 44 | # - Check accessors that return non-pointer member variables are
|
---|
| 45 | # declared const
|
---|
| 46 | # - Check accessors that return non-const pointer member vars are
|
---|
| 47 | # *not* declared const
|
---|
| 48 | # - Check for using public includes for testing
|
---|
| 49 | # - Check for spaces between brackets in one-line inline method
|
---|
| 50 | # - Check for no assert()
|
---|
| 51 | # - Check for spaces surrounding operators
|
---|
| 52 | # - Check for 0 in pointer context (should be NULL)
|
---|
| 53 | # - Check for 0 in char context (should be '\0')
|
---|
| 54 | # - Check for camel-case method name conventions for methods
|
---|
| 55 | # that are not simple inline getters and setters
|
---|
| 56 | # - Check that base classes have virtual destructors
|
---|
| 57 | # put " // namespace" after } that closes a namespace, with
|
---|
| 58 | # namespace's name after 'namespace' if it is named.
|
---|
| 59 | # - Do not indent namespace contents
|
---|
| 60 | # - Avoid inlining non-trivial constructors in header files
|
---|
| 61 | # include base/basictypes.h if DISALLOW_EVIL_CONSTRUCTORS is used
|
---|
| 62 | # - Check for old-school (void) cast for call-sites of functions
|
---|
| 63 | # ignored return value
|
---|
| 64 | # - Check gUnit usage of anonymous namespace
|
---|
| 65 | # - Check for class declaration order (typedefs, consts, enums,
|
---|
| 66 | # ctor(s?), dtor, friend declarations, methods, member vars)
|
---|
| 67 | #
|
---|
| 68 |
|
---|
| 69 | """Does google-lint on c++ files.
|
---|
| 70 |
|
---|
| 71 | The goal of this script is to identify places in the code that *may*
|
---|
| 72 | be in non-compliance with google style. It does not attempt to fix
|
---|
| 73 | up these problems -- the point is to educate. It does also not
|
---|
| 74 | attempt to find all problems, or to ensure that everything it does
|
---|
| 75 | find is legitimately a problem.
|
---|
| 76 |
|
---|
| 77 | In particular, we can get very confused by /* and // inside strings!
|
---|
| 78 | We do a small hack, which is to ignore //'s with "'s after them on the
|
---|
| 79 | same line, but it is far from perfect (in either direction).
|
---|
| 80 | """
|
---|
| 81 |
|
---|
| 82 | import codecs
|
---|
| 83 | import getopt
|
---|
| 84 | import math # for log
|
---|
| 85 | import os
|
---|
| 86 | import re
|
---|
| 87 | import sre_compile
|
---|
| 88 | import string
|
---|
| 89 | import sys
|
---|
| 90 | import unicodedata
|
---|
| 91 |
|
---|
| 92 | EXTENSIONS = ['c', 'cc', 'cpp', 'cxx', 'c++',
|
---|
| 93 | 'h', 'hpp', 'hxx', 'h++']
|
---|
| 94 |
|
---|
| 95 | _USAGE = """
|
---|
| 96 | Syntax: cpplint.py [--verbose=#] [--output=vs7] [--filter=-x,+y,...]
|
---|
| 97 | [--counting=total|toplevel|detailed]
|
---|
| 98 | <file> [file] ...
|
---|
| 99 |
|
---|
| 100 | The style guidelines this tries to follow are those in
|
---|
| 101 | http://google-styleguide.googlecode.com/svn/trunk/cppguide.xml
|
---|
| 102 |
|
---|
| 103 | Every problem is given a confidence score from 1-5, with 5 meaning we are
|
---|
| 104 | certain of the problem, and 1 meaning it could be a legitimate construct.
|
---|
| 105 | This will miss some errors, and is not a substitute for a code review.
|
---|
| 106 |
|
---|
| 107 | To suppress false-positive errors of a certain category, add a
|
---|
| 108 | 'NOLINT(category)' comment to the line. NOLINT or NOLINT(*)
|
---|
| 109 | suppresses errors of all categories on that line.
|
---|
| 110 |
|
---|
| 111 | The files passed in will be linted; at least one file must be provided.
|
---|
| 112 | Linted extensions are %s. Other file types will be ignored.
|
---|
| 113 |
|
---|
| 114 | Flags:
|
---|
| 115 |
|
---|
| 116 | output=vs7
|
---|
| 117 | By default, the output is formatted to ease emacs parsing. Visual Studio
|
---|
| 118 | compatible output (vs7) may also be used. Other formats are unsupported.
|
---|
| 119 |
|
---|
| 120 | verbose=#
|
---|
| 121 | Specify a number 0-5 to restrict errors to certain verbosity levels.
|
---|
| 122 |
|
---|
| 123 | filter=-x,+y,...
|
---|
| 124 | Specify a comma-separated list of category-filters to apply: only
|
---|
| 125 | error messages whose category names pass the filters will be printed.
|
---|
| 126 | (Category names are printed with the message and look like
|
---|
| 127 | "[whitespace/indent]".) Filters are evaluated left to right.
|
---|
| 128 | "-FOO" and "FOO" means "do not print categories that start with FOO".
|
---|
| 129 | "+FOO" means "do print categories that start with FOO".
|
---|
| 130 |
|
---|
| 131 | Examples: --filter=-whitespace,+whitespace/braces
|
---|
| 132 | --filter=whitespace,runtime/printf,+runtime/printf_format
|
---|
| 133 | --filter=-,+build/include_what_you_use
|
---|
| 134 |
|
---|
| 135 | To see a list of all the categories used in cpplint, pass no arg:
|
---|
| 136 | --filter=
|
---|
| 137 |
|
---|
| 138 | counting=total|toplevel|detailed
|
---|
| 139 | The total number of errors found is always printed. If
|
---|
| 140 | 'toplevel' is provided, then the count of errors in each of
|
---|
| 141 | the top-level categories like 'build' and 'whitespace' will
|
---|
| 142 | also be printed. If 'detailed' is provided, then a count
|
---|
| 143 | is provided for each category like 'build/class'.
|
---|
| 144 | """ % (EXTENSIONS)
|
---|
| 145 |
|
---|
| 146 | # We categorize each error message we print. Here are the categories.
|
---|
| 147 | # We want an explicit list so we can list them all in cpplint --filter=.
|
---|
| 148 | # If you add a new error message with a new category, add it to the list
|
---|
| 149 | # here! cpplint_unittest.py should tell you if you forget to do this.
|
---|
| 150 | # \ used for clearer layout -- pylint: disable-msg=C6013
|
---|
| 151 | _ERROR_CATEGORIES = [
|
---|
| 152 | 'build/class',
|
---|
| 153 | 'build/deprecated',
|
---|
| 154 | 'build/endif_comment',
|
---|
| 155 | 'build/explicit_make_pair',
|
---|
| 156 | 'build/forward_decl',
|
---|
| 157 | 'build/header_guard',
|
---|
| 158 | 'build/include',
|
---|
| 159 | 'build/include_alpha',
|
---|
| 160 | 'build/include_order',
|
---|
| 161 | 'build/include_what_you_use',
|
---|
| 162 | 'build/namespaces',
|
---|
| 163 | 'build/printf_format',
|
---|
| 164 | 'build/storage_class',
|
---|
| 165 | 'legal/copyright',
|
---|
| 166 | 'readability/braces',
|
---|
| 167 | 'readability/casting',
|
---|
| 168 | 'readability/check',
|
---|
| 169 | 'readability/constructors',
|
---|
| 170 | 'readability/fn_size',
|
---|
| 171 | 'readability/function',
|
---|
| 172 | 'readability/multiline_comment',
|
---|
| 173 | 'readability/multiline_string',
|
---|
| 174 | 'readability/nolint',
|
---|
| 175 | 'readability/streams',
|
---|
| 176 | 'readability/todo',
|
---|
| 177 | 'readability/utf8',
|
---|
| 178 | 'runtime/arrays',
|
---|
| 179 | 'runtime/casting',
|
---|
| 180 | 'runtime/explicit',
|
---|
| 181 | 'runtime/int',
|
---|
| 182 | 'runtime/init',
|
---|
| 183 | 'runtime/invalid_increment',
|
---|
| 184 | 'runtime/member_string_references',
|
---|
| 185 | 'runtime/memset',
|
---|
| 186 | 'runtime/operator',
|
---|
| 187 | 'runtime/printf',
|
---|
| 188 | 'runtime/printf_format',
|
---|
| 189 | 'runtime/references',
|
---|
| 190 | 'runtime/rtti',
|
---|
| 191 | 'runtime/sizeof',
|
---|
| 192 | 'runtime/string',
|
---|
| 193 | 'runtime/threadsafe_fn',
|
---|
| 194 | 'runtime/virtual',
|
---|
| 195 | 'whitespace/blank_line',
|
---|
| 196 | 'whitespace/braces',
|
---|
| 197 | 'whitespace/comma',
|
---|
| 198 | 'whitespace/comments',
|
---|
| 199 | 'whitespace/end_of_line',
|
---|
| 200 | 'whitespace/ending_newline',
|
---|
| 201 | 'whitespace/indent',
|
---|
| 202 | 'whitespace/labels',
|
---|
| 203 | 'whitespace/line_length',
|
---|
| 204 | 'whitespace/newline',
|
---|
| 205 | 'whitespace/operators',
|
---|
| 206 | 'whitespace/parens',
|
---|
| 207 | 'whitespace/semicolon',
|
---|
| 208 | 'whitespace/tab',
|
---|
| 209 | 'whitespace/todo'
|
---|
| 210 | ]
|
---|
| 211 |
|
---|
| 212 | # The default state of the category filter. This is overrided by the --filter=
|
---|
| 213 | # flag. By default all errors are on, so only add here categories that should be
|
---|
| 214 | # off by default (i.e., categories that must be enabled by the --filter= flags).
|
---|
| 215 | # All entries here should start with a '-' or '+', as in the --filter= flag.
|
---|
| 216 | _DEFAULT_FILTERS = ['-build/include_alpha']
|
---|
| 217 |
|
---|
| 218 | # We used to check for high-bit characters, but after much discussion we
|
---|
| 219 | # decided those were OK, as long as they were in UTF-8 and didn't represent
|
---|
| 220 | # hard-coded international strings, which belong in a separate i18n file.
|
---|
| 221 |
|
---|
| 222 | # Headers that we consider STL headers.
|
---|
| 223 | _STL_HEADERS = frozenset([
|
---|
| 224 | 'algobase.h', 'algorithm', 'alloc.h', 'bitset', 'deque', 'exception',
|
---|
| 225 | 'function.h', 'functional', 'hash_map', 'hash_map.h', 'hash_set',
|
---|
| 226 | 'hash_set.h', 'iterator', 'list', 'list.h', 'map', 'memory', 'new',
|
---|
| 227 | 'pair.h', 'pthread_alloc', 'queue', 'set', 'set.h', 'sstream', 'stack',
|
---|
| 228 | 'stl_alloc.h', 'stl_relops.h', 'type_traits.h',
|
---|
| 229 | 'utility', 'vector', 'vector.h',
|
---|
| 230 | ])
|
---|
| 231 |
|
---|
| 232 |
|
---|
| 233 | # Non-STL C++ system headers.
|
---|
| 234 | _CPP_HEADERS = frozenset([
|
---|
| 235 | 'algo.h', 'builtinbuf.h', 'bvector.h', 'cassert', 'cctype',
|
---|
| 236 | 'cerrno', 'cfloat', 'ciso646', 'climits', 'clocale', 'cmath',
|
---|
| 237 | 'complex', 'complex.h', 'csetjmp', 'csignal', 'cstdarg', 'cstddef',
|
---|
| 238 | 'cstdio', 'cstdlib', 'cstring', 'ctime', 'cwchar', 'cwctype',
|
---|
| 239 | 'defalloc.h', 'deque.h', 'editbuf.h', 'exception', 'fstream',
|
---|
| 240 | 'fstream.h', 'hashtable.h', 'heap.h', 'indstream.h', 'iomanip',
|
---|
| 241 | 'iomanip.h', 'ios', 'iosfwd', 'iostream', 'iostream.h', 'istream',
|
---|
| 242 | 'istream.h', 'iterator.h', 'limits', 'map.h', 'multimap.h', 'multiset.h',
|
---|
| 243 | 'numeric', 'ostream', 'ostream.h', 'parsestream.h', 'pfstream.h',
|
---|
| 244 | 'PlotFile.h', 'procbuf.h', 'pthread_alloc.h', 'rope', 'rope.h',
|
---|
| 245 | 'ropeimpl.h', 'SFile.h', 'slist', 'slist.h', 'stack.h', 'stdexcept',
|
---|
| 246 | 'stdiostream.h', 'streambuf.h', 'stream.h', 'strfile.h', 'string',
|
---|
| 247 | 'strstream', 'strstream.h', 'tempbuf.h', 'tree.h', 'typeinfo', 'valarray',
|
---|
| 248 | ])
|
---|
| 249 |
|
---|
| 250 |
|
---|
| 251 | # Assertion macros. These are defined in base/logging.h and
|
---|
| 252 | # testing/base/gunit.h. Note that the _M versions need to come first
|
---|
| 253 | # for substring matching to work.
|
---|
| 254 | _CHECK_MACROS = [
|
---|
| 255 | 'DCHECK', 'CHECK',
|
---|
| 256 | 'EXPECT_TRUE_M', 'EXPECT_TRUE',
|
---|
| 257 | 'ASSERT_TRUE_M', 'ASSERT_TRUE',
|
---|
| 258 | 'EXPECT_FALSE_M', 'EXPECT_FALSE',
|
---|
| 259 | 'ASSERT_FALSE_M', 'ASSERT_FALSE',
|
---|
| 260 | ]
|
---|
| 261 |
|
---|
| 262 | # Replacement macros for CHECK/DCHECK/EXPECT_TRUE/EXPECT_FALSE
|
---|
| 263 | _CHECK_REPLACEMENT = dict([(m, {}) for m in _CHECK_MACROS])
|
---|
| 264 |
|
---|
| 265 | for op, replacement in [('==', 'EQ'), ('!=', 'NE'),
|
---|
| 266 | ('>=', 'GE'), ('>', 'GT'),
|
---|
| 267 | ('<=', 'LE'), ('<', 'LT')]:
|
---|
| 268 | _CHECK_REPLACEMENT['DCHECK'][op] = 'DCHECK_%s' % replacement
|
---|
| 269 | _CHECK_REPLACEMENT['CHECK'][op] = 'CHECK_%s' % replacement
|
---|
| 270 | _CHECK_REPLACEMENT['EXPECT_TRUE'][op] = 'EXPECT_%s' % replacement
|
---|
| 271 | _CHECK_REPLACEMENT['ASSERT_TRUE'][op] = 'ASSERT_%s' % replacement
|
---|
| 272 | _CHECK_REPLACEMENT['EXPECT_TRUE_M'][op] = 'EXPECT_%s_M' % replacement
|
---|
| 273 | _CHECK_REPLACEMENT['ASSERT_TRUE_M'][op] = 'ASSERT_%s_M' % replacement
|
---|
| 274 |
|
---|
| 275 | for op, inv_replacement in [('==', 'NE'), ('!=', 'EQ'),
|
---|
| 276 | ('>=', 'LT'), ('>', 'LE'),
|
---|
| 277 | ('<=', 'GT'), ('<', 'GE')]:
|
---|
| 278 | _CHECK_REPLACEMENT['EXPECT_FALSE'][op] = 'EXPECT_%s' % inv_replacement
|
---|
| 279 | _CHECK_REPLACEMENT['ASSERT_FALSE'][op] = 'ASSERT_%s' % inv_replacement
|
---|
| 280 | _CHECK_REPLACEMENT['EXPECT_FALSE_M'][op] = 'EXPECT_%s_M' % inv_replacement
|
---|
| 281 | _CHECK_REPLACEMENT['ASSERT_FALSE_M'][op] = 'ASSERT_%s_M' % inv_replacement
|
---|
| 282 |
|
---|
| 283 |
|
---|
| 284 | # These constants define types of headers for use with
|
---|
| 285 | # _IncludeState.CheckNextIncludeOrder().
|
---|
| 286 | _C_SYS_HEADER = 1
|
---|
| 287 | _CPP_SYS_HEADER = 2
|
---|
| 288 | _LIKELY_MY_HEADER = 3
|
---|
| 289 | _POSSIBLE_MY_HEADER = 4
|
---|
| 290 | _OTHER_HEADER = 5
|
---|
| 291 |
|
---|
| 292 |
|
---|
| 293 | _regexp_compile_cache = {}
|
---|
| 294 |
|
---|
| 295 | # Finds occurrences of NOLINT or NOLINT(...).
|
---|
| 296 | _RE_SUPPRESSION = re.compile(r'\bNOLINT\b(\([^)]*\))?')
|
---|
| 297 |
|
---|
| 298 | # {str, set(int)}: a map from error categories to sets of linenumbers
|
---|
| 299 | # on which those errors are expected and should be suppressed.
|
---|
| 300 | _error_suppressions = {}
|
---|
| 301 |
|
---|
| 302 |
|
---|
| 303 | if sys.version_info < (3,):
|
---|
| 304 | def u(x):
|
---|
| 305 | return codecs.unicode_escape_decode(x)[0]
|
---|
| 306 | TEXT_TYPE = unicode
|
---|
| 307 | # BINARY_TYPE = str
|
---|
| 308 | range = xrange
|
---|
| 309 | itervalues = dict.itervalues
|
---|
| 310 | iteritems = dict.iteritems
|
---|
| 311 | else:
|
---|
| 312 | def u(x):
|
---|
| 313 | return x
|
---|
| 314 | TEXT_TYPE = str
|
---|
| 315 | # BINARY_TYPE = bytes
|
---|
| 316 | itervalues = dict.values
|
---|
| 317 | iteritems = dict.items
|
---|
| 318 |
|
---|
| 319 | def ParseNolintSuppressions(filename, raw_line, linenum, error):
|
---|
| 320 | """Updates the global list of error-suppressions.
|
---|
| 321 |
|
---|
| 322 | Parses any NOLINT comments on the current line, updating the global
|
---|
| 323 | error_suppressions store. Reports an error if the NOLINT comment
|
---|
| 324 | was malformed.
|
---|
| 325 |
|
---|
| 326 | Args:
|
---|
| 327 | filename: str, the name of the input file.
|
---|
| 328 | raw_line: str, the line of input text, with comments.
|
---|
| 329 | linenum: int, the number of the current line.
|
---|
| 330 | error: function, an error handler.
|
---|
| 331 | """
|
---|
| 332 | # FIXME(adonovan): "NOLINT(" is misparsed as NOLINT(*).
|
---|
| 333 | matched = _RE_SUPPRESSION.search(raw_line)
|
---|
| 334 | if matched:
|
---|
| 335 | category = matched.group(1)
|
---|
| 336 | if category in (None, '(*)'): # => "suppress all"
|
---|
| 337 | _error_suppressions.setdefault(None, set()).add(linenum)
|
---|
| 338 | else:
|
---|
| 339 | if category.startswith('(') and category.endswith(')'):
|
---|
| 340 | category = category[1:-1]
|
---|
| 341 | if category in _ERROR_CATEGORIES:
|
---|
| 342 | _error_suppressions.setdefault(category, set()).add(linenum)
|
---|
| 343 | else:
|
---|
| 344 | error(filename, linenum, 'readability/nolint', 5,
|
---|
| 345 | 'Unknown NOLINT error category: %s' % category)
|
---|
| 346 |
|
---|
| 347 |
|
---|
| 348 | def ResetNolintSuppressions():
|
---|
| 349 | "Resets the set of NOLINT suppressions to empty."
|
---|
| 350 | _error_suppressions.clear()
|
---|
| 351 |
|
---|
| 352 |
|
---|
| 353 | def IsErrorSuppressedByNolint(category, linenum):
|
---|
| 354 | """Returns true if the specified error category is suppressed on this line.
|
---|
| 355 |
|
---|
| 356 | Consults the global error_suppressions map populated by
|
---|
| 357 | ParseNolintSuppressions/ResetNolintSuppressions.
|
---|
| 358 |
|
---|
| 359 | Args:
|
---|
| 360 | category: str, the category of the error.
|
---|
| 361 | linenum: int, the current line number.
|
---|
| 362 | Returns:
|
---|
| 363 | bool, True iff the error should be suppressed due to a NOLINT comment.
|
---|
| 364 | """
|
---|
| 365 | return (linenum in _error_suppressions.get(category, set()) or
|
---|
| 366 | linenum in _error_suppressions.get(None, set()))
|
---|
| 367 |
|
---|
| 368 | def Match(pattern, s):
|
---|
| 369 | """Matches the string with the pattern, caching the compiled regexp."""
|
---|
| 370 | # The regexp compilation caching is inlined in both Match and Search for
|
---|
| 371 | # performance reasons; factoring it out into a separate function turns out
|
---|
| 372 | # to be noticeably expensive.
|
---|
| 373 | if not pattern in _regexp_compile_cache:
|
---|
| 374 | _regexp_compile_cache[pattern] = sre_compile.compile(pattern)
|
---|
| 375 | return _regexp_compile_cache[pattern].match(s)
|
---|
| 376 |
|
---|
| 377 |
|
---|
| 378 | def Search(pattern, s):
|
---|
| 379 | """Searches the string for the pattern, caching the compiled regexp."""
|
---|
| 380 | if not pattern in _regexp_compile_cache:
|
---|
| 381 | _regexp_compile_cache[pattern] = sre_compile.compile(pattern)
|
---|
| 382 | return _regexp_compile_cache[pattern].search(s)
|
---|
| 383 |
|
---|
| 384 |
|
---|
| 385 | class _IncludeState(dict):
|
---|
| 386 | """Tracks line numbers for includes, and the order in which includes appear.
|
---|
| 387 |
|
---|
| 388 | As a dict, an _IncludeState object serves as a mapping between include
|
---|
| 389 | filename and line number on which that file was included.
|
---|
| 390 |
|
---|
| 391 | Call CheckNextIncludeOrder() once for each header in the file, passing
|
---|
| 392 | in the type constants defined above. Calls in an illegal order will
|
---|
| 393 | raise an _IncludeError with an appropriate error message.
|
---|
| 394 |
|
---|
| 395 | """
|
---|
| 396 | # self._section will move monotonically through this set. If it ever
|
---|
| 397 | # needs to move backwards, CheckNextIncludeOrder will raise an error.
|
---|
| 398 | _INITIAL_SECTION = 0
|
---|
| 399 | _MY_H_SECTION = 1
|
---|
| 400 | _C_SECTION = 2
|
---|
| 401 | _CPP_SECTION = 3
|
---|
| 402 | _OTHER_H_SECTION = 4
|
---|
| 403 |
|
---|
| 404 | _TYPE_NAMES = {
|
---|
| 405 | _C_SYS_HEADER: 'C system header',
|
---|
| 406 | _CPP_SYS_HEADER: 'C++ system header',
|
---|
| 407 | _LIKELY_MY_HEADER: 'header this file implements',
|
---|
| 408 | _POSSIBLE_MY_HEADER: 'header this file may implement',
|
---|
| 409 | _OTHER_HEADER: 'other header',
|
---|
| 410 | }
|
---|
| 411 | _SECTION_NAMES = {
|
---|
| 412 | _INITIAL_SECTION: "... nothing. (This can't be an error.)",
|
---|
| 413 | _MY_H_SECTION: 'a header this file implements',
|
---|
| 414 | _C_SECTION: 'C system header',
|
---|
| 415 | _CPP_SECTION: 'C++ system header',
|
---|
| 416 | _OTHER_H_SECTION: 'other header',
|
---|
| 417 | }
|
---|
| 418 |
|
---|
| 419 | def __init__(self):
|
---|
| 420 | dict.__init__(self)
|
---|
| 421 | # The name of the current section.
|
---|
| 422 | self._section = self._INITIAL_SECTION
|
---|
| 423 | # The path of last found header.
|
---|
| 424 | self._last_header = ''
|
---|
| 425 |
|
---|
| 426 | def CanonicalizeAlphabeticalOrder(self, header_path):
|
---|
| 427 | """Returns a path canonicalized for alphabetical comparison.
|
---|
| 428 |
|
---|
| 429 | - replaces "-" with "_" so they both cmp the same.
|
---|
| 430 | - removes '-inl' since we don't require them to be after the main header.
|
---|
| 431 | - lowercase everything, just in case.
|
---|
| 432 |
|
---|
| 433 | Args:
|
---|
| 434 | header_path: Path to be canonicalized.
|
---|
| 435 |
|
---|
| 436 | Returns:
|
---|
| 437 | Canonicalized path.
|
---|
| 438 | """
|
---|
| 439 | return header_path.replace('-inl.h', '.h').replace('-', '_').lower()
|
---|
| 440 |
|
---|
| 441 | def IsInAlphabeticalOrder(self, header_path):
|
---|
| 442 | """Check if a header is in alphabetical order with the previous header.
|
---|
| 443 |
|
---|
| 444 | Args:
|
---|
| 445 | header_path: Header to be checked.
|
---|
| 446 |
|
---|
| 447 | Returns:
|
---|
| 448 | Returns true if the header is in alphabetical order.
|
---|
| 449 | """
|
---|
| 450 | canonical_header = self.CanonicalizeAlphabeticalOrder(header_path)
|
---|
| 451 | if self._last_header > canonical_header:
|
---|
| 452 | return False
|
---|
| 453 | self._last_header = canonical_header
|
---|
| 454 | return True
|
---|
| 455 |
|
---|
| 456 | def CheckNextIncludeOrder(self, header_type):
|
---|
| 457 | """Returns a non-empty error message if the next header is out of order.
|
---|
| 458 |
|
---|
| 459 | This function also updates the internal state to be ready to check
|
---|
| 460 | the next include.
|
---|
| 461 |
|
---|
| 462 | Args:
|
---|
| 463 | header_type: One of the _XXX_HEADER constants defined above.
