3 # Copyright (C) 2013-2020 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.py --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.py --this-year libitm
39 # would run the script on just libitm/.
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 ([
84 # Skip auto- and libtool-related files
105 # Skip FSF mission statement, etc.
110 # Skip imported texinfo files.
115 def get_line_filter (self
, dir, filename
):
116 if filename
.startswith ('ChangeLog'):
117 # Ignore references to copyright in changelog entries.
118 return re
.compile ('\t')
122 def skip_file (self
, dir, filename
):
123 if filename
in self
.skip_files
:
126 (base
, extension
) = os
.path
.splitext (os
.path
.join (dir, filename
))
127 if extension
in self
.skip_extensions
:
130 if extension
== '.in':
131 # Skip .in files produced by automake.
132 if os
.path
.exists (base
+ '.am'):
135 # Skip files produced by autogen
136 if (os
.path
.exists (base
+ '.def')
137 and os
.path
.exists (base
+ '.tpl')):
140 # Skip configure files produced by autoconf
141 if filename
== 'configure':
142 if os
.path
.exists (base
+ '.ac'):
144 if os
.path
.exists (base
+ '.in'):
149 def skip_dir (self
, dir, subdir
):
150 return subdir
in self
.skip_dirs
152 def is_fossilised_file (self
, dir, filename
):
153 if filename
in self
.fossilised_files
:
155 # Only touch current current ChangeLogs.
156 if filename
!= 'ChangeLog' and filename
.find ('ChangeLog') >= 0:
160 def by_package_author (self
, dir, filename
):
161 return filename
in self
.own_files
164 def __init__ (self
, errors
):
167 # Characters in a range of years. Include '.' for typos.
168 ranges
= '[0-9](?:[-0-9.,\s]|\s+and\s+)*[0-9]'
170 # Non-whitespace characters in a copyright holder's name.
174 self
.year_re
= re
.compile ('[0-9]+')
176 # Matches part of a year or copyright holder.
177 self
.continuation_re
= re
.compile (ranges
+ '|' + name
)
179 # Matches a full copyright notice:
180 self
.copyright_re
= re
.compile (
181 # 1: 'Copyright (C)', etc.
183 '|[Cc]opyright\s+\([Cc]\)'
185 '|[Cc]opyright\s+©'
186 '|[Cc]opyright\s+@copyright{}'
188 '|@set\s+copyright[\w-]+)'
190 # 2: the years. Include the whitespace in the year, so that
191 # we can remove any excess.
192 '(\s*(?:' + ranges
+ ',?'
193 '|@value\{[^{}]*\})\s*)'
198 # 4: the copyright holder. Don't allow multiple consecutive
199 # spaces, so that right-margin gloss doesn't get caught
200 # (e.g. gnat_ugn.texi).
201 '(' + name
+ '(?:\s?' + name
+ ')*)?')
203 # A regexp for notices that might have slipped by. Just matching
204 # 'copyright' is too noisy, and 'copyright.*[0-9]' falls foul of
205 # HTML header markers, so check for 'copyright' and two digits.
206 self
.other_copyright_re
= re
.compile ('copyright.*[0-9][0-9]',
208 self
.comment_re
= re
.compile('#+|[*]+|;+|%+|//+|@c |dnl ')
209 self
.holders
= { '@copying': '@copying' }
210 self
.holder_prefixes
= set()
212 # True to 'quilt add' files before changing them.
213 self
.use_quilt
= False
215 # If set, force all notices to include this year.
218 # Goes after the year(s). Could be ', '.
