Debian Codegraph @

debian/scripts/update-copyright

raw
1 #!/usr/bin/python3
2
3 import os
4 import re
5 import sys
6
7 verbose = '--verbose' in sys.argv
8
9 default_copyright = 'The Qt Company Ltd.'
10 default_licenses = (
11 'BSD-3-clause',
12 'LGPL-3 or GPL-2',
13 'GPL-3 with Qt-1.0 exception',
14 'GFDL-NIV-1.3',
15 )
16 header_re = re.compile(r"\$QT_BEGIN_LICENSE:([\w\-]+)\$")
17 copyright_re = re.compile(r"Copyright \(C\) (\d\d.+)")
18
19 author_map = {
20 re.compile('^BlackBerry|^Research [Ii]n Motion'): 'BlackBerry Limited (formerly Research In Motion)',
21 re.compile('^BogDan Vatra'): 'BogDan Vatra <[email protected]>',
22 re.compile('^Canonical'): 'Canonical, Ltd.',
23 re.compile('^David Faure'): 'David Faure <[email protected]>',
24 re.compile('^Giuseppe D\'Angelo'): 'Giuseppe D\'Angelo <[email protected]>',
25 re.compile('^Governikus GmbH & Co. KG'): 'Governikus GmbH & Co. KG.',
26 re.compile('^Green Hills Software'): 'Green Hills Software',
27 re.compile('^Intel Corporation'): 'Intel Corporation',
28 re.compile('^Ivan Komissarov'): 'Ivan Komissarov <[email protected]>',
29 re.compile('KDAB'): 'Klarälvdalens Datakonsult AB, a KDAB Group company',
30 re.compile('^Konstantin Ritt'): 'Konstantin Ritt <[email protected]>',
31 re.compile('^Lorn Potter'): 'Lorn Potter',
32 re.compile(r'Martsum .*tmartsum\[at\]gmail.com'): 'Thorbjørn Lund Martsum <[email protected]>',
33 re.compile('^Olivier Goffart'): 'Olivier Goffart <[email protected]>',
34 re.compile('^Richard J. Moore'): 'Richard J. Moore <[email protected]>',
35 re.compile('^Robin Burchell'): 'Robin Burchell <[email protected]>',
36 re.compile('^Samuel Gaist'): 'Samuel Gaist <[email protected]>',
37 re.compile('^Stephen Kelly'): 'Stephen Kelly <[email protected]>',
38 re.compile('^The Qt Company'): 'The Qt Company Ltd.',
39 }
40
41 licenses_map = {
42 'BSD': 'BSD-3-clause',
43 'FDL': 'GFDL-NIV-1.3',
44 'GPL': 'GPL-3',
45 'GPL-EXCEPT': 'GPL-3 with Qt-1.0 exception',
46 'LGPL': 'LGPL-3 or GPL-2',
47 'LGPL21': 'LGPL-2.1-or-3 with Qt-1.1 exception',
48 'LGPL3': 'LGPL-3 or GPL-2+',
49 'LGPL3-COMM': 'LGPL-3',
50 'MIT': 'Expat',
51 }
52
53 exclude_prefixes = (
54 'header',
55 '.git',
56 )
57
58 start_header = '## BEGIN AUTO GENERATED BLOCK'
59 end_header = '## END AUTO GENERATED BLOCK'
60
61
62 class CopyrightInfo():
63 def __init__(self):
64 self.min_years = {}
65 self.max_years = {}
66 self.files = []
67
68 def add_file(self, authors, file):
69 for min_year, max_year, author in authors:
70 if author in self.min_years:
71 self.min_years[author] = min(self.min_years[author], min_year)
72 else:
73 self.min_years[author] = min_year
74 if author in self.max_years:
75 self.max_years[author] = max(self.max_years[author], max_year)
76 else:
77 self.max_years[author] = max_year
78 self.files.append(file)
79
80 def get_strings(self, authors):
81 for author in authors:
82 min_year = self.min_years[author]
83 max_year = self.max_years[author]
84 if min_year == max_year:
85 yield '%d %s' % (min_year, author)
86 else:
87 yield '%d-%d %s' % (min_year, max_year, author)
88
89
90 def canonicalize_author_name(author):
91 for regex, replacement in author_map.items():
92 if regex.search(author):
93 return replacement
94 return author
95
96
97 def parse_file(filename):
98 license = None
99 authors = []
100 with open(filename) as file:
101 try:
102 data = file.readlines(500)
103 except UnicodeDecodeError:
104 data = []
105 authors = None
106 for line in data:
107 match = copyright_re.search(line)
108 if match:
109 copyright = match.group(1)
110 max_year = min_year = int(copyright[:4])
111 if copyright[4] == '-':
112 max_year = int(copyright[5:9])
113 author = copyright[10:]
114 elif copyright[4:7] == ' - ':
115 max_year = int(copyright[7:11])
116 author = copyright[12:]
117 else:
118 author = copyright[5:]
119 author = canonicalize_author_name(author)
120 authors.append((min_year, max_year, author))
121 match = header_re.search(line)
122 if match:
123 license = licenses_map[match.group(1)]
124 if license and not authors:
125 print(f'{filename} ({license}): No authors!', file=sys.stderr)
126 elif verbose:
127 if authors is None:
128 print(f'{filename} (binary)')
129 elif license is None:
130 print(f'{filename} (unknown)')
131 else:
132 print(f'{filename} ({license})')
133 return license, authors
134
135
136 def get_source_files(root_directory):
137 for dirpath, dirnames, filenames in os.walk(root_directory):
138 for filename in filenames:
139 full_path = os.path.join(dirpath, filename)
140 if full_path.startswith('./'):
141 full_path = full_path[2:]
142 if any(full_path.startswith(prefix) for prefix in exclude_prefixes):
143 continue
144 yield full_path
145
146
147 def format_list(title, strings):
148 return title + ('\n' + ' ' * len(title)).join(strings)
149
150
151 def main(root_directory):
152 with open('debian/copyright') as copyright_file:
153 current_copyright = copyright_file.read()
154 start_pos = current_copyright.find(start_header) + len(start_header) + 1
155 start_data = current_copyright[:start_pos]
156 end_pos = current_copyright.find(end_header) - 1
157 end_data = current_copyright[end_pos:]
158
159 data = read_input(root_directory)
160
161 with open('debian/copyright', 'w') as output_file:
162 output_file.write(start_data)
163 write_output(data, output_file)
164 output_file.write(end_data)
165
166
167 def read_input(root_directory):
168 data = {}
169
170 for filename in get_source_files(root_directory):
171 license, authors = parse_file(filename)
172 if license is None:
173 continue
174 if license not in data:
175 data[license] = {}
176 license_dict = data[license]
177 authors_tuple = tuple(sorted({author[2] for author in authors}))
178 if authors_tuple not in license_dict:
179 license_dict[authors_tuple] = CopyrightInfo()
180 license_dict[authors_tuple].add_file(authors, filename)
181 return data
182
183
184 def write_output(data, output_file):
185 for license in sorted(data.keys()):
186 output_file.write('\n## ' + license + '\n')
187 license_dict = data[license]
188 for authors in sorted(license_dict.keys()):
189 if authors == (default_copyright,) and license in default_licenses:
190 continue
191 copyright_info = license_dict[authors]
192 output_file.write('\n')
193 output_file.write(format_list('Files: ', sorted(copyright_info.files)) + '\n')
194 output_file.write(format_list('Copyright: ', copyright_info.get_strings(authors)) + '\n')
195 output_file.write('License: ' + license + '\n')
196
197
198 if __name__ == '__main__':
199 main('.')