|
---|
| 464 |
|
---|
| 465 | Returns:
|
---|
| 466 | The empty string if the header is in the right order, or an
|
---|
| 467 | error message describing what's wrong.
|
---|
| 468 |
|
---|
| 469 | """
|
---|
| 470 | error_message = ('Found %s after %s' %
|
---|
| 471 | (self._TYPE_NAMES[header_type],
|
---|
| 472 | self._SECTION_NAMES[self._section]))
|
---|
| 473 |
|
---|
| 474 | last_section = self._section
|
---|
| 475 |
|
---|
| 476 | if header_type == _C_SYS_HEADER:
|
---|
| 477 | if self._section <= self._C_SECTION:
|
---|
| 478 | self._section = self._C_SECTION
|
---|
| 479 | else:
|
---|
| 480 | self._last_header = ''
|
---|
| 481 | return error_message
|
---|
| 482 | elif header_type == _CPP_SYS_HEADER:
|
---|
| 483 | if self._section <= self._CPP_SECTION:
|
---|
| 484 | self._section = self._CPP_SECTION
|
---|
| 485 | else:
|
---|
| 486 | self._last_header = ''
|
---|
| 487 | return error_message
|
---|
| 488 | elif header_type == _LIKELY_MY_HEADER:
|
---|
| 489 | if self._section <= self._MY_H_SECTION:
|
---|
| 490 | self._section = self._MY_H_SECTION
|
---|
| 491 | else:
|
---|
| 492 | self._section = self._OTHER_H_SECTION
|
---|
| 493 | elif header_type == _POSSIBLE_MY_HEADER:
|
---|
| 494 | if self._section <= self._MY_H_SECTION:
|
---|
| 495 | self._section = self._MY_H_SECTION
|
---|
| 496 | else:
|
---|
| 497 | # This will always be the fallback because we're not sure
|
---|
| 498 | # enough that the header is associated with this file.
|
---|
| 499 | self._section = self._OTHER_H_SECTION
|
---|
| 500 | else:
|
---|
| 501 | assert header_type == _OTHER_HEADER
|
---|
| 502 | self._section = self._OTHER_H_SECTION
|
---|
| 503 |
|
---|
| 504 | if last_section != self._section:
|
---|
| 505 | self._last_header = ''
|
---|
| 506 |
|
---|
| 507 | return ''
|
---|
| 508 |
|
---|
| 509 |
|
---|
| 510 | class _CppLintState(object):
|
---|
| 511 | """Maintains module-wide state.."""
|
---|
| 512 |
|
---|
| 513 | def __init__(self):
|
---|
| 514 | self.verbose_level = 1 # global setting.
|
---|
| 515 | self.error_count = 0 # global count of reported errors
|
---|
| 516 | # filters to apply when emitting error messages
|
---|
| 517 | self.filters = _DEFAULT_FILTERS[:]
|
---|
| 518 | self.counting = 'total' # In what way are we counting errors?
|
---|
| 519 | self.errors_by_category = {} # string to int dict storing error counts
|
---|
| 520 |
|
---|
| 521 | # output format:
|
---|
| 522 | # "emacs" - format that emacs can parse (default)
|
---|
| 523 | # "vs7" - format that Microsoft Visual Studio 7 can parse
|
---|
| 524 | self.output_format = 'emacs'
|
---|
| 525 |
|
---|
| 526 | def SetOutputFormat(self, output_format):
|
---|
| 527 | """Sets the output format for errors."""
|
---|
| 528 | self.output_format = output_format
|
---|
| 529 |
|
---|
| 530 | def SetVerboseLevel(self, level):
|
---|
| 531 | """Sets the module's verbosity, and returns the previous setting."""
|
---|
| 532 | last_verbose_level = self.verbose_level
|
---|
| 533 | self.verbose_level = level
|
---|
| 534 | return last_verbose_level
|
---|
| 535 |
|
---|
| 536 | def SetCountingStyle(self, counting_style):
|
---|
| 537 | """Sets the module's counting options."""
|
---|
| 538 | self.counting = counting_style
|
---|
| 539 |
|
---|
| 540 | def SetFilters(self, filters):
|
---|
| 541 | """Sets the error-message filters.
|
---|
| 542 |
|
---|
| 543 | These filters are applied when deciding whether to emit a given
|
---|
| 544 | error message.
|
---|
| 545 |
|
---|
| 546 | Args:
|
---|
| 547 | filters: A string of comma-separated filters (eg "+whitespace/indent").
|
---|
| 548 | Each filter should start with + or -; else we die.
|
---|
| 549 |
|
---|
| 550 | Raises:
|
---|
| 551 | ValueError: The comma-separated filters did not all start with '+' or '-'.
|
---|
| 552 | E.g. "-,+whitespace,-whitespace/indent,whitespace/badfilter"
|
---|
| 553 | """
|
---|
| 554 | # Default filters always have less priority than the flag ones.
|
---|
| 555 | self.filters = _DEFAULT_FILTERS[:]
|
---|
| 556 | for filt in filters.split(','):
|
---|
| 557 | clean_filt = filt.strip()
|
---|
| 558 | if clean_filt:
|
---|
| 559 | self.filters.append(clean_filt)
|
---|
| 560 | for filt in self.filters:
|
---|
| 561 | if not (filt.startswith('+') or filt.startswith('-')):
|
---|
| 562 | raise ValueError('Every filter in --filters must start with + or -'
|
---|
| 563 | ' (%s does not)' % filt)
|
---|
| 564 |
|
---|
| 565 | def ResetErrorCounts(self):
|
---|
| 566 | """Sets the module's error statistic back to zero."""
|
---|
| 567 | self.error_count = 0
|
---|
| 568 | self.errors_by_category = {}
|
---|
| 569 |
|
---|
| 570 | def IncrementErrorCount(self, category):
|
---|
| 571 | """Bumps the module's error statistic."""
|
---|
| 572 | self.error_count += 1
|
---|
| 573 | if self.counting in ('toplevel', 'detailed'):
|
---|
| 574 | if self.counting != 'detailed':
|
---|
| 575 | category = category.split('/')[0]
|
---|
| 576 | if category not in self.errors_by_category:
|
---|
| 577 | self.errors_by_category[category] = 0
|
---|
| 578 | self.errors_by_category[category] += 1
|
---|
| 579 |
|
---|
| 580 | def PrintErrorCounts(self):
|
---|
| 581 | """Print a summary of errors by category, and the total."""
|
---|
| 582 | for category, count in iteritems(self.errors_by_category):
|
---|
| 583 | sys.stderr.write('Category \'%s\' errors found: %d\n' %
|
---|
| 584 | (category, count))
|
---|
| 585 | sys.stderr.write('Total errors found: %d\n' % self.error_count)
|
---|
| 586 |
|
---|
| 587 | _cpplint_state = _CppLintState()
|
---|
| 588 |
|
---|
| 589 |
|
---|
| 590 | def _OutputFormat():
|
---|
| 591 | """Gets the module's output format."""
|
---|
| 592 | return _cpplint_state.output_format
|
---|
| 593 |
|
---|
| 594 |
|
---|
| 595 | def _SetOutputFormat(output_format):
|
---|
| 596 | """Sets the module's output format."""
|
---|
| 597 | _cpplint_state.SetOutputFormat(output_format)
|
---|
| 598 |
|
---|
| 599 |
|
---|
| 600 | def _VerboseLevel():
|
---|
| 601 | """Returns the module's verbosity setting."""
|
---|
| 602 | return _cpplint_state.verbose_level
|
---|
| 603 |
|
---|
| 604 |
|
---|
| 605 | def _SetVerboseLevel(level):
|
---|
| 606 | """Sets the module's verbosity, and returns the previous setting."""
|
---|
| 607 | return _cpplint_state.SetVerboseLevel(level)
|
---|
| 608 |
|
---|
| 609 |
|
---|
| 610 | def _SetCountingStyle(level):
|
---|
| 611 | """Sets the module's counting options."""
|
---|
| 612 | _cpplint_state.SetCountingStyle(level)
|
---|
| 613 |
|
---|
| 614 |
|
---|
| 615 | def _Filters():
|
---|
| 616 | """Returns the module's list of output filters, as a list."""
|
---|
| 617 | return _cpplint_state.filters
|
---|
| 618 |
|
---|
| 619 |
|
---|
| 620 | def _SetFilters(filters):
|
---|
| 621 | """Sets the module's error-message filters.
|
---|
| 622 |
|
---|
| 623 | These filters are applied when deciding whether to emit a given
|
---|
| 624 | error message.
|
---|
| 625 |
|
---|
| 626 | Args:
|
---|
| 627 | filters: A string of comma-separated filters (eg "whitespace/indent").
|
---|
| 628 | Each filter should start with + or -; else we die.
|
---|
| 629 | """
|
---|
| 630 | _cpplint_state.SetFilters(filters)
|
---|
| 631 |
|
---|
| 632 |
|
---|
| 633 | class _FunctionState(object):
|
---|
| 634 | """Tracks current function name and the number of lines in its body."""
|
---|
| 635 |
|
---|
| 636 | _NORMAL_TRIGGER = 250 # for --v=0, 500 for --v=1, etc.
|
---|
| 637 | _TEST_TRIGGER = 400 # about 50% more than _NORMAL_TRIGGER.
|
---|
| 638 |
|
---|
| 639 | def __init__(self):
|
---|
| 640 | self.in_a_function = False
|
---|
| 641 | self.lines_in_function = 0
|
---|
| 642 | self.current_function = ''
|
---|
| 643 |
|
---|
| 644 | def Begin(self, function_name):
|
---|
| 645 | """Start analyzing function body.
|
---|
| 646 |
|
---|
| 647 | Args:
|
---|
| 648 | function_name: The name of the function being tracked.
|
---|
| 649 | """
|
---|
| 650 | self.in_a_function = True
|
---|
| 651 | self.lines_in_function = 0
|
---|
| 652 | self.current_function = function_name
|
---|
| 653 |
|
---|
| 654 | def Count(self):
|
---|
| 655 | """Count line in current function body."""
|
---|
| 656 | if self.in_a_function:
|
---|
| 657 | self.lines_in_function += 1
|
---|
| 658 |
|
---|
| 659 | def Check(self, error, filename, linenum):
|
---|
| 660 | """Report if too many lines in function body.
|
---|
| 661 |
|
---|
| 662 | Args:
|
---|
| 663 | error: The function to call with any errors found.
|
---|
| 664 | filename: The name of the current file.
|
---|
| 665 | linenum: The number of the line to check.
|
---|
| 666 | """
|
---|
| 667 | if Match(r'T(EST|est)', self.current_function):
|
---|
| 668 | base_trigger = self._TEST_TRIGGER
|
---|
| 669 | else:
|
---|
| 670 | base_trigger = self._NORMAL_TRIGGER
|
---|
| 671 | trigger = base_trigger * 2**_VerboseLevel()
|
---|
| 672 |
|
---|
| 673 | if self.lines_in_function > trigger:
|
---|
| 674 | error_level = int(math.log(self.lines_in_function / base_trigger, 2))
|
---|
| 675 | # 50 => 0, 100 => 1, 200 => 2, 400 => 3, 800 => 4, 1600 => 5, ...
|
---|
| 676 | if error_level > 5:
|
---|
| 677 | error_level = 5
|
---|
| 678 | error(filename, linenum, 'readability/fn_size', error_level,
|
---|
| 679 | 'Small and focused functions are preferred:'
|
---|
| 680 | ' %s has %d non-comment lines'
|
---|
| 681 | ' (error triggered by exceeding %d lines).' % (
|
---|
| 682 | self.current_function, self.lines_in_function, trigger))
|
---|
| 683 |
|
---|
| 684 | def End(self):
|
---|
| 685 | """Stop analyzing function body."""
|
---|
| 686 | self.in_a_function = False
|
---|
| 687 |
|
---|
| 688 |
|
---|
| 689 | class _IncludeError(Exception):
|
---|
| 690 | """Indicates a problem with the include order in a file."""
|
---|
| 691 | pass
|
---|
| 692 |
|
---|
| 693 |
|
---|
| 694 | class FileInfo:
|
---|
| 695 | """Provides utility functions for filenames.
|
---|
| 696 |
|
---|
| 697 | FileInfo provides easy access to the components of a file's path
|
---|
| 698 | relative to the project root.
|
---|
| 699 | """
|
---|
| 700 |
|
---|
| 701 | def __init__(self, filename):
|
---|
| 702 | self._filename = filename
|
---|
| 703 |
|
---|
| 704 | def FullName(self):
|
---|
| 705 | """Make Windows paths like Unix."""
|
---|
| 706 | return os.path.abspath(self._filename).replace('\\', '/')
|
---|
| 707 |
|
---|
| 708 | def RepositoryName(self):
|
---|
| 709 | """FullName after removing the local path to the repository.
|
---|
| 710 |
|
---|
| 711 | If we have a real absolute path name here we can try to do something smart:
|
---|
| 712 | detecting the root of the checkout and truncating /path/to/checkout from
|
---|
| 713 | the name so that we get header guards that don't include things like
|
---|
| 714 | "C:\Documents and Settings\..." or "/home/username/..." in them and thus
|
---|
| 715 | people on different computers who have checked the source out to different
|
---|
| 716 | locations won't see bogus errors.
|
---|
| 717 | """
|
---|
| 718 | fullname = self.FullName()
|
---|
| 719 |
|
---|
| 720 | if os.path.exists(fullname):
|
---|
| 721 | project_dir = os.path.dirname(fullname)
|
---|
| 722 |
|
---|
| 723 | if os.path.exists(os.path.join(project_dir, ".svn")):
|
---|
| 724 | # If there's a .svn file in the current directory, we recursively look
|
---|
| 725 | # up the directory tree for the top of the SVN checkout
|
---|
| 726 | root_dir = project_dir
|
---|
| 727 | one_up_dir = os.path.dirname(root_dir)
|
---|
| 728 | while os.path.exists(os.path.join(one_up_dir, ".svn")):
|
---|
| 729 | root_dir = os.path.dirname(root_dir)
|
---|
| 730 | one_up_dir = os.path.dirname(one_up_dir)
|
---|
| 731 |
|
---|
| 732 | prefix = os.path.commonprefix([root_dir, project_dir])
|
---|
| 733 | return fullname[len(prefix) + 1:]
|
---|
| 734 |
|
---|
| 735 | # Not SVN <= 1.6? Try to find a git, hg, or svn top level directory by
|
---|
| 736 | # searching up from the current path.
|
---|
| 737 | root_dir = os.path.dirname(fullname)
|
---|
| 738 | while (root_dir != os.path.dirname(root_dir) and
|
---|
| 739 | not os.path.exists(os.path.join(root_dir, ".git")) and
|
---|
| 740 | not os.path.exists(os.path.join(root_dir, ".hg")) and
|
---|
| 741 | not os.path.exists(os.path.join(root_dir, ".svn"))):
|
---|
| 742 | root_dir = os.path.dirname(root_dir)
|
---|
| 743 |
|
---|
| 744 | if (os.path.exists(os.path.join(root_dir, ".git")) or
|
---|
| 745 | os.path.exists(os.path.join(root_dir, ".hg")) or
|
---|
| 746 | os.path.exists(os.path.join(root_dir, ".svn"))):
|
---|
| 747 | prefix = os.path.commonprefix([root_dir, project_dir])
|
---|
| 748 | return fullname[len(prefix) + 1:]
|
---|
| 749 |
|
---|
| 750 | # Don't know what to do; header guard warnings may be wrong...
|
---|
| 751 | return fullname
|
---|
| 752 |
|
---|
| 753 | def Split(self):
|
---|
| 754 | """Splits the file into the directory, basename, and extension.
|
---|
| 755 |
|
---|
| 756 | For 'chrome/browser/browser.cc', Split() would
|
---|
| 757 | return ('chrome/browser', 'browser', '.cc')
|
---|
| 758 |
|
---|
| 759 | Returns:
|
---|
| 760 | A tuple of (directory, basename, extension).
|
---|
| 761 | """
|
---|
| 762 |
|
---|
| 763 | googlename = self.RepositoryName()
|
---|
| 764 | project, rest = os.path.split(googlename)
|
---|
| 765 | return (project,) + os.path.splitext(rest)
|
---|
| 766 |
|
---|
| 767 | def BaseName(self):
|
---|
| 768 | """File base name - text after the final slash, before the final period."""
|
---|
| 769 | return self.Split()[1]
|
---|
| 770 |
|
---|
| 771 | def Extension(self):
|
---|
| 772 | """File extension - text following the final period."""
|
---|
| 773 | return self.Split()[2]
|
---|
| 774 |
|
---|
| 775 | def NoExtension(self):
|
---|
| 776 | """File has no source file extension."""
|
---|
| 777 | return '/'.join(self.Split()[0:2])
|
---|
| 778 |
|
---|
| 779 | def IsSource(self):
|
---|
| 780 | """File has a source file extension."""
|
---|
| 781 | return self.Extension()[1:] in EXTENSIONS
|
---|
| 782 |
|
---|
| 783 |
|
---|
| 784 | def _ShouldPrintError(category, confidence, linenum):
|
---|
| 785 | """If confidence >= verbose, category passes filter and is not suppressed."""
|
---|
| 786 |
|
---|
| 787 | # There are three ways we might decide not to print an error message:
|
---|
| 788 | # a "NOLINT(category)" comment appears in the source,
|
---|
| 789 | # the verbosity level isn't high enough, or the filters filter it out.
|
---|
| 790 | if IsErrorSuppressedByNolint(category, linenum):
|
---|
| 791 | return False
|
---|
| 792 | if confidence < _cpplint_state.verbose_level:
|
---|
| 793 | return False
|
---|
| 794 |
|
---|
| 795 | is_filtered = False
|
---|
| 796 | for one_filter in _Filters():
|
---|
| 797 | if one_filter.startswith('-'):
|
---|
| 798 | if category.startswith(one_filter[1:]):
|
---|
| 799 | is_filtered = True
|
---|
| 800 | elif one_filter.startswith('+'):
|
---|
| 801 | if category.startswith(one_filter[1:]):
|
---|
| 802 | is_filtered = False
|
---|
| 803 | else:
|
---|
| 804 | assert False # should have been checked for in SetFilter.
|
---|
| 805 | if is_filtered:
|
---|
| 806 | return False
|
---|
| 807 |
|
---|
| 808 | return True
|
---|
| 809 |
|
---|
| 810 |
|
---|
| 811 | def Error(filename, linenum, category, confidence, message):
|
---|
| 812 | """Logs the fact we've found a lint error.
|
---|
| 813 |
|
---|
| 814 | We log where the error was found, and also our confidence in the error,
|
---|
| 815 | that is, how certain we are this is a legitimate style regression, and
|
---|
| 816 | not a misidentification or a use that's sometimes justified.
|
---|
| 817 |
|
---|
| 818 | False positives can be suppressed by the use of
|
---|
| 819 | "cpplint(category)" comments on the offending line. These are
|
---|
| 820 | parsed into _error_suppressions.
|
---|
| 821 |
|
---|
| 822 | Args:
|
---|
| 823 | filename: The name of the file containing the error.
|
---|
| 824 | linenum: The number of the line containing the error.
|
---|
| 825 | category: A string used to describe the "category" this bug
|
---|
| 826 | falls under: "whitespace", say, or "runtime". Categories
|
---|
| 827 | may have a hierarchy separated by slashes: "whitespace/indent".
|
---|
| 828 | confidence: A number from 1-5 representing a confidence score for
|
---|
| 829 | the error, with 5 meaning that we are certain of the problem,
|
---|
| 830 | and 1 meaning that it could be a legitimate construct.
|
---|
| 831 | message: The error message.
|
---|
| 832 | """
|
---|
| 833 | if _ShouldPrintError(category, confidence, linenum):
|
---|
| 834 | _cpplint_state.IncrementErrorCount(category)
|
---|
| 835 | if _cpplint_state.output_format == 'vs7':
|
---|
| 836 | sys.stderr.write('%s(%s): %s [%s] [%d]\n' % (
|
---|
| 837 | filename, linenum, message, category, confidence))
|
---|
| 838 | else:
|
---|
| 839 | m = '%s:%s: %s [%s] [%d]\n' % (
|
---|
| 840 | filename, linenum, message, category, confidence)
|
---|
| 841 | sys.stderr.write(m)
|
---|
| 842 |
|
---|
| 843 | # Matches standard C++ escape esequences per 2.13.2.3 of the C++ standard.
|
---|
| 844 | _RE_PATTERN_CLEANSE_LINE_ESCAPES = re.compile(
|
---|
| 845 | r'\\([abfnrtv?"\\\']|\d+|x[0-9a-fA-F]+)')
|
---|
| 846 | # Matches strings. Escape codes should already be removed by ESCAPES.
|
---|
| 847 | _RE_PATTERN_CLEANSE_LINE_DOUBLE_QUOTES = re.compile(r'"[^"]*"')
|
---|
| 848 | # Matches characters. Escape codes should already be removed by ESCAPES.
|
---|
| 849 | _RE_PATTERN_CLEANSE_LINE_SINGLE_QUOTES = re.compile(r"'.'")
|
---|
| 850 | # Matches multi-line C++ comments.
|
---|
| 851 | # This RE is a little bit more complicated than one might expect, because we
|
---|
| 852 | # have to take care of space removals tools so we can handle comments inside
|
---|
| 853 | # statements better.
|
---|
| 854 | # The current rule is: We only clear spaces from both sides when we're at the
|
---|
| 855 | # end of the line. Otherwise, we try to remove spaces from the right side,
|
---|
| 856 | # if this doesn't work we try on left side but only if there's a non-character
|
---|
| 857 | # on the right.
|
---|
| 858 | _RE_PATTERN_CLEANSE_LINE_C_COMMENTS = re.compile(
|
---|
| 859 | r"""(\s*/\*.*\*/\s*$|
|
---|
| 860 | /\*.*\*/\s+|
|
---|
| 861 | \s+/\*.*\*/(?=\W)|
|
---|
| 862 | /\*.*\*/)""", re.VERBOSE)
|
---|
| 863 |
|
---|
| 864 |
|
---|
| 865 | def IsCppString(line):
|
---|
| 866 | """Does line terminate so, that the next symbol is in string constant.
|
---|
| 867 |
|
---|
| 868 | This function does not consider single-line nor multi-line comments.
|
---|
| 869 |
|
---|
| 870 | Args:
|
---|
| 871 | line: is a partial line of code starting from the 0..n.
|
---|
| 872 |
|
---|
| 873 | Returns:
|
---|
| 874 | True, if next character appended to 'line' is inside a
|
---|
| 875 | string constant.
|
---|
| 876 | """
|
---|
| 877 |
|
---|
| 878 | line = line.replace(r'\\', 'XX') # after this, \\" does not match to \"
|
---|
| 879 | return ((line.count('"') - line.count(r'\"') - line.count("'\"'")) & 1) == 1
|
---|
| 880 |
|
---|
| 881 |
|
---|
| 882 | def FindNextMultiLineCommentStart(lines, lineix):
|
---|
| 883 | """Find the beginning marker for a multiline comment."""
|
---|
| 884 | while lineix < len(lines):
|
---|
| 885 | if lines[lineix].strip().startswith('/*'):
|
---|
| 886 | # Only return this marker if the comment goes beyond this line
|
---|
| 887 | if lines[lineix].strip().find('*/', 2) < 0:
|
---|
| 888 | return lineix
|
---|
| 889 | lineix += 1
|
---|
| 890 | return len(lines)
|
---|
| 891 |
|
---|
| 892 |
|
---|
| 893 | def FindNextMultiLineCommentEnd(lines, lineix):
|
---|
| 894 | """We are inside a comment, find the end marker."""
|
---|
| 895 | while lineix < len(lines):
|
---|
| 896 | if lines[lineix].strip().endswith('*/'):
|
---|
| 897 | return lineix
|
---|
| 898 | lineix += 1
|
---|
| 899 | return len(lines)
|
---|
| 900 |
|
---|
| 901 |
|
---|
| 902 | def RemoveMultiLineCommentsFromRange(lines, begin, end):
|
---|
| 903 | """Clears a range of lines for multi-line comments."""
|
---|
| 904 | # Having // dummy comments makes the lines non-empty, so we will not get
|
---|
| 905 | # unnecessary blank line warnings later in the code.
|
---|
| 906 | for i in range(begin, end):
|
---|
| 907 | lines[i] = '// dummy'
|
---|
| 908 |
|
---|
| 909 |
|
---|
| 910 | def RemoveMultiLineComments(filename, lines, error):
|
---|
| 911 | """Removes multiline (c-style) comments from lines."""
|
---|
| 912 | lineix = 0
|
---|
| 913 | while lineix < len(lines):
|
---|
| 914 | lineix_begin = FindNextMultiLineCommentStart(lines, lineix)
|
---|
| 915 | if lineix_begin >= len(lines):
|
---|
| 916 | return
|
---|
| 917 | lineix_end = FindNextMultiLineCommentEnd(lines, lineix_begin)
|
---|
| 918 | if lineix_end >= len(lines):
|
---|
| 919 | error(filename, lineix_begin + 1, 'readability/multiline_comment', 5,
|
---|
| 920 | 'Could not find end of multi-line comment')
|
---|
| 921 | return
|
---|
| 922 | RemoveMultiLineCommentsFromRange(lines, lineix_begin, lineix_end + 1)
|
---|
| 923 | lineix = lineix_end + 1
|
---|
| 924 |
|
---|
| 925 |
|
---|
| 926 | def CleanseComments(line):
|
---|
| 927 | """Removes //-comments and single-line C-style /* */ comments.
|
---|
| 928 |
|
---|
| 929 | Args:
|
---|
| 930 | line: A line of C++ source.
|
---|
| 931 |
|
---|
| 932 | Returns:
|
---|
| 933 | The line with single-line comments removed.
