util/style.py

   1 #! /usr/bin/env python
   2 # Copyright (c) 2006 The Regents of The University of Michigan
   3 # Copyright (c) 2007 The Hewlett-Packard Development Company
   4 # All rights reserved.
   5 #
   6 # Redistribution and use in source and binary forms, with or without
   7 # modification, are permitted provided that the following conditions are
   8 # met: redistributions of source code must retain the above copyright
   9 # notice, this list of conditions and the following disclaimer;
  10 # redistributions in binary form must reproduce the above copyright
  11 # notice, this list of conditions and the following disclaimer in the
  12 # documentation and/or other materials provided with the distribution;
  13 # neither the name of the copyright holders nor the names of its
  14 # contributors may be used to endorse or promote products derived from
  15 # this software without specific prior written permission.
  16 #
  17 # THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
  18 # "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
  19 # LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
  20 # A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
  21 # OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
  22 # SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
  23 # LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
  24 # DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
  25 # THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
  26 # (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
  27 # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
  28 #
  29 # Authors: Nathan Binkert
  30
  31 import re
  32 import os
  33 import sys
  34
  35 sys.path.insert(0, os.path.dirname(__file__))
  36
  37 from file_types import lang_type
  38
  39 tabsize = 8
  40 lead = re.compile(r'^([ \t]+)')
  41 trail = re.compile(r'([ \t]+)$')
  42 any_control = re.compile(r'\b(if|while|for)[ \t]*[(]')
  43 good_control = re.compile(r'\b(if|while|for) [(]')
  44
  45 whitespace_types = set(('C', 'C++', 'swig', 'python', 'asm', 'isa', 'scons'))
  46 format_types = set(('C', 'C++'))
  47
  48 def checkwhite_line(line):
  49     match = lead.search(line)
  50     if match and match.group(1).find('\t') != -1:
  51         return False
  52
  53     match = trail.search(line)
  54     if match:
  55         return False
  56
  57     return True
  58
  59 def checkwhite(filename):
  60     if lang_type(filename) not in whitespace_types:
  61         return
  62
  63     try:
  64         f = file(filename, 'r+')
  65     except OSError, msg:
  66         print 'could not open file %s: %s' % (filename, msg)
  67         return
  68
  69     for num,line in enumerate(f):
  70         if not checkwhite_line(line):
  71             yield line,num + 1
  72
  73 def fixwhite_line(line):
  74     if lead.search(line):
  75         newline = ''
  76         for i,c in enumerate(line):
  77             if c == ' ':
  78                 newline += ' '
  79             elif c == '\t':
  80                 newline += ' ' * (tabsize - len(newline) % tabsize)
  81             else:
  82                 newline += line[i:]
  83                 break
  84
  85         line = newline
  86
  87     return line.rstrip() + '\n'
  88
  89 def fixwhite(filename, fixonly=None):
  90     if lang_type(filename) not in whitespace_types:
  91         return
  92
  93     try:
  94         f = file(filename, 'r+')
  95     except OSError, msg:
  96         print 'could not open file %s: %s' % (filename, msg)
  97         return
  98
  99     lines = list(f)
 100
 101     f.seek(0)
 102     f.truncate()
 103
 104     for i,line in enumerate(lines):
 105         if fixonly is None or i in fixonly:
 106             line = fixwhite_line(line)
 107
 108         print >>f, line,
 109
 110 def linelen(line):
 111     tabs = line.count('\t')
 112     if not tabs:
 113         return len(line)
 114
 115     count = 0
 116     for c in line:
 117         if c == '\t':
 118             count += tabsize - count % tabsize
 119         else:
 120             count += 1
 121
 122     return count
 123
 124 class ValidationStats(object):
 125     def __init__(self):
 126         self.toolong = 0
 127         self.toolong80 = 0
 128         self.leadtabs = 0
 129         self.trailwhite = 0
 130         self.badcontrol = 0
 131         self.cret = 0
 132
 133     def dump(self):
 134         print '''\
 135 %d violations of lines over 79 chars. %d of which are 80 chars exactly.
 136 %d cases of whitespace at the end of a line.
 137 %d cases of tabs to indent.
 138 %d bad parens after if/while/for.
 139 %d carriage returns found.
