3 # Copyright (C) 2013 Free Software Foundation, Inc.
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)
10 # This script adjusts the copyright notices at the top of source files
11 # so that they have the form:
13 # Copyright XXXX-YYYY Free Software Foundation, Inc.
15 # It doesn't change code that is known to be maintained elsewhere or
16 # that carries a non-FSF copyright.
18 # The script also doesn't change testsuite files, except those in
19 # libstdc++-v3. This is because libstdc++-v3 has a conformance testsuite,
20 # while most tests in other directories are just things that failed at some
23 # Pass --this-year to the script if you want it to add the current year
24 # to all applicable notices. Pass --quilt if you are using quilt and
25 # want files to be added to the quilt before being changed.
27 # By default the script will update all directories for which the
28 # output has been vetted. You can instead pass the names of individual
29 # directories, including those that haven't been approved. So:
31 # update-copyright.pl --this-year
33 # is the command that would be used at the beginning of a year to update
34 # all copyright notices (and possibly at other times to check whether
35 # new files have been added with old years). On the other hand:
37 # update-copyright.pl --this-year libjava
39 # would run the script on just libjava/.
41 # Note that things like --version output strings must be updated before
42 # this script is run. There's already a separate procedure for that.
54 def report (self, filename, string):
56 string = filename + ': ' + string
57 sys.stderr.write (string + '\n')
61 return self.num_errors == 0
65 self.skip_files = set()
66 self.skip_dirs = set()
67 self.skip_extensions = set()
68 self.fossilised_files = set()
69 self.own_files = set()
71 self.skip_files |= set ([
83 # Skip auto- and libtool-related files
104 # Skip FSF mission statement, etc.
109 # Skip imported texinfo files.
114 def get_line_filter (self, dir, filename):
115 if filename.startswith ('ChangeLog'):
116 # Ignore references to copyright in changelog entries.
117 return re.compile ('\t')
121 def skip_file (self, dir, filename):
122 if filename in self.skip_files:
125 (base, extension) = os.path.splitext (os.path.join (dir, filename))
126 if extension in self.skip_extensions:
129 if extension == '.in':
130 # Skip .in files produced by automake.
131 if os.path.exists (base + '.am'):
134 # Skip files produced by autogen
135 if (os.path.exists (base + '.def')
136 and os.path.exists (base + '.tpl')):
139 # Skip configure files produced by autoconf
140 if filename == 'configure':
141 if os.path.exists (base + '.ac'):
143 if os.path.exists (base + '.in'):
148 def skip_dir (self, dir, subdir):
149 return subdir in self.skip_dirs
151 def is_fossilised_file (self, dir, filename):
152 if filename in self.fossilised_files:
154 # Only touch current current ChangeLogs.
155 if filename != 'ChangeLog' and filename.find ('ChangeLog') >= 0:
159 def by_package_author (self, dir, filename):
160 return filename in self.own_files
163 def __init__ (self, errors):
166 # Characters in a range of years. Include '.' for typos.
167 ranges = '[0-9](?:[-0-9.,\s]|\s+and\s+)*[0-9]'
169 # Non-whitespace characters in a copyright holder's name.
173 self.year_re = re.compile ('[0-9]+')
175 # Matches part of a year or copyright holder.
176 self.continuation_re = re.compile (ranges + '|' + name)
178 # Matches a full copyright notice:
179 self.copyright_re = re.compile (
180 # 1: 'Copyright (C)', etc.
182 '|[Cc]opyright\s+\([Cc]\)'
184 '|[Cc]opyright\s+©'
185 '|[Cc]opyright\s+@copyright{}'
186 '|@set\s+copyright[\w-]+)'
188 # 2: the years. Include the whitespace in the year, so that
189 # we can remove any excess.
190 '(\s*(?:' + ranges + ',?'
191 '|@value\{[^{}]*\})\s*)'
196 # 4: the copyright holder. Don't allow multiple consecutive
197 # spaces, so that right-margin gloss doesn't get caught
198 # (e.g. gnat_ugn.texi).
