etc/update-copyright.py

   1 #!/usr/bin/env python3
   2 #
   3 # Copyright (C) 2013-2024 Free Software Foundation, Inc.
   4 #
   5 # This script is free software; you can redistribute it and/or modify
   6 # it under the terms of the GNU General Public License as published by
   7 # the Free Software Foundation; either version 3, or (at your option)
   8 # any later version.
   9
  10 # This script adjusts the copyright notices at the top of source files
  11 # so that they have the form:
  12 #
  13 #   Copyright XXXX-YYYY Free Software Foundation, Inc.
  14 #
  15 # It doesn't change code that is known to be maintained elsewhere or
  16 # that carries a non-FSF copyright.
  17 #
  18 # Pass --this-year to the script if you want it to add the current year
  19 # to all applicable notices.  Pass --quilt if you are using quilt and
  20 # want files to be added to the quilt before being changed.
  21 #
  22 # By default the script will update all directories for which the
  23 # output has been vetted.  You can instead pass the names of individual
  24 # directories, including those that haven't been approved.  So:
  25 #
  26 #    update-copyright.py --this-year
  27 #
  28 # is the command that would be used at the beginning of a year to update
  29 # all copyright notices (and possibly at other times to check whether
  30 # new files have been added with old years).  On the other hand:
  31 #
  32 #    update-copyright.py --this-year libiberty
  33 #
  34 # would run the script on just libiberty/.
  35 #
  36 # This script was copied from gcc's contrib/ and modified to suit
  37 # binutils.  In contrast to the gcc script, this one will update
  38 # the testsuite and --version output strings too.
  39
  40 import os
  41 import re
  42 import sys
  43 import time
  44 import subprocess
  45
  46 class Errors:
  47     def __init__ (self):
  48         self.num_errors = 0
  49
  50     def report (self, filename, string):
  51         if filename:
  52             string = filename + ': ' + string
  53         sys.stderr.write (string + '\n')
  54         self.num_errors += 1
  55
  56     def ok (self):
  57         return self.num_errors == 0
  58
  59 class GenericFilter:
  60     def __init__ (self):
  61         self.skip_files = set()
  62         self.skip_dirs = set()
  63         self.skip_extensions = set([
  64                 '.png',
  65                 '.pyc',
  66                 ])
  67         self.fossilised_files = set()
  68         self.own_files = set()
  69
  70         self.skip_files |= set ([
  71                 # Skip licence files.
  72                 'COPYING',
  73                 'COPYING.LIB',
  74                 'COPYING3',
  75                 'COPYING3.LIB',
  76                 'COPYING.LIBGLOSS',
  77                 'COPYING.NEWLIB',
  78                 'LICENSE',
  79                 'fdl.texi',
  80                 'gpl_v3.texi',
  81                 'fdl-1.3.xml',
  82                 'gpl-3.0.xml',
  83
  84                 # Skip auto- and libtool-related files
  85                 'aclocal.m4',
  86                 'compile',
  87                 'config.guess',
  88                 'config.sub',
  89                 'depcomp',
  90                 'install-sh',
  91                 'libtool.m4',
  92                 'ltmain.sh',
  93                 'ltoptions.m4',
  94                 'ltsugar.m4',
  95                 'ltversion.m4',
  96                 'lt~obsolete.m4',
  97                 'missing',
  98                 'mkdep',
  99                 'mkinstalldirs',
 100                 'move-if-change',
 101                 'shlibpath.m4',
 102                 'symlink-tree',
 103                 'ylwrap',
 104
 105                 # Skip FSF mission statement, etc.
 106                 'gnu.texi',
 107                 'funding.texi',
 108                 'appendix_free.xml',
 109
 110                 # Skip imported texinfo files.
 111                 'texinfo.tex',
 112                 ])
 113
 114         self.skip_extensions |= set ([
 115                 # Maintained by the translation project.
 116                 '.po',
 117
 118                 # Automatically-generated.