|
---|
| 934 | """
|
---|
| 935 | commentpos = line.find('//')
|
---|
| 936 | if commentpos != -1 and not IsCppString(line[:commentpos]):
|
---|
| 937 | line = line[:commentpos].rstrip()
|
---|
| 938 | # get rid of /* ... */
|
---|
| 939 | return _RE_PATTERN_CLEANSE_LINE_C_COMMENTS.sub('', line)
|
---|
| 940 |
|
---|
| 941 |
|
---|
| 942 | class CleansedLines(object):
|
---|
| 943 | """Holds 3 copies of all lines with different preprocessing applied to them.
|
---|
| 944 |
|
---|
| 945 | 1) elided member contains lines without strings and comments,
|
---|
| 946 | 2) lines member contains lines without comments, and
|
---|
| 947 | 3) raw member contains all the lines without processing.
|
---|
| 948 | All these three members are of <type 'list'>, and of the same length.
|
---|
| 949 | """
|
---|
| 950 |
|
---|
| 951 | def __init__(self, lines):
|
---|
| 952 | self.elided = []
|
---|
| 953 | self.lines = []
|
---|
| 954 | self.raw_lines = lines
|
---|
| 955 | self.num_lines = len(lines)
|
---|
| 956 | for linenum in range(len(lines)):
|
---|
| 957 | self.lines.append(CleanseComments(lines[linenum]))
|
---|
| 958 | elided = self._CollapseStrings(lines[linenum])
|
---|
| 959 | self.elided.append(CleanseComments(elided))
|
---|
| 960 |
|
---|
| 961 | def NumLines(self):
|
---|
| 962 | """Returns the number of lines represented."""
|
---|
| 963 | return self.num_lines
|
---|
| 964 |
|
---|
| 965 | @staticmethod
|
---|
| 966 | def _CollapseStrings(elided):
|
---|
| 967 | """Collapses strings and chars on a line to simple "" or '' blocks.
|
---|
| 968 |
|
---|
| 969 | We nix strings first so we're not fooled by text like '"http://"'
|
---|
| 970 |
|
---|
| 971 | Args:
|
---|
| 972 | elided: The line being processed.
|
---|
| 973 |
|
---|
| 974 | Returns:
|
---|
| 975 | The line with collapsed strings.
|
---|
| 976 | """
|
---|
| 977 | if not _RE_PATTERN_INCLUDE.match(elided):
|
---|
| 978 | # Remove escaped characters first to make quote/single quote collapsing
|
---|
| 979 | # basic. Things that look like escaped characters shouldn't occur
|
---|
| 980 | # outside of strings and chars.
|
---|
| 981 | elided = _RE_PATTERN_CLEANSE_LINE_ESCAPES.sub('', elided)
|
---|
| 982 | elided = _RE_PATTERN_CLEANSE_LINE_SINGLE_QUOTES.sub("''", elided)
|
---|
| 983 | elided = _RE_PATTERN_CLEANSE_LINE_DOUBLE_QUOTES.sub('""', elided)
|
---|
| 984 | return elided
|
---|
| 985 |
|
---|
| 986 |
|
---|
| 987 | def CloseExpression(clean_lines, linenum, pos):
|
---|
| 988 | """If input points to ( or { or [, finds the position that closes it.
|
---|
| 989 |
|
---|
| 990 | If lines[linenum][pos] points to a '(' or '{' or '[', finds the
|
---|
| 991 | linenum/pos that correspond to the closing of the expression.
|
---|
| 992 |
|
---|
| 993 | Args:
|
---|
| 994 | clean_lines: A CleansedLines instance containing the file.
|
---|
| 995 | linenum: The number of the line to check.
|
---|
| 996 | pos: A position on the line.
|
---|
| 997 |
|
---|
| 998 | Returns:
|
---|
| 999 | A tuple (line, linenum, pos) pointer *past* the closing brace, or
|
---|
| 1000 | (line, len(lines), -1) if we never find a close. Note we ignore
|
---|
| 1001 | strings and comments when matching; and the line we return is the
|
---|
| 1002 | 'cleansed' line at linenum.
|
---|
| 1003 | """
|
---|
| 1004 |
|
---|
| 1005 | line = clean_lines.elided[linenum]
|
---|
| 1006 | startchar = line[pos]
|
---|
| 1007 | if startchar not in '({[':
|
---|
| 1008 | return (line, clean_lines.NumLines(), -1)
|
---|
| 1009 | if startchar == '(': endchar = ')'
|
---|
| 1010 | if startchar == '[': endchar = ']'
|
---|
| 1011 | if startchar == '{': endchar = '}'
|
---|
| 1012 |
|
---|
| 1013 | num_open = line.count(startchar) - line.count(endchar)
|
---|
| 1014 | while linenum < clean_lines.NumLines() and num_open > 0:
|
---|
| 1015 | linenum += 1
|
---|
| 1016 | line = clean_lines.elided[linenum]
|
---|
| 1017 | num_open += line.count(startchar) - line.count(endchar)
|
---|
| 1018 | # OK, now find the endchar that actually got us back to even
|
---|
| 1019 | endpos = len(line)
|
---|
| 1020 | while num_open >= 0:
|
---|
| 1021 | endpos = line.rfind(')', 0, endpos)
|
---|
| 1022 | num_open -= 1 # chopped off another )
|
---|
| 1023 | return (line, linenum, endpos + 1)
|
---|
| 1024 |
|
---|
| 1025 |
|
---|
| 1026 | def CheckForCopyright(filename, lines, error):
|
---|
| 1027 | """Logs an error if no Copyright message appears at the top of the file."""
|
---|
| 1028 |
|
---|
| 1029 | # We'll say it should occur by line 10. Don't forget there's a
|
---|
| 1030 | # dummy line at the front.
|
---|
| 1031 | for line in range(1, min(len(lines), 11)):
|
---|
| 1032 | if re.search(r'Copyright', lines[line], re.I):
|
---|
| 1033 | break
|
---|
| 1034 | else: # means no copyright line was found
|
---|
| 1035 | error(filename, 0, 'legal/copyright', 5,
|
---|
| 1036 | 'No copyright message found. '
|
---|
| 1037 | 'You should have a line: "Copyright [year] <Copyright Owner>"')
|
---|
| 1038 |
|
---|
| 1039 |
|
---|
| 1040 | def GetHeaderGuardCPPVariable(filename):
|
---|
| 1041 | """Returns the CPP variable that should be used as a header guard.
|
---|
| 1042 |
|
---|
| 1043 | Args:
|
---|
| 1044 | filename: The name of a C++ header file.
|
---|
| 1045 |
|
---|
| 1046 | Returns:
|
---|
| 1047 | The CPP variable that should be used as a header guard in the
|
---|
| 1048 | named file.
|
---|
| 1049 |
|
---|
| 1050 | """
|
---|
| 1051 |
|
---|
| 1052 | # Restores original filename in case that cpplint is invoked from Emacs's
|
---|
| 1053 | # flymake.
|
---|
| 1054 | filename = re.sub(r'_flymake\.h$', '.h', filename)
|
---|
| 1055 |
|
---|
| 1056 | fileinfo = FileInfo(filename)
|
---|
| 1057 | return re.sub(r'[-./\s]', '_', fileinfo.RepositoryName()).upper() + '_'
|
---|
| 1058 |
|
---|
| 1059 |
|
---|
| 1060 | def CheckForHeaderGuard(filename, lines, error):
|
---|
| 1061 | """Checks that the file contains a header guard.
|
---|
| 1062 |
|
---|
| 1063 | Logs an error if no #ifndef header guard is present. For other
|
---|
| 1064 | headers, checks that the full pathname is used.
|
---|
| 1065 |
|
---|
| 1066 | Args:
|
---|
| 1067 | filename: The name of the C++ header file.
|
---|
| 1068 | lines: An array of strings, each representing a line of the file.
|
---|
| 1069 | error: The function to call with any errors found.
|
---|
| 1070 | """
|
---|
| 1071 |
|
---|
| 1072 | cppvar = GetHeaderGuardCPPVariable(filename)
|
---|
| 1073 |
|
---|
| 1074 | ifndef = None
|
---|
| 1075 | ifndef_linenum = 0
|
---|
| 1076 | define = None
|
---|
| 1077 | endif = None
|
---|
| 1078 | endif_linenum = 0
|
---|
| 1079 | for linenum, line in enumerate(lines):
|
---|
| 1080 | linesplit = line.split()
|
---|
| 1081 | if len(linesplit) >= 2:
|
---|
| 1082 | # find the first occurrence of #ifndef and #define, save arg
|
---|
| 1083 | if not ifndef and linesplit[0] == '#ifndef':
|
---|
| 1084 | # set ifndef to the header guard presented on the #ifndef line.
|
---|
| 1085 | ifndef = linesplit[1]
|
---|
| 1086 | ifndef_linenum = linenum
|
---|
| 1087 | if not define and linesplit[0] == '#define':
|
---|
| 1088 | define = linesplit[1]
|
---|
| 1089 | # find the last occurrence of #endif, save entire line
|
---|
| 1090 | if line.startswith('#endif'):
|
---|
| 1091 | endif = line
|
---|
| 1092 | endif_linenum = linenum
|
---|
| 1093 |
|
---|
| 1094 | if not ifndef:
|
---|
| 1095 | error(filename, 0, 'build/header_guard', 5,
|
---|
| 1096 | 'No #ifndef header guard found, suggested CPP variable is: %s' %
|
---|
| 1097 | cppvar)
|
---|
| 1098 | return
|
---|
| 1099 |
|
---|
| 1100 | if not define:
|
---|
| 1101 | error(filename, 0, 'build/header_guard', 5,
|
---|
| 1102 | 'No #define header guard found, suggested CPP variable is: %s' %
|
---|
| 1103 | cppvar)
|
---|
| 1104 | return
|
---|
| 1105 |
|
---|
| 1106 | # The guard should be PATH_FILE_H_, but we also allow PATH_FILE_H__
|
---|
| 1107 | # for backward compatibility.
|
---|
| 1108 | if ifndef != cppvar:
|
---|
| 1109 | error_level = 0
|
---|
| 1110 | if ifndef != cppvar + '_':
|
---|
| 1111 | error_level = 5
|
---|
| 1112 |
|
---|
| 1113 | ParseNolintSuppressions(filename, lines[ifndef_linenum], ifndef_linenum,
|
---|
| 1114 | error)
|
---|
| 1115 | error(filename, ifndef_linenum, 'build/header_guard', error_level,
|
---|
| 1116 | '#ifndef header guard has wrong style, please use: %s' % cppvar)
|
---|
| 1117 |
|
---|
| 1118 | if define != ifndef:
|
---|
| 1119 | error(filename, 0, 'build/header_guard', 5,
|
---|
| 1120 | '#ifndef and #define don\'t match, suggested CPP variable is: %s' %
|
---|
| 1121 | cppvar)
|
---|
| 1122 | return
|
---|
| 1123 |
|
---|
| 1124 | if endif != ('#endif // %s' % cppvar):
|
---|
| 1125 | error_level = 0
|
---|
| 1126 | if endif != ('#endif // %s' % (cppvar + '_')):
|
---|
| 1127 | error_level = 5
|
---|
| 1128 |
|
---|
| 1129 | ParseNolintSuppressions(filename, lines[endif_linenum], endif_linenum,
|
---|
| 1130 | error)
|
---|
| 1131 | error(filename, endif_linenum, 'build/header_guard', error_level,
|
---|
| 1132 | '#endif line should be "#endif // %s"' % cppvar)
|
---|
| 1133 |
|
---|
| 1134 |
|
---|
| 1135 | def CheckForUnicodeReplacementCharacters(filename, lines, error):
|
---|
| 1136 | """Logs an error for each line containing Unicode replacement characters.
|
---|
| 1137 |
|
---|
| 1138 | These indicate that either the file contained invalid UTF-8 (likely)
|
---|
| 1139 | or Unicode replacement characters (which it shouldn't). Note that
|
---|
| 1140 | it's possible for this to throw off line numbering if the invalid
|
---|
| 1141 | UTF-8 occurred adjacent to a newline.
|
---|
| 1142 |
|
---|
| 1143 | Args:
|
---|
| 1144 | filename: The name of the current file.
|
---|
| 1145 | lines: An array of strings, each representing a line of the file.
|
---|
| 1146 | error: The function to call with any errors found.
|
---|
| 1147 | """
|
---|
| 1148 | for linenum, line in enumerate(lines):
|
---|
| 1149 | if u('\ufffd') in line:
|
---|
| 1150 | error(filename, linenum, 'readability/utf8', 5,
|
---|
| 1151 | 'Line contains invalid UTF-8 (or Unicode replacement character).')
|
---|
| 1152 |
|
---|
| 1153 |
|
---|
| 1154 | def CheckForNewlineAtEOF(filename, lines, error):
|
---|
| 1155 | """Logs an error if there is no newline char at the end of the file.
|
---|
| 1156 |
|
---|
| 1157 | Args:
|
---|
| 1158 | filename: The name of the current file.
|
---|
| 1159 | lines: An array of strings, each representing a line of the file.
|
---|
| 1160 | error: The function to call with any errors found.
|
---|
| 1161 | """
|
---|
| 1162 |
|
---|
| 1163 | # The array lines() was created by adding two newlines to the
|
---|
| 1164 | # original file (go figure), then splitting on \n.
|
---|
| 1165 | # To verify that the file ends in \n, we just have to make sure the
|
---|
| 1166 | # last-but-two element of lines() exists and is empty.
|
---|
| 1167 | if len(lines) < 3 or lines[-2]:
|
---|
| 1168 | error(filename, len(lines) - 2, 'whitespace/ending_newline', 5,
|
---|
| 1169 | 'Could not find a newline character at the end of the file.')
|
---|
| 1170 |
|
---|
| 1171 |
|
---|
| 1172 | def CheckForMultilineCommentsAndStrings(filename, clean_lines, linenum, error):
|
---|
| 1173 | """Logs an error if we see /* ... */ or "..." that extend past one line.
|
---|
| 1174 |
|
---|
| 1175 | /* ... */ comments are legit inside macros, for one line.
|
---|
| 1176 | Otherwise, we prefer // comments, so it's ok to warn about the
|
---|
| 1177 | other. Likewise, it's ok for strings to extend across multiple
|
---|
| 1178 | lines, as long as a line continuation character (backslash)
|
---|
| 1179 | terminates each line. Although not currently prohibited by the C++
|
---|
| 1180 | style guide, it's ugly and unnecessary. We don't do well with either
|
---|
| 1181 | in this lint program, so we warn about both.
|
---|
| 1182 |
|
---|
| 1183 | Args:
|
---|
| 1184 | filename: The name of the current file.
|
---|
| 1185 | clean_lines: A CleansedLines instance containing the file.
|
---|
| 1186 | linenum: The number of the line to check.
|
---|
| 1187 | error: The function to call with any errors found.
|
---|
| 1188 | """
|
---|
| 1189 | line = clean_lines.elided[linenum]
|
---|
| 1190 |
|
---|
| 1191 | # Remove all \\ (escaped backslashes) from the line. They are OK, and the
|
---|
| 1192 | # second (escaped) slash may trigger later \" detection erroneously.
|
---|
| 1193 | line = line.replace('\\\\', '')
|
---|
| 1194 |
|
---|
| 1195 | if line.count('/*') > line.count('*/'):
|
---|
| 1196 | error(filename, linenum, 'readability/multiline_comment', 5,
|
---|
| 1197 | 'Complex multi-line /*...*/-style comment found. '
|
---|
| 1198 | 'Lint may give bogus warnings. '
|
---|
| 1199 | 'Consider replacing these with //-style comments, '
|
---|
| 1200 | 'with #if 0...#endif, '
|
---|
| 1201 | 'or with more clearly structured multi-line comments.')
|
---|
| 1202 |
|
---|
| 1203 | if (line.count('"') - line.count('\\"')) % 2:
|
---|
| 1204 | error(filename, linenum, 'readability/multiline_string', 5,
|
---|
| 1205 | 'Multi-line string ("...") found. This lint script doesn\'t '
|
---|
| 1206 | 'do well with such strings, and may give bogus warnings. They\'re '
|
---|
| 1207 | 'ugly and unnecessary, and you should use concatenation instead".')
|
---|
| 1208 |
|
---|
| 1209 |
|
---|
| 1210 | threading_list = (
|
---|
| 1211 | ('asctime(', 'asctime_r('),
|
---|
| 1212 | ('ctime(', 'ctime_r('),
|
---|
| 1213 | ('getgrgid(', 'getgrgid_r('),
|
---|
| 1214 | ('getgrnam(', 'getgrnam_r('),
|
---|
| 1215 | ('getlogin(', 'getlogin_r('),
|
---|
| 1216 | ('getpwnam(', 'getpwnam_r('),
|
---|
| 1217 | ('getpwuid(', 'getpwuid_r('),
|
---|
| 1218 | ('gmtime(', 'gmtime_r('),
|
---|
| 1219 | ('localtime(', 'localtime_r('),
|
---|
| 1220 | ('rand(', 'rand_r('),
|
---|
| 1221 | ('readdir(', 'readdir_r('),
|
---|
| 1222 | ('strtok(', 'strtok_r('),
|
---|
| 1223 | ('ttyname(', 'ttyname_r('),
|
---|
| 1224 | )
|
---|
| 1225 |
|
---|
| 1226 |
|
---|
| 1227 | def CheckPosixThreading(filename, clean_lines, linenum, error):
|
---|
| 1228 | """Checks for calls to thread-unsafe functions.
|
---|
| 1229 |
|
---|
| 1230 | Much code has been originally written without consideration of
|
---|
| 1231 | multi-threading. Also, engineers are relying on their old experience;
|
---|
| 1232 | they have learned posix before threading extensions were added. These
|
---|
| 1233 | tests guide the engineers to use thread-safe functions (when using
|
---|
| 1234 | posix directly).
|
---|
| 1235 |
|
---|
| 1236 | Args:
|
---|
| 1237 | filename: The name of the current file.
|
---|
| 1238 | clean_lines: A CleansedLines instance containing the file.
|
---|
| 1239 | linenum: The number of the line to check.
|
---|
| 1240 | error: The function to call with any errors found.
|
---|
| 1241 | """
|
---|
| 1242 | line = clean_lines.elided[linenum]
|
---|
| 1243 | for single_thread_function, multithread_safe_function in threading_list:
|
---|
| 1244 | ix = line.find(single_thread_function)
|
---|
| 1245 | # Comparisons made explicit for clarity -- pylint: disable-msg=C6403
|
---|
| 1246 | if ix >= 0 and (ix == 0 or (not line[ix - 1].isalnum() and
|
---|
| 1247 | line[ix - 1] not in ('_', '.', '>'))):
|
---|
| 1248 | error(filename, linenum, 'runtime/threadsafe_fn', 2,
|
---|
| 1249 | 'Consider using ' + multithread_safe_function +
|
---|
| 1250 | '...) instead of ' + single_thread_function +
|
---|
| 1251 | '...) for improved thread safety.')
|
---|
| 1252 |
|
---|
| 1253 |
|
---|
| 1254 | # Matches invalid increment: *count++, which moves pointer instead of
|
---|
| 1255 | # incrementing a value.
|
---|
| 1256 | _RE_PATTERN_INVALID_INCREMENT = re.compile(
|
---|
| 1257 | r'^\s*\*\w+(\+\+|--);')
|
---|
| 1258 |
|
---|
| 1259 |
|
---|
| 1260 | def CheckInvalidIncrement(filename, clean_lines, linenum, error):
|
---|
| 1261 | """Checks for invalid increment *count++.
|
---|
| 1262 |
|
---|
| 1263 | For example following function:
|
---|
| 1264 | void increment_counter(int* count) {
|
---|
| 1265 | *count++;
|
---|
| 1266 | }
|
---|
| 1267 | is invalid, because it effectively does count++, moving pointer, and should
|
---|
| 1268 | be replaced with ++*count, (*count)++ or *count += 1.
|
---|
| 1269 |
|
---|
| 1270 | Args:
|
---|
| 1271 | filename: The name of the current file.
|
---|
| 1272 | clean_lines: A CleansedLines instance containing the file.
|
---|
| 1273 | linenum: The number of the line to check.
|
---|
| 1274 | error: The function to call with any errors found.
|
---|
| 1275 | """
|
---|
| 1276 | line = clean_lines.elided[linenum]
|
---|
| 1277 | if _RE_PATTERN_INVALID_INCREMENT.match(line):
|
---|
| 1278 | error(filename, linenum, 'runtime/invalid_increment', 5,
|
---|
| 1279 | 'Changing pointer instead of value (or unused value of operator*).')
|
---|
| 1280 |
|
---|
| 1281 |
|
---|
| 1282 | class _ClassInfo(object):
|
---|
| 1283 | """Stores information about a class."""
|
---|
| 1284 |
|
---|
| 1285 | def __init__(self, name, clean_lines, linenum):
|
---|
| 1286 | self.name = name
|
---|
| 1287 | self.linenum = linenum
|
---|
| 1288 | self.seen_open_brace = False
|
---|
| 1289 | self.is_derived = False
|
---|
| 1290 | self.virtual_method_linenumber = None
|
---|
| 1291 | self.has_virtual_destructor = False
|
---|
| 1292 | self.brace_depth = 0
|
---|
| 1293 |
|
---|
| 1294 | # Try to find the end of the class. This will be confused by things like:
|
---|
| 1295 | # class A {
|
---|
| 1296 | # } *x = { ...
|
---|
| 1297 | #
|
---|
| 1298 | # But it's still good enough for CheckSectionSpacing.
|
---|
| 1299 | self.last_line = 0
|
---|
| 1300 | depth = 0
|
---|
| 1301 | for i in range(linenum, clean_lines.NumLines()):
|
---|
| 1302 | line = clean_lines.lines[i]
|
---|
| 1303 | depth += line.count('{') - line.count('}')
|
---|
| 1304 | if not depth:
|
---|
| 1305 | self.last_line = i
|
---|
| 1306 | break
|
---|
| 1307 |
|
---|
| 1308 |
|
---|
| 1309 | class _ClassState(object):
|
---|
| 1310 | """Holds the current state of the parse relating to class declarations.
|
---|
| 1311 |
|
---|
| 1312 | It maintains a stack of _ClassInfos representing the parser's guess
|
---|
| 1313 | as to the current nesting of class declarations. The innermost class
|
---|
| 1314 | is at the top (back) of the stack. Typically, the stack will either
|
---|
| 1315 | be empty or have exactly one entry.
|
---|
| 1316 | """
|
---|
| 1317 |
|
---|
| 1318 | def __init__(self):
|
---|
| 1319 | self.classinfo_stack = []
|
---|
| 1320 |
|
---|
| 1321 | def CheckFinished(self, filename, error):
|
---|
| 1322 | """Checks that all classes have been completely parsed.
|
---|
| 1323 |
|
---|
| 1324 | Call this when all lines in a file have been processed.
|
---|
| 1325 | Args:
|
---|
| 1326 | filename: The name of the current file.
|
---|
| 1327 | error: The function to call with any errors found.
|
---|
| 1328 | """
|
---|
| 1329 | if self.classinfo_stack:
|
---|
| 1330 | # Note: This test can result in false positives if #ifdef constructs
|
---|
| 1331 | # get in the way of brace matching. See the testBuildClass test in
|
---|
| 1332 | # cpplint_unittest.py for an example of this.
|
---|
| 1333 | error(filename, self.classinfo_stack[0].linenum, 'build/class', 5,
|
---|
| 1334 | 'Failed to find complete declaration of class %s' %
|
---|
| 1335 | self.classinfo_stack[0].name)
|
---|
| 1336 |
|
---|
| 1337 |
|
---|
| 1338 | def CheckForNonStandardConstructs(filename, clean_lines, linenum,
|
---|
| 1339 | class_state, error):
|
---|
| 1340 | """Logs an error if we see certain non-ANSI constructs ignored by gcc-2.
|
---|
| 1341 |
|
---|
| 1342 | Complain about several constructs which gcc-2 accepts, but which are
|
---|
| 1343 | not standard C++. Warning about these in lint is one way to ease the
|
---|
| 1344 | transition to new compilers.
|
---|
| 1345 | - put storage class first (e.g. "static const" instead of "const static").
|
---|
| 1346 | - "%lld" instead of %qd" in printf-type functions.
|
---|
| 1347 | - "%1$d" is non-standard in printf-type functions.
|
---|
| 1348 | - "\%" is an undefined character escape sequence.
|
---|
| 1349 | - text after #endif is not allowed.
|
---|
| 1350 | - invalid inner-style forward declaration.
|
---|
| 1351 | - >? and <? operators, and their >?= and <?= cousins.
|
---|
| 1352 | - classes with virtual methods need virtual destructors (compiler warning
|
---|
| 1353 | available, but not turned on yet.)
|
---|
| 1354 |
|
---|
| 1355 | Additionally, check for constructor/destructor style violations and reference
|
---|
| 1356 | members, as it is very convenient to do so while checking for
|
---|
| 1357 | gcc-2 compliance.
|
---|
| 1358 |
|
---|
| 1359 | Args:
|
---|
| 1360 | filename: The name of the current file.
|
---|
| 1361 | clean_lines: A CleansedLines instance containing the file.
|
---|
| 1362 | linenum: The number of the line to check.
|
---|
| 1363 | class_state: A _ClassState instance which maintains information about
|
---|
| 1364 | the current stack of nested class declarations being parsed.
|
---|
| 1365 | error: A callable to which errors are reported, which takes 4 arguments:
|
---|
| 1366 | filename, line number, error level, and message
|
---|
| 1367 | """
|
---|
| 1368 |
|
---|
| 1369 | # Remove comments from the line, but leave in strings for now.
|
---|
| 1370 | line = clean_lines.lines[linenum]
|
---|
| 1371 |
|
---|
| 1372 | if Search(r'printf\s*\(.*".*%[-+ ]?\d*q', line):
|
---|
| 1373 | error(filename, linenum, 'runtime/printf_format', 3,
|
---|
| 1374 | '%q in format strings is deprecated. Use %ll instead.')