199 '(' + name + '(?:\s?' + name + ')*)?')
201 # A regexp for notices that might have slipped by. Just matching
202 # 'copyright' is too noisy, and 'copyright.*[0-9]' falls foul of
203 # HTML header markers, so check for 'copyright' and two digits.
204 self.other_copyright_re = re.compile ('copyright.*[0-9][0-9]',
206 self.comment_re = re.compile('#+|[*]+|;+|%+|//+|@c |dnl ')
207 self.holders = { '@copying': '@copying' }
208 self.holder_prefixes = set()
210 # True to 'quilt add' files before changing them.
211 self.use_quilt = False
213 # If set, force all notices to include this year.
216 # Goes after the year(s). Could be ', '.
219 def add_package_author (self, holder, canon_form = None):
222 self.holders[holder] = canon_form
223 index = holder.find (' ')
225 self.holder_prefixes.add (holder[:index])
226 index = holder.find (' ', index + 1)
228 def add_external_author (self, holder):
229 self.holders[holder] = None
232 def __init__ (self, year):
236 return 'unrecognised year: ' + self.year
238 def parse_year (self, string):
240 if len (string) == 2:
243 elif len (string) == 4:
245 raise self.BadYear (string)
247 def year_range (self, years):
248 year_list = [self.parse_year (year)
249 for year in self.year_re.findall (years)]
250 assert len (year_list) > 0
251 return (min (year_list), max (year_list))
253 def set_use_quilt (self, use_quilt):
254 self.use_quilt = use_quilt
256 def include_year (self, year):
257 assert not self.max_year
260 def canonicalise_years (self, dir, filename, filter, years):
261 # Leave texinfo variables alone.
262 if years.startswith ('@value'):
265 (min_year, max_year) = self.year_range (years)
267 # Update the upper bound, if enabled.
268 if self.max_year and not filter.is_fossilised_file (dir, filename):
269 max_year = max (max_year, self.max_year)
272 if min_year == max_year:
273 return '%d' % min_year
275 return '%d-%d' % (min_year, max_year)
277 def strip_continuation (self, line):
279 match = self.comment_re.match (line)
281 line = line[match.end():].lstrip()
284 def is_complete (self, match):
285 holder = match.group (4)
287 and (holder not in self.holder_prefixes
288 or holder in self.holders))
290 def update_copyright (self, dir, filename, filter, file, line, match):
293 pathname = os.path.join (dir, filename)
295 intro = match.group (1)
296 if intro.startswith ('@set'):
297 # Texinfo year variables should always be on one line
298 after_years = line[match.end (2):].strip()
299 if after_years != '':
300 self.errors.report (pathname,
301 'trailing characters in @set: '
303 return (False, orig_line, next_line)
305 # If it looks like the copyright is incomplete, add the next line.
306 while not self.is_complete (match):
308 next_line = file.next()
309 except StopIteration:
312 # If the next line doesn't look like a proper continuation,
313 # assume that what we've got is complete.
314 continuation = self.strip_continuation (next_line)
315 if not self.continuation_re.match (continuation):
318 # Merge the lines for matching purposes.
319 orig_line += next_line
320 line = line.rstrip() + ' ' + continuation
323 # Rematch with the longer line, at the original position.
324 match = self.copyright_re.match (line, match.start())
327 holder = match.group (4)
329 # Use the filter to test cases where markup is getting in the way.
330 if filter.by_package_author (dir, filename):
331 assert holder not in self.holders
334 self.errors.report (pathname, 'missing copyright holder')
335 return (False, orig_line, next_line)
337 elif holder not in self.holders:
338 self.errors.report (pathname,
339 'unrecognised copyright holder: ' + holder)
340 return (False, orig_line, next_line)
343 # See whether the copyright is associated with the package
345 canon_form = self.holders[holder]
347 return (False, orig_line, next_line)
349 # Make sure the author is given in a consistent way.