 119                 '.pot',
 120                 ])
 121
 122         self.skip_dirs |= set ([
 123                 'autom4te.cache',
 124                 ])
 125
 126
 127     def get_line_filter (self, dir, filename):
 128         if filename.startswith ('ChangeLog'):
 129             # Ignore references to copyright in changelog entries.
 130             return re.compile ('\t')
 131
 132         return None
 133
 134     def skip_file (self, dir, filename):
 135         if filename in self.skip_files:
 136             return True
 137
 138         (base, extension) = os.path.splitext (os.path.join (dir, filename))
 139         if extension in self.skip_extensions:
 140             return True
 141
 142         if extension == '.in':
 143             # Skip .in files produced by automake.
 144             if os.path.exists (base + '.am'):
 145                 return True
 146
 147             # Skip files produced by autogen
 148             if (os.path.exists (base + '.def')
 149                 and os.path.exists (base + '.tpl')):
 150                 return True
 151
 152         # Skip configure files produced by autoconf
 153         if filename == 'configure':
 154             if os.path.exists (base + '.ac'):
 155                 return True
 156             if os.path.exists (base + '.in'):
 157                 return True
 158
 159         return False
 160
 161     def skip_dir (self, dir, subdir):
 162         return subdir in self.skip_dirs
 163
 164     def is_fossilised_file (self, dir, filename):
 165         if filename in self.fossilised_files:
 166             return True
 167         # Only touch current current ChangeLogs.
 168         if filename != 'ChangeLog' and filename.find ('ChangeLog') >= 0:
 169             return True
 170         return False
 171
 172     def by_package_author (self, dir, filename):
 173         return filename in self.own_files
 174
 175 class Copyright:
 176     def __init__ (self, errors):
 177         self.errors = errors
 178
 179         # Characters in a range of years.  Include '.' for typos.
 180         ranges = '[0-9](?:[-0-9.,\s]|\s+and\s+)*[0-9]'
 181
 182         # Non-whitespace characters in a copyright holder's name.
 183         name = '[\w.,-]'
 184
 185         # Matches one year.
 186         self.year_re = re.compile ('[0-9]+')
 187
 188         # Matches part of a year or copyright holder.
 189         self.continuation_re = re.compile (ranges + '|' + name)
 190
 191         # Matches a full copyright notice:
 192         self.copyright_re = re.compile (
 193             # 1: 'Copyright (C)', etc.
 194             '([Cc]opyright'
 195             '|[Cc]opyright\s+\([Cc]\)'
 196             '|[Cc]opyright\s+%s'
 197             '|[Cc]opyright\s+&copy;'
 198             '|[Cc]opyright\s+@copyright{}'
 199             '|@set\s+copyright[\w-]+)'
 200
 201             # 2: the years.  Include the whitespace in the year, so that
 202             # we can remove any excess.
 203             '(\s*(?:' + ranges + ',?'
 204             '|@value\{[^{}]*\})\s*)'
 205
 206             # 3: 'by ', if used
 207             '(by\s+)?'
 208
 209             # 4: the copyright holder.  Don't allow multiple consecutive
 210             # spaces, so that right-margin gloss doesn't get caught
 211             # (e.g. gnat_ugn.texi).
 212             '(' + name + '(?:\s?' + name + ')*)?')
 213
 214         # A regexp for notices that might have slipped by.  Just matching
 215         # 'copyright' is too noisy, and 'copyright.*[0-9]' falls foul of
 216         # HTML header markers, so check for 'copyright' and two digits.
 217         self.other_copyright_re = re.compile ('(^|[^\._])copyright[^=]*[0-9][0-9]',
 218                                               re.IGNORECASE)
 219         self.comment_re = re.compile('#+|[*]+|;+|%+|//+|@c |dnl ')
 220         self.holders = { '@copying': '@copying' }
 221         self.holder_prefixes = set()
 222
 223         # True to 'quilt add' files before changing them.
