1#! /usr/bin/env python
2# Copyright (c) 2014 ARM Limited
3# All rights reserved
4#
5# The license below extends only to copyright in the software and shall
6# not be construed as granting a license to any other intellectual
7# property including but not limited to intellectual property relating
8# to a hardware implementation of the functionality of the software
9# licensed hereunder. You may use the software subject to the license
10# terms below provided that you ensure that this notice is replicated
11# unmodified and in its entirety in all distributions of the software,
12# modified or unmodified, in source code or in binary form.
13#
14# Copyright (c) 2006 The Regents of The University of Michigan
15# Copyright (c) 2007,2011 The Hewlett-Packard Development Company
16# All rights reserved.
17#
18# Redistribution and use in source and binary forms, with or without
19# modification, are permitted provided that the following conditions are
20# met: redistributions of source code must retain the above copyright
21# notice, this list of conditions and the following disclaimer;
22# redistributions in binary form must reproduce the above copyright
23# notice, this list of conditions and the following disclaimer in the
24# documentation and/or other materials provided with the distribution;
25# neither the name of the copyright holders nor the names of its
26# contributors may be used to endorse or promote products derived from
27# this software without specific prior written permission.
28#
29# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
30# "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
31# LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
32# A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
33# OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
34# SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
35# LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
36# DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
37# THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
38# (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
39# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
40#
41# Authors: Nathan Binkert
42
43import heapq
44import os
45import re
46import sys
47
48from os.path import dirname, join as joinpath
49from itertools import count
50from mercurial import bdiff, mdiff, commands
51
52current_dir = dirname(__file__)
53sys.path.insert(0, current_dir)
54sys.path.insert(1, joinpath(dirname(current_dir), 'src', 'python'))
55
56from m5.util import neg_inf, pos_inf, Region, Regions
57import sort_includes
58from file_types import lang_type
59
60all_regions = Regions(Region(neg_inf, pos_inf))
61
62tabsize = 8
63lead = re.compile(r'^([ \t]+)')
64trail = re.compile(r'([ \t]+)$')
65any_control = re.compile(r'\b(if|while|for)[ \t]*[(]')
66good_control = re.compile(r'\b(if|while|for) [(]')
67
68format_types = set(('C', 'C++'))
69
70
71def re_ignore(expr):
72 """Helper function to create regular expression ignore file
73 matcher functions"""
74
75 rex = re.compile(expr)
76 def match_re(fname):
77 return rex.match(fname)
78 return match_re
79
80# This list contains a list of functions that are called to determine
81# if a file should be excluded from the style matching rules or
82# not. The functions are called with the file name relative to the
83# repository root (without a leading slash) as their argument. A file
84# is excluded if any function in the list returns true.
85style_ignores = [
86 # Ignore external projects as they are unlikely to follow the gem5
87 # coding convention.
88 re_ignore("^ext/"),
89]
90
91def check_ignores(fname):
92 """Check if a file name matches any of the ignore rules"""
93
94 for rule in style_ignores:
95 if rule(fname):
96 return True
97
98 return False
99
100
101def modified_regions(old_data, new_data):
102 regions = Regions()
103 beg = None
104 for pbeg, pend, fbeg, fend in bdiff.blocks(old_data, new_data):
105 if beg is not None and beg != fbeg:
106 regions.append(beg, fbeg)
107 beg = fend
108 return regions
109
110def modregions(wctx, fname):
111 fctx = wctx.filectx(fname)
112 pctx = fctx.parents()
113
114 file_data = fctx.data()
115 lines = mdiff.splitnewlines(file_data)
116 if len(pctx) in (1, 2):
117 mod_regions = modified_regions(pctx[0].data(), file_data)
118 if len(pctx) == 2:
119 m2 = modified_regions(pctx[1].data(), file_data)
120 # only the lines that are new in both
121 mod_regions &= m2
122 else:
123 mod_regions = Regions()
124 mod_regions.append(0, len(lines))
125
126 return mod_regions
127
128class UserInterface(object):
129 def __init__(self, verbose=False, auto=False):
130 self.auto = auto
131 self.verbose = verbose
132
133 def prompt(self, prompt, results, default):
134 if self.auto:
135 return self.auto
136
137 while True:
138 result = self.do_prompt(prompt, results, default)
139 if result in results:
140 return result
141
142class MercurialUI(UserInterface):
143 def __init__(self, ui, *args, **kwargs):
144 super(MercurialUI, self).__init__(*args, **kwargs)
145 self.ui = ui
146
147 def do_prompt(self, prompt, results, default):
148 return self.ui.prompt(prompt, default=default)
149
150 def write(self, string):
151 self.ui.write(string)
152
153class StdioUI(UserInterface):
154 def do_prompt(self, prompt, results, default):
155 return raw_input(prompt) or default
156
157 def write(self, string):
158 sys.stdout.write(string)
159
160class Verifier(object):
161 def __init__(self, ui, repo=None):
162 self.ui = ui
163 self.repo = repo
164 if repo is None:
165 self.wctx = None
166
167 def __getattr__(self, attr):
168 if attr in ('prompt', 'write'):
169 return getattr(self.ui, attr)
170
171 if attr == 'wctx':
172 try:
173 wctx = repo.workingctx()
174 except:
175 from mercurial import context
176 wctx = context.workingctx(repo)
177 self.wctx = wctx
178 return wctx
179
180 raise AttributeError
181
182 def open(self, filename, mode):
183 if self.repo:
184 filename = self.repo.wjoin(filename)
185
186 try:
187 f = file(filename, mode)
188 except OSError, msg:
189 print 'could not open file %s: %s' % (filename, msg)
190 return None
191
192 return f
193
194 def skip(self, filename):
195 # We never want to handle symlinks, so always skip them: If the location
196 # pointed to is a directory, skip it. If the location is a file inside
197 # the gem5 directory, it will be checked as a file, so symlink can be
198 # skipped. If the location is a file outside gem5, we don't want to
199 # check it anyway.