221 def add_package_author (self
, holder
, canon_form
= None):
224 self
.holders
[holder
] = canon_form
225 index
= holder
.find (' ')
227 self
.holder_prefixes
.add (holder
[:index
])
228 index
= holder
.find (' ', index
+ 1)
230 def add_external_author (self
, holder
):
231 self
.holders
[holder
] = None
234 def __init__ (self
, year
):
238 return 'unrecognised year: ' + self
.year
240 def parse_year (self
, string
):
242 if len (string
) == 2:
245 elif len (string
) == 4:
247 raise self
.BadYear (string
)
249 def year_range (self
, years
):
250 year_list
= [self
.parse_year (year
)
251 for year
in self
.year_re
.findall (years
)]
252 assert len (year_list
) > 0
253 return (min (year_list
), max (year_list
))
255 def set_use_quilt (self
, use_quilt
):
256 self
.use_quilt
= use_quilt
258 def include_year (self
, year
):
259 assert not self
.max_year
262 def canonicalise_years (self
, dir, filename
, filter, years
):
263 # Leave texinfo variables alone.
264 if years
.startswith ('@value'):
267 (min_year
, max_year
) = self
.year_range (years
)
269 # Update the upper bound, if enabled.
270 if self
.max_year
and not filter.is_fossilised_file (dir, filename
):
271 max_year
= max (max_year
, self
.max_year
)
274 if min_year
== max_year
:
275 return '%d' % min_year
277 return '%d-%d' % (min_year
, max_year
)
279 def strip_continuation (self
, line
):
281 match
= self
.comment_re
.match (line
)
283 line
= line
[match
.end():].lstrip()
286 def is_complete (self
, match
):
287 holder
= match
.group (4)
289 and (holder
not in self
.holder_prefixes
290 or holder
in self
.holders
))
292 def update_copyright (self
, dir, filename
, filter, file, line
, match
):
295 pathname
= os
.path
.join (dir, filename
)
297 intro
= match
.group (1)
298 if intro
.startswith ('@set'):
299 # Texinfo year variables should always be on one line
300 after_years
= line
[match
.end (2):].strip()
301 if after_years
!= '':
302 self
.errors
.report (pathname
,
303 'trailing characters in @set: '
305 return (False, orig_line
, next_line
)
307 # If it looks like the copyright is incomplete, add the next line.
308 while not self
.is_complete (match
):
310 next_line
= file.next()
311 except StopIteration:
314 # If the next line doesn't look like a proper continuation,
315 # assume that what we've got is complete.
316 continuation
= self
.strip_continuation (next_line
)
317 if not self
.continuation_re
.match (continuation
):
320 # Merge the lines for matching purposes.
321 orig_line
+= next_line
322 line
= line
.rstrip() + ' ' + continuation
325 # Rematch with the longer line, at the original position.
326 match
= self
.copyright_re
.match (line
, match
.start())
329 holder
= match
.group (4)
331 # Use the filter to test cases where markup is getting in the way.
332 if filter.by_package_author (dir, filename
):
333 assert holder
not in self
.holders
336 self
.errors
.report (pathname
, 'missing copyright holder')
337 return (False, orig_line
, next_line
)
339 elif holder
not in self
.holders
:
340 self
.errors
.report (pathname
,
341 'unrecognised copyright holder: ' + holder
)
342 return (False, orig_line
, next_line
)
345 # See whether the copyright is associated with the package
347 canon_form
= self
.holders
[holder
]
349 return (False, orig_line
, next_line
)
351 # Make sure the author is given in a consistent way.
352 line
= (line
[:match
.start (4)]
354 + line
[match
.end (4):])
357 line
= line
[:match
.start (3)] + line
[match
.end (3):]
359 # Update the copyright years.
360 years
= match
.group (2).strip()
362 canon_form
= self
.canonicalise_years (dir, filename
, filter, years
)
363 except self
.BadYear
as e
:
364 self
.errors
.report (pathname
, str (e
))
365 return (False, orig_line
, next_line
)
367 line
= (line
[:match
.start (2)]
368 + ('' if intro
.startswith ('copyright = ') else ' ')
369 + canon_form
+ self
.separator
370 + line
[match
.end (2):])
372 # Use the standard (C) form.
373 if intro
.endswith ('right'):
375 elif intro
.endswith ('(c)'):
376 intro
= intro
[:-3] + '(C)'
377 line
= line
[:match
.start (1)] + intro
+ line
[match
.end (1):]
379 # Strip trailing whitespace
380 line
= line
.rstrip() + '\n'
382 return (line
!= orig_line
, line
, next_line
)
384 def process_file (self
, dir, filename
, filter):
385 pathname
= os
.path
.join (dir, filename
)
386 if filename
.endswith ('.tmp'):
387 # Looks like something we tried to create before.