350 line = (line[:match.start (4)]
352 + line[match.end (4):])
355 line = line[:match.start (3)] + line[match.end (3):]
357 # Update the copyright years.
358 years = match.group (2).strip()
360 canon_form = self.canonicalise_years (dir, filename, filter, years)
361 except self.BadYear as e:
362 self.errors.report (pathname, str (e))
363 return (False, orig_line, next_line)
365 line = (line[:match.start (2)]
366 + ' ' + canon_form + self.separator
367 + line[match.end (2):])
369 # Use the standard (C) form.
370 if intro.endswith ('right'):
372 elif intro.endswith ('(c)'):
373 intro = intro[:-3] + '(C)'
374 line = line[:match.start (1)] + intro + line[match.end (1):]
376 # Strip trailing whitespace
377 line = line.rstrip() + '\n'
379 return (line != orig_line, line, next_line)
381 def process_file (self, dir, filename, filter):
382 pathname = os.path.join (dir, filename)
383 if filename.endswith ('.tmp'):
384 # Looks like something we tried to create before.
393 line_filter = filter.get_line_filter (dir, filename)
394 with open (pathname, 'r') as file:
399 # Leave filtered-out lines alone.
400 if not (line_filter and line_filter.match (line)):
401 match = self.copyright_re.search (line)
403 res = self.update_copyright (dir, filename, filter,
405 (this_changed, line, next_line) = res
406 changed = changed or this_changed
408 # Check for copyright lines that might have slipped by.
409 elif self.other_copyright_re.search (line):
410 self.errors.report (pathname,
411 'unrecognised copyright: %s'
416 # If something changed, write the new file out.
417 if changed and self.errors.ok():
418 tmp_pathname = pathname + '.tmp'
419 with open (tmp_pathname, 'w') as file:
423 subprocess.call (['quilt', 'add', pathname])
424 os.rename (tmp_pathname, pathname)
426 def process_tree (self, tree, filter):
427 for (dir, subdirs, filenames) in os.walk (tree):
428 # Don't recurse through directories that should be skipped.
429 for i in xrange (len (subdirs) - 1, -1, -1):
430 if filter.skip_dir (dir, subdirs[i]):
433 # Handle the files in this directory.
434 for filename in filenames:
435 if filter.skip_file (dir, filename):
436 sys.stdout.write ('Skipping %s\n'
437 % os.path.join (dir, filename))
439 self.process_file (dir, filename, filter)
442 def __init__ (self, copyright = Copyright):
443 self.errors = Errors()
444 self.copyright = copyright (self.errors)
446 self.default_dirs = []
447 self.chosen_dirs = []
448 self.option_handlers = dict()
449 self.option_help = []
451 self.add_option ('--help', 'Print this help', self.o_help)
452 self.add_option ('--quilt', '"quilt add" files before changing them',
454 self.add_option ('--this-year', 'Add the current year to every notice',
457 def add_option (self, name, help, handler):
458 self.option_help.append ((name, help))
459 self.option_handlers[name] = handler
461 def add_dir (self, dir, filter = GenericFilter()):
462 self.dirs.append ((dir, filter))
464 def o_help (self, option = None):
465 sys.stdout.write ('Usage: %s [options] dir1 dir2...\n\n'
466 'Options:\n' % sys.argv[0])
467 format = '%-15s %s\n'
468 for (what, help) in self.option_help:
469 sys.stdout.write (format % (what, help))
470 sys.stdout.write ('\nDirectories:\n')
474 for (dir, filter) in self.dirs:
476 if i % 3 == 0 or i == len (self.dirs):
477 sys.stdout.write (dir + '\n')
479 sys.stdout.write (format % dir)
482 def o_quilt (self, option):
483 self.copyright.set_use_quilt (True)
485 def o_this_year (self, option):
486 self.copyright.include_year (time.localtime().tm_year)
489 for arg in sys.argv[1:]:
491 self.chosen_dirs.append (arg)
492 elif arg in self.option_handlers:
493 self.option_handlers[arg] (arg)
495 self.errors.report (None, 'unrecognised option: ' + arg)
497 if len (self.chosen_dirs) == 0:
498 self.chosen_dirs = self.default_dirs
499 if len (self.chosen_dirs) == 0:
502 for chosen_dir in self.chosen_dirs:
503 canon_dir = os.path.join (chosen_dir, '')
505 for (dir, filter) in self.dirs:
506 if (dir + os.sep).startswith (canon_dir):
508 self.copyright.process_tree (dir, filter)
510 self.errors.report (None, 'unrecognised directory: '
512 sys.exit (0 if self.errors.ok() else 1)
514 #----------------------------------------------------------------------------
516 class TopLevelFilter (GenericFilter):
517 def skip_dir (self, dir, subdir):
520 class ConfigFilter (GenericFilter):
522 GenericFilter.__init__ (self)
524 def skip_file (self, dir, filename):
525 if filename.endswith ('.m4'):
526 pathname = os.path.join (dir, filename)
527 with open (pathname) as file:
528 # Skip files imported from gettext.