200 if os.path.islink(filename):
201 return True
202 return lang_type(filename) not in self.languages
203
204 def check(self, filename, regions=all_regions):
205 f = self.open(filename, 'r')
206
207 errors = 0
208 for num,line in enumerate(f):
209 if num not in regions:
210 continue
211 if not self.check_line(line):
212 self.write("invalid %s in %s:%d\n" % \
213 (self.test_name, filename, num + 1))
214 if self.ui.verbose:
215 self.write(">>%s<<\n" % line[-1])
216 errors += 1
217 return errors
218
219 def fix(self, filename, regions=all_regions):
220 f = self.open(filename, 'r+')
221
222 lines = list(f)
223
224 f.seek(0)
225 f.truncate()
226
227 for i,line in enumerate(lines):
228 if i in regions:
229 line = self.fix_line(line)
230
231 f.write(line)
232 f.close()
233
234 def apply(self, filename, prompt, regions=all_regions):
235 if not self.skip(filename):
236 errors = self.check(filename, regions)
237 if errors:
238 if prompt(filename, self.fix, regions):
239 return True
240 return False
241
242
243class Whitespace(Verifier):
244 languages = set(('C', 'C++', 'swig', 'python', 'asm', 'isa', 'scons'))
245 test_name = 'whitespace'
246 def check_line(self, line):
247 match = lead.search(line)
248 if match and match.group(1).find('\t') != -1:
249 return False
250
251 match = trail.search(line)
252 if match:
253 return False
254
255 return True
256
257 def fix_line(self, line):
258 if lead.search(line):
259 newline = ''
260 for i,c in enumerate(line):
261 if c == ' ':
262 newline += ' '
263 elif c == '\t':
264 newline += ' ' * (tabsize - len(newline) % tabsize)
265 else:
266 newline += line[i:]
267 break
268
269 line = newline
270
271 return line.rstrip() + '\n'
272
273class SortedIncludes(Verifier):
274 languages = sort_includes.default_languages
275 def __init__(self, *args, **kwargs):
276 super(SortedIncludes, self).__init__(*args, **kwargs)
277 self.sort_includes = sort_includes.SortIncludes()
278
279 def check(self, filename, regions=all_regions):
280 f = self.open(filename, 'r')
281
282 lines = [ l.rstrip('\n') for l in f.xreadlines() ]
283 old = ''.join(line + '\n' for line in lines)
284 f.close()
285
286 if len(lines) == 0:
287 return 0
288
289 language = lang_type(filename, lines[0])
290 sort_lines = list(self.sort_includes(lines, filename, language))
291 new = ''.join(line + '\n' for line in sort_lines)
292
293 mod = modified_regions(old, new)
294 modified = mod & regions
295
296 if modified:
297 self.write("invalid sorting of includes in %s\n" % (filename))
298 if self.ui.verbose:
299 for start, end in modified.regions:
300 self.write("bad region [%d, %d)\n" % (start, end))
301 return 1
302
303 return 0
304
305 def fix(self, filename, regions=all_regions):
306 f = self.open(filename, 'r+')
307
308 old = f.readlines()
309 lines = [ l.rstrip('\n') for l in old ]
310 language = lang_type(filename, lines[0])
311 sort_lines = list(self.sort_includes(lines, filename, language))
312 new = ''.join(line + '\n' for line in sort_lines)
313
314 f.seek(0)
315 f.truncate()
316
317 for i,line in enumerate(sort_lines):
318 f.write(line)
319 f.write('\n')
320 f.close()
321
322def linelen(line):
323 tabs = line.count('\t')
324 if not tabs:
325 return len(line)
326
327 count = 0
328 for c in line:
329 if c == '\t':
330 count += tabsize - count % tabsize
331 else:
332 count += 1
333
334 return count
335
336class ValidationStats(object):
337 def __init__(self):
338 self.toolong = 0
339 self.toolong80 = 0
340 self.leadtabs = 0
341 self.trailwhite = 0
342 self.badcontrol = 0
343 self.cret = 0
344
345 def dump(self):
346 print '''\
347%d violations of lines over 79 chars. %d of which are 80 chars exactly.