396 line_filter
= filter.get_line_filter (dir, filename
)
398 with
open (pathname
, 'r') as file:
400 mode
= os
.fstat (file.fileno()).st_mode
404 # Leave filtered-out lines alone.
405 if not (line_filter
and line_filter
.match (line
)):
406 match
= self
.copyright_re
.search (line
)
408 res
= self
.update_copyright (dir, filename
, filter,
410 (this_changed
, line
, next_line
) = res
411 changed
= changed
or this_changed
413 # Check for copyright lines that might have slipped by.
414 elif self
.other_copyright_re
.search (line
):
415 self
.errors
.report (pathname
,
416 'unrecognised copyright: %s'
421 # If something changed, write the new file out.
422 if changed
and self
.errors
.ok():
423 tmp_pathname
= pathname
+ '.tmp'
424 with
open (tmp_pathname
, 'w') as file:
427 os
.fchmod (file.fileno(), mode
)
429 subprocess
.call (['quilt', 'add', pathname
])
430 os
.rename (tmp_pathname
, pathname
)
432 def process_tree (self
, tree
, filter):
433 for (dir, subdirs
, filenames
) in os
.walk (tree
):
434 # Don't recurse through directories that should be skipped.
435 for i
in xrange (len (subdirs
) - 1, -1, -1):
436 if filter.skip_dir (dir, subdirs
[i
]):
439 # Handle the files in this directory.
440 for filename
in filenames
:
441 if filter.skip_file (dir, filename
):
442 sys
.stdout
.write ('Skipping %s\n'
443 % os
.path
.join (dir, filename
))
445 self
.process_file (dir, filename
, filter)
448 def __init__ (self
, copyright
= Copyright
):
449 self
.errors
= Errors()
450 self
.copyright
= copyright (self
.errors
)
452 self
.default_dirs
= []
453 self
.chosen_dirs
= []
454 self
.option_handlers
= dict()
455 self
.option_help
= []
457 self
.add_option ('--help', 'Print this help', self
.o_help
)
458 self
.add_option ('--quilt', '"quilt add" files before changing them',
460 self
.add_option ('--this-year', 'Add the current year to every notice',
463 def add_option (self
, name
, help, handler
):
464 self
.option_help
.append ((name
, help))
465 self
.option_handlers
[name
] = handler
467 def add_dir (self
, dir, filter = GenericFilter()):
468 self
.dirs
.append ((dir, filter))
470 def o_help (self
, option
= None):
471 sys
.stdout
.write ('Usage: %s [options] dir1 dir2...\n\n'
472 'Options:\n' % sys
.argv
[0])
473 format
= '%-15s %s\n'
474 for (what
, help) in self
.option_help
:
475 sys
.stdout
.write (format
% (what
, help))
476 sys
.stdout
.write ('\nDirectories:\n')
480 for (dir, filter) in self
.dirs
:
482 if i
% 3 == 0 or i
== len (self
.dirs
):
483 sys
.stdout
.write (dir + '\n')
485 sys
.stdout
.write (format
% dir)
488 def o_quilt (self
, option
):
489 self
.copyright
.set_use_quilt (True)
491 def o_this_year (self
, option
):
492 self
.copyright
.include_year (time
.localtime().tm_year
)
495 for arg
in sys
.argv
[1:]:
497 self
.chosen_dirs
.append (arg
)
498 elif arg
in self
.option_handlers
:
499 self
.option_handlers
[arg
] (arg
)
501 self
.errors
.report (None, 'unrecognised option: ' + arg
)
503 if len (self
.chosen_dirs
) == 0:
504 self
.chosen_dirs
= self
.default_dirs
505 if len (self
.chosen_dirs
) == 0:
508 for chosen_dir
in self
.chosen_dirs
:
509 canon_dir
= os
.path
.join (chosen_dir
, '')
511 for (dir, filter) in self
.dirs
:
512 if (dir + os
.sep
).startswith (canon_dir
):
514 self
.copyright
.process_tree (dir, filter)
516 self
.errors
.report (None, 'unrecognised directory: '
518 sys
.exit (0 if self
.errors
.ok() else 1)
520 #----------------------------------------------------------------------------
522 class TopLevelFilter (GenericFilter
):
523 def skip_dir (self
, dir, subdir
):
526 class ConfigFilter (GenericFilter
):
528 GenericFilter
.__init
__ (self
)
530 def skip_file (self
, dir, filename
):
531 if filename
.endswith ('.m4'):
532 pathname
= os
.path
.join (dir, filename
)
533 with
open (pathname
) as file:
534 # Skip files imported from gettext.