|
---|
| 1375 |
|
---|
| 1376 | if Search(r'printf\s*\(.*".*%\d+\$', line):
|
---|
| 1377 | error(filename, linenum, 'runtime/printf_format', 2,
|
---|
| 1378 | '%N$ formats are unconventional. Try rewriting to avoid them.')
|
---|
| 1379 |
|
---|
| 1380 | # Remove escaped backslashes before looking for undefined escapes.
|
---|
| 1381 | line = line.replace('\\\\', '')
|
---|
| 1382 |
|
---|
| 1383 | if Search(r'("|\').*\\(%|\[|\(|{)', line):
|
---|
| 1384 | error(filename, linenum, 'build/printf_format', 3,
|
---|
| 1385 | '%, [, (, and { are undefined character escapes. Unescape them.')
|
---|
| 1386 |
|
---|
| 1387 | # For the rest, work with both comments and strings removed.
|
---|
| 1388 | line = clean_lines.elided[linenum]
|
---|
| 1389 |
|
---|
| 1390 | if Search(r'\b(const|volatile|void|char|short|int|long'
|
---|
| 1391 | r'|float|double|signed|unsigned'
|
---|
| 1392 | r'|schar|u?int8|u?int16|u?int32|u?int64)'
|
---|
| 1393 | r'\s+(auto|register|static|extern|typedef)\b',
|
---|
| 1394 | line):
|
---|
| 1395 | error(filename, linenum, 'build/storage_class', 5,
|
---|
| 1396 | 'Storage class (static, extern, typedef, etc) should be first.')
|
---|
| 1397 |
|
---|
| 1398 | if Match(r'\s*#\s*endif\s*[^/\s]+', line):
|
---|
| 1399 | error(filename, linenum, 'build/endif_comment', 5,
|
---|
| 1400 | 'Uncommented text after #endif is non-standard. Use a comment.')
|
---|
| 1401 |
|
---|
| 1402 | if Match(r'\s*class\s+(\w+\s*::\s*)+\w+\s*;', line):
|
---|
| 1403 | error(filename, linenum, 'build/forward_decl', 5,
|
---|
| 1404 | 'Inner-style forward declarations are invalid. Remove this line.')
|
---|
| 1405 |
|
---|
| 1406 | if Search(r'(\w+|[+-]?\d+(\.\d*)?)\s*(<|>)\?=?\s*(\w+|[+-]?\d+)(\.\d*)?',
|
---|
| 1407 | line):
|
---|
| 1408 | error(filename, linenum, 'build/deprecated', 3,
|
---|
| 1409 | '>? and <? (max and min) operators are non-standard and deprecated.')
|
---|
| 1410 |
|
---|
| 1411 | if Search(r'^\s*const\s*string\s*&\s*\w+\s*;', line):
|
---|
| 1412 | # TODO(unknown): Could it be expanded safely to arbitrary references,
|
---|
| 1413 | # without triggering too many false positives? The first
|
---|
| 1414 | # attempt triggered 5 warnings for mostly benign code in the regtest, hence
|
---|
| 1415 | # the restriction.
|
---|
| 1416 | # Here's the original regexp, for the reference:
|
---|
| 1417 | # type_name = r'\w+((\s*::\s*\w+)|(\s*<\s*\w+?\s*>))?'
|
---|
| 1418 | # r'\s*const\s*' + type_name + '\s*&\s*\w+\s*;'
|
---|
| 1419 | error(filename, linenum, 'runtime/member_string_references', 2,
|
---|
| 1420 | 'const string& members are dangerous. It is much better to use '
|
---|
| 1421 | 'alternatives, such as pointers or simple constants.')
|
---|
| 1422 |
|
---|
| 1423 | # Track class entry and exit, and attempt to find cases within the
|
---|
| 1424 | # class declaration that don't meet the C++ style
|
---|
| 1425 | # guidelines. Tracking is very dependent on the code matching Google
|
---|
| 1426 | # style guidelines, but it seems to perform well enough in testing
|
---|
| 1427 | # to be a worthwhile addition to the checks.
|
---|
| 1428 | classinfo_stack = class_state.classinfo_stack
|
---|
| 1429 | # Look for a class declaration. The regexp accounts for decorated classes
|
---|
| 1430 | # such as in:
|
---|
| 1431 | # class LOCKABLE API Object {
|
---|
| 1432 | # };
|
---|
| 1433 | class_decl_match = Match(
|
---|
| 1434 | r'\s*(template\s*<[\w\s<>,:]*>\s*)?'
|
---|
| 1435 | '(class|struct)\s+([A-Z_]+\s+)*(\w+(::\w+)*)', line)
|
---|
| 1436 | if class_decl_match:
|
---|
| 1437 | classinfo_stack.append(_ClassInfo(
|
---|
| 1438 | class_decl_match.group(4), clean_lines, linenum))
|
---|
| 1439 |
|
---|
| 1440 | # Everything else in this function uses the top of the stack if it's
|
---|
| 1441 | # not empty.
|
---|
| 1442 | if not classinfo_stack:
|
---|
| 1443 | return
|
---|
| 1444 |
|
---|
| 1445 | classinfo = classinfo_stack[-1]
|
---|
| 1446 |
|
---|
| 1447 | # If the opening brace hasn't been seen look for it and also
|
---|
| 1448 | # parent class declarations.
|
---|
| 1449 | if not classinfo.seen_open_brace:
|
---|
| 1450 | # If the line has a ';' in it, assume it's a forward declaration or
|
---|
| 1451 | # a single-line class declaration, which we won't process.
|
---|
| 1452 | if line.find(';') != -1:
|
---|
| 1453 | classinfo_stack.pop()
|
---|
| 1454 | return
|
---|
| 1455 | classinfo.seen_open_brace = (line.find('{') != -1)
|
---|
| 1456 | # Look for a bare ':'
|
---|
| 1457 | if Search('(^|[^:]):($|[^:])', line):
|
---|
| 1458 | classinfo.is_derived = True
|
---|
| 1459 | if not classinfo.seen_open_brace:
|
---|
| 1460 | return # Everything else in this function is for after open brace
|
---|
| 1461 |
|
---|
| 1462 | # The class may have been declared with namespace or classname qualifiers.
|
---|
| 1463 | # The constructor and destructor will not have those qualifiers.
|
---|
| 1464 | base_classname = classinfo.name.split('::')[-1]
|
---|
| 1465 |
|
---|
| 1466 | # Look for single-argument constructors that aren't marked explicit.
|
---|
| 1467 | # Technically a valid construct, but against style.
|
---|
| 1468 | args = Match(r'\s+(?:inline\s+)?%s\s*\(([^,()]+)\)'
|
---|
| 1469 | % re.escape(base_classname),
|
---|
| 1470 | line)
|
---|
| 1471 | if (args and
|
---|
| 1472 | args.group(1) != 'void' and
|
---|
| 1473 | not Match(r'(const\s+)?%s\s*(?:<\w+>\s*)?&' % re.escape(base_classname),
|
---|
| 1474 | args.group(1).strip())):
|
---|
| 1475 | error(filename, linenum, 'runtime/explicit', 5,
|
---|
| 1476 | 'Single-argument constructors should be marked explicit.')
|
---|
| 1477 |
|
---|
| 1478 | # Look for methods declared virtual.
|
---|
| 1479 | if Search(r'\bvirtual\b', line):
|
---|
| 1480 | classinfo.virtual_method_linenumber = linenum
|
---|
| 1481 | # Only look for a destructor declaration on the same line. It would
|
---|
| 1482 | # be extremely unlikely for the destructor declaration to occupy
|
---|
| 1483 | # more than one line.
|
---|
| 1484 | if Search(r'~%s\s*\(' % base_classname, line):
|
---|
| 1485 | classinfo.has_virtual_destructor = True
|
---|
| 1486 |
|
---|
| 1487 | # Look for class end.
|
---|
| 1488 | brace_depth = classinfo.brace_depth
|
---|
| 1489 | brace_depth = brace_depth + line.count('{') - line.count('}')
|
---|
| 1490 | if brace_depth <= 0:
|
---|
| 1491 | classinfo = classinfo_stack.pop()
|
---|
| 1492 | # Try to detect missing virtual destructor declarations.
|
---|
| 1493 | # For now, only warn if a non-derived class with virtual methods lacks
|
---|
| 1494 | # a virtual destructor. This is to make it less likely that people will
|
---|
| 1495 | # declare derived virtual destructors without declaring the base
|
---|
| 1496 | # destructor virtual.
|
---|
| 1497 | if ((classinfo.virtual_method_linenumber is not None) and
|
---|
| 1498 | (not classinfo.has_virtual_destructor) and
|
---|
| 1499 | (not classinfo.is_derived)): # Only warn for base classes
|
---|
| 1500 | error(filename, classinfo.linenum, 'runtime/virtual', 4,
|
---|
| 1501 | 'The class %s probably needs a virtual destructor due to '
|
---|
| 1502 | 'having virtual method(s), one declared at line %d.'
|
---|
| 1503 | % (classinfo.name, classinfo.virtual_method_linenumber))
|
---|
| 1504 | else:
|
---|
| 1505 | classinfo.brace_depth = brace_depth
|
---|
| 1506 |
|
---|
| 1507 |
|
---|
| 1508 | def CheckSpacingForFunctionCall(filename, line, linenum, error):
|
---|
| 1509 | """Checks for the correctness of various spacing around function calls.
|
---|
| 1510 |
|
---|
| 1511 | Args:
|
---|
| 1512 | filename: The name of the current file.
|
---|
| 1513 | line: The text of the line to check.
|
---|
| 1514 | linenum: The number of the line to check.
|
---|
| 1515 | error: The function to call with any errors found.
|
---|
| 1516 | """
|
---|
| 1517 |
|
---|
| 1518 | # Since function calls often occur inside if/for/while/switch
|
---|
| 1519 | # expressions - which have their own, more liberal conventions - we
|
---|
| 1520 | # first see if we should be looking inside such an expression for a
|
---|
| 1521 | # function call, to which we can apply more strict standards.
|
---|
| 1522 | fncall = line # if there's no control flow construct, look at whole line
|
---|
| 1523 | for pattern in (r'\bif\s*\((.*)\)\s*{',
|
---|
| 1524 | r'\bfor\s*\((.*)\)\s*{',
|
---|
| 1525 | r'\bwhile\s*\((.*)\)\s*[{;]',
|
---|
| 1526 | r'\bswitch\s*\((.*)\)\s*{'):
|
---|
| 1527 | match = Search(pattern, line)
|
---|
| 1528 | if match:
|
---|
| 1529 | fncall = match.group(1) # look inside the parens for function calls
|
---|
| 1530 | break
|
---|
| 1531 |
|
---|
| 1532 | # Except in if/for/while/switch, there should never be space
|
---|
| 1533 | # immediately inside parens (eg "f( 3, 4 )"). We make an exception
|
---|
| 1534 | # for nested parens ( (a+b) + c ). Likewise, there should never be
|
---|
| 1535 | # a space before a ( when it's a function argument. I assume it's a
|
---|
| 1536 | # function argument when the char before the whitespace is legal in
|
---|
| 1537 | # a function name (alnum + _) and we're not starting a macro. Also ignore
|
---|
| 1538 | # pointers and references to arrays and functions coz they're too tricky:
|
---|
| 1539 | # we use a very simple way to recognize these:
|
---|
| 1540 | # " (something)(maybe-something)" or
|
---|
| 1541 | # " (something)(maybe-something," or
|
---|
| 1542 | # " (something)[something]"
|
---|
| 1543 | # Note that we assume the contents of [] to be short enough that
|
---|
| 1544 | # they'll never need to wrap.
|
---|
| 1545 | if ( # Ignore control structures.
|
---|
| 1546 | not Search(r'\b(if|for|while|switch|return|delete)\b', fncall) and
|
---|
| 1547 | # Ignore pointers/references to functions.
|
---|
| 1548 | not Search(r' \([^)]+\)\([^)]*(\)|,$)', fncall) and
|
---|
| 1549 | # Ignore pointers/references to arrays.
|
---|
| 1550 | not Search(r' \([^)]+\)\[[^\]]+\]', fncall)):
|
---|
| 1551 | if Search(r'\w\s*\(\s(?!\s*\\$)', fncall): # a ( used for a fn call
|
---|
| 1552 | error(filename, linenum, 'whitespace/parens', 4,
|
---|
| 1553 | 'Extra space after ( in function call')
|
---|
| 1554 | elif Search(r'\(\s+(?!(\s*\\)|\()', fncall):
|
---|
| 1555 | error(filename, linenum, 'whitespace/parens', 2,
|
---|
| 1556 | 'Extra space after (')
|
---|
| 1557 | if (Search(r'\w\s+\(', fncall) and
|
---|
| 1558 | not Search(r'#\s*define|typedef', fncall)):
|
---|
| 1559 | error(filename, linenum, 'whitespace/parens', 4,
|
---|
| 1560 | 'Extra space before ( in function call')
|
---|
| 1561 | # If the ) is followed only by a newline or a { + newline, assume it's
|
---|
| 1562 | # part of a control statement (if/while/etc), and don't complain
|
---|
| 1563 | if Search(r'[^)]\s+\)\s*[^{\s]', fncall):
|
---|
| 1564 | # If the closing parenthesis is preceded by only whitespaces,
|
---|
| 1565 | # try to give a more descriptive error message.
|
---|
| 1566 | if Search(r'^\s+\)', fncall):
|
---|
| 1567 | error(filename, linenum, 'whitespace/parens', 2,
|
---|
| 1568 | 'Closing ) should be moved to the previous line')
|
---|
| 1569 | else:
|
---|
| 1570 | error(filename, linenum, 'whitespace/parens', 2,
|
---|
| 1571 | 'Extra space before )')
|
---|
| 1572 |
|
---|
| 1573 |
|
---|
| 1574 | def IsBlankLine(line):
|
---|
| 1575 | """Returns true if the given line is blank.
|
---|
| 1576 |
|
---|
| 1577 | We consider a line to be blank if the line is empty or consists of
|
---|
| 1578 | only white spaces.
|
---|
| 1579 |
|
---|
| 1580 | Args:
|
---|
| 1581 | line: A line of a string.
|
---|
| 1582 |
|
---|
| 1583 | Returns:
|
---|
| 1584 | True, if the given line is blank.
|
---|
| 1585 | """
|
---|
| 1586 | return not line or line.isspace()
|
---|
| 1587 |
|
---|
| 1588 |
|
---|
| 1589 | def CheckForFunctionLengths(filename, clean_lines, linenum,
|
---|
| 1590 | function_state, error):
|
---|
| 1591 | """Reports for long function bodies.
|
---|
| 1592 |
|
---|
| 1593 | For an overview why this is done, see:
|
---|
| 1594 | http://google-styleguide.googlecode.com/svn/trunk/cppguide.xml#Write_Short_Functions
|
---|
| 1595 |
|
---|
| 1596 | Uses a simplistic algorithm assuming other style guidelines
|
---|
| 1597 | (especially spacing) are followed.
|
---|
| 1598 | Only checks unindented functions, so class members are unchecked.
|
---|
| 1599 | Trivial bodies are unchecked, so constructors with huge initializer lists
|
---|
| 1600 | may be missed.
|
---|
| 1601 | Blank/comment lines are not counted so as to avoid encouraging the removal
|
---|
| 1602 | of vertical space and comments just to get through a lint check.
|
---|
| 1603 | NOLINT *on the last line of a function* disables this check.
|
---|
| 1604 |
|
---|
| 1605 | Args:
|
---|
| 1606 | filename: The name of the current file.
|
---|
| 1607 | clean_lines: A CleansedLines instance containing the file.
|
---|
| 1608 | linenum: The number of the line to check.
|
---|
| 1609 | function_state: Current function name and lines in body so far.
|
---|
| 1610 | error: The function to call with any errors found.
|
---|
| 1611 | """
|
---|
| 1612 | lines = clean_lines.lines
|
---|
| 1613 | line = lines[linenum]
|
---|
| 1614 | raw = clean_lines.raw_lines
|
---|
| 1615 | raw_line = raw[linenum]
|
---|
| 1616 | joined_line = ''
|
---|
| 1617 |
|
---|
| 1618 | starting_func = False
|
---|
| 1619 | regexp = r'(\w(\w|::|\*|\&|\s)*)\(' # decls * & space::name( ...
|
---|
| 1620 | match_result = Match(regexp, line)
|
---|
| 1621 | if match_result:
|
---|
| 1622 | # If the name is all caps and underscores, figure it's a macro and
|
---|
| 1623 | # ignore it, unless it's TEST or TEST_F.
|
---|
| 1624 | function_name = match_result.group(1).split()[-1]
|
---|
| 1625 | if function_name == 'TEST' or function_name == 'TEST_F' or (
|
---|
| 1626 | not Match(r'[A-Z_]+$', function_name)):
|
---|
| 1627 | starting_func = True
|
---|
| 1628 |
|
---|
| 1629 | if starting_func:
|
---|
| 1630 | body_found = False
|
---|
| 1631 | for start_linenum in range(linenum, clean_lines.NumLines()):
|
---|
| 1632 | start_line = lines[start_linenum]
|
---|
| 1633 | joined_line += ' ' + start_line.lstrip()
|
---|
| 1634 | if Search(r'(;|})', start_line): # Declarations and trivial functions
|
---|
| 1635 | body_found = True
|
---|
| 1636 | break # ... ignore
|
---|
| 1637 | elif Search(r'{', start_line):
|
---|
| 1638 | body_found = True
|
---|
| 1639 | function = Search(r'((\w|:)*)\(', line).group(1)
|
---|
| 1640 | if Match(r'TEST', function): # Handle TEST... macros
|
---|
| 1641 | parameter_regexp = Search(r'(\(.*\))', joined_line)
|
---|
| 1642 | if parameter_regexp: # Ignore bad syntax
|
---|
| 1643 | function += parameter_regexp.group(1)
|
---|
| 1644 | else:
|
---|
| 1645 | function += '()'
|
---|
| 1646 | function_state.Begin(function)
|
---|
| 1647 | break
|
---|
| 1648 | if not body_found:
|
---|
| 1649 | # No body for the function (or evidence of a non-function) was found.
|
---|
| 1650 | error(filename, linenum, 'readability/fn_size', 5,
|
---|
| 1651 | 'Lint failed to find start of function body.')
|
---|
| 1652 | elif Match(r'^\}\s*$', line): # function end
|
---|
| 1653 | function_state.Check(error, filename, linenum)
|
---|
| 1654 | function_state.End()
|
---|
| 1655 | elif not Match(r'^\s*$', line):
|
---|
| 1656 | function_state.Count() # Count non-blank/non-comment lines.
|
---|
| 1657 |
|
---|
| 1658 |
|
---|
| 1659 | _RE_PATTERN_TODO = re.compile(r'^//(\s*)TODO(\(.+?\))?:?(\s|$)?')
|
---|
| 1660 |
|
---|
| 1661 |
|
---|
| 1662 | def CheckComment(comment, filename, linenum, error):
|
---|
| 1663 | """Checks for common mistakes in TODO comments.
|
---|
| 1664 |
|
---|
| 1665 | Args:
|
---|
| 1666 | comment: The text of the comment from the line in question.
|
---|
| 1667 | filename: The name of the current file.
|
---|
| 1668 | linenum: The number of the line to check.
|
---|
| 1669 | error: The function to call with any errors found.
|
---|
| 1670 | """
|
---|
| 1671 | match = _RE_PATTERN_TODO.match(comment)
|
---|
| 1672 | if match:
|
---|
| 1673 | # One whitespace is correct; zero whitespace is handled elsewhere.
|
---|
| 1674 | leading_whitespace = match.group(1)
|
---|
| 1675 | if len(leading_whitespace) > 1:
|
---|
| 1676 | error(filename, linenum, 'whitespace/todo', 2,
|
---|
| 1677 | 'Too many spaces before TODO')
|
---|
| 1678 |
|
---|
| 1679 | username = match.group(2)
|
---|
| 1680 | if not username:
|
---|
| 1681 | error(filename, linenum, 'readability/todo', 2,
|
---|
| 1682 | 'Missing username in TODO; it should look like '
|
---|
| 1683 | '"// TODO(my_username): Stuff."')
|
---|
| 1684 |
|
---|
| 1685 | middle_whitespace = match.group(3)
|
---|
| 1686 | # Comparisons made explicit for correctness -- pylint: disable-msg=C6403
|
---|
| 1687 | if middle_whitespace != ' ' and middle_whitespace != '':
|
---|
| 1688 | error(filename, linenum, 'whitespace/todo', 2,
|
---|
| 1689 | 'TODO(my_username) should be followed by a space')
|
---|
| 1690 |
|
---|
| 1691 |
|
---|
| 1692 | def CheckSpacing(filename, clean_lines, linenum, error):
|
---|
| 1693 | """Checks for the correctness of various spacing issues in the code.
|
---|
| 1694 |
|
---|
| 1695 | Things we check for: spaces around operators, spaces after
|
---|
| 1696 | if/for/while/switch, no spaces around parens in function calls, two
|
---|
| 1697 | spaces between code and comment, don't start a block with a blank
|
---|
| 1698 | line, don't end a function with a blank line, don't add a blank line
|
---|
| 1699 | after public/protected/private, don't have too many blank lines in a row.
|
---|
| 1700 |
|
---|
| 1701 | Args:
|
---|
| 1702 | filename: The name of the current file.
|
---|
| 1703 | clean_lines: A CleansedLines instance containing the file.
|
---|
| 1704 | linenum: The number of the line to check.
|
---|
| 1705 | error: The function to call with any errors found.
|
---|
| 1706 | """
|
---|
| 1707 |
|
---|
| 1708 | raw = clean_lines.raw_lines
|
---|
| 1709 | line = raw[linenum]
|
---|
| 1710 |
|
---|
| 1711 | # Before nixing comments, check if the line is blank for no good
|
---|
| 1712 | # reason. This includes the first line after a block is opened, and
|
---|
| 1713 | # blank lines at the end of a function (ie, right before a line like '}'
|
---|
| 1714 | if IsBlankLine(line):
|
---|
| 1715 | elided = clean_lines.elided
|
---|
| 1716 | prev_line = elided[linenum - 1]
|
---|
| 1717 | prevbrace = prev_line.rfind('{')
|
---|
| 1718 | # TODO(unknown): Don't complain if line before blank line, and line after,
|
---|
| 1719 | # both start with alnums and are indented the same amount.
|
---|
| 1720 | # This ignores whitespace at the start of a namespace block
|
---|
| 1721 | # because those are not usually indented.
|
---|
| 1722 | if (prevbrace != -1 and prev_line[prevbrace:].find('}') == -1
|
---|
| 1723 | and prev_line[:prevbrace].find('namespace') == -1):
|
---|
| 1724 | # OK, we have a blank line at the start of a code block. Before we
|
---|
| 1725 | # complain, we check if it is an exception to the rule: The previous
|
---|
| 1726 | # non-empty line has the parameters of a function header that are indented
|
---|
| 1727 | # 4 spaces (because they did not fit in a 80 column line when placed on
|
---|
| 1728 | # the same line as the function name). We also check for the case where
|
---|
| 1729 | # the previous line is indented 6 spaces, which may happen when the
|
---|
| 1730 | # initializers of a constructor do not fit into a 80 column line.
|
---|
| 1731 | exception = False
|
---|
| 1732 | if Match(r' {6}\w', prev_line): # Initializer list?
|
---|
| 1733 | # We are looking for the opening column of initializer list, which
|
---|
| 1734 | # should be indented 4 spaces to cause 6 space indentation afterwards.
|
---|
| 1735 | search_position = linenum-2
|
---|
| 1736 | while (search_position >= 0
|
---|
| 1737 | and Match(r' {6}\w', elided[search_position])):
|
---|
| 1738 | search_position -= 1
|
---|
| 1739 | exception = (search_position >= 0
|
---|
| 1740 | and elided[search_position][:5] == ' :')
|
---|
| 1741 | else:
|
---|
| 1742 | # Search for the function arguments or an initializer list. We use a
|
---|
| 1743 | # simple heuristic here: If the line is indented 4 spaces; and we have a
|
---|
| 1744 | # closing paren, without the opening paren, followed by an opening brace
|
---|
| 1745 | # or colon (for initializer lists) we assume that it is the last line of
|
---|
| 1746 | # a function header. If we have a colon indented 4 spaces, it is an
|
---|
| 1747 | # initializer list.
|
---|
| 1748 | exception = (Match(r' {4}\w[^\(]*\)\s*(const\s*)?(\{\s*$|:)',
|
---|
| 1749 | prev_line)
|
---|
| 1750 | or Match(r' {4}:', prev_line))
|
---|
| 1751 |
|
---|
| 1752 | if not exception:
|
---|
| 1753 | error(filename, linenum, 'whitespace/blank_line', 2,
|
---|
| 1754 | 'Blank line at the start of a code block. Is this needed?')
|
---|
| 1755 | # This doesn't ignore whitespace at the end of a namespace block
|
---|
| 1756 | # because that is too hard without pairing open/close braces;
|
---|
| 1757 | # however, a special exception is made for namespace closing
|
---|
| 1758 | # brackets which have a comment containing "namespace".
|
---|
| 1759 | #
|
---|
| 1760 | # Also, ignore blank lines at the end of a block in a long if-else
|
---|
| 1761 | # chain, like this:
|
---|
| 1762 | # if (condition1) {
|
---|
| 1763 | # // Something followed by a blank line
|
---|
| 1764 | #
|
---|
| 1765 | # } else if (condition2) {
|
---|
| 1766 | # // Something else
|
---|
| 1767 | # }
|
---|
| 1768 | if linenum + 1 < clean_lines.NumLines():
|
---|
| 1769 | next_line = raw[linenum + 1]
|
---|
| 1770 | if (next_line
|
---|
| 1771 | and Match(r'\s*}', next_line)
|
---|
| 1772 | and next_line.find('namespace') == -1
|
---|
| 1773 | and next_line.find('} else ') == -1):
|
---|
| 1774 | error(filename, linenum, 'whitespace/blank_line', 3,
|
---|
| 1775 | 'Blank line at the end of a code block. Is this needed?')