 224         self.use_quilt = False
 225
 226         # If set, force all notices to include this year.
 227         self.max_year = None
 228
 229         # Goes after the year(s).  Could be ', '.
 230         self.separator = ' '
 231
 232     def add_package_author (self, holder, canon_form = None):
 233         if not canon_form:
 234             canon_form = holder
 235         self.holders[holder] = canon_form
 236         index = holder.find (' ')
 237         while index >= 0:
 238             self.holder_prefixes.add (holder[:index])
 239             index = holder.find (' ', index + 1)
 240
 241     def add_external_author (self, holder):
 242         self.holders[holder] = None
 243
 244     class BadYear (Exception):
 245         def __init__ (self, year):
 246             self.year = year
 247
 248         def __str__ (self):
 249             return 'unrecognised year: ' + self.year
 250
 251     def parse_year (self, string):
 252         year = int (string)
 253         if len (string) == 2:
 254             if year > 70:
 255                 return year + 1900
 256         elif len (string) == 4:
 257             return year
 258         raise self.BadYear (string)
 259
 260     def year_range (self, years):
 261         year_list = [self.parse_year (year)
 262                      for year in self.year_re.findall (years)]
 263         assert len (year_list) > 0
 264         return (min (year_list), max (year_list))
 265
 266     def set_use_quilt (self, use_quilt):
 267         self.use_quilt = use_quilt
 268
 269     def include_year (self, year):
 270         assert not self.max_year
 271         self.max_year = year
 272
 273     def canonicalise_years (self, dir, filename, filter, years):
 274         # Leave texinfo variables alone.
 275         if years.startswith ('@value'):
 276             return years
 277
 278         (min_year, max_year) = self.year_range (years)
 279
 280         # Update the upper bound, if enabled.
 281         if self.max_year and not filter.is_fossilised_file (dir, filename):
 282             max_year = max (max_year, self.max_year)
 283
 284         # Use a range.
 285         if min_year == max_year:
 286             return '%d' % min_year
 287         else:
 288             return '%d-%d' % (min_year, max_year)
 289
 290     def strip_continuation (self, line):
 291         line = line.lstrip()
 292         match = self.comment_re.match (line)
 293         if match:
 294             line = line[match.end():].lstrip()
 295         return line
 296
 297     def is_complete (self, match):
 298         holder = match.group (4)
 299         return (holder
 300                 and (holder not in self.holder_prefixes
 301                      or holder in self.holders))
 302
 303     def update_copyright (self, dir, filename, filter, file, line, match):
 304         orig_line = line
 305         next_line = None
 306         pathname = os.path.join (dir, filename)
 307
 308         intro = match.group (1)
 309         if intro.startswith ('@set'):
 310             # Texinfo year variables should always be on one line
 311             after_years = line[match.end (2):].strip()
 312             if after_years != '':
 313                 self.errors.report (pathname,
 314                                     'trailing characters in @set: '
 315                                     + after_years)
 316                 return (False, orig_line, next_line)
 317         else:
 318             # If it looks like the copyright is incomplete, add the next line.
 319             while not self.is_complete (match):
 320                 try:
 321                     next_line = file.readline()
 322                 except StopIteration:
 323                     break
 324
 325                 # If the next line doesn't look like a proper continuation,
 326                 # assume that what we've got is complete.
 327                 continuation = self.strip_continuation (next_line)
 328                 if not self.continuation_re.match (continuation):
 329                     break
 330
 331                 # Merge the lines for matching purposes.
 332                 orig_line += next_line
 333                 line = line.rstrip() + ' ' + continuation
 334                 next_line = None
 335
 336                 # Rematch with the longer line, at the original position.
 337                 match = self.copyright_re.match (line, match.start())
 338                 assert match
 339
 340             holder = match.group (4)
 341
 342             # Use the filter to test cases where markup is getting in the way.