535 if file.readline().find ('gettext-') >= 0:
537 return GenericFilter
.skip_file (self
, dir, filename
)
539 class GCCFilter (GenericFilter
):
541 GenericFilter
.__init
__ (self
)
543 self
.skip_files |
= set ([
548 self
.skip_dirs |
= set ([
549 # Better not create a merge nightmare for the GNAT folks.
552 # Handled separately.
556 self
.skip_extensions |
= set ([
557 # Maintained by the translation project.
560 # Automatically-generated.
564 self
.fossilised_files |
= set ([
565 # Old news won't be updated.
569 class TestsuiteFilter (GenericFilter
):
571 GenericFilter
.__init
__ (self
)
573 self
.skip_extensions |
= set ([
574 # Don't change the tests, which could be woend by anyone.
588 def skip_file (self
, dir, filename
):
589 # g++.niklas/README contains historical copyright information
591 if filename
== 'README' and os
.path
.basename (dir) == 'g++.niklas':
593 # Similarly params/README.
594 if filename
== 'README' and os
.path
.basename (dir) == 'params':
596 if filename
== 'pdt_5.f03' and os
.path
.basename (dir) == 'gfortran.dg':
598 return GenericFilter
.skip_file (self
, dir, filename
)
600 class LibCppFilter (GenericFilter
):
602 GenericFilter
.__init
__ (self
)
604 self
.skip_extensions |
= set ([
605 # Maintained by the translation project.
608 # Automatically-generated.
612 class LibGCCFilter (GenericFilter
):
614 GenericFilter
.__init
__ (self
)
616 self
.skip_dirs |
= set ([
617 # Imported from GLIBC.
621 class LibPhobosFilter (GenericFilter
):
623 GenericFilter
.__init
__ (self
)
625 self
.skip_files |
= set ([
626 # Source module imported from upstream.
630 self
.skip_dirs |
= set ([
631 # Contains sources imported from upstream.
640 class LibStdCxxFilter (GenericFilter
):
642 GenericFilter
.__init
__ (self
)
644 self
.skip_files |
= set ([
645 # Contains no copyright of its own, but quotes the GPL.
649 self
.skip_dirs |
= set ([
650 # Contains automatically-generated sources.
653 # The testsuite data files shouldn't be changed.
656 # Contains imported images
660 self
.own_files |
= set ([
661 # Contains markup around the copyright owner.
665 def get_line_filter (self
, dir, filename
):
666 if filename
== 'boost_concept_check.h':
667 return re
.compile ('// \(C\) Copyright Jeremy Siek')
668 return GenericFilter
.get_line_filter (self
, dir, filename
)
670 class GCCCopyright (Copyright
):
671 def __init__ (self
, errors
):
672 Copyright
.__init
__ (self
, errors
)
674 canon_fsf
= 'Free Software Foundation, Inc.'
675 self
.add_package_author ('Free Software Foundation', canon_fsf
)
676 self
.add_package_author ('Free Software Foundation.', canon_fsf
)
677 self
.add_package_author ('Free Software Foundation Inc.', 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 ('The Free Software Foundation', canon_fsf
)
681 self
.add_package_author ('The Free Software Foundation, Inc.', canon_fsf
)
682 self
.add_package_author ('Software Foundation, Inc.', canon_fsf
)
684 self
.add_external_author ('ARM')
685 self
.add_external_author ('AdaCore')
686 self
.add_external_author ('Ami Tavory and Vladimir Dreizin, IBM-HRL.')