|
---|
| 1776 |
|
---|
| 1777 | matched = Match(r'\s*(public|protected|private):', prev_line)
|
---|
| 1778 | if matched:
|
---|
| 1779 | error(filename, linenum, 'whitespace/blank_line', 3,
|
---|
| 1780 | 'Do not leave a blank line after "%s:"' % matched.group(1))
|
---|
| 1781 |
|
---|
| 1782 | # Next, we complain if there's a comment too near the text
|
---|
| 1783 | commentpos = line.find('//')
|
---|
| 1784 | if commentpos != -1:
|
---|
| 1785 | # Check if the // may be in quotes. If so, ignore it
|
---|
| 1786 | # Comparisons made explicit for clarity -- pylint: disable-msg=C6403
|
---|
| 1787 | if (line.count('"', 0, commentpos) -
|
---|
| 1788 | line.count('\\"', 0, commentpos)) % 2 == 0: # not in quotes
|
---|
| 1789 | # Allow one space for new scopes, two spaces otherwise:
|
---|
| 1790 | if (not Match(r'^\s*{ //', line) and
|
---|
| 1791 | ((commentpos >= 1 and
|
---|
| 1792 | line[commentpos-1] not in string.whitespace) or
|
---|
| 1793 | (commentpos >= 2 and
|
---|
| 1794 | line[commentpos-2] not in string.whitespace))):
|
---|
| 1795 | error(filename, linenum, 'whitespace/comments', 2,
|
---|
| 1796 | 'At least two spaces is best between code and comments')
|
---|
| 1797 | # There should always be a space between the // and the comment
|
---|
| 1798 | commentend = commentpos + 2
|
---|
| 1799 | if commentend < len(line) and not line[commentend] == ' ':
|
---|
| 1800 | # but some lines are exceptions -- e.g. if they're big
|
---|
| 1801 | # comment delimiters like:
|
---|
| 1802 | # //----------------------------------------------------------
|
---|
| 1803 | # or are an empty C++ style Doxygen comment, like:
|
---|
| 1804 | # ///
|
---|
| 1805 | # or they begin with multiple slashes followed by a space:
|
---|
| 1806 | # //////// Header comment
|
---|
| 1807 | match = (Search(r'[=/-]{4,}\s*$', line[commentend:]) or
|
---|
| 1808 | Search(r'^/$', line[commentend:]) or
|
---|
| 1809 | Search(r'^/+ ', line[commentend:]))
|
---|
| 1810 | if not match:
|
---|
| 1811 | error(filename, linenum, 'whitespace/comments', 4,
|
---|
| 1812 | 'Should have a space between // and comment')
|
---|
| 1813 | CheckComment(line[commentpos:], filename, linenum, error)
|
---|
| 1814 |
|
---|
| 1815 | line = clean_lines.elided[linenum] # get rid of comments and strings
|
---|
| 1816 |
|
---|
| 1817 | # Don't try to do spacing checks for operator methods
|
---|
| 1818 | line = re.sub(r'operator(==|!=|<|<<|<=|>=|>>|>)\(', 'operator\(', line)
|
---|
| 1819 |
|
---|
| 1820 | # We allow no-spaces around = within an if: "if ( (a=Foo()) == 0 )".
|
---|
| 1821 | # Otherwise not. Note we only check for non-spaces on *both* sides;
|
---|
| 1822 | # sometimes people put non-spaces on one side when aligning ='s among
|
---|
| 1823 | # many lines (not that this is behavior that I approve of...)
|
---|
| 1824 | if Search(r'[\w.]=[\w.]', line) and not Search(r'\b(if|while) ', line):
|
---|
| 1825 | error(filename, linenum, 'whitespace/operators', 4,
|
---|
| 1826 | 'Missing spaces around =')
|
---|
| 1827 |
|
---|
| 1828 | # It's ok not to have spaces around binary operators like + - * /, but if
|
---|
| 1829 | # there's too little whitespace, we get concerned. It's hard to tell,
|
---|
| 1830 | # though, so we punt on this one for now. TODO.
|
---|
| 1831 |
|
---|
| 1832 | # You should always have whitespace around binary operators.
|
---|
| 1833 | # Alas, we can't test < or > because they're legitimately used sans spaces
|
---|
| 1834 | # (a->b, vector<int> a). The only time we can tell is a < with no >, and
|
---|
| 1835 | # only if it's not template params list spilling into the next line.
|
---|
| 1836 | match = Search(r'[^<>=!\s](==|!=|<=|>=)[^<>=!\s]', line)
|
---|
| 1837 | if not match:
|
---|
| 1838 | # Note that while it seems that the '<[^<]*' term in the following
|
---|
| 1839 | # regexp could be simplified to '<.*', which would indeed match
|
---|
| 1840 | # the same class of strings, the [^<] means that searching for the
|
---|
| 1841 | # regexp takes linear rather than quadratic time.
|
---|
| 1842 | if not Search(r'<[^<]*,\s*$', line): # template params spill
|
---|
| 1843 | match = Search(r'[^<>=!\s](<)[^<>=!\s]([^>]|->)*$', line)
|
---|
| 1844 | if match:
|
---|
| 1845 | error(filename, linenum, 'whitespace/operators', 3,
|
---|
| 1846 | 'Missing spaces around %s' % match.group(1))
|
---|
| 1847 | # We allow no-spaces around << and >> when used like this: 10<<20, but
|
---|
| 1848 | # not otherwise (particularly, not when used as streams)
|
---|
| 1849 | match = Search(r'[^0-9\s](<<|>>)[^0-9\s]', line)
|
---|
| 1850 | if match:
|
---|
| 1851 | error(filename, linenum, 'whitespace/operators', 3,
|
---|
| 1852 | 'Missing spaces around %s' % match.group(1))
|
---|
| 1853 |
|
---|
| 1854 | # There shouldn't be space around unary operators
|
---|
| 1855 | match = Search(r'(!\s|~\s|[\s]--[\s;]|[\s]\+\+[\s;])', line)
|
---|
| 1856 | if match:
|
---|
| 1857 | error(filename, linenum, 'whitespace/operators', 4,
|
---|
| 1858 | 'Extra space for operator %s' % match.group(1))
|
---|
| 1859 |
|
---|
| 1860 | # A pet peeve of mine: no spaces after an if, while, switch, or for
|
---|
| 1861 | match = Search(r' (if\(|for\(|while\(|switch\()', line)
|
---|
| 1862 | if match:
|
---|
| 1863 | error(filename, linenum, 'whitespace/parens', 5,
|
---|
| 1864 | 'Missing space before ( in %s' % match.group(1))
|
---|
| 1865 |
|
---|
| 1866 | # For if/for/while/switch, the left and right parens should be
|
---|
| 1867 | # consistent about how many spaces are inside the parens, and
|
---|
| 1868 | # there should either be zero or one spaces inside the parens.
|
---|
| 1869 | # We don't want: "if ( foo)" or "if ( foo )".
|
---|
| 1870 | # Exception: "for ( ; foo; bar)" and "for (foo; bar; )" are allowed.
|
---|
| 1871 | match = Search(r'\b(if|for|while|switch)\s*'
|
---|
| 1872 | r'\(([ ]*)(.).*[^ ]+([ ]*)\)\s*{\s*$',
|
---|
| 1873 | line)
|
---|
| 1874 | if match:
|
---|
| 1875 | if len(match.group(2)) != len(match.group(4)):
|
---|
| 1876 | if not (match.group(3) == ';' and
|
---|
| 1877 | len(match.group(2)) == 1 + len(match.group(4)) or
|
---|
| 1878 | not match.group(2) and Search(r'\bfor\s*\(.*; \)', line)):
|
---|
| 1879 | error(filename, linenum, 'whitespace/parens', 5,
|
---|
| 1880 | 'Mismatching spaces inside () in %s' % match.group(1))
|
---|
| 1881 | if not len(match.group(2)) in [0, 1]:
|
---|
| 1882 | error(filename, linenum, 'whitespace/parens', 5,
|
---|
| 1883 | 'Should have zero or one spaces inside ( and ) in %s' %
|
---|
| 1884 | match.group(1))
|
---|
| 1885 |
|
---|
| 1886 | # You should always have a space after a comma (either as fn arg or operator)
|
---|
| 1887 | if Search(r',[^\s]', line):
|
---|
| 1888 | error(filename, linenum, 'whitespace/comma', 3,
|
---|
| 1889 | 'Missing space after ,')
|
---|
| 1890 |
|
---|
| 1891 | # You should always have a space after a semicolon
|
---|
| 1892 | # except for few corner cases
|
---|
| 1893 | # TODO(unknown): clarify if 'if (1) { return 1;}' is requires one more
|
---|
| 1894 | # space after ;
|
---|
| 1895 | if Search(r';[^\s};\\)/]', line):
|
---|
| 1896 | error(filename, linenum, 'whitespace/semicolon', 3,
|
---|
| 1897 | 'Missing space after ;')
|
---|
| 1898 |
|
---|
| 1899 | # Next we will look for issues with function calls.
|
---|
| 1900 | CheckSpacingForFunctionCall(filename, line, linenum, error)
|
---|
| 1901 |
|
---|
| 1902 | # Except after an opening paren, or after another opening brace (in case of
|
---|
| 1903 | # an initializer list, for instance), you should have spaces before your
|
---|
| 1904 | # braces. And since you should never have braces at the beginning of a line,
|
---|
| 1905 | # this is an easy test.
|
---|
| 1906 | if Search(r'[^ ({]{', line):
|
---|
| 1907 | error(filename, linenum, 'whitespace/braces', 5,
|
---|
| 1908 | 'Missing space before {')
|
---|
| 1909 |
|
---|
| 1910 | # Make sure '} else {' has spaces.
|
---|
| 1911 | if Search(r'}else', line):
|
---|
| 1912 | error(filename, linenum, 'whitespace/braces', 5,
|
---|
| 1913 | 'Missing space before else')
|
---|
| 1914 |
|
---|
| 1915 | # You shouldn't have spaces before your brackets, except maybe after
|
---|
| 1916 | # 'delete []' or 'new char * []'.
|
---|
| 1917 | if Search(r'\w\s+\[', line) and not Search(r'delete\s+\[', line):
|
---|
| 1918 | error(filename, linenum, 'whitespace/braces', 5,
|
---|
| 1919 | 'Extra space before [')
|
---|
| 1920 |
|
---|
| 1921 | # You shouldn't have a space before a semicolon at the end of the line.
|
---|
| 1922 | # There's a special case for "for" since the style guide allows space before
|
---|
| 1923 | # the semicolon there.
|
---|
| 1924 | if Search(r':\s*;\s*$', line):
|
---|
| 1925 | error(filename, linenum, 'whitespace/semicolon', 5,
|
---|
| 1926 | 'Semicolon defining empty statement. Use { } instead.')
|
---|
| 1927 | elif Search(r'^\s*;\s*$', line):
|
---|
| 1928 | error(filename, linenum, 'whitespace/semicolon', 5,
|
---|
| 1929 | 'Line contains only semicolon. If this should be an empty statement, '
|
---|
| 1930 | 'use { } instead.')
|
---|
| 1931 | elif (Search(r'\s+;\s*$', line) and
|
---|
| 1932 | not Search(r'\bfor\b', line)):
|
---|
| 1933 | error(filename, linenum, 'whitespace/semicolon', 5,
|
---|
| 1934 | 'Extra space before last semicolon. If this should be an empty '
|
---|
| 1935 | 'statement, use { } instead.')
|
---|
| 1936 |
|
---|
| 1937 |
|
---|
| 1938 | def CheckSectionSpacing(filename, clean_lines, class_info, linenum, error):
|
---|
| 1939 | """Checks for additional blank line issues related to sections.
|
---|
| 1940 |
|
---|
| 1941 | Currently the only thing checked here is blank line before protected/private.
|
---|
| 1942 |
|
---|
| 1943 | Args:
|
---|
| 1944 | filename: The name of the current file.
|
---|
| 1945 | clean_lines: A CleansedLines instance containing the file.
|
---|
| 1946 | class_info: A _ClassInfo objects.
|
---|
| 1947 | linenum: The number of the line to check.
|
---|
| 1948 | error: The function to call with any errors found.
|
---|
| 1949 | """
|
---|
| 1950 | # Skip checks if the class is small, where small means 25 lines or less.
|
---|
| 1951 | # 25 lines seems like a good cutoff since that's the usual height of
|
---|
| 1952 | # terminals, and any class that can't fit in one screen can't really
|
---|
| 1953 | # be considered "small".
|
---|
| 1954 | #
|
---|
| 1955 | # Also skip checks if we are on the first line. This accounts for
|
---|
| 1956 | # classes that look like
|
---|
| 1957 | # class Foo { public: ... };
|
---|
| 1958 | #
|
---|
| 1959 | # If we didn't find the end of the class, last_line would be zero,
|
---|
| 1960 | # and the check will be skipped by the first condition.
|
---|
| 1961 | if (class_info.last_line - class_info.linenum <= 24 or
|
---|
| 1962 | linenum <= class_info.linenum):
|
---|
| 1963 | return
|
---|
| 1964 |
|
---|
| 1965 | matched = Match(r'\s*(public|protected|private):', clean_lines.lines[linenum])
|
---|
| 1966 | if matched:
|
---|
| 1967 | # Issue warning if the line before public/protected/private was
|
---|
| 1968 | # not a blank line, but don't do this if the previous line contains
|
---|
| 1969 | # "class" or "struct". This can happen two ways:
|
---|
| 1970 | # - We are at the beginning of the class.
|
---|
| 1971 | # - We are forward-declaring an inner class that is semantically
|
---|
| 1972 | # private, but needed to be public for implementation reasons.
|
---|
| 1973 | prev_line = clean_lines.lines[linenum - 1]
|
---|
| 1974 | if (not IsBlankLine(prev_line) and
|
---|
| 1975 | not Search(r'\b(class|struct)\b', prev_line)):
|
---|
| 1976 | # Try a bit harder to find the beginning of the class. This is to
|
---|
| 1977 | # account for multi-line base-specifier lists, e.g.:
|
---|
| 1978 | # class Derived
|
---|
| 1979 | # : public Base {
|
---|
| 1980 | end_class_head = class_info.linenum
|
---|
| 1981 | for i in range(class_info.linenum, linenum):
|
---|
| 1982 | if Search(r'\{\s*$', clean_lines.lines[i]):
|
---|
| 1983 | end_class_head = i
|
---|
| 1984 | break
|
---|
| 1985 | if end_class_head < linenum - 1:
|
---|
| 1986 | error(filename, linenum, 'whitespace/blank_line', 3,
|
---|
| 1987 | '"%s:" should be preceded by a blank line' % matched.group(1))
|
---|
| 1988 |
|
---|
| 1989 |
|
---|
| 1990 | def GetPreviousNonBlankLine(clean_lines, linenum):
|
---|
| 1991 | """Return the most recent non-blank line and its line number.
|
---|
| 1992 |
|
---|
| 1993 | Args:
|
---|
| 1994 | clean_lines: A CleansedLines instance containing the file contents.
|
---|
| 1995 | linenum: The number of the line to check.
|
---|
| 1996 |
|
---|
| 1997 | Returns:
|
---|
| 1998 | A tuple with two elements. The first element is the contents of the last
|
---|
| 1999 | non-blank line before the current line, or the empty string if this is the
|
---|
| 2000 | first non-blank line. The second is the line number of that line, or -1
|
---|
| 2001 | if this is the first non-blank line.
|
---|
| 2002 | """
|
---|
| 2003 |
|
---|
| 2004 | prevlinenum = linenum - 1
|
---|
| 2005 | while prevlinenum >= 0:
|
---|
| 2006 | prevline = clean_lines.elided[prevlinenum]
|
---|
| 2007 | if not IsBlankLine(prevline): # if not a blank line...
|
---|
| 2008 | return (prevline, prevlinenum)
|
---|
| 2009 | prevlinenum -= 1
|
---|
| 2010 | return ('', -1)
|
---|
| 2011 |
|
---|
| 2012 |
|
---|
| 2013 | def CheckBraces(filename, clean_lines, linenum, error):
|
---|
| 2014 | """Looks for misplaced braces (e.g. at the end of line).
|
---|
| 2015 |
|
---|
| 2016 | Args:
|
---|
| 2017 | filename: The name of the current file.
|
---|
| 2018 | clean_lines: A CleansedLines instance containing the file.
|
---|
| 2019 | linenum: The number of the line to check.
|
---|
| 2020 | error: The function to call with any errors found.
|
---|
| 2021 | """
|
---|
| 2022 |
|
---|
| 2023 | line = clean_lines.elided[linenum] # get rid of comments and strings
|
---|
| 2024 |
|
---|
| 2025 | if Match(r'\s*{\s*$', line):
|
---|
| 2026 | # We allow an open brace to start a line in the case where someone
|
---|
| 2027 | # is using braces in a block to explicitly create a new scope,
|
---|
| 2028 | # which is commonly used to control the lifetime of
|
---|
| 2029 | # stack-allocated variables. We don't detect this perfectly: we
|
---|
| 2030 | # just don't complain if the last non-whitespace character on the
|
---|
| 2031 | # previous non-blank line is ';', ':', '{', or '}'.
|
---|
| 2032 | prevline = GetPreviousNonBlankLine(clean_lines, linenum)[0]
|
---|
| 2033 | if not Search(r'[;:}{]\s*$', prevline):
|
---|
| 2034 | error(filename, linenum, 'whitespace/braces', 4,
|
---|
| 2035 | '{ should almost always be at the end of the previous line')
|
---|
| 2036 |
|
---|
| 2037 | # An else clause should be on the same line as the preceding closing brace.
|
---|
| 2038 | if Match(r'\s*else\s*', line):
|
---|
| 2039 | prevline = GetPreviousNonBlankLine(clean_lines, linenum)[0]
|
---|
| 2040 | if Match(r'\s*}\s*$', prevline):
|
---|
| 2041 | error(filename, linenum, 'whitespace/newline', 4,
|
---|
| 2042 | 'An else should appear on the same line as the preceding }')
|
---|
| 2043 |
|
---|
| 2044 | # If braces come on one side of an else, they should be on both.
|
---|
| 2045 | # However, we have to worry about "else if" that spans multiple lines!
|
---|
| 2046 | if Search(r'}\s*else[^{]*$', line) or Match(r'[^}]*else\s*{', line):
|
---|
| 2047 | if Search(r'}\s*else if([^{]*)$', line): # could be multi-line if
|
---|
| 2048 | # find the ( after the if
|
---|
| 2049 | pos = line.find('else if')
|
---|
| 2050 | pos = line.find('(', pos)
|
---|
| 2051 | if pos > 0:
|
---|
| 2052 | (endline, _, endpos) = CloseExpression(clean_lines, linenum, pos)
|
---|
| 2053 | if endline[endpos:].find('{') == -1: # must be brace after if
|
---|
| 2054 | error(filename, linenum, 'readability/braces', 5,
|
---|
| 2055 | 'If an else has a brace on one side, it should have it on both')
|
---|
| 2056 | else: # common case: else not followed by a multi-line if
|
---|
| 2057 | error(filename, linenum, 'readability/braces', 5,
|
---|
| 2058 | 'If an else has a brace on one side, it should have it on both')
|
---|
| 2059 |
|
---|
| 2060 | # Likewise, an else should never have the else clause on the same line
|
---|
| 2061 | if Search(r'\belse [^\s{]', line) and not Search(r'\belse if\b', line):
|
---|
| 2062 | error(filename, linenum, 'whitespace/newline', 4,
|
---|
| 2063 | 'Else clause should never be on same line as else (use 2 lines)')
|
---|
| 2064 |
|
---|
| 2065 | # In the same way, a do/while should never be on one line
|
---|
| 2066 | if Match(r'\s*do [^\s{]', line):
|
---|
| 2067 | error(filename, linenum, 'whitespace/newline', 4,
|
---|
| 2068 | 'do/while clauses should not be on a single line')
|
---|
| 2069 |
|
---|
| 2070 | # Braces shouldn't be followed by a ; unless they're defining a struct
|
---|
| 2071 | # or initializing an array.
|
---|
| 2072 | # We can't tell in general, but we can for some common cases.
|
---|
| 2073 | prevlinenum = linenum
|
---|
| 2074 | while True:
|
---|
| 2075 | (prevline, prevlinenum) = GetPreviousNonBlankLine(clean_lines, prevlinenum)
|
---|
| 2076 | if Match(r'\s+{.*}\s*;', line) and not prevline.count(';'):
|
---|
| 2077 | line = prevline + line
|
---|
| 2078 | else:
|
---|
| 2079 | break
|
---|
| 2080 | if (Search(r'{.*}\s*;', line) and
|
---|
| 2081 | line.count('{') == line.count('}') and
|
---|
| 2082 | not Search(r'struct|class|enum|\s*=\s*{', line)):
|
---|
| 2083 | error(filename, linenum, 'readability/braces', 4,
|
---|
| 2084 | "You don't need a ; after a }")
|
---|
| 2085 |
|
---|
| 2086 |
|
---|
| 2087 | def ReplaceableCheck(operator, macro, line):
|
---|
| 2088 | """Determine whether a basic CHECK can be replaced with a more specific one.
|
---|
| 2089 |
|
---|
| 2090 | For example suggest using CHECK_EQ instead of CHECK(a == b) and
|
---|
| 2091 | similarly for CHECK_GE, CHECK_GT, CHECK_LE, CHECK_LT, CHECK_NE.
|
---|
| 2092 |
|
---|
| 2093 | Args:
|
---|
| 2094 | operator: The C++ operator used in the CHECK.
|
---|
| 2095 | macro: The CHECK or EXPECT macro being called.
|
---|
| 2096 | line: The current source line.
|
---|
| 2097 |
|
---|
| 2098 | Returns:
|
---|
| 2099 | True if the CHECK can be replaced with a more specific one.
|
---|
| 2100 | """
|
---|
| 2101 |
|
---|
| 2102 | # This matches decimal and hex integers, strings, and chars (in that order).
|
---|
| 2103 | match_constant = r'([-+]?(\d+|0[xX][0-9a-fA-F]+)[lLuU]{0,3}|".*"|\'.*\')'
|
---|
| 2104 |
|
---|
| 2105 | # Expression to match two sides of the operator with something that
|
---|
| 2106 | # looks like a literal, since CHECK(x == iterator) won't compile.
|
---|
| 2107 | # This means we can't catch all the cases where a more specific
|
---|
| 2108 | # CHECK is possible, but it's less annoying than dealing with
|
---|
| 2109 | # extraneous warnings.
|
---|
| 2110 | match_this = (r'\s*' + macro + r'\((\s*' +
|
---|
| 2111 | match_constant + r'\s*' + operator + r'[^<>].*|'
|
---|
| 2112 | r'.*[^<>]' + operator + r'\s*' + match_constant +
|
---|
| 2113 | r'\s*\))')
|
---|
| 2114 |
|
---|
| 2115 | # Don't complain about CHECK(x == NULL) or similar because
|
---|
| 2116 | # CHECK_EQ(x, NULL) won't compile (requires a cast).
|
---|
| 2117 | # Also, don't complain about more complex boolean expressions
|
---|
| 2118 | # involving && or || such as CHECK(a == b || c == d).
|
---|
| 2119 | return Match(match_this, line) and not Search(r'NULL|&&|\|\|', line)
|
---|
| 2120 |
|
---|
| 2121 |
|
---|
| 2122 | def CheckCheck(filename, clean_lines, linenum, error):
|
---|
| 2123 | """Checks the use of CHECK and EXPECT macros.
|
---|
| 2124 |
|
---|
| 2125 | Args:
|
---|
| 2126 | filename: The name of the current file.
|
---|
| 2127 | clean_lines: A CleansedLines instance containing the file.
|
---|
| 2128 | linenum: The number of the line to check.
|
---|
| 2129 | error: The function to call with any errors found.
|
---|
| 2130 | """
|
---|
| 2131 |
|
---|
| 2132 | # Decide the set of replacement macros that should be suggested
|
---|
| 2133 | raw_lines = clean_lines.raw_lines
|
---|
| 2134 | current_macro = ''
|
---|
| 2135 | for macro in _CHECK_MACROS:
|
---|
| 2136 | if raw_lines[linenum].find(macro) >= 0:
|
---|
| 2137 | current_macro = macro
|
---|
| 2138 | break
|
---|
| 2139 | if not current_macro:
|
---|
| 2140 | # Don't waste time here if line doesn't contain 'CHECK' or 'EXPECT'
|
---|
| 2141 | return
|
---|
| 2142 |
|
---|
| 2143 | line = clean_lines.elided[linenum] # get rid of comments and strings
|
---|
| 2144 |
|
---|
| 2145 | # Encourage replacing plain CHECKs with CHECK_EQ/CHECK_NE/etc.
|
---|
| 2146 | for operator in ['==', '!=', '>=', '>', '<=', '<']:
|
---|
| 2147 | if ReplaceableCheck(operator, current_macro, line):
|
---|
| 2148 | error(filename, linenum, 'readability/check', 2,
|
---|
| 2149 | 'Consider using %s instead of %s(a %s b)' % (
|
---|
| 2150 | _CHECK_REPLACEMENT[current_macro][operator],
|
---|
| 2151 | current_macro, operator))
|
---|
| 2152 | break
|
---|
| 2153 |
|
---|
| 2154 |
|
---|
| 2155 | def GetLineWidth(line):
|
---|
| 2156 | """Determines the width of the line in column positions.
|
---|
| 2157 |
|
---|
| 2158 | Args:
|
---|
| 2159 | line: A string, which may be a Unicode string.
|
---|
| 2160 |
|
---|
| 2161 | Returns:
|
---|
| 2162 | The width of the line in column positions, accounting for Unicode
|
---|
| 2163 | combining characters and wide characters.