 343             if filter.by_package_author (dir, filename):
 344                 assert holder not in self.holders
 345
 346             elif not holder:
 347                 self.errors.report (pathname, 'missing copyright holder')
 348                 return (False, orig_line, next_line)
 349
 350             elif holder not in self.holders:
 351                 self.errors.report (pathname,
 352                                     'unrecognised copyright holder: ' + holder)
 353                 return (False, orig_line, next_line)
 354
 355             else:
 356                 # See whether the copyright is associated with the package
 357                 # author.
 358                 canon_form = self.holders[holder]
 359                 if not canon_form:
 360                     return (False, orig_line, next_line)
 361
 362                 # Make sure the author is given in a consistent way.
 363                 line = (line[:match.start (4)]
 364                         + canon_form
 365                         + line[match.end (4):])
 366
 367                 # Remove any 'by'
 368                 line = line[:match.start (3)] + line[match.end (3):]
 369
 370         # Update the copyright years.
 371         years = match.group (2).strip()
 372         if (self.max_year
 373             and match.start(0) > 0 and line[match.start(0)-1] == '"'
 374             and not filter.is_fossilised_file (dir, filename)):
 375             # A printed copyright date consists of the current year
 376             canon_form = '%d' % self.max_year
 377         else:
 378             try:
 379                 canon_form = self.canonicalise_years (dir, filename, filter, years)
 380             except self.BadYear as e:
 381                 self.errors.report (pathname, str (e))
 382                 return (False, orig_line, next_line)
 383
 384         line = (line[:match.start (2)]
 385                 + ' ' + canon_form + self.separator
 386                 + line[match.end (2):])
 387
 388         # Use the standard (C) form.
 389         if intro.endswith ('right'):
 390             intro += ' (C)'
 391         elif intro.endswith ('(c)'):
 392             intro = intro[:-3] + '(C)'
 393         line = line[:match.start (1)] + intro + line[match.end (1):]
 394
 395         # Strip trailing whitespace
 396         line = line.rstrip() + '\n'
 397
 398         return (line != orig_line, line, next_line)
 399
 400     def guess_encoding (self, pathname):
 401         for encoding in ('utf8', 'iso8859'):
 402             try:
 403                 open(pathname, 'r', encoding=encoding).read()
 404                 return encoding
 405             except UnicodeDecodeError:
 406                 pass
 407         return None
 408
 409     def process_file (self, dir, filename, filter):
 410         pathname = os.path.join (dir, filename)
 411         if filename.endswith ('.tmp'):
 412             # Looks like something we tried to create before.
 413             try:
 414                 os.remove (pathname)
 415             except OSError:
 416                 pass
 417             return
 418
 419         lines = []
 420         changed = False
 421         line_filter = filter.get_line_filter (dir, filename)
 422         mode = None
 423         encoding = self.guess_encoding(pathname)
 424         with open (pathname, 'r', encoding=encoding) as file:
 425             prev = None
 426             mode = os.fstat (file.fileno()).st_mode
 427             for line in file:
 428                 while line:
 429                     next_line = None
 430                     # Leave filtered-out lines alone.
 431                     if not (line_filter and line_filter.match (line)):
 432                         match = self.copyright_re.search (line)
 433                         if match:
 434                             res = self.update_copyright (dir, filename, filter,
 435                                                          file, line, match)
 436                             (this_changed, line, next_line) = res
 437                             changed = changed or this_changed
 438
 439                         # Check for copyright lines that might have slipped by.
 440                         elif self.other_copyright_re.search (line):
 441                             self.errors.report (pathname,
 442                                                 'unrecognised copyright: %s'
 443                                                 % line.strip())
 444                     lines.append (line)
 445                     line = next_line
 446
 447         # If something changed, write the new file out.
 448         if changed and self.errors.ok():
 449             tmp_pathname = pathname + '.tmp'
 450             with open (tmp_pathname, 'w', encoding=encoding) as file:
 451                 for line in lines:
 452                     file.write (line)
 453                 os.fchmod (file.fileno(), mode)
 454             if self.use_quilt:
 455                 subprocess.call (['quilt', 'add', pathname])
 456             os.rename (tmp_pathname, pathname)
 457
 458     def process_tree (self, tree, filter):
 459         for (dir, subdirs, filenames) in os.walk (tree):
 460             # Don't recurse through directories that should be skipped.