348%d cases of whitespace at the end of a line.
349%d cases of tabs to indent.
350%d bad parens after if/while/for.
351%d carriage returns found.
352''' % (self.toolong, self.toolong80, self.trailwhite, self.leadtabs,
353 self.badcontrol, self.cret)
354
355 def __nonzero__(self):
356 return self.toolong or self.toolong80 or self.leadtabs or \
357 self.trailwhite or self.badcontrol or self.cret
358
359def validate(filename, stats, verbose, exit_code):
360 if lang_type(filename) not in format_types:
361 return
362
363 def msg(lineno, line, message):
364 print '%s:%d>' % (filename, lineno + 1), message
365 if verbose > 2:
366 print line
367
368 def bad():
369 if exit_code is not None:
370 sys.exit(exit_code)
371
372 try:
373 f = file(filename, 'r')
374 except OSError:
375 if verbose > 0:
376 print 'could not open file %s' % filename
377 bad()
378 return
379
380 for i,line in enumerate(f):
381 line = line.rstrip('\n')
382
383 # no carriage returns
384 if line.find('\r') != -1:
385 self.cret += 1
386 if verbose > 1:
387 msg(i, line, 'carriage return found')
388 bad()
389
390 # lines max out at 79 chars
391 llen = linelen(line)
392 if llen > 79:
393 stats.toolong += 1
394 if llen == 80:
395 stats.toolong80 += 1
396 if verbose > 1:
397 msg(i, line, 'line too long (%d chars)' % llen)
398 bad()
399
400 # no tabs used to indent
401 match = lead.search(line)
402 if match and match.group(1).find('\t') != -1:
403 stats.leadtabs += 1
404 if verbose > 1:
405 msg(i, line, 'using tabs to indent')
406 bad()
407
408 # no trailing whitespace
409 if trail.search(line):
410 stats.trailwhite +=1
411 if verbose > 1:
412 msg(i, line, 'trailing whitespace')
413 bad()
414
415 # for c++, exactly one space betwen if/while/for and (
416 if cpp:
417 match = any_control.search(line)
418 if match and not good_control.search(line):
419 stats.badcontrol += 1
420 if verbose > 1:
421 msg(i, line, 'improper spacing after %s' % match.group(1))
422 bad()
423
424
425def do_check_style(hgui, repo, *pats, **opts):
426 """check files for proper m5 style guidelines
427
428 Without an argument, checks all modified and added files for gem5
429 coding style violations. A list of files can be specified to limit
430 the checker to a subset of the repository. The style rules are
431 normally applied on a diff of the repository state (i.e., added
432 files are checked in their entirety while only modifications of
433 modified files are checked).
434
435 The --all option can be specified to include clean files and check
436 modified files in their entirety.
437 """
438 from mercurial import mdiff, util
439
440 opt_fix_white = opts.get('fix_white', False)
441 opt_all = opts.get('all', False)
442 opt_no_ignore = opts.get('no_ignore', False)
443 ui = MercurialUI(hgui, hgui.verbose, opt_fix_white)
444
445 def prompt(name, func, regions=all_regions):
446 result = ui.prompt("(a)bort, (i)gnore, or (f)ix?", 'aif', 'a')
447 if result == 'a':
448 return True
449 elif result == 'f':
450 func(repo.wjoin(name), regions)
451
452 return False
453
454
455 # Import the match (repository file name matching helper)
456 # function. Different versions of Mercurial keep it in different
457 # modules and implement them differently.