529 if file.readline().find ('gettext-') >= 0:
531 return GenericFilter.skip_file (self, dir, filename)
533 class GCCFilter (GenericFilter):
535 GenericFilter.__init__ (self)
537 self.skip_files |= set ([
542 self.skip_dirs |= set ([
543 # Better not create a merge nightmare for the GNAT folks.
546 # Handled separately.
550 self.skip_extensions |= set ([
551 # Maintained by the translation project.
554 # Automatically-generated.
558 self.fossilised_files |= set ([
559 # Old news won't be updated.
563 class TestsuiteFilter (GenericFilter):
565 GenericFilter.__init__ (self)
567 self.skip_extensions |= set ([
568 # Don't change the tests, which could be woend by anyone.
581 def skip_file (self, dir, filename):
582 # g++.niklas/README contains historical copyright information
584 if filename == 'README' and os.path.basename (dir) == 'g++.niklas':
586 return GenericFilter.skip_file (self, dir, filename)
588 class LibCppFilter (GenericFilter):
590 GenericFilter.__init__ (self)
592 self.skip_extensions |= set ([
593 # Maintained by the translation project.
596 # Automatically-generated.
600 class LibGCCFilter (GenericFilter):
602 GenericFilter.__init__ (self)
604 self.skip_dirs |= set ([
605 # Imported from GLIBC.
609 class LibJavaFilter (GenericFilter):
611 GenericFilter.__init__ (self)
613 self.skip_dirs |= set ([
614 # Handled separately.
617 # Not really part of the library
620 # Imported from upstream
625 def get_line_filter (self, dir, filename):
626 if filename == 'NameDecoder.h':
627 return re.compile ('.*NAME_COPYRIGHT')
628 if filename == 'ICC_Profile.h':
629 return re.compile ('.*icSigCopyrightTag')
630 return GenericFilter.get_line_filter (self, dir, filename)
632 class LibMudflapFilter (GenericFilter):
634 GenericFilter.__init__ (self)
636 self.skip_dirs |= set ([
637 # Handled separately.
641 class LibStdCxxFilter (GenericFilter):
643 GenericFilter.__init__ (self)
645 self.skip_files |= set ([
646 # Contains no copyright of its own, but quotes the GPL.
650 self.skip_dirs |= set ([
651 # Contains automatically-generated sources.
654 # The testsuite data files shouldn't be changed.
657 # Contains imported images
661 self.own_files |= set ([
662 # Contains markup around the copyright owner.
666 def get_line_filter (self, dir, filename):
667 if filename == 'boost_concept_check.h':
668 return re.compile ('// \(C\) Copyright Jeremy Siek')
669 return GenericFilter.get_line_filter (self, dir, filename)
671 class GCCCopyright (Copyright):
672 def __init__ (self, errors):
673 Copyright.__init__ (self, errors)
675 canon_fsf = 'Free Software Foundation, Inc.'