 461             for i in range (len (subdirs) - 1, -1, -1):
 462                 if filter.skip_dir (dir, subdirs[i]):
 463                     del subdirs[i]
 464
 465             # Handle the files in this directory.
 466             for filename in filenames:
 467                 if filter.skip_file (dir, filename):
 468                     sys.stdout.write ('Skipping %s\n'
 469                                       % os.path.join (dir, filename))
 470                 else:
 471                     self.process_file (dir, filename, filter)
 472
 473 class CmdLine:
 474     def __init__ (self, copyright = Copyright):
 475         self.errors = Errors()
 476         self.copyright = copyright (self.errors)
 477         self.dirs = []
 478         self.default_dirs = []
 479         self.chosen_dirs = []
 480         self.option_handlers = dict()
 481         self.option_help = []
 482
 483         self.add_option ('--help', 'Print this help', self.o_help)
 484         self.add_option ('--quilt', '"quilt add" files before changing them',
 485                          self.o_quilt)
 486         self.add_option ('--this-year', 'Add the current year to every notice',
 487                          self.o_this_year)
 488
 489     def add_option (self, name, help, handler):
 490         self.option_help.append ((name, help))
 491         self.option_handlers[name] = handler
 492
 493     def add_dir (self, dir, filter = GenericFilter()):
 494         self.dirs.append ((dir, filter))
 495
 496     def o_help (self, option = None):
 497         sys.stdout.write ('Usage: %s [options] dir1 dir2...\n\n'
 498                           'Options:\n' % sys.argv[0])
 499         format = '%-15s %s\n'
 500         for (what, help) in self.option_help:
 501             sys.stdout.write (format % (what, help))
 502         sys.stdout.write ('\nDirectories:\n')
 503
 504         format = '%-25s'
 505         i = 0
 506         for (dir, filter) in self.dirs:
 507             i += 1
 508             if i % 3 == 0 or i == len (self.dirs):
 509                 sys.stdout.write (dir + '\n')
 510             else:
 511                 sys.stdout.write (format % dir)
 512         sys.exit (0)
 513
 514     def o_quilt (self, option):
 515         self.copyright.set_use_quilt (True)
 516
 517     def o_this_year (self, option):
 518         self.copyright.include_year (time.localtime().tm_year)
 519
 520     def main (self):
 521         for arg in sys.argv[1:]:
 522             if arg[:1] != '-':
 523                 self.chosen_dirs.append (arg)
 524             elif arg in self.option_handlers:
 525                 self.option_handlers[arg] (arg)
 526             else:
 527                 self.errors.report (None, 'unrecognised option: ' + arg)
 528         if self.errors.ok():
 529             if len (self.chosen_dirs) == 0:
 530                 self.chosen_dirs = self.default_dirs
 531             if len (self.chosen_dirs) == 0:
 532                 self.o_help()
 533             else:
 534                 for chosen_dir in self.chosen_dirs:
 535                     canon_dir = os.path.join (chosen_dir, '')
 536                     count = 0
 537                     for (dir, filter) in self.dirs:
 538                         if (dir + os.sep).startswith (canon_dir):
 539                             count += 1
 540                             self.copyright.process_tree (dir, filter)
 541                     if count == 0:
 542                         self.errors.report (None, 'unrecognised directory: '
 543                                             + chosen_dir)
 544         sys.exit (0 if self.errors.ok() else 1)
 545
 546 #----------------------------------------------------------------------------
 547
 548 class TopLevelFilter (GenericFilter):
 549     def skip_dir (self, dir, subdir):
 550         return True
 551
 552 class ConfigFilter (GenericFilter):
 553     def __init__ (self):
 554         GenericFilter.__init__ (self)
 555
 556     def skip_file (self, dir, filename):
 557         if filename.endswith ('.m4'):
 558             pathname = os.path.join (dir, filename)
 559             with open (pathname) as file:
 560                 # Skip files imported from gettext.