|
---|
| 2164 | """
|
---|
| 2165 | if isinstance(line, TEXT_TYPE):
|
---|
| 2166 | width = 0
|
---|
| 2167 | for uc in unicodedata.normalize('NFC', line):
|
---|
| 2168 | if unicodedata.east_asian_width(uc) in ('W', 'F'):
|
---|
| 2169 | width += 2
|
---|
| 2170 | elif not unicodedata.combining(uc):
|
---|
| 2171 | width += 1
|
---|
| 2172 | return width
|
---|
| 2173 | else:
|
---|
| 2174 | return len(line)
|
---|
| 2175 |
|
---|
| 2176 |
|
---|
| 2177 | def CheckStyle(filename, clean_lines, linenum, file_extension, class_state,
|
---|
| 2178 | error):
|
---|
| 2179 | """Checks rules from the 'C++ style rules' section of cppguide.html.
|
---|
| 2180 |
|
---|
| 2181 | Most of these rules are hard to test (naming, comment style), but we
|
---|
| 2182 | do what we can. In particular we check for 2-space indents, line lengths,
|
---|
| 2183 | tab usage, spaces inside code, etc.
|
---|
| 2184 |
|
---|
| 2185 | Args:
|
---|
| 2186 | filename: The name of the current file.
|
---|
| 2187 | clean_lines: A CleansedLines instance containing the file.
|
---|
| 2188 | linenum: The number of the line to check.
|
---|
| 2189 | file_extension: The extension (without the dot) of the filename.
|
---|
| 2190 | error: The function to call with any errors found.
|
---|
| 2191 | """
|
---|
| 2192 |
|
---|
| 2193 | raw_lines = clean_lines.raw_lines
|
---|
| 2194 | line = raw_lines[linenum]
|
---|
| 2195 |
|
---|
| 2196 | if line.find('\t') != -1:
|
---|
| 2197 | error(filename, linenum, 'whitespace/tab', 1,
|
---|
| 2198 | 'Tab found; better to use spaces')
|
---|
| 2199 |
|
---|
| 2200 | # One or three blank spaces at the beginning of the line is weird; it's
|
---|
| 2201 | # hard to reconcile that with 2-space indents.
|
---|
| 2202 | # NOTE: here are the conditions rob pike used for his tests. Mine aren't
|
---|
| 2203 | # as sophisticated, but it may be worth becoming so: RLENGTH==initial_spaces
|
---|
| 2204 | # if(RLENGTH > 20) complain = 0;
|
---|
| 2205 | # if(match($0, " +(error|private|public|protected):")) complain = 0;
|
---|
| 2206 | # if(match(prev, "&& *$")) complain = 0;
|
---|
| 2207 | # if(match(prev, "\\|\\| *$")) complain = 0;
|
---|
| 2208 | # if(match(prev, "[\",=><] *$")) complain = 0;
|
---|
| 2209 | # if(match($0, " <<")) complain = 0;
|
---|
| 2210 | # if(match(prev, " +for \\(")) complain = 0;
|
---|
| 2211 | # if(prevodd && match(prevprev, " +for \\(")) complain = 0;
|
---|
| 2212 | initial_spaces = 0
|
---|
| 2213 | cleansed_line = clean_lines.elided[linenum]
|
---|
| 2214 | while initial_spaces < len(line) and line[initial_spaces] == ' ':
|
---|
| 2215 | initial_spaces += 1
|
---|
| 2216 | if line and line[-1].isspace():
|
---|
| 2217 | error(filename, linenum, 'whitespace/end_of_line', 4,
|
---|
| 2218 | 'Line ends in whitespace. Consider deleting these extra spaces.')
|
---|
| 2219 | # There are certain situations we allow one space, notably for labels
|
---|
| 2220 | elif ((initial_spaces == 1 or initial_spaces == 3) and
|
---|
| 2221 | not Match(r'\s*\w+\s*:\s*$', cleansed_line)):
|
---|
| 2222 | error(filename, linenum, 'whitespace/indent', 3,
|
---|
| 2223 | 'Weird number of spaces at line-start. '
|
---|
| 2224 | 'Are you using a 2-space indent?')
|
---|
| 2225 | # Labels should always be indented at least one space.
|
---|
| 2226 | elif not initial_spaces and line[:2] != '//' and Search(r'[^:]:\s*$',
|
---|
| 2227 | line):
|
---|
| 2228 | error(filename, linenum, 'whitespace/labels', 4,
|
---|
| 2229 | 'Labels should always be indented at least one space. '
|
---|
| 2230 | 'If this is a member-initializer list in a constructor or '
|
---|
| 2231 | 'the base class list in a class definition, the colon should '
|
---|
| 2232 | 'be on the following line.')
|
---|
| 2233 |
|
---|
| 2234 |
|
---|
| 2235 | # Check if the line is a header guard.
|
---|
| 2236 | is_header_guard = False
|
---|
| 2237 | if file_extension == 'h':
|
---|
| 2238 | cppvar = GetHeaderGuardCPPVariable(filename)
|
---|
| 2239 | if (line.startswith('#ifndef %s' % cppvar) or
|
---|
| 2240 | line.startswith('#define %s' % cppvar) or
|
---|
| 2241 | line.startswith('#endif // %s' % cppvar)):
|
---|
| 2242 | is_header_guard = True
|
---|
| 2243 | # #include lines and header guards can be long, since there's no clean way to
|
---|
| 2244 | # split them.
|
---|
| 2245 | #
|
---|
| 2246 | # URLs can be long too. It's possible to split these, but it makes them
|
---|
| 2247 | # harder to cut&paste.
|
---|
| 2248 | #
|
---|
| 2249 | # The "$Id:...$" comment may also get very long without it being the
|
---|
| 2250 | # developers fault.
|
---|
| 2251 | if (not line.startswith('#include') and not is_header_guard and
|
---|
| 2252 | not Match(r'^\s*//.*http(s?)://\S*$', line) and
|
---|
| 2253 | not Match(r'^// \$Id:.*#[0-9]+ \$$', line)):
|
---|
| 2254 | line_width = GetLineWidth(line)
|
---|
| 2255 | if line_width > 100:
|
---|
| 2256 | error(filename, linenum, 'whitespace/line_length', 4,
|
---|
| 2257 | 'Lines should very rarely be longer than 100 characters')
|
---|
| 2258 | elif line_width > 80:
|
---|
| 2259 | error(filename, linenum, 'whitespace/line_length', 2,
|
---|
| 2260 | 'Lines should be <= 80 characters long')
|
---|
| 2261 |
|
---|
| 2262 | if (cleansed_line.count(';') > 1 and
|
---|
| 2263 | # for loops are allowed two ;'s (and may run over two lines).
|
---|
| 2264 | cleansed_line.find('for') == -1 and
|
---|
| 2265 | (GetPreviousNonBlankLine(clean_lines, linenum)[0].find('for') == -1 or
|
---|
| 2266 | GetPreviousNonBlankLine(clean_lines, linenum)[0].find(';') != -1) and
|
---|
| 2267 | # It's ok to have many commands in a switch case that fits in 1 line
|
---|
| 2268 | not ((cleansed_line.find('case ') != -1 or
|
---|
| 2269 | cleansed_line.find('default:') != -1) and
|
---|
| 2270 | cleansed_line.find('break;') != -1)):
|
---|
| 2271 | error(filename, linenum, 'whitespace/newline', 4,
|
---|
| 2272 | 'More than one command on the same line')
|
---|
| 2273 |
|
---|
| 2274 | # Some more style checks
|
---|
| 2275 | CheckBraces(filename, clean_lines, linenum, error)
|
---|
| 2276 | CheckSpacing(filename, clean_lines, linenum, error)
|
---|
| 2277 | CheckCheck(filename, clean_lines, linenum, error)
|
---|
| 2278 | if class_state and class_state.classinfo_stack:
|
---|
| 2279 | CheckSectionSpacing(filename, clean_lines,
|
---|
| 2280 | class_state.classinfo_stack[-1], linenum, error)
|
---|
| 2281 |
|
---|
| 2282 |
|
---|
| 2283 | _RE_PATTERN_INCLUDE_NEW_STYLE = re.compile(r'#include +"[^/]+\.h"')
|
---|
| 2284 | _RE_PATTERN_INCLUDE = re.compile(r'^\s*#\s*include\s*([<"])([^>"]*)[>"].*$')
|
---|
| 2285 | # Matches the first component of a filename delimited by -s and _s. That is:
|
---|
| 2286 | # _RE_FIRST_COMPONENT.match('foo').group(0) == 'foo'
|
---|
| 2287 | # _RE_FIRST_COMPONENT.match('foo.cc').group(0) == 'foo'
|
---|
| 2288 | # _RE_FIRST_COMPONENT.match('foo-bar_baz.cc').group(0) == 'foo'
|
---|
| 2289 | # _RE_FIRST_COMPONENT.match('foo_bar-baz.cc').group(0) == 'foo'
|
---|
| 2290 | _RE_FIRST_COMPONENT = re.compile(r'^[^-_.]+')
|
---|
| 2291 |
|
---|
| 2292 |
|
---|
| 2293 | def _DropCommonSuffixes(filename):
|
---|
| 2294 | """Drops common suffixes like _test.cc or -inl.h from filename.
|
---|
| 2295 |
|
---|
| 2296 | For example:
|
---|
| 2297 | >>> _DropCommonSuffixes('foo/foo-inl.h')
|
---|
| 2298 | 'foo/foo'
|
---|
| 2299 | >>> _DropCommonSuffixes('foo/bar/foo.cc')
|
---|
| 2300 | 'foo/bar/foo'
|
---|
| 2301 | >>> _DropCommonSuffixes('foo/foo_internal.h')
|
---|
| 2302 | 'foo/foo'
|
---|
| 2303 | >>> _DropCommonSuffixes('foo/foo_unusualinternal.h')
|
---|
| 2304 | 'foo/foo_unusualinternal'
|
---|
| 2305 |
|
---|
| 2306 | Args:
|
---|
| 2307 | filename: The input filename.
|
---|
| 2308 |
|
---|
| 2309 | Returns:
|
---|
| 2310 | The filename with the common suffix removed.
|
---|
| 2311 | """
|
---|
| 2312 | for suffix in ('test.cc', 'regtest.cc', 'unittest.cc',
|
---|
| 2313 | 'inl.h', 'impl.h', 'internal.h'):
|
---|
| 2314 | if (filename.endswith(suffix) and len(filename) > len(suffix) and
|
---|
| 2315 | filename[-len(suffix) - 1] in ('-', '_')):
|
---|
| 2316 | return filename[:-len(suffix) - 1]
|
---|
| 2317 | return os.path.splitext(filename)[0]
|
---|
| 2318 |
|
---|
| 2319 |
|
---|
| 2320 | def _IsTestFilename(filename):
|
---|
| 2321 | """Determines if the given filename has a suffix that identifies it as a test.
|
---|
| 2322 |
|
---|
| 2323 | Args:
|
---|
| 2324 | filename: The input filename.
|
---|
| 2325 |
|
---|
| 2326 | Returns:
|
---|
| 2327 | True if 'filename' looks like a test, False otherwise.
|
---|
| 2328 | """
|
---|
| 2329 | if (filename.endswith('_test.cc') or
|
---|
| 2330 | filename.endswith('_unittest.cc') or
|
---|
| 2331 | filename.endswith('_regtest.cc')):
|
---|
| 2332 | return True
|
---|
| 2333 | else:
|
---|
| 2334 | return False
|
---|
| 2335 |
|
---|
| 2336 |
|
---|
| 2337 | def _ClassifyInclude(fileinfo, include, is_system):
|
---|
| 2338 | """Figures out what kind of header 'include' is.
|
---|
| 2339 |
|
---|
| 2340 | Args:
|
---|
| 2341 | fileinfo: The current file cpplint is running over. A FileInfo instance.
|
---|
| 2342 | include: The path to a #included file.
|
---|
| 2343 | is_system: True if the #include used <> rather than "".
|
---|
| 2344 |
|
---|
| 2345 | Returns:
|
---|
| 2346 | One of the _XXX_HEADER constants.
|
---|
| 2347 |
|
---|
| 2348 | For example:
|
---|
| 2349 | >>> _ClassifyInclude(FileInfo('foo/foo.cc'), 'stdio.h', True)
|
---|
| 2350 | _C_SYS_HEADER
|
---|
| 2351 | >>> _ClassifyInclude(FileInfo('foo/foo.cc'), 'string', True)
|
---|
| 2352 | _CPP_SYS_HEADER
|
---|
| 2353 | >>> _ClassifyInclude(FileInfo('foo/foo.cc'), 'foo/foo.h', False)
|
---|
| 2354 | _LIKELY_MY_HEADER
|
---|
| 2355 | >>> _ClassifyInclude(FileInfo('foo/foo_unknown_extension.cc'),
|
---|
| 2356 | ... 'bar/foo_other_ext.h', False)
|
---|
| 2357 | _POSSIBLE_MY_HEADER
|
---|
| 2358 | >>> _ClassifyInclude(FileInfo('foo/foo.cc'), 'foo/bar.h', False)
|
---|
| 2359 | _OTHER_HEADER
|
---|
| 2360 | """
|
---|
| 2361 | # This is a list of all standard c++ header files, except
|
---|
| 2362 | # those already checked for above.
|
---|
| 2363 | is_stl_h = include in _STL_HEADERS
|
---|
| 2364 | is_cpp_h = is_stl_h or include in _CPP_HEADERS
|
---|
| 2365 |
|
---|
| 2366 | if is_system:
|
---|
| 2367 | if is_cpp_h:
|
---|
| 2368 | return _CPP_SYS_HEADER
|
---|
| 2369 | else:
|
---|
| 2370 | return _C_SYS_HEADER
|
---|
| 2371 |
|
---|
| 2372 | # If the target file and the include we're checking share a
|
---|
| 2373 | # basename when we drop common extensions, and the include
|
---|
| 2374 | # lives in . , then it's likely to be owned by the target file.
|
---|
| 2375 | target_dir, target_base = (
|
---|
| 2376 | os.path.split(_DropCommonSuffixes(fileinfo.RepositoryName())))
|
---|
| 2377 | include_dir, include_base = os.path.split(_DropCommonSuffixes(include))
|
---|
| 2378 | if target_base == include_base and (
|
---|
| 2379 | include_dir == target_dir or
|
---|
| 2380 | include_dir == os.path.normpath(target_dir + '/../public')):
|
---|
| 2381 | return _LIKELY_MY_HEADER
|
---|
| 2382 |
|
---|
| 2383 | # If the target and include share some initial basename
|
---|
| 2384 | # component, it's possible the target is implementing the
|
---|
| 2385 | # include, so it's allowed to be first, but we'll never
|
---|
| 2386 | # complain if it's not there.
|
---|
| 2387 | target_first_component = _RE_FIRST_COMPONENT.match(target_base)
|
---|
| 2388 | include_first_component = _RE_FIRST_COMPONENT.match(include_base)
|
---|
| 2389 | if (target_first_component and include_first_component and
|
---|
| 2390 | target_first_component.group(0) ==
|
---|
| 2391 | include_first_component.group(0)):
|
---|
| 2392 | return _POSSIBLE_MY_HEADER
|
---|
| 2393 |
|
---|
| 2394 | return _OTHER_HEADER
|
---|
| 2395 |
|
---|
| 2396 |
|
---|
| 2397 |
|
---|
| 2398 | def CheckIncludeLine(filename, clean_lines, linenum, include_state, error):
|
---|
| 2399 | """Check rules that are applicable to #include lines.
|
---|
| 2400 |
|
---|
| 2401 | Strings on #include lines are NOT removed from elided line, to make
|
---|
| 2402 | certain tasks easier. However, to prevent false positives, checks
|
---|
| 2403 | applicable to #include lines in CheckLanguage must be put here.
|
---|
| 2404 |
|
---|
| 2405 | Args:
|
---|
| 2406 | filename: The name of the current file.
|
---|
| 2407 | clean_lines: A CleansedLines instance containing the file.
|
---|
| 2408 | linenum: The number of the line to check.
|
---|
| 2409 | include_state: An _IncludeState instance in which the headers are inserted.
|
---|
| 2410 | error: The function to call with any errors found.
|
---|
| 2411 | """
|
---|
| 2412 | fileinfo = FileInfo(filename)
|
---|
| 2413 |
|
---|
| 2414 | line = clean_lines.lines[linenum]
|
---|
| 2415 |
|
---|
| 2416 | # "include" should use the new style "foo/bar.h" instead of just "bar.h"
|
---|
| 2417 | if _RE_PATTERN_INCLUDE_NEW_STYLE.search(line):
|
---|
| 2418 | error(filename, linenum, 'build/include', 4,
|
---|
| 2419 | 'Include the directory when naming .h files')
|
---|
| 2420 |
|
---|
| 2421 | # we shouldn't include a file more than once. actually, there are a
|
---|
| 2422 | # handful of instances where doing so is okay, but in general it's
|
---|
| 2423 | # not.
|
---|
| 2424 | match = _RE_PATTERN_INCLUDE.search(line)
|
---|
| 2425 | if match:
|
---|
| 2426 | include = match.group(2)
|
---|
| 2427 | is_system = (match.group(1) == '<')
|
---|
| 2428 | if include in include_state:
|
---|
| 2429 | error(filename, linenum, 'build/include', 4,
|
---|
| 2430 | '"%s" already included at %s:%s' %
|
---|
| 2431 | (include, filename, include_state[include]))
|
---|
| 2432 | else:
|
---|
| 2433 | include_state[include] = linenum
|
---|
| 2434 |
|
---|
| 2435 | # We want to ensure that headers appear in the right order:
|
---|
| 2436 | # 1) for foo.cc, foo.h (preferred location)
|
---|
| 2437 | # 2) c system files
|
---|
| 2438 | # 3) cpp system files
|
---|
| 2439 | # 4) for foo.cc, foo.h (deprecated location)
|
---|
| 2440 | # 5) other google headers
|
---|
| 2441 | #
|
---|
| 2442 | # We classify each include statement as one of those 5 types
|
---|
| 2443 | # using a number of techniques. The include_state object keeps
|
---|
| 2444 | # track of the highest type seen, and complains if we see a
|
---|
| 2445 | # lower type after that.
|
---|
| 2446 | error_message = include_state.CheckNextIncludeOrder(
|
---|
| 2447 | _ClassifyInclude(fileinfo, include, is_system))
|
---|
| 2448 | if error_message:
|
---|
| 2449 | error(filename, linenum, 'build/include_order', 4,
|
---|
| 2450 | '%s. Should be: %s.h, c system, c++ system, other.' %
|
---|
| 2451 | (error_message, fileinfo.BaseName()))
|
---|
| 2452 | if not include_state.IsInAlphabeticalOrder(include):
|
---|
| 2453 | error(filename, linenum, 'build/include_alpha', 4,
|
---|
| 2454 | 'Include "%s" not in alphabetical order' % include)
|
---|
| 2455 |
|
---|
| 2456 | # Look for any of the stream classes that are part of standard C++.
|
---|
| 2457 | match = _RE_PATTERN_INCLUDE.match(line)
|
---|
| 2458 | if match:
|
---|
| 2459 | include = match.group(2)
|
---|
| 2460 | if Match(r'(f|ind|io|i|o|parse|pf|stdio|str|)?stream$', include):
|
---|
| 2461 | # Many unit tests use cout, so we exempt them.
|
---|
| 2462 | if not _IsTestFilename(filename):
|
---|
| 2463 | error(filename, linenum, 'readability/streams', 3,
|
---|
| 2464 | 'Streams are highly discouraged.')
|
---|
| 2465 |
|
---|
| 2466 |
|
---|
| 2467 | def _GetTextInside(text, start_pattern):
|
---|
| 2468 | """Retrieves all the text between matching open and close parentheses.
|
---|
| 2469 |
|
---|
| 2470 | Given a string of lines and a regular expression string, retrieve all the text
|
---|
| 2471 | following the expression and between opening punctuation symbols like
|
---|
| 2472 | (, [, or {, and the matching close-punctuation symbol. This properly nested
|
---|
| 2473 | occurrences of the punctuations, so for the text like
|
---|
| 2474 | printf(a(), b(c()));
|
---|
| 2475 | a call to _GetTextInside(text, r'printf\(') will return 'a(), b(c())'.
|
---|
| 2476 | start_pattern must match string having an open punctuation symbol at the end.
|
---|
| 2477 |
|
---|
| 2478 | Args:
|
---|
| 2479 | text: The lines to extract text. Its comments and strings must be elided.
|
---|
| 2480 | It can be single line and can span multiple lines.
|
---|
| 2481 | start_pattern: The regexp string indicating where to start extracting
|
---|
| 2482 | the text.
|
---|
| 2483 | Returns:
|
---|
| 2484 | The extracted text.
|
---|
| 2485 | None if either the opening string or ending punctuation could not be found.
|
---|
| 2486 | """
|
---|
| 2487 | # TODO(sugawarayu): Audit cpplint.py to see what places could be profitably
|
---|
| 2488 | # rewritten to use _GetTextInside (and use inferior regexp matching today).
|
---|
| 2489 |
|
---|
| 2490 | # Give opening punctuations to get the matching close-punctuations.
|
---|
| 2491 | matching_punctuation = {'(': ')', '{': '}', '[': ']'}
|
---|
| 2492 | closing_punctuation = set(itervalues(matching_punctuation))
|
---|
| 2493 |
|
---|
| 2494 | # Find the position to start extracting text.
|
---|
| 2495 | match = re.search(start_pattern, text, re.M)
|
---|
| 2496 | if not match: # start_pattern not found in text.
|
---|
| 2497 | return None
|
---|
| 2498 | start_position = match.end(0)
|
---|
| 2499 |
|
---|
| 2500 | assert start_position > 0, (
|
---|
| 2501 | 'start_pattern must ends with an opening punctuation.')
|
---|
| 2502 | assert text[start_position - 1] in matching_punctuation, (
|
---|
| 2503 | 'start_pattern must ends with an opening punctuation.')
|
---|
| 2504 | # Stack of closing punctuations we expect to have in text after position.
|
---|
| 2505 | punctuation_stack = [matching_punctuation[text[start_position - 1]]]
|
---|
| 2506 | position = start_position
|
---|
| 2507 | while punctuation_stack and position < len(text):
|
---|
| 2508 | if text[position] == punctuation_stack[-1]:
|
---|
| 2509 | punctuation_stack.pop()
|
---|
| 2510 | elif text[position] in closing_punctuation:
|
---|
| 2511 | # A closing punctuation without matching opening punctuations.
|
---|
| 2512 | return None
|
---|
| 2513 | elif text[position] in matching_punctuation:
|
---|
| 2514 | punctuation_stack.append(matching_punctuation[text[position]])
|
---|
| 2515 | position += 1
|
---|
| 2516 | if punctuation_stack:
|
---|
| 2517 | # Opening punctuations left without matching close-punctuations.
|
---|
| 2518 | return None
|
---|
| 2519 | # punctuations match.
|
---|
| 2520 | return text[start_position:position - 1]
|
---|
| 2521 |
|
---|
| 2522 |
|
---|
| 2523 | def CheckLanguage(filename, clean_lines, linenum, file_extension, include_state,
|
---|
| 2524 | error):
|
---|
| 2525 | """Checks rules from the 'C++ language rules' section of cppguide.html.
|
---|
| 2526 |
|
---|
| 2527 | Some of these rules are hard to test (function overloading, using
|
---|
| 2528 | uint32 inappropriately), but we do the best we can.
|
---|
| 2529 |
|
---|
| 2530 | Args:
|
---|
| 2531 | filename: The name of the current file.
|
---|
| 2532 | clean_lines: A CleansedLines instance containing the file.
|
---|
| 2533 | linenum: The number of the line to check.
|
---|
| 2534 | file_extension: The extension (without the dot) of the filename.
|
---|
| 2535 | include_state: An _IncludeState instance in which the headers are inserted.
|
---|
| 2536 | error: The function to call with any errors found.
|
---|
| 2537 | """
|
---|
| 2538 | # If the line is empty or consists of entirely a comment, no need to
|
---|
| 2539 | # check it.
|
---|
| 2540 | line = clean_lines.elided[linenum]
|
---|
| 2541 | if not line:
|
---|
| 2542 | return
|
---|
| 2543 |
|
---|
| 2544 | match = _RE_PATTERN_INCLUDE.search(line)
|
---|
| 2545 | if match:
|
---|
| 2546 | CheckIncludeLine(filename, clean_lines, linenum, include_state, error)
|
---|
| 2547 | return
|
---|
| 2548 |
|
---|
| 2549 | # Create an extended_line, which is the concatenation of the current and
|
---|
| 2550 | # next lines, for more effective checking of code that may span more than one
|
---|
| 2551 | # line.
|
---|
| 2552 | if linenum + 1 < clean_lines.NumLines():
|
---|
| 2553 | extended_line = line + clean_lines.elided[linenum + 1]
|
---|
| 2554 | else:
|
---|
| 2555 | extended_line = line
|
---|
| 2556 |
|
---|
| 2557 | # Make Windows paths like Unix.
|
---|
| 2558 | fullname = os.path.abspath(filename).replace('\\', '/')
|
---|
| 2559 |
|
---|
| 2560 | # TODO(unknown): figure out if they're using default arguments in fn proto.
|
---|
| 2561 |
|
---|
| 2562 | # Check for non-const references in functions. This is tricky because &
|
---|
| 2563 | # is also used to take the address of something. We allow <> for templates,
|
---|
| 2564 | # (ignoring whatever is between the braces) and : for classes.
|
---|
| 2565 | # These are complicated re's. They try to capture the following:
|
---|
| 2566 | # paren (for fn-prototype start), typename, &, varname. For the const
|
---|
| 2567 | # version, we're willing for const to be before typename or after
|
---|
| 2568 | # Don't check the implementation on same line.
|
---|
| 2569 | fnline = line.split('{', 1)[0]
|
---|
| 2570 | if (len(re.findall(r'\([^()]*\b(?:[\w:]|<[^()]*>)+(\s?&|&\s?)\w+', fnline)) >
|
---|
| 2571 | len(re.findall(r'\([^()]*\bconst\s+(?:typename\s+)?(?:struct\s+)?'