687 self
.add_external_author ('Cavium Networks.')
688 self
.add_external_author ('Faraday Technology Corp.')
689 self
.add_external_author ('Florida State University')
690 self
.add_external_author ('Gerard Jungman')
691 self
.add_external_author ('Greg Colvin and Beman Dawes.')
692 self
.add_external_author ('Hewlett-Packard Company')
693 self
.add_external_author ('Intel Corporation')
694 self
.add_external_author ('Information Technology Industry Council.')
695 self
.add_external_author ('James Theiler, Brian Gough')
696 self
.add_external_author ('Makoto Matsumoto and Takuji Nishimura,')
697 self
.add_external_author ('Mentor Graphics Corporation')
698 self
.add_external_author ('National Research Council of Canada.')
699 self
.add_external_author ('NVIDIA Corporation')
700 self
.add_external_author ('Peter Dimov and Multi Media Ltd.')
701 self
.add_external_author ('Peter Dimov')
702 self
.add_external_author ('Pipeline Associates, Inc.')
703 self
.add_external_author ('Regents of the University of California.')
704 self
.add_external_author ('Silicon Graphics Computer Systems, Inc.')
705 self
.add_external_author ('Silicon Graphics')
706 self
.add_external_author ('Stephen L. Moshier')
707 self
.add_external_author ('Sun Microsystems, Inc. All rights reserved.')
708 self
.add_external_author ('The D Language Foundation, All Rights Reserved')
709 self
.add_external_author ('The Go Authors. All rights reserved.')
710 self
.add_external_author ('The Go Authors. All rights reserved.')
711 self
.add_external_author ('The Go Authors.')
712 self
.add_external_author ('The Regents of the University of California.')
713 self
.add_external_author ('Unicode, Inc.')
714 self
.add_external_author ('University of Toronto.')
715 self
.add_external_author ('Yoshinori Sato')
717 class GCCCmdLine (CmdLine
):
719 CmdLine
.__init
__ (self
, GCCCopyright
)
721 self
.add_dir ('.', TopLevelFilter())
722 # boehm-gc is imported from upstream.
723 self
.add_dir ('config', ConfigFilter())
724 # contrib isn't really part of GCC.
725 self
.add_dir ('fixincludes')
726 self
.add_dir ('gcc', GCCFilter())
727 self
.add_dir (os
.path
.join ('gcc', 'testsuite'), TestsuiteFilter())
728 self
.add_dir ('gnattools')
729 self
.add_dir ('gotools')
730 self
.add_dir ('include')
731 # intl is imported from upstream.
732 self
.add_dir ('libada')
733 self
.add_dir ('libatomic')
734 self
.add_dir ('libbacktrace')
735 self
.add_dir ('libcc1')
736 self
.add_dir ('libcpp', LibCppFilter())
737 self
.add_dir ('libdecnumber')
738 # libffi is imported from upstream.
739 self
.add_dir ('libgcc', LibGCCFilter())
740 self
.add_dir ('libgfortran')
741 # libgo is imported from upstream.
742 self
.add_dir ('libgomp')
743 self
.add_dir ('libhsail-rt')
744 self
.add_dir ('libiberty')
745 self
.add_dir ('libitm')
746 self
.add_dir ('libobjc')
747 # liboffloadmic is imported from upstream.
748 self
.add_dir ('libphobos', LibPhobosFilter())
749 self
.add_dir ('libquadmath')
750 # libsanitizer is imported from upstream.
751 self
.add_dir ('libssp')
752 self
.add_dir ('libstdc++-v3', LibStdCxxFilter())
753 self
.add_dir ('libvtv')
754 self
.add_dir ('lto-plugin')
755 # maintainer-scripts maintainer-scripts
756 # zlib is imported from upstream.
758 self
.default_dirs
= [