 561                 if file.readline().find ('gettext-') >= 0:
 562                     return True
 563         return GenericFilter.skip_file (self, dir, filename)
 564
 565 class LdFilter (GenericFilter):
 566     def __init__ (self):
 567         GenericFilter.__init__ (self)
 568
 569         self.skip_extensions |= set ([
 570                 # ld testsuite output match files.
 571                 '.ro',
 572                 ])
 573
 574 class BinutilsCopyright (Copyright):
 575     def __init__ (self, errors):
 576         Copyright.__init__ (self, errors)
 577
 578         canon_fsf = 'Free Software Foundation, Inc.'
 579         self.add_package_author ('Free Software Foundation', canon_fsf)
 580         self.add_package_author ('Free Software Foundation.', canon_fsf)
 581         self.add_package_author ('Free Software Foundation Inc.', canon_fsf)
 582         self.add_package_author ('Free Software Foundation, Inc', canon_fsf)
 583         self.add_package_author ('Free Software Foundation, Inc.', canon_fsf)
 584         self.add_package_author ('The Free Software Foundation', canon_fsf)
 585         self.add_package_author ('The Free Software Foundation, Inc.', canon_fsf)
 586         self.add_package_author ('Software Foundation, Inc.', canon_fsf)
 587
 588         self.add_external_author ('Carnegie Mellon University')
 589         self.add_external_author ('John D. Polstra.')
 590         self.add_external_author ('Innovative Computing Labs')
 591         self.add_external_author ('Linaro Ltd.')
 592         self.add_external_author ('MIPS Computer Systems, Inc.')
 593         self.add_external_author ('Red Hat Inc.')
 594         self.add_external_author ('Regents of the University of California.')
 595         self.add_external_author ('The Regents of the University of California.')
 596         self.add_external_author ('Third Eye Software, Inc.')
 597         self.add_external_author ('Ulrich Drepper')
 598         self.add_external_author ('Synopsys Inc.')
 599         self.add_external_author ('Alan Woodland')
 600         self.add_external_author ('Diego Elio Petteno')
 601
 602 class BinutilsCmdLine (CmdLine):
 603     def __init__ (self):
 604         CmdLine.__init__ (self, BinutilsCopyright)
 605
 606         self.add_dir ('.', TopLevelFilter())
 607         self.add_dir ('bfd')
 608         self.add_dir ('binutils')
 609         self.add_dir ('config', ConfigFilter())
 610         self.add_dir ('cpu')
 611         self.add_dir ('elfcpp')
 612         self.add_dir ('etc')
 613         self.add_dir ('gas')
 614         self.add_dir ('gdb')
 615         self.add_dir ('gdbserver')
 616         self.add_dir ('gdbsupport')
 617         self.add_dir ('gold')
 618         self.add_dir ('gprof')
 619         self.add_dir ('gprofng')
 620         self.add_dir ('include')
 621         self.add_dir ('ld', LdFilter())
 622         self.add_dir ('libbacktrace')
 623         self.add_dir ('libctf')
 624         self.add_dir ('libdecnumber')
 625         self.add_dir ('libiberty')
 626         self.add_dir ('libsframe')
 627         self.add_dir ('opcodes')
 628         self.add_dir ('readline')
 629         self.add_dir ('sim')
 630
 631         self.default_dirs = [
 632             'bfd',
 633             'binutils',
 634             'elfcpp',
 635             'etc',
 636             'gas',
 637             'gold',
 638             'gprof',
 639             'gprofng',
 640             'include',
 641             'ld',
 642             'libctf',
 643             'libiberty',
 644             'libsframe',
 645             'opcodes',
 646             ]
 647
 648 BinutilsCmdLine().main()