458 try:
459 from mercurial import scmutil
460 m = scmutil.match(repo[None], pats, opts)
461 except ImportError:
462 from mercurial import cmdutil
463 m = cmdutil.match(repo, pats, opts)
464
465 modified, added, removed, deleted, unknown, ignore, clean = \
466 repo.status(match=m, clean=opt_all)
467 if not opt_all:
468 try:
469 wctx = repo.workingctx()
470 except:
471 from mercurial import context
472 wctx = context.workingctx(repo)
473
474 files = [ (fn, all_regions) for fn in added ] + \
475 [ (fn, modregions(wctx, fn)) for fn in modified ]
476 else:
477 files = [ (fn, all_regions) for fn in added + modified + clean ]
478
479 whitespace = Whitespace(ui)
480 sorted_includes = SortedIncludes(ui)
481 for fname, mod_regions in files:
482 if not opt_no_ignore and check_ignores(fname):
483 continue
484
485 fpath = joinpath(repo.root, fname)
486
487 if whitespace.apply(fpath, prompt, mod_regions):
488 return True
489
490 if sorted_includes.apply(fpath, prompt, mod_regions):
491 return True
492
493 return False
494
495def do_check_format(hgui, repo, **args):
496 ui = MercurialUI(hgui, hgui.verbose, auto)
497
498 modified, added, removed, deleted, unknown, ignore, clean = repo.status()
499
500 verbose = 0
501 stats = ValidationStats()
502 for f in modified + added:
503 validate(joinpath(repo.root, f), stats, verbose, None)
504
505 if stats:
506 stats.dump()
507 result = ui.prompt("invalid formatting\n(i)gnore or (a)bort?",
508 'ai', 'a')
509 if result == 'a':
510 return True
511
512 return False
513
514def check_hook(hooktype):
515 if hooktype not in ('pretxncommit', 'pre-qrefresh'):
516 raise AttributeError, \
517 "This hook is not meant for %s" % hooktype
518
519def check_style(ui, repo, hooktype, **kwargs):
520 check_hook(hooktype)
521 args = {}
522
523 try:
524 return do_check_style(ui, repo, **args)
525 except Exception, e:
526 import traceback
527 traceback.print_exc()
528 return True
529
530def check_format(ui, repo, hooktype, **kwargs):
531 check_hook(hooktype)
532 args = {}
533
534 try:
535 return do_check_format(ui, repo, **args)
536 except Exception, e:
537 import traceback
538 traceback.print_exc()
539 return True
540
541try:
542 from mercurial.i18n import _
543except ImportError:
544 def _(arg):
545 return arg
546
547cmdtable = {
548 '^m5style' : (
549 do_check_style, [
550 ('w', 'fix-white', False, _("automatically fix whitespace")),
551 ('a', 'all', False,
552 _("include clean files and unmodified parts of modified files")),
553 ('', 'no-ignore', False, _("ignore the style ignore list")),
554 ] + commands.walkopts,
555 _('hg m5style [-a] [FILE]...')),
556 '^m5format' :
557 ( do_check_format,
558 [ ],
559 _('hg m5format [FILE]...')),
560}
561
562if __name__ == '__main__':
563 import getopt
564
565 progname = sys.argv[0]
566 if len(sys.argv) < 2:
567 sys.exit('usage: %s <command> [<command args>]' % progname)
568
569 fixwhite_usage = '%s fixwhite [-t <tabsize> ] <path> [...] \n' % progname
570 chkformat_usage = '%s chkformat <path> [...] \n' % progname
571 chkwhite_usage = '%s chkwhite <path> [...] \n' % progname
572
573 command = sys.argv[1]
574 if command == 'fixwhite':
575 flags = 't:'
576 usage = fixwhite_usage
577 elif command == 'chkwhite':
578 flags = 'nv'
579 usage = chkwhite_usage
580 elif command == 'chkformat':
581 flags = 'nv'
582 usage = chkformat_usage
583 else:
584 sys.exit(fixwhite_usage + chkwhite_usage + chkformat_usage)
585
586 opts, args = getopt.getopt(sys.argv[2:], flags)
587
588 code = 1
589 verbose = 1
590 for opt,arg in opts:
591 if opt == '-n':
592 code = None
593 if opt == '-t':
594 tabsize = int(arg)
595 if opt == '-v':
596 verbose += 1
597
598 if command == 'fixwhite':
599 for filename in args:
600 fixwhite(filename, tabsize)
601 elif command == 'chkwhite':
602 for filename in args:
603 for line,num in checkwhite(filename):
604 print 'invalid whitespace: %s:%d' % (filename, num)
605 if verbose:
606 print '>>%s<<' % line[:-1]
607 elif command == 'chkformat':
608 stats = ValidationStats()
609 for filename in args:
610 validate(filename, stats=stats, verbose=verbose, exit_code=code)
611
612 if verbose > 0:
613 stats.dump()
614 else:
615 sys.exit("command '%s' not found" % command)