|
---|
| 2572 | r'(?:[\w:]|<[^()]*>)+(\s?&|&\s?)\w+', fnline)) +
|
---|
| 2573 | len(re.findall(r'\([^()]*\b(?:[\w:]|<[^()]*>)+\s+const(\s?&|&\s?)[\w]+',
|
---|
| 2574 | fnline))):
|
---|
| 2575 |
|
---|
| 2576 | # We allow non-const references in a few standard places, like functions
|
---|
| 2577 | # called "swap()" or iostream operators like "<<" or ">>".
|
---|
| 2578 | if not Search(
|
---|
| 2579 | r'(swap|Swap|operator[<>][<>])\s*\(\s*(?:[\w:]|<.*>)+\s*&',
|
---|
| 2580 | fnline):
|
---|
| 2581 | error(filename, linenum, 'runtime/references', 2,
|
---|
| 2582 | 'Is this a non-const reference? '
|
---|
| 2583 | 'If so, make const or use a pointer.')
|
---|
| 2584 |
|
---|
| 2585 | # Check to see if they're using an conversion function cast.
|
---|
| 2586 | # I just try to capture the most common basic types, though there are more.
|
---|
| 2587 | # Parameterless conversion functions, such as bool(), are allowed as they are
|
---|
| 2588 | # probably a member operator declaration or default constructor.
|
---|
| 2589 | match = Search(
|
---|
| 2590 | r'(\bnew\s+)?\b' # Grab 'new' operator, if it's there
|
---|
| 2591 | r'(int|float|double|bool|char|int32|uint32|int64|uint64)\([^)]', line)
|
---|
| 2592 | if match:
|
---|
| 2593 | # gMock methods are defined using some variant of MOCK_METHODx(name, type)
|
---|
| 2594 | # where type may be float(), int(string), etc. Without context they are
|
---|
| 2595 | # virtually indistinguishable from int(x) casts. Likewise, gMock's
|
---|
| 2596 | # MockCallback takes a template parameter of the form return_type(arg_type),
|
---|
| 2597 | # which looks much like the cast we're trying to detect.
|
---|
| 2598 | if (match.group(1) is None and # If new operator, then this isn't a cast
|
---|
| 2599 | not (Match(r'^\s*MOCK_(CONST_)?METHOD\d+(_T)?\(', line) or
|
---|
| 2600 | Match(r'^\s*MockCallback<.*>', line))):
|
---|
| 2601 | error(filename, linenum, 'readability/casting', 4,
|
---|
| 2602 | 'Using deprecated casting style. '
|
---|
| 2603 | 'Use static_cast<%s>(...) instead' %
|
---|
| 2604 | match.group(2))
|
---|
| 2605 |
|
---|
| 2606 | CheckCStyleCast(filename, linenum, line, clean_lines.raw_lines[linenum],
|
---|
| 2607 | 'static_cast',
|
---|
| 2608 | r'\((int|float|double|bool|char|u?int(16|32|64))\)', error)
|
---|
| 2609 |
|
---|
| 2610 | # This doesn't catch all cases. Consider (const char * const)"hello".
|
---|
| 2611 | #
|
---|
| 2612 | # (char *) "foo" should always be a const_cast (reinterpret_cast won't
|
---|
| 2613 | # compile).
|
---|
| 2614 | if CheckCStyleCast(filename, linenum, line, clean_lines.raw_lines[linenum],
|
---|
| 2615 | 'const_cast', r'\((char\s?\*+\s?)\)\s*"', error):
|
---|
| 2616 | pass
|
---|
| 2617 | else:
|
---|
| 2618 | # Check pointer casts for other than string constants
|
---|
| 2619 | CheckCStyleCast(filename, linenum, line, clean_lines.raw_lines[linenum],
|
---|
| 2620 | 'reinterpret_cast', r'\((\w+\s?\*+\s?)\)', error)
|
---|
| 2621 |
|
---|
| 2622 | # In addition, we look for people taking the address of a cast. This
|
---|
| 2623 | # is dangerous -- casts can assign to temporaries, so the pointer doesn't
|
---|
| 2624 | # point where you think.
|
---|
| 2625 | if Search(
|
---|
| 2626 | r'(&\([^)]+\)[\w(])|(&(static|dynamic|reinterpret)_cast\b)', line):
|
---|
| 2627 | error(filename, linenum, 'runtime/casting', 4,
|
---|
| 2628 | ('Are you taking an address of a cast? '
|
---|
| 2629 | 'This is dangerous: could be a temp var. '
|
---|
| 2630 | 'Take the address before doing the cast, rather than after'))
|
---|
| 2631 |
|
---|
| 2632 | # Check for people declaring static/global STL strings at the top level.
|
---|
| 2633 | # This is dangerous because the C++ language does not guarantee that
|
---|
| 2634 | # globals with constructors are initialized before the first access.
|
---|
| 2635 | match = Match(
|
---|
| 2636 | r'((?:|static +)(?:|const +))string +([a-zA-Z0-9_:]+)\b(.*)',
|
---|
| 2637 | line)
|
---|
| 2638 | # Make sure it's not a function.
|
---|
| 2639 | # Function template specialization looks like: "string foo<Type>(...".
|
---|
| 2640 | # Class template definitions look like: "string Foo<Type>::Method(...".
|
---|
| 2641 | if match and not Match(r'\s*(<.*>)?(::[a-zA-Z0-9_]+)?\s*\(([^"]|$)',
|
---|
| 2642 | match.group(3)):
|
---|
| 2643 | error(filename, linenum, 'runtime/string', 4,
|
---|
| 2644 | 'For a static/global string constant, use a C style string instead: '
|
---|
| 2645 | '"%schar %s[]".' %
|
---|
| 2646 | (match.group(1), match.group(2)))
|
---|
| 2647 |
|
---|
| 2648 | # Check that we're not using RTTI outside of testing code.
|
---|
| 2649 | if Search(r'\bdynamic_cast<', line) and not _IsTestFilename(filename):
|
---|
| 2650 | error(filename, linenum, 'runtime/rtti', 5,
|
---|
| 2651 | 'Do not use dynamic_cast<>. If you need to cast within a class '
|
---|
| 2652 | "hierarchy, use static_cast<> to upcast. Google doesn't support "
|
---|
| 2653 | 'RTTI.')
|
---|
| 2654 |
|
---|
| 2655 | if Search(r'\b([A-Za-z0-9_]*_)\(\1\)', line):
|
---|
| 2656 | error(filename, linenum, 'runtime/init', 4,
|
---|
| 2657 | 'You seem to be initializing a member variable with itself.')
|
---|
| 2658 |
|
---|
| 2659 | if file_extension == 'h':
|
---|
| 2660 | # TODO(unknown): check that 1-arg constructors are explicit.
|
---|
| 2661 | # How to tell it's a constructor?
|
---|
| 2662 | # (handled in CheckForNonStandardConstructs for now)
|
---|
| 2663 | # TODO(unknown): check that classes have DISALLOW_EVIL_CONSTRUCTORS
|
---|
| 2664 | # (level 1 error)
|
---|
| 2665 | pass
|
---|
| 2666 |
|
---|
| 2667 | # Check if people are using the verboten C basic types. The only exception
|
---|
| 2668 | # we regularly allow is "unsigned short port" for port.
|
---|
| 2669 | if Search(r'\bshort port\b', line):
|
---|
| 2670 | if not Search(r'\bunsigned short port\b', line):
|
---|
| 2671 | error(filename, linenum, 'runtime/int', 4,
|
---|
| 2672 | 'Use "unsigned short" for ports, not "short"')
|
---|
| 2673 | else:
|
---|
| 2674 | match = Search(r'\b(short|long(?! +double)|long long)\b', line)
|
---|
| 2675 | if match:
|
---|
| 2676 | error(filename, linenum, 'runtime/int', 4,
|
---|
| 2677 | 'Use int16/int64/etc, rather than the C type %s' % match.group(1))
|
---|
| 2678 |
|
---|
| 2679 | # When snprintf is used, the second argument shouldn't be a literal.
|
---|
| 2680 | match = Search(r'snprintf\s*\(([^,]*),\s*([0-9]*)\s*,', line)
|
---|
| 2681 | if match and match.group(2) != '0':
|
---|
| 2682 | # If 2nd arg is zero, snprintf is used to calculate size.
|
---|
| 2683 | error(filename, linenum, 'runtime/printf', 3,
|
---|
| 2684 | 'If you can, use sizeof(%s) instead of %s as the 2nd arg '
|
---|
| 2685 | 'to snprintf.' % (match.group(1), match.group(2)))
|
---|
| 2686 |
|
---|
| 2687 | # Check if some verboten C functions are being used.
|
---|
| 2688 | if Search(r'\bsprintf\b', line):
|
---|
| 2689 | error(filename, linenum, 'runtime/printf', 5,
|
---|
| 2690 | 'Never use sprintf. Use snprintf instead.')
|
---|
| 2691 | match = Search(r'\b(strcpy|strcat)\b', line)
|
---|
| 2692 | if match:
|
---|
| 2693 | error(filename, linenum, 'runtime/printf', 4,
|
---|
| 2694 | 'Almost always, snprintf is better than %s' % match.group(1))
|
---|
| 2695 |
|
---|
| 2696 | if Search(r'\bsscanf\b', line):
|
---|
| 2697 | error(filename, linenum, 'runtime/printf', 1,
|
---|
| 2698 | 'sscanf can be ok, but is slow and can overflow buffers.')
|
---|
| 2699 |
|
---|
| 2700 | # Check if some verboten operator overloading is going on
|
---|
| 2701 | # TODO(unknown): catch out-of-line unary operator&:
|
---|
| 2702 | # class X {};
|
---|
| 2703 | # int operator&(const X& x) { return 42; } // unary operator&
|
---|
| 2704 | # The trick is it's hard to tell apart from binary operator&:
|
---|
| 2705 | # class Y { int operator&(const Y& x) { return 23; } }; // binary operator&
|
---|
| 2706 | if Search(r'\boperator\s*&\s*\(\s*\)', line):
|
---|
| 2707 | error(filename, linenum, 'runtime/operator', 4,
|
---|
| 2708 | 'Unary operator& is dangerous. Do not use it.')
|
---|
| 2709 |
|
---|
| 2710 | # Check for suspicious usage of "if" like
|
---|
| 2711 | # } if (a == b) {
|
---|
| 2712 | if Search(r'\}\s*if\s*\(', line):
|
---|
| 2713 | error(filename, linenum, 'readability/braces', 4,
|
---|
| 2714 | 'Did you mean "else if"? If not, start a new line for "if".')
|
---|
| 2715 |
|
---|
| 2716 | # Check for potential format string bugs like printf(foo).
|
---|
| 2717 | # We constrain the pattern not to pick things like DocidForPrintf(foo).
|
---|
| 2718 | # Not perfect but it can catch printf(foo.c_str()) and printf(foo->c_str())
|
---|
| 2719 | # TODO(sugawarayu): Catch the following case. Need to change the calling
|
---|
| 2720 | # convention of the whole function to process multiple line to handle it.
|
---|
| 2721 | # printf(
|
---|
| 2722 | # boy_this_is_a_really_long_variable_that_cannot_fit_on_the_prev_line);
|
---|
| 2723 | printf_args = _GetTextInside(line, r'(?i)\b(string)?printf\s*\(')
|
---|
| 2724 | if printf_args:
|
---|
| 2725 | match = Match(r'([\w.\->()]+)$', printf_args)
|
---|
| 2726 | if match:
|
---|
| 2727 | function_name = re.search(r'\b((?:string)?printf)\s*\(',
|
---|
| 2728 | line, re.I).group(1)
|
---|
| 2729 | error(filename, linenum, 'runtime/printf', 4,
|
---|
| 2730 | 'Potential format string bug. Do %s("%%s", %s) instead.'
|
---|
| 2731 | % (function_name, match.group(1)))
|
---|
| 2732 |
|
---|
| 2733 | # Check for potential memset bugs like memset(buf, sizeof(buf), 0).
|
---|
| 2734 | match = Search(r'memset\s*\(([^,]*),\s*([^,]*),\s*0\s*\)', line)
|
---|
| 2735 | if match and not Match(r"^''|-?[0-9]+|0x[0-9A-Fa-f]$", match.group(2)):
|
---|
| 2736 | error(filename, linenum, 'runtime/memset', 4,
|
---|
| 2737 | 'Did you mean "memset(%s, 0, %s)"?'
|
---|
| 2738 | % (match.group(1), match.group(2)))
|
---|
| 2739 |
|
---|
| 2740 | if Search(r'\busing namespace\b', line):
|
---|
| 2741 | error(filename, linenum, 'build/namespaces', 5,
|
---|
| 2742 | 'Do not use namespace using-directives. '
|
---|
| 2743 | 'Use using-declarations instead.')
|
---|
| 2744 |
|
---|
| 2745 | # Detect variable-length arrays.
|
---|
| 2746 | match = Match(r'\s*(.+::)?(\w+) [a-z]\w*\[(.+)];', line)
|
---|
| 2747 | if (match and match.group(2) != 'return' and match.group(2) != 'delete' and
|
---|
| 2748 | match.group(3).find(']') == -1):
|
---|
| 2749 | # Split the size using space and arithmetic operators as delimiters.
|
---|
| 2750 | # If any of the resulting tokens are not compile time constants then
|
---|
| 2751 | # report the error.
|
---|
| 2752 | tokens = re.split(r'\s|\+|\-|\*|\/|<<|>>]', match.group(3))
|
---|
| 2753 | is_const = True
|
---|
| 2754 | skip_next = False
|
---|
| 2755 | for tok in tokens:
|
---|
| 2756 | if skip_next:
|
---|
| 2757 | skip_next = False
|
---|
| 2758 | continue
|
---|
| 2759 |
|
---|
| 2760 | if Search(r'sizeof\(.+\)', tok): continue
|
---|
| 2761 | if Search(r'arraysize\(\w+\)', tok): continue
|
---|
| 2762 |
|
---|
| 2763 | tok = tok.lstrip('(')
|
---|
| 2764 | tok = tok.rstrip(')')
|
---|
| 2765 | if not tok: continue
|
---|
| 2766 | if Match(r'\d+', tok): continue
|
---|
| 2767 | if Match(r'0[xX][0-9a-fA-F]+', tok): continue
|
---|
| 2768 | if Match(r'k[A-Z0-9]\w*', tok): continue
|
---|
| 2769 | if Match(r'(.+::)?k[A-Z0-9]\w*', tok): continue
|
---|
| 2770 | if Match(r'(.+::)?[A-Z][A-Z0-9_]*', tok): continue
|
---|
| 2771 | # A catch all for tricky sizeof cases, including 'sizeof expression',
|
---|
| 2772 | # 'sizeof(*type)', 'sizeof(const type)', 'sizeof(struct StructName)'
|
---|
| 2773 | # requires skipping the next token because we split on ' ' and '*'.
|
---|
| 2774 | if tok.startswith('sizeof'):
|
---|
| 2775 | skip_next = True
|
---|
| 2776 | continue
|
---|
| 2777 | is_const = False
|
---|
| 2778 | break
|
---|
| 2779 | if not is_const:
|
---|
| 2780 | error(filename, linenum, 'runtime/arrays', 1,
|
---|
| 2781 | 'Do not use variable-length arrays. Use an appropriately named '
|
---|
| 2782 | "('k' followed by CamelCase) compile-time constant for the size.")
|
---|
| 2783 |
|
---|
| 2784 | # If DISALLOW_EVIL_CONSTRUCTORS, DISALLOW_COPY_AND_ASSIGN, or
|
---|
| 2785 | # DISALLOW_IMPLICIT_CONSTRUCTORS is present, then it should be the last thing
|
---|
| 2786 | # in the class declaration.
|
---|
| 2787 | match = Match(
|
---|
| 2788 | (r'\s*'
|
---|
| 2789 | r'(DISALLOW_(EVIL_CONSTRUCTORS|COPY_AND_ASSIGN|IMPLICIT_CONSTRUCTORS))'
|
---|
| 2790 | r'\(.*\);$'),
|
---|
| 2791 | line)
|
---|
| 2792 | if match and linenum + 1 < clean_lines.NumLines():
|
---|
| 2793 | next_line = clean_lines.elided[linenum + 1]
|
---|
| 2794 | # We allow some, but not all, declarations of variables to be present
|
---|
| 2795 | # in the statement that defines the class. The [\w\*,\s]* fragment of
|
---|
| 2796 | # the regular expression below allows users to declare instances of
|
---|
| 2797 | # the class or pointers to instances, but not less common types such
|
---|
| 2798 | # as function pointers or arrays. It's a tradeoff between allowing
|
---|
| 2799 | # reasonable code and avoiding trying to parse more C++ using regexps.
|
---|
| 2800 | if not Search(r'^\s*}[\w\*,\s]*;', next_line):
|
---|
| 2801 | error(filename, linenum, 'readability/constructors', 3,
|
---|
| 2802 | match.group(1) + ' should be the last thing in the class')
|
---|
| 2803 |
|
---|
| 2804 | # Check for use of unnamed namespaces in header files. Registration
|
---|
| 2805 | # macros are typically OK, so we allow use of "namespace {" on lines
|
---|
| 2806 | # that end with backslashes.
|
---|
| 2807 | if (file_extension == 'h'
|
---|
| 2808 | and Search(r'\bnamespace\s*{', line)
|
---|
| 2809 | and line[-1] != '\\'):
|
---|
| 2810 | error(filename, linenum, 'build/namespaces', 4,
|
---|
| 2811 | 'Do not use unnamed namespaces in header files. See '
|
---|
| 2812 | 'http://google-styleguide.googlecode.com/svn/trunk/cppguide.xml#Namespaces'
|
---|
| 2813 | ' for more information.')
|
---|
| 2814 |
|
---|
| 2815 |
|
---|
| 2816 | def CheckCStyleCast(filename, linenum, line, raw_line, cast_type, pattern,
|
---|
| 2817 | error):
|
---|
| 2818 | """Checks for a C-style cast by looking for the pattern.
|
---|
| 2819 |
|
---|
| 2820 | This also handles sizeof(type) warnings, due to similarity of content.
|
---|
| 2821 |
|
---|
| 2822 | Args:
|
---|
| 2823 | filename: The name of the current file.
|
---|
| 2824 | linenum: The number of the line to check.
|
---|
| 2825 | line: The line of code to check.
|
---|
| 2826 | raw_line: The raw line of code to check, with comments.
|
---|
| 2827 | cast_type: The string for the C++ cast to recommend. This is either
|
---|
| 2828 | reinterpret_cast, static_cast, or const_cast, depending.
|
---|
| 2829 | pattern: The regular expression used to find C-style casts.
|
---|
| 2830 | error: The function to call with any errors found.
|
---|
| 2831 |
|
---|
| 2832 | Returns:
|
---|
| 2833 | True if an error was emitted.
|
---|
| 2834 | False otherwise.
|
---|
| 2835 | """
|
---|
| 2836 | match = Search(pattern, line)
|
---|
| 2837 | if not match:
|
---|
| 2838 | return False
|
---|
| 2839 |
|
---|
| 2840 | # e.g., sizeof(int)
|
---|
| 2841 | sizeof_match = Match(r'.*sizeof\s*$', line[0:match.start(1) - 1])
|
---|
| 2842 | if sizeof_match:
|
---|
| 2843 | error(filename, linenum, 'runtime/sizeof', 1,
|
---|
| 2844 | 'Using sizeof(type). Use sizeof(varname) instead if possible')
|
---|
| 2845 | return True
|
---|
| 2846 |
|
---|
| 2847 | remainder = line[match.end(0):]
|
---|
| 2848 |
|
---|
| 2849 | # The close paren is for function pointers as arguments to a function.
|
---|
| 2850 | # eg, void foo(void (*bar)(int));
|
---|
| 2851 | # The semicolon check is a more basic function check; also possibly a
|
---|
| 2852 | # function pointer typedef.
|
---|
| 2853 | # eg, void foo(int); or void foo(int) const;
|
---|
| 2854 | # The equals check is for function pointer assignment.
|
---|
| 2855 | # eg, void *(*foo)(int) = ...
|
---|
| 2856 | # The > is for MockCallback<...> ...
|
---|
| 2857 | #
|
---|
| 2858 | # Right now, this will only catch cases where there's a single argument, and
|
---|
| 2859 | # it's unnamed. It should probably be expanded to check for multiple
|
---|
| 2860 | # arguments with some unnamed.
|
---|
| 2861 | function_match = Match(r'\s*(\)|=|(const)?\s*(;|\{|throw\(\)|>))', remainder)
|
---|
| 2862 | if function_match:
|
---|
| 2863 | if (not function_match.group(3) or
|
---|
| 2864 | function_match.group(3) == ';' or
|
---|
| 2865 | ('MockCallback<' not in raw_line and
|
---|
| 2866 | '/*' not in raw_line)):
|
---|
| 2867 | error(filename, linenum, 'readability/function', 3,
|
---|
| 2868 | 'All parameters should be named in a function')
|
---|
| 2869 | return True
|
---|
| 2870 |
|
---|
| 2871 | # At this point, all that should be left is actual casts.
|
---|
| 2872 | error(filename, linenum, 'readability/casting', 4,
|
---|
| 2873 | 'Using C-style cast. Use %s<%s>(...) instead' %
|
---|
| 2874 | (cast_type, match.group(1)))
|
---|
| 2875 |
|
---|
| 2876 | return True
|
---|
| 2877 |
|
---|
| 2878 |
|
---|
| 2879 | _HEADERS_CONTAINING_TEMPLATES = (
|
---|
| 2880 | ('<deque>', ('deque',)),
|
---|
| 2881 | ('<functional>', ('unary_function', 'binary_function',
|
---|
| 2882 | 'plus', 'minus', 'multiplies', 'divides', 'modulus',
|
---|
| 2883 | 'negate',
|
---|
| 2884 | 'equal_to', 'not_equal_to', 'greater', 'less',
|
---|
| 2885 | 'greater_equal', 'less_equal',
|
---|
| 2886 | 'logical_and', 'logical_or', 'logical_not',
|
---|
| 2887 | 'unary_negate', 'not1', 'binary_negate', 'not2',
|
---|
| 2888 | 'bind1st', 'bind2nd',
|
---|
| 2889 | 'pointer_to_unary_function',
|
---|
| 2890 | 'pointer_to_binary_function',
|
---|
| 2891 | 'ptr_fun',
|
---|
| 2892 | 'mem_fun_t', 'mem_fun', 'mem_fun1_t', 'mem_fun1_ref_t',
|
---|
| 2893 | 'mem_fun_ref_t',
|
---|
| 2894 | 'const_mem_fun_t', 'const_mem_fun1_t',
|
---|
| 2895 | 'const_mem_fun_ref_t', 'const_mem_fun1_ref_t',
|
---|
| 2896 | 'mem_fun_ref',
|
---|
| 2897 | )),
|
---|
| 2898 | ('<limits>', ('numeric_limits',)),
|
---|
| 2899 | ('<list>', ('list',)),
|
---|
| 2900 | ('<map>', ('map', 'multimap',)),
|
---|
| 2901 | ('<memory>', ('allocator',)),
|
---|
| 2902 | ('<queue>', ('queue', 'priority_queue',)),
|
---|
| 2903 | ('<set>', ('set', 'multiset',)),
|
---|
| 2904 | ('<stack>', ('stack',)),
|
---|
| 2905 | ('<string>', ('char_traits', 'basic_string',)),
|
---|
| 2906 | ('<utility>', ('pair',)),
|
---|
| 2907 | ('<vector>', ('vector',)),
|
---|
| 2908 |
|
---|
| 2909 | # gcc extensions.
|
---|
| 2910 | # Note: std::hash is their hash, ::hash is our hash
|
---|
| 2911 | ('<hash_map>', ('hash_map', 'hash_multimap',)),
|
---|
| 2912 | ('<hash_set>', ('hash_set', 'hash_multiset',)),
|
---|
| 2913 | ('<slist>', ('slist',)),
|
---|
| 2914 | )
|
---|
| 2915 |
|
---|
| 2916 | _RE_PATTERN_STRING = re.compile(r'\bstring\b')
|
---|
| 2917 |
|
---|
| 2918 | _re_pattern_algorithm_header = []
|
---|
| 2919 | for _template in ('copy', 'max', 'min', 'min_element', 'sort', 'swap',
|
---|
| 2920 | 'transform'):
|
---|
| 2921 | # Match max<type>(..., ...), max(..., ...), but not foo->max, foo.max or
|
---|
| 2922 | # type::max().
|
---|
| 2923 | _re_pattern_algorithm_header.append(
|
---|
| 2924 | (re.compile(r'[^>.]\b' + _template + r'(<.*?>)?\([^\)]'),
|
---|
| 2925 | _template,
|
---|
| 2926 | '<algorithm>'))
|
---|
| 2927 |
|
---|
| 2928 | _re_pattern_templates = []
|
---|
| 2929 | for _header, _templates in _HEADERS_CONTAINING_TEMPLATES:
|
---|
| 2930 | for _template in _templates:
|
---|
| 2931 | _re_pattern_templates.append(
|
---|
| 2932 | (re.compile(r'(\<|\b)' + _template + r'\s*\<'),
|
---|
| 2933 | _template + '<>',
|
---|
| 2934 | _header))
|
---|
| 2935 |
|
---|
| 2936 |
|
---|
| 2937 | def FilesBelongToSameModule(filename_cc, filename_h):
|
---|
| 2938 | """Check if these two filenames belong to the same module.
|
---|
| 2939 |
|
---|
| 2940 | The concept of a 'module' here is a as follows:
|
---|
| 2941 | foo.h, foo-inl.h, foo.cc, foo_test.cc and foo_unittest.cc belong to the
|
---|
| 2942 | same 'module' if they are in the same directory.
|
---|
| 2943 | some/path/public/xyzzy and some/path/internal/xyzzy are also considered
|
---|
| 2944 | to belong to the same module here.