 140 ''' % (self.toolong, self.toolong80, self.trailwhite, self.leadtabs,
 141        self.badcontrol, self.cret)
 142
 143     def __nonzero__(self):
 144         return self.toolong or self.toolong80 or self.leadtabs or \
 145                self.trailwhite or self.badcontrol or self.cret
 146
 147 def validate(filename, stats, verbose, exit_code):
 148     if lang_type(filename) not in format_types:
 149         return
 150
 151     def msg(lineno, line, message):
 152         print '%s:%d>' % (filename, lineno + 1), message
 153         if verbose > 2:
 154             print line
 155
 156     def bad():
 157         if exit_code is not None:
 158             sys.exit(exit_code)
 159
 160     try:
 161         f = file(filename, 'r')
 162     except OSError:
 163         if verbose > 0:
 164             print 'could not open file %s' % filename
 165         bad()
 166         return
 167
 168     for i,line in enumerate(f):
 169         line = line.rstrip('\n')
 170
 171         # no carriage returns
 172         if line.find('\r') != -1:
 173             self.cret += 1
 174             if verbose > 1:
 175                 msg(i, line, 'carriage return found')
 176             bad()
 177
 178         # lines max out at 79 chars
 179         llen = linelen(line)
 180         if llen > 79:
 181             stats.toolong += 1
 182             if llen == 80:
 183                 stats.toolong80 += 1
 184             if verbose > 1:
 185                 msg(i, line, 'line too long (%d chars)' % llen)
 186             bad()
 187
 188         # no tabs used to indent
 189         match = lead.search(line)
 190         if match and match.group(1).find('\t') != -1:
 191             stats.leadtabs += 1
 192             if verbose > 1:
 193                 msg(i, line, 'using tabs to indent')
 194             bad()
 195
 196         # no trailing whitespace
 197         if trail.search(line):
 198             stats.trailwhite +=1
 199             if verbose > 1:
 200                 msg(i, line, 'trailing whitespace')
 201             bad()
 202
 203         # for c++, exactly one space betwen if/while/for and (
 204         if cpp:
 205             match = any_control.search(line)
 206             if match and not good_control.search(line):
 207                 stats.badcontrol += 1
 208                 if verbose > 1:
 209                     msg(i, line, 'improper spacing after %s' % match.group(1))
 210                 bad()
 211
 212 def modified_lines(old_data, new_data, max_lines):
 213     from itertools import count
 214     from mercurial import bdiff, mdiff
 215
 216     modified = set()
 217     counter = count()
 218     for pbeg, pend, fbeg, fend in bdiff.blocks(old_data, new_data):
 219         for i in counter:
 220             if i < fbeg:
 221                 modified.add(i)
 222             elif i + 1 >= fend:
 223                 break
 224             elif i > max_lines:
 225                 break
 226     return modified
 227
 228 def do_check_style(ui, repo, *files, **args):
 229     """check files for proper m5 style guidelines"""
 230     from mercurial import mdiff, util
 231
 232     if files:
 233         files = frozenset(files)
 234
 235     def skip(name):
 236         return files and name in files
 237
 238     def prompt(name, func, fixonly=None):
 239         if args.get('auto', False):
 240             result = 'f'
 241         else:
 242             while True:
 243                 result = ui.prompt("(a)bort, (i)gnore, or (f)ix?", default='a')
 244                 if result in 'aif':
 245                     break
 246
 247         if result == 'a':
 248             return True
 249         elif result == 'f':
 250             func(repo.wjoin(name), fixonly)
 251
 252         return False
 253
 254     modified, added, removed, deleted, unknown, ignore, clean = repo.status()
 255
 256     for fname in added:
 257         if skip(fname):
 258             continue
 259
 260         ok = True
 261         for line,num in checkwhite(repo.wjoin(fname)):
 262             ui.write("invalid whitespace in %s:%d\n" % (fname, num))
 263             if ui.verbose:
 264                 ui.write(">>%s<<\n" % line[-1])
 265             ok = False
 266
 267         if not ok:
 268             if prompt(fname, fixwhite):
 269                 return True
 270
 271     try:
 272         wctx = repo.workingctx()
 273     except:
 274         from mercurial import context
 275         wctx = context.workingctx(repo)
 276
 277     for fname in modified:
 278         if skip(fname):
 279             continue
 280
 281         if lang_type(fname) not in whitespace_types:
 282             continue
 283
 284         fctx = wctx.filectx(fname)
 285         pctx = fctx.parents()
 286
 287         file_data = fctx.data()