676 self.add_package_author ('Free Software Foundation', canon_fsf)
677 self.add_package_author ('Free Software Foundation.', canon_fsf)
678 self.add_package_author ('Free Software Foundation Inc.', canon_fsf)
679 self.add_package_author ('Free Software Foundation, Inc', canon_fsf)
680 self.add_package_author ('Free Software Foundation, Inc.', canon_fsf)
681 self.add_package_author ('The Free Software Foundation', canon_fsf)
682 self.add_package_author ('The Free Software Foundation, Inc.', canon_fsf)
683 self.add_package_author ('Software Foundation, Inc.', canon_fsf)
685 self.add_external_author ('ARM')
686 self.add_external_author ('AdaCore')
687 self.add_external_author ('Ami Tavory and Vladimir Dreizin, IBM-HRL.')
688 self.add_external_author ('Cavium Networks.')
689 self.add_external_author ('Faraday Technology Corp.')
690 self.add_external_author ('Florida State University')
691 self.add_external_author ('Greg Colvin and Beman Dawes.')
692 self.add_external_author ('Hewlett-Packard Company')
693 self.add_external_author ('Information Technology Industry Council.')
694 self.add_external_author ('James Theiler, Brian Gough')
695 self.add_external_author ('Makoto Matsumoto and Takuji Nishimura,')
696 self.add_external_author ('National Research Council of Canada.')
697 self.add_external_author ('Peter Dimov and Multi Media Ltd.')
698 self.add_external_author ('Peter Dimov')
699 self.add_external_author ('Pipeline Associates, Inc.')
700 self.add_external_author ('Regents of the University of California.')
701 self.add_external_author ('Silicon Graphics Computer Systems, Inc.')
702 self.add_external_author ('Silicon Graphics')
703 self.add_external_author ('Stephen L. Moshier')
704 self.add_external_author ('Sun Microsystems, Inc. All rights reserved.')
705 self.add_external_author ('The Go Authors. All rights reserved.')
706 self.add_external_author ('The Go Authors. All rights reserved.')
707 self.add_external_author ('The Go Authors.')
708 self.add_external_author ('The Regents of the University of California.')
709 self.add_external_author ('Unicode, Inc.')
710 self.add_external_author ('University of Toronto.')
712 class GCCCmdLine (CmdLine):
714 CmdLine.__init__ (self, GCCCopyright)
716 self.add_dir ('.', TopLevelFilter())
717 # boehm-gc is imported from upstream.
718 self.add_dir ('config', ConfigFilter())
719 # contrib isn't really part of GCC.
720 self.add_dir ('fixincludes')
721 self.add_dir ('gcc', GCCFilter())
722 self.add_dir (os.path.join ('gcc', 'testsuite'), TestsuiteFilter())
723 self.add_dir ('gnattools')
724 self.add_dir ('include')
725 self.add_dir ('libada')
726 self.add_dir ('libatomic')
727 self.add_dir ('libbacktrace')
728 self.add_dir ('libcpp', LibCppFilter())
729 self.add_dir ('libdecnumber')
730 # libffi is imported from upstream.
731 self.add_dir ('libgcc', LibGCCFilter())
732 self.add_dir ('libgfortran')
733 self.add_dir ('libgomp')
734 self.add_dir ('libiberty')
735 self.add_dir ('libitm')
736 self.add_dir ('libjava', LibJavaFilter())
737 self.add_dir (os.path.join ('libjava', 'testsuite'), TestsuiteFilter())
738 self.add_dir ('libmudflap', LibMudflapFilter())
739 self.add_dir (os.path.join ('libmudflap', 'testsuite'),
741 self.add_dir ('libobjc')
742 self.add_dir ('libquadmath')
743 # libsanitiser is imported from upstream.
744 self.add_dir ('libssp')
745 self.add_dir ('libstdc++-v3', LibStdCxxFilter())
746 self.add_dir ('lto-plugin')
747 # zlib is imported from upstream.
749 self.default_dirs = [