#!/usr/bin/python # # Copyright (C) 2009 W. Trevor King # # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2 of the License, or # (at your option) any later version. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License along # with this program; if not, write to the Free Software Foundation, Inc., # 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA. import os.path import re import sys import time import os import sys import select from threading import Thread from libbe.subproc import Pipe COPYRIGHT_TEXT="""# # This program is free software; you can redistribute it and/or modify # it under the terms of the GNU General Public License as published by # the Free Software Foundation; either version 2 of the License, or # (at your option) any later version. # # This program is distributed in the hope that it will be useful, # but WITHOUT ANY WARRANTY; without even the implied warranty of # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the # GNU General Public License for more details. # # You should have received a copy of the GNU General Public License along # with this program; if not, write to the Free Software Foundation, Inc., # 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.""" COPYRIGHT_TAG='-xyz-COPYRIGHT-zyx-' # unlikely to occur in the wild :p ALIASES = [ ['Ben Finney ', 'Ben Finney ', 'John Doe '], ['Chris Ball ', 'Chris Ball '], ['Gianluca Montecchi ', 'gian ', 'gianluca '], ['W. Trevor King ', 'wking '], [None, 'j^ '], ] COPYRIGHT_ALIASES = [ ['Aaron Bentley and Panometrics, Inc.', 'Aaron Bentley '], ] EXCLUDES = [ ['Aaron Bentley and Panometrics, Inc.', 'Aaron Bentley ',] ] IGNORED_PATHS = ['./.be/', './.bzr/', './build/'] IGNORED_FILES = ['COPYING', 'update_copyright.py', 'catmutt'] def _strip_email(*args): """ >>> _strip_email('J Doe ') ['J Doe'] >>> _strip_email('J Doe ', 'JJJ Smith ') ['J Doe', 'JJJ Smith'] """ args = list(args) for i,arg in enumerate(args): if arg == None: continue index = arg.find('<') if index > 0: args[i] = arg[:index].rstrip() return args def _replace_aliases(authors, with_email=True, aliases=None, excludes=None): """ >>> aliases = [['J Doe and C, Inc.', 'J Doe '], ... ['J Doe ', 'Johnny '], ... ['JJJ Smith ', 'Jingly '], ... [None, 'Anonymous ']] >>> excludes = [['J Doe and C, Inc.', 'J Doe ']] >>> _replace_aliases(['JJJ Smith ', 'Johnny ', ... 'Jingly ', 'Anonymous '], ... with_email=True, aliases=aliases, excludes=excludes) ['J Doe ', 'JJJ Smith '] >>> _replace_aliases(['JJJ Smith', 'Johnny', 'Jingly', 'Anonymous'], ... with_email=False, aliases=aliases, excludes=excludes) ['J Doe', 'JJJ Smith'] >>> _replace_aliases(['JJJ Smith ', 'Johnny ', ... 'Jingly ', 'J Doe '], ... with_email=True, aliases=aliases, excludes=excludes) ['J Doe and C, Inc.', 'JJJ Smith '] """ if aliases == None: aliases = ALIASES if excludes == None: excludes = EXCLUDES if with_email == False: aliases = [_strip_email(*alias) for alias in aliases] exclude = [_strip_email(*exclude) for exclude in excludes] for i,author in enumerate(authors): for alias in aliases: if author in alias[1:]: authors[i] = alias[0] break for i,author in enumerate(authors): for exclude in excludes: if author in exclude[1:] and exclude[0] in authors: authors[i] = None authors = sorted(set(authors)) if None in authors: authors.remove(None) return authors def authors_list(): p = Pipe([['bzr', 'log', '-n0'], ['grep', '^ *committer\|^ *author'], ['cut', '-d:', '-f2'], ['sed', 's/ <.*//;s/^ *//'], ['sort'], ['uniq']]) assert p.status == 0, p.statuses authors = p.stdout.rstrip().split('\n') return _replace_aliases(authors, with_email=False) def update_authors(verbose=True): print "updating AUTHORS" f = file('AUTHORS', 'w') authors_text = 'Bugs Everywhere was written by:\n%s\n' % '\n'.join(authors_list()) f.write(authors_text) f.close() def ignored_file(filename, ignored_paths=None, ignored_files=None): """ >>> ignored_paths = ['./a/', './b/'] >>> ignored_files = ['x', 'y'] >>> ignored_file('./a/z', ignored_paths, ignored_files) True >>> ignored_file('./ab/z', ignored_paths, ignored_files) False >>> ignored_file('./ab/x', ignored_paths, ignored_files) True >>> ignored_file('./ab/xy', ignored_paths, ignored_files) False >>> ignored_file('./z', ignored_paths, ignored_files) False """ if ignored_paths == None: ignored_paths = IGNORED_PATHS if ignored_files == None: ignored_files = IGNORED_FILES for path in ignored_paths: if filename.startswith(path): return True if os.path.basename(filename) in ignored_files: return True if os.path.abspath(filename) != os.path.realpath(filename): return True # symink somewhere in path... return False def _copyright_string(orig_year, final_year, authors): """ >>> print _copyright_string(orig_year=2005, ... final_year=2005, ... authors=['A ', 'B '] ... ) # doctest: +ELLIPSIS # Copyright (C) 2005 A # B # # This program... >>> print _copyright_string(orig_year=2005, ... final_year=2009, ... authors=['A ', 'B '] ... ) # doctest: +ELLIPSIS # Copyright (C) 2005-2009 A # B # # This program... """ if orig_year == final_year: date_range = '%s' % orig_year else: date_range = '%s-%s' % (orig_year, final_year) lines = ['# Copyright (C) %s %s' % (date_range, authors[0])] for author in authors[1:]: lines.append('#' + ' '*(len(' Copyright (C) ')+len(date_range)+1) + author) return '%s\n%s' % ('\n'.join(lines), COPYRIGHT_TEXT) def _tag_copyright(contents): """ >>> contents = '''Some file ... bla bla ... # Copyright (copyright begins) ... # (copyright continues) ... # bla bla bla ... (copyright ends) ... bla bla bla ... ''' >>> print _tag_copyright(contents), Some file bla bla -xyz-COPYRIGHT-zyx- (copyright ends) bla bla bla """ lines = [] incopy = False for line in contents.splitlines(): if incopy == False and line.startswith('# Copyright'): incopy = True lines.append(COPYRIGHT_TAG) elif incopy == True and not line.startswith('#'): incopy = False if incopy == False: lines.append(line.rstrip('\n')) return '\n'.join(lines)+'\n' def _update_copyright(contents, orig_year, authors): current_year = time.gmtime()[0] copyright_string = _copyright_string(orig_year, current_year, authors) contents = _tag_copyright(contents) return contents.replace(COPYRIGHT_TAG, copyright_string) def update_file(filename, verbose=True): if verbose == True: print "updating", filename contents = file(filename, 'r').read() p = Pipe([['bzr', 'log', '-n0', filename], ['grep', '^ *timestamp: '], ['tail', '-n1'], ['sed', 's/^ *//;'], ['cut', '-b', '16-19']]) if p.status != 0: assert p.statuses[0] == 3, p.statuses return # bzr doesn't version that file assert p.status == 0, p.statuses orig_year = int(p.stdout.strip()) p = Pipe([['bzr', 'log', '-n0', filename], ['grep', '^ *author: \|^ *committer: '], ['cut', '-d:', '-f2'], ['sed', 's/^ *//;s/ *$//'], ['sort'], ['uniq']]) assert p.status == 0, p.statuses authors = p.stdout.rstrip().split('\n') authors = _replace_aliases(authors, with_email=True, aliases=ALIASES+COPYRIGHT_ALIASES) contents = _update_copyright(contents, orig_year, authors) f = file(filename, 'w') f.write(contents) f.close() def update_files(files=None): if files == None or len(files) == 0: p = Pipe([['grep', '-rc', '# Copyright', '.'], ['grep', '-v', ':0$'], ['cut', '-d:', '-f1']]) assert p.status == 0 files = p.stdout.rstrip().split('\n') for filename in files: if ignored_file(filename) == True: continue update_file(filename) def test(): import doctest doctest.testmod() if __name__ == '__main__': import optparse usage = """%prog [options] [file ...] Update copyright information in source code with information from the bzr repository. Run from the BE repository root. Replaces every line starting with '^# Copyright' and continuing with '^#' with an auto-generated copyright blurb. If you want to add #-commented material after a copyright blurb, please insert a blank line between the blurb and your comment (as in this file), so the next run of update_copyright.py doesn't clobber your comment. If no files are given, a list of files to update is generated automatically. """ p = optparse.OptionParser(usage) p.add_option('--test', dest='test', default=False, action='store_true', help='Run internal tests and exit') options,args = p.parse_args() if options.test == True: test() sys.exit(0) update_authors() update_files(files=args)