 288         lines = mdiff.splitnewlines(file_data)
 289         if len(pctx) in (1, 2):
 290             mod_lines = modified_lines(pctx[0].data(), file_data, len(lines))
 291             if len(pctx) == 2:
 292                 m2 = modified_lines(pctx[1].data(), file_data, len(lines))
 293                 # only the lines that are new in both
 294                 mod_lines = mod_lines & m2
 295         else:
 296             mod_lines = xrange(0, len(lines))
 297
 298         fixonly = set()
 299         for i,line in enumerate(lines):
 300             if i not in mod_lines:
 301                 continue
 302
 303             if checkwhite_line(line):
 304                 continue
 305
 306             ui.write("invalid whitespace: %s:%d\n" % (fname, i+1))
 307             if ui.verbose:
 308                 ui.write(">>%s<<\n" % line[:-1])
 309             fixonly.add(i)
 310
 311         if fixonly:
 312             if prompt(fname, fixwhite, fixonly):
 313                 return True
 314
 315 def do_check_format(ui, repo, **args):
 316     modified, added, removed, deleted, unknown, ignore, clean = repo.status()
 317
 318     verbose = 0
 319     stats = ValidationStats()
 320     for f in modified + added:
 321         validate(f, stats, verbose, None)
 322
 323     if stats:
 324         stats.dump()
 325         result = ui.prompt("invalid formatting\n(i)gnore or (a)bort?",
 326                            "^[ia]$", "a")
 327         if result.startswith('i'):
 328             pass
 329         elif result.startswith('a'):
 330             return True
 331         else:
 332             raise util.Abort(_("Invalid response: '%s'") % result)
 333
 334     return False
 335
 336 def check_hook(hooktype):
 337     if hooktype not in ('pretxncommit', 'pre-qrefresh'):
 338         raise AttributeError, \
 339               "This hook is not meant for %s" % hooktype
 340
 341 def check_style(ui, repo, hooktype, **kwargs):
 342     check_hook(hooktype)
 343     args = {}
 344
 345     try:
 346         return do_check_style(ui, repo, **args)
 347     except Exception, e:
 348         import traceback
 349         traceback.print_exc()
 350         return True
 351
 352 def check_format(ui, repo, hooktype, **kwargs):
 353     check_hook(hooktype)
 354     args = {}
 355
 356     try:
 357         return do_check_format(ui, repo, **args)
 358     except Exception, e:
 359         import traceback
 360         traceback.print_exc()
 361         return True
 362
 363 try:
 364     from mercurial.i18n import _
 365 except ImportError:
 366     def _(arg):
 367         return arg
 368
 369 cmdtable = {
 370     '^m5style' :
 371     ( do_check_style,
 372       [ ('a', 'auto', False, _("automatically fix whitespace")) ],
 373       _('hg m5style [-a] [FILE]...')),
 374     '^m5format' :
 375     ( do_check_format,
 376       [ ],
 377       _('hg m5format [FILE]...')),
 378 }
 379
 380 if __name__ == '__main__':
 381     import getopt
 382
 383     progname = sys.argv[0]
 384     if len(sys.argv) < 2:
 385         sys.exit('usage: %s <command> [<command args>]' % progname)
 386
 387     fixwhite_usage = '%s fixwhite [-t <tabsize> ] <path> [...] \n' % progname
 388     chkformat_usage = '%s chkformat <path> [...] \n' % progname
 389     chkwhite_usage = '%s chkwhite <path> [...] \n' % progname
 390
 391     command = sys.argv[1]
 392     if command == 'fixwhite':
 393         flags = 't:'
 394         usage = fixwhite_usage
 395     elif command == 'chkwhite':
 396         flags = 'nv'
 397         usage = chkwhite_usage
 398     elif command == 'chkformat':
 399         flags = 'nv'
 400         usage = chkformat_usage
 401     else:
 402         sys.exit(fixwhite_usage + chkwhite_usage + chkformat_usage)
 403
 404     opts, args = getopt.getopt(sys.argv[2:], flags)
 405
 406     code = 1
 407     verbose = 1
 408     for opt,arg in opts:
 409         if opt == '-n':
 410             code = None
 411         if opt == '-t':
 412             tabsize = int(arg)
 413         if opt == '-v':
 414             verbose += 1
 415
 416     if command == 'fixwhite':
 417         for filename in args:
 418             fixwhite(filename, tabsize)
 419     elif command == 'chkwhite':
 420         for filename in args:
 421             for line,num in checkwhite(filename):
 422                 print 'invalid whitespace: %s:%d' % (filename, num)
 423                 if verbose:
 424                     print '>>%s<<' % line[:-1]
 425     elif command == 'chkformat':
 426         stats = ValidationStats()
 427         for filename in args:
 428             validate(filename, stats=stats, verbose=verbose, exit_code=code)
 429
 430         if verbose > 0:
 431             stats.dump()
 432     else:
 433         sys.exit("command '%s' not found" % command)