|
---|
| 2945 |
|
---|
| 2946 | If the filename_cc contains a longer path than the filename_h, for example,
|
---|
| 2947 | '/absolute/path/to/base/sysinfo.cc', and this file would include
|
---|
| 2948 | 'base/sysinfo.h', this function also produces the prefix needed to open the
|
---|
| 2949 | header. This is used by the caller of this function to more robustly open the
|
---|
| 2950 | header file. We don't have access to the real include paths in this context,
|
---|
| 2951 | so we need this guesswork here.
|
---|
| 2952 |
|
---|
| 2953 | Known bugs: tools/base/bar.cc and base/bar.h belong to the same module
|
---|
| 2954 | according to this implementation. Because of this, this function gives
|
---|
| 2955 | some false positives. This should be sufficiently rare in practice.
|
---|
| 2956 |
|
---|
| 2957 | Args:
|
---|
| 2958 | filename_cc: is the path for the .cc file
|
---|
| 2959 | filename_h: is the path for the header path
|
---|
| 2960 |
|
---|
| 2961 | Returns:
|
---|
| 2962 | Tuple with a bool and a string:
|
---|
| 2963 | bool: True if filename_cc and filename_h belong to the same module.
|
---|
| 2964 | string: the additional prefix needed to open the header file.
|
---|
| 2965 | """
|
---|
| 2966 |
|
---|
| 2967 | if not filename_cc.endswith('.cc'):
|
---|
| 2968 | return (False, '')
|
---|
| 2969 | filename_cc = filename_cc[:-len('.cc')]
|
---|
| 2970 | if filename_cc.endswith('_unittest'):
|
---|
| 2971 | filename_cc = filename_cc[:-len('_unittest')]
|
---|
| 2972 | elif filename_cc.endswith('_test'):
|
---|
| 2973 | filename_cc = filename_cc[:-len('_test')]
|
---|
| 2974 | filename_cc = filename_cc.replace('/public/', '/')
|
---|
| 2975 | filename_cc = filename_cc.replace('/internal/', '/')
|
---|
| 2976 |
|
---|
| 2977 | if not filename_h.endswith('.h'):
|
---|
| 2978 | return (False, '')
|
---|
| 2979 | filename_h = filename_h[:-len('.h')]
|
---|
| 2980 | if filename_h.endswith('-inl'):
|
---|
| 2981 | filename_h = filename_h[:-len('-inl')]
|
---|
| 2982 | filename_h = filename_h.replace('/public/', '/')
|
---|
| 2983 | filename_h = filename_h.replace('/internal/', '/')
|
---|
| 2984 |
|
---|
| 2985 | files_belong_to_same_module = filename_cc.endswith(filename_h)
|
---|
| 2986 | common_path = ''
|
---|
| 2987 | if files_belong_to_same_module:
|
---|
| 2988 | common_path = filename_cc[:-len(filename_h)]
|
---|
| 2989 | return files_belong_to_same_module, common_path
|
---|
| 2990 |
|
---|
| 2991 |
|
---|
| 2992 | def UpdateIncludeState(filename, include_state, io=codecs):
|
---|
| 2993 | """Fill up the include_state with new includes found from the file.
|
---|
| 2994 |
|
---|
| 2995 | Args:
|
---|
| 2996 | filename: the name of the header to read.
|
---|
| 2997 | include_state: an _IncludeState instance in which the headers are inserted.
|
---|
| 2998 | io: The io factory to use to read the file. Provided for testability.
|
---|
| 2999 |
|
---|
| 3000 | Returns:
|
---|
| 3001 | True if a header was succesfully added. False otherwise.
|
---|
| 3002 | """
|
---|
| 3003 | headerfile = None
|
---|
| 3004 | try:
|
---|
| 3005 | headerfile = io.open(filename, 'r', 'utf8', 'replace')
|
---|
| 3006 | except IOError:
|
---|
| 3007 | return False
|
---|
| 3008 | linenum = 0
|
---|
| 3009 | for line in headerfile:
|
---|
| 3010 | linenum += 1
|
---|
| 3011 | clean_line = CleanseComments(line)
|
---|
| 3012 | match = _RE_PATTERN_INCLUDE.search(clean_line)
|
---|
| 3013 | if match:
|
---|
| 3014 | include = match.group(2)
|
---|
| 3015 | # The value formatting is cute, but not really used right now.
|
---|
| 3016 | # What matters here is that the key is in include_state.
|
---|
| 3017 | include_state.setdefault(include, '%s:%d' % (filename, linenum))
|
---|
| 3018 | return True
|
---|
| 3019 |
|
---|
| 3020 |
|
---|
| 3021 | def CheckForIncludeWhatYouUse(filename, clean_lines, include_state, error,
|
---|
| 3022 | io=codecs):
|
---|
| 3023 | """Reports for missing stl includes.
|
---|
| 3024 |
|
---|
| 3025 | This function will output warnings to make sure you are including the headers
|
---|
| 3026 | necessary for the stl containers and functions that you use. We only give one
|
---|
| 3027 | reason to include a header. For example, if you use both equal_to<> and
|
---|
| 3028 | less<> in a .h file, only one (the latter in the file) of these will be
|
---|
| 3029 | reported as a reason to include the <functional>.
|
---|
| 3030 |
|
---|
| 3031 | Args:
|
---|
| 3032 | filename: The name of the current file.
|
---|
| 3033 | clean_lines: A CleansedLines instance containing the file.
|
---|
| 3034 | include_state: An _IncludeState instance.
|
---|
| 3035 | error: The function to call with any errors found.
|
---|
| 3036 | io: The IO factory to use to read the header file. Provided for unittest
|
---|
| 3037 | injection.
|
---|
| 3038 | """
|
---|
| 3039 | required = {} # A map of header name to linenumber and the template entity.
|
---|
| 3040 | # Example of required: { '<functional>': (1219, 'less<>') }
|
---|
| 3041 |
|
---|
| 3042 | for linenum in range(clean_lines.NumLines()):
|
---|
| 3043 | line = clean_lines.elided[linenum]
|
---|
| 3044 | if not line or line[0] == '#':
|
---|
| 3045 | continue
|
---|
| 3046 |
|
---|
| 3047 | # String is special -- it is a non-templatized type in STL.
|
---|
| 3048 | matched = _RE_PATTERN_STRING.search(line)
|
---|
| 3049 | if matched:
|
---|
| 3050 | # Don't warn about strings in non-STL namespaces:
|
---|
| 3051 | # (We check only the first match per line; good enough.)
|
---|
| 3052 | prefix = line[:matched.start()]
|
---|
| 3053 | if prefix.endswith('std::') or not prefix.endswith('::'):
|
---|
| 3054 | required['<string>'] = (linenum, 'string')
|
---|
| 3055 |
|
---|
| 3056 | for pattern, template, header in _re_pattern_algorithm_header:
|
---|
| 3057 | if pattern.search(line):
|
---|
| 3058 | required[header] = (linenum, template)
|
---|
| 3059 |
|
---|
| 3060 | # The following function is just a speed up, no semantics are changed.
|
---|
| 3061 | if not '<' in line: # Reduces the cpu time usage by skipping lines.
|
---|
| 3062 | continue
|
---|
| 3063 |
|
---|
| 3064 | for pattern, template, header in _re_pattern_templates:
|
---|
| 3065 | if pattern.search(line):
|
---|
| 3066 | required[header] = (linenum, template)
|
---|
| 3067 |
|
---|
| 3068 | # The policy is that if you #include something in foo.h you don't need to
|
---|
| 3069 | # include it again in foo.cc. Here, we will look at possible includes.
|
---|
| 3070 | # Let's copy the include_state so it is only messed up within this function.
|
---|
| 3071 | include_state = include_state.copy()
|
---|
| 3072 |
|
---|
| 3073 | # Did we find the header for this file (if any) and succesfully load it?
|
---|
| 3074 | header_found = False
|
---|
| 3075 |
|
---|
| 3076 | # Use the absolute path so that matching works properly.
|
---|
| 3077 | abs_filename = FileInfo(filename).FullName()
|
---|
| 3078 |
|
---|
| 3079 | # For Emacs's flymake.
|
---|
| 3080 | # If cpplint is invoked from Emacs's flymake, a temporary file is generated
|
---|
| 3081 | # by flymake and that file name might end with '_flymake.cc'. In that case,
|
---|
| 3082 | # restore original file name here so that the corresponding header file can be
|
---|
| 3083 | # found.
|
---|
| 3084 | # e.g. If the file name is 'foo_flymake.cc', we should search for 'foo.h'
|
---|
| 3085 | # instead of 'foo_flymake.h'
|
---|
| 3086 | abs_filename = re.sub(r'_flymake\.cc$', '.cc', abs_filename)
|
---|
| 3087 |
|
---|
| 3088 | # include_state is modified during iteration, so we iterate over a copy of
|
---|
| 3089 | # the keys.
|
---|
| 3090 | header_keys = list(include_state.keys())
|
---|
| 3091 | for header in header_keys:
|
---|
| 3092 | (same_module, common_path) = FilesBelongToSameModule(abs_filename, header)
|
---|
| 3093 | fullpath = common_path + header
|
---|
| 3094 | if same_module and UpdateIncludeState(fullpath, include_state, io):
|
---|
| 3095 | header_found = True
|
---|
| 3096 |
|
---|
| 3097 | # If we can't find the header file for a .cc, assume it's because we don't
|
---|
| 3098 | # know where to look. In that case we'll give up as we're not sure they
|
---|
| 3099 | # didn't include it in the .h file.
|
---|
| 3100 | # TODO(unknown): Do a better job of finding .h files so we are confident that
|
---|
| 3101 | # not having the .h file means there isn't one.
|
---|
| 3102 | if filename.endswith('.cc') and not header_found:
|
---|
| 3103 | return
|
---|
| 3104 |
|
---|
| 3105 | # All the lines have been processed, report the errors found.
|
---|
| 3106 | for required_header_unstripped in required:
|
---|
| 3107 | template = required[required_header_unstripped][1]
|
---|
| 3108 | if required_header_unstripped.strip('<>"') not in include_state:
|
---|
| 3109 | error(filename, required[required_header_unstripped][0],
|
---|
| 3110 | 'build/include_what_you_use', 4,
|
---|
| 3111 | 'Add #include ' + required_header_unstripped + ' for ' + template)
|
---|
| 3112 |
|
---|
| 3113 |
|
---|
| 3114 | _RE_PATTERN_EXPLICIT_MAKEPAIR = re.compile(r'\bmake_pair\s*<')
|
---|
| 3115 |
|
---|
| 3116 |
|
---|
| 3117 | def CheckMakePairUsesDeduction(filename, clean_lines, linenum, error):
|
---|
| 3118 | """Check that make_pair's template arguments are deduced.
|
---|
| 3119 |
|
---|
| 3120 | G++ 4.6 in C++0x mode fails badly if make_pair's template arguments are
|
---|
| 3121 | specified explicitly, and such use isn't intended in any case.
|
---|
| 3122 |
|
---|
| 3123 | Args:
|
---|
| 3124 | filename: The name of the current file.
|
---|
| 3125 | clean_lines: A CleansedLines instance containing the file.
|
---|
| 3126 | linenum: The number of the line to check.
|
---|
| 3127 | error: The function to call with any errors found.
|
---|
| 3128 | """
|
---|
| 3129 | raw = clean_lines.raw_lines
|
---|
| 3130 | line = raw[linenum]
|
---|
| 3131 | match = _RE_PATTERN_EXPLICIT_MAKEPAIR.search(line)
|
---|
| 3132 | if match:
|
---|
| 3133 | error(filename, linenum, 'build/explicit_make_pair',
|
---|
| 3134 | 4, # 4 = high confidence
|
---|
| 3135 | 'Omit template arguments from make_pair OR use pair directly OR'
|
---|
| 3136 | ' if appropriate, construct a pair directly')
|
---|
| 3137 |
|
---|
| 3138 |
|
---|
| 3139 | def ProcessLine(filename, file_extension,
|
---|
| 3140 | clean_lines, line, include_state, function_state,
|
---|
| 3141 | class_state, error, extra_check_functions=[]):
|
---|
| 3142 | """Processes a single line in the file.
|
---|
| 3143 |
|
---|
| 3144 | Args:
|
---|
| 3145 | filename: Filename of the file that is being processed.
|
---|
| 3146 | file_extension: The extension (dot not included) of the file.
|
---|
| 3147 | clean_lines: An array of strings, each representing a line of the file,
|
---|
| 3148 | with comments stripped.
|
---|
| 3149 | line: Number of line being processed.
|
---|
| 3150 | include_state: An _IncludeState instance in which the headers are inserted.
|
---|
| 3151 | function_state: A _FunctionState instance which counts function lines, etc.
|
---|
| 3152 | class_state: A _ClassState instance which maintains information about
|
---|
| 3153 | the current stack of nested class declarations being parsed.
|
---|
| 3154 | error: A callable to which errors are reported, which takes 4 arguments:
|
---|
| 3155 | filename, line number, error level, and message
|
---|
| 3156 | extra_check_functions: An array of additional check functions that will be
|
---|
| 3157 | run on each source line. Each function takes 4
|
---|
| 3158 | arguments: filename, clean_lines, line, error
|
---|
| 3159 | """
|
---|
| 3160 | raw_lines = clean_lines.raw_lines
|
---|
| 3161 | ParseNolintSuppressions(filename, raw_lines[line], line, error)
|
---|
| 3162 | CheckForFunctionLengths(filename, clean_lines, line, function_state, error)
|
---|
| 3163 | CheckForMultilineCommentsAndStrings(filename, clean_lines, line, error)
|
---|
| 3164 | CheckStyle(filename, clean_lines, line, file_extension, class_state, error)
|
---|
| 3165 | CheckLanguage(filename, clean_lines, line, file_extension, include_state,
|
---|
| 3166 | error)
|
---|
| 3167 | CheckForNonStandardConstructs(filename, clean_lines, line,
|
---|
| 3168 | class_state, error)
|
---|
| 3169 | CheckPosixThreading(filename, clean_lines, line, error)
|
---|
| 3170 | CheckInvalidIncrement(filename, clean_lines, line, error)
|
---|
| 3171 | CheckMakePairUsesDeduction(filename, clean_lines, line, error)
|
---|
| 3172 | for check_fn in extra_check_functions:
|
---|
| 3173 | check_fn(filename, clean_lines, line, error)
|
---|
| 3174 |
|
---|
| 3175 | def ProcessFileData(filename, file_extension, lines, error,
|
---|
| 3176 | extra_check_functions=[]):
|
---|
| 3177 | """Performs lint checks and reports any errors to the given error function.
|
---|
| 3178 |
|
---|
| 3179 | Args:
|
---|
| 3180 | filename: Filename of the file that is being processed.
|
---|
| 3181 | file_extension: The extension (dot not included) of the file.
|
---|
| 3182 | lines: An array of strings, each representing a line of the file, with the
|
---|
| 3183 | last element being empty if the file is terminated with a newline.
|
---|
| 3184 | error: A callable to which errors are reported, which takes 4 arguments:
|
---|
| 3185 | filename, line number, error level, and message
|
---|
| 3186 | extra_check_functions: An array of additional check functions that will be
|
---|
| 3187 | run on each source line. Each function takes 4
|
---|
| 3188 | arguments: filename, clean_lines, line, error
|
---|
| 3189 | """
|
---|
| 3190 | lines = (['// marker so line numbers and indices both start at 1'] + lines +
|
---|
| 3191 | ['// marker so line numbers end in a known way'])
|
---|
| 3192 |
|
---|
| 3193 | include_state = _IncludeState()
|
---|
| 3194 | function_state = _FunctionState()
|
---|
| 3195 | class_state = _ClassState()
|
---|
| 3196 |
|
---|
| 3197 | ResetNolintSuppressions()
|
---|
| 3198 |
|
---|
| 3199 | CheckForCopyright(filename, lines, error)
|
---|
| 3200 |
|
---|
| 3201 | if file_extension == 'h':
|
---|
| 3202 | CheckForHeaderGuard(filename, lines, error)
|
---|
| 3203 |
|
---|
| 3204 | RemoveMultiLineComments(filename, lines, error)
|
---|
| 3205 | clean_lines = CleansedLines(lines)
|
---|
| 3206 | for line in range(clean_lines.NumLines()):
|
---|
| 3207 | ProcessLine(filename, file_extension, clean_lines, line,
|
---|
| 3208 | include_state, function_state, class_state, error,
|
---|
| 3209 | extra_check_functions)
|
---|
| 3210 | class_state.CheckFinished(filename, error)
|
---|
| 3211 |
|
---|
| 3212 | CheckForIncludeWhatYouUse(filename, clean_lines, include_state, error)
|
---|
| 3213 |
|
---|
| 3214 | # We check here rather than inside ProcessLine so that we see raw
|
---|
| 3215 | # lines rather than "cleaned" lines.
|
---|
| 3216 | CheckForUnicodeReplacementCharacters(filename, lines, error)
|
---|
| 3217 |
|
---|
| 3218 | CheckForNewlineAtEOF(filename, lines, error)
|
---|
| 3219 |
|
---|
| 3220 | def ProcessFile(filename, vlevel, extra_check_functions=[]):
|
---|
| 3221 | """Does google-lint on a single file.
|
---|
| 3222 |
|
---|
| 3223 | Args:
|
---|
| 3224 | filename: The name of the file to parse.
|
---|
| 3225 |
|
---|
| 3226 | vlevel: The level of errors to report. Every error of confidence
|
---|
| 3227 | >= verbose_level will be reported. 0 is a good default.
|
---|
| 3228 |
|
---|
| 3229 | extra_check_functions: An array of additional check functions that will be
|
---|
| 3230 | run on each source line. Each function takes 4
|
---|
| 3231 | arguments: filename, clean_lines, line, error
|
---|
| 3232 | """
|
---|
| 3233 |
|
---|
| 3234 | _SetVerboseLevel(vlevel)
|
---|
| 3235 |
|
---|
| 3236 | try:
|
---|
| 3237 | # Support the UNIX convention of using "-" for stdin. Note that
|
---|
| 3238 | # we are not opening the file with universal newline support
|
---|
| 3239 | # (which codecs doesn't support anyway), so the resulting lines do
|
---|
| 3240 | # contain trailing '\r' characters if we are reading a file that
|
---|
| 3241 | # has CRLF endings.
|
---|
| 3242 | # If after the split a trailing '\r' is present, it is removed
|
---|
| 3243 | # below. If it is not expected to be present (i.e. os.linesep !=
|
---|
| 3244 | # '\r\n' as in Windows), a warning is issued below if this file
|
---|
| 3245 | # is processed.
|
---|
| 3246 |
|
---|
| 3247 | if filename == '-':
|
---|
| 3248 | lines = codecs.StreamReaderWriter(sys.stdin,
|
---|
| 3249 | codecs.getreader('utf8'),
|
---|
| 3250 | codecs.getwriter('utf8'),
|
---|
| 3251 | 'replace').read().split('\n')
|
---|
| 3252 | else:
|
---|
| 3253 | lines = codecs.open(filename, 'r', 'utf8', 'replace').read().split('\n')
|
---|
| 3254 |
|
---|
| 3255 | carriage_return_found = False
|
---|
| 3256 | # Remove trailing '\r'.
|
---|
| 3257 | for linenum in range(len(lines)):
|
---|
| 3258 | if lines[linenum].endswith('\r'):
|
---|
| 3259 | lines[linenum] = lines[linenum].rstrip('\r')
|
---|
| 3260 | carriage_return_found = True
|
---|
| 3261 |
|
---|
| 3262 | except IOError:
|
---|
| 3263 | sys.stderr.write(
|
---|
| 3264 | "Skipping input '%s': Can't open for reading\n" % filename)
|
---|
| 3265 | return
|
---|
| 3266 |
|
---|
| 3267 | # Note, if no dot is found, this will give the entire filename as the ext.
|
---|
| 3268 | file_extension = filename[filename.rfind('.') + 1:]
|
---|
| 3269 |
|
---|
| 3270 | # When reading from stdin, the extension is unknown, so no cpplint tests
|
---|
| 3271 | # should rely on the extension.
|
---|
| 3272 | if (filename != '-' and file_extension not in EXTENSIONS):
|
---|
| 3273 | sys.stderr.write('Ignoring %s; extension not in %s\n' % (filename, EXTENSIONS))
|
---|
| 3274 | else:
|
---|
| 3275 | ProcessFileData(filename, file_extension, lines, Error,
|
---|
| 3276 | extra_check_functions)
|
---|
| 3277 | if carriage_return_found and os.linesep != '\r\n':
|
---|
| 3278 | # Use 0 for linenum since outputting only one error for potentially
|
---|
| 3279 | # several lines.
|
---|
| 3280 | Error(filename, 0, 'whitespace/newline', 1,
|
---|
| 3281 | 'One or more unexpected \\r (^M) found;'
|
---|
| 3282 | 'better to use only a \\n')
|
---|
| 3283 |
|
---|
| 3284 | sys.stderr.write('Done processing %s\n' % filename)
|
---|
| 3285 |
|
---|
| 3286 |
|
---|
| 3287 | def PrintUsage(message):
|
---|
| 3288 | """Prints a brief usage string and exits, optionally with an error message.
|
---|
| 3289 |
|
---|
| 3290 | Args:
|
---|
| 3291 | message: The optional error message.
|
---|
| 3292 | """
|
---|
| 3293 | sys.stderr.write(_USAGE)
|
---|
| 3294 |
|
---|
| 3295 | if message:
|
---|
| 3296 | sys.exit('\nFATAL ERROR: ' + message)
|
---|
| 3297 | else:
|
---|
| 3298 | sys.exit(1)
|
---|
| 3299 |
|
---|
| 3300 |
|
---|
| 3301 | def PrintCategories():
|
---|
| 3302 | """Prints a list of all the error-categories used by error messages.
|
---|
| 3303 |
|
---|
| 3304 | These are the categories used to filter messages via --filter.
|
---|
| 3305 | """
|
---|
| 3306 | sys.stderr.write(''.join(' %s\n' % cat for cat in _ERROR_CATEGORIES))
|
---|
| 3307 | sys.exit(0)
|
---|
| 3308 |
|
---|
| 3309 |
|
---|
| 3310 | def ParseArguments(args):
|
---|
| 3311 | """Parses the command line arguments.
|
---|
| 3312 |
|
---|
| 3313 | This may set the output format and verbosity level as side-effects.
|
---|
| 3314 |
|
---|
| 3315 | Args:
|
---|
| 3316 | args: The command line arguments:
|
---|
| 3317 |
|
---|
| 3318 | Returns:
|
---|
| 3319 | The list of filenames to lint.
|
---|
| 3320 | """
|
---|
| 3321 | try:
|
---|
| 3322 | (opts, filenames) = getopt.getopt(args, '', ['help', 'output=', 'verbose=',
|
---|
| 3323 | 'counting=',
|
---|
| 3324 | 'filter='])
|
---|
| 3325 | except getopt.GetoptError:
|
---|
| 3326 | PrintUsage('Invalid arguments.')
|
---|
| 3327 |
|
---|
| 3328 | verbosity = _VerboseLevel()
|
---|
| 3329 | output_format = _OutputFormat()
|
---|
| 3330 | filters = ''
|
---|
| 3331 | counting_style = ''
|
---|
| 3332 |
|
---|
| 3333 | for (opt, val) in opts:
|
---|
| 3334 | if opt == '--help':
|
---|
| 3335 | PrintUsage(None)
|
---|
| 3336 | elif opt == '--output':
|
---|
| 3337 | if not val in ('emacs', 'vs7'):
|
---|
| 3338 | PrintUsage('The only allowed output formats are emacs and vs7.')
|
---|
| 3339 | output_format = val
|
---|
| 3340 | elif opt == '--verbose':
|
---|
| 3341 | verbosity = int(val)
|
---|
| 3342 | elif opt == '--filter':
|
---|
| 3343 | filters = val
|
---|
| 3344 | if not filters:
|
---|
| 3345 | PrintCategories()
|
---|
| 3346 | elif opt == '--counting':
|
---|
| 3347 | if val not in ('total', 'toplevel', 'detailed'):
|
---|
| 3348 | PrintUsage('Valid counting options are total, toplevel, and detailed')
|
---|
| 3349 | counting_style = val
|
---|
| 3350 |
|
---|
| 3351 | if not filenames:
|
---|
| 3352 | PrintUsage('No files were specified.')
|
---|
| 3353 |
|
---|
| 3354 | _SetOutputFormat(output_format)
|
---|
| 3355 | _SetVerboseLevel(verbosity)
|
---|
| 3356 | _SetFilters(filters)
|
---|
| 3357 | _SetCountingStyle(counting_style)
|
---|
| 3358 |
|
---|
| 3359 | return filenames
|
---|
| 3360 |
|
---|
| 3361 |
|
---|
| 3362 | def main():
|
---|
| 3363 | filenames = ParseArguments(sys.argv[1:])
|
---|
| 3364 | backup_err = sys.stderr
|
---|
| 3365 | try:
|
---|
| 3366 | # Change stderr to write with replacement characters so we don't die
|
---|
| 3367 | # if we try to print something containing non-ASCII characters.
|
---|
| 3368 | sys.stderr = codecs.StreamReader(sys.stderr,
|
---|
| 3369 | 'replace')
|
---|
| 3370 | _cpplint_state.ResetErrorCounts()
|
---|
| 3371 | for filename in filenames:
|
---|
| 3372 | ProcessFile(filename, _cpplint_state.verbose_level)
|
---|
| 3373 | _cpplint_state.PrintErrorCounts()
|
---|
| 3374 | finally:
|
---|
| 3375 | sys.stderr = backup_err
|
---|
| 3376 |
|
---|
| 3377 | sys.exit(_cpplint_state.error_count > 0)
|
---|
| 3378 |
|
---|
| 3379 |
|
---|
| 3380 | if __name__ == '__main__':
|
---|
| 3381 | main()
|
---|