projects/linux·blob v5.5 · refs

293 lines · 9.9 KiB

1#!/usr/bin/env python
2# SPDX-License-Identifier: GPL-2.0
3# Copyright Thomas Gleixner <[email protected]>
4
5from argparse import ArgumentParser
6from ply import lex, yacc
7import locale
8import traceback
9import sys
10import git
11import re
12import os
13
14class ParserException(Exception):
15 def __init__(self, tok, txt):
16 self.tok = tok
17 self.txt = txt
18
19class SPDXException(Exception):
20 def __init__(self, el, txt):
21 self.el = el
22 self.txt = txt
23
24class SPDXdata(object):
25 def __init__(self):
26 self.license_files = 0
27 self.exception_files = 0
28 self.licenses = [ ]
29 self.exceptions = { }
30
31# Read the spdx data from the LICENSES directory
32def read_spdxdata(repo):
33
34 # The subdirectories of LICENSES in the kernel source
35 # Note: exceptions needs to be parsed as last directory.
36 license_dirs = [ "preferred", "dual", "deprecated", "exceptions" ]
37 lictree = repo.head.commit.tree['LICENSES']
38
39 spdx = SPDXdata()
40
41 for d in license_dirs:
42 for el in lictree[d].traverse():
43 if not os.path.isfile(el.path):
44 continue
45
46 exception = None
47 for l in open(el.path).readlines():
48 if l.startswith('Valid-License-Identifier:'):
49 lid = l.split(':')[1].strip().upper()
50 if lid in spdx.licenses:
51 raise SPDXException(el, 'Duplicate License Identifier: %s' %lid)
52 else:
53 spdx.licenses.append(lid)
54
55 elif l.startswith('SPDX-Exception-Identifier:'):
56 exception = l.split(':')[1].strip().upper()
57 spdx.exceptions[exception] = []
58
59 elif l.startswith('SPDX-Licenses:'):
60 for lic in l.split(':')[1].upper().strip().replace(' ', '').replace('\t', '').split(','):
61 if not lic in spdx.licenses:
62 raise SPDXException(None, 'Exception %s missing license %s' %(exception, lic))
63 spdx.exceptions[exception].append(lic)
64
65 elif l.startswith("License-Text:"):
66 if exception:
67 if not len(spdx.exceptions[exception]):
68 raise SPDXException(el, 'Exception %s is missing SPDX-Licenses' %exception)
69 spdx.exception_files += 1
70 else:
71 spdx.license_files += 1
72 break
73 return spdx
74
75class id_parser(object):
76
77 reserved = [ 'AND', 'OR', 'WITH' ]
78 tokens = [ 'LPAR', 'RPAR', 'ID', 'EXC' ] + reserved
79
80 precedence = ( ('nonassoc', 'AND', 'OR'), )
81
82 t_ignore = ' \t'
83
84 def __init__(self, spdx):
85 self.spdx = spdx
86 self.lasttok = None
87 self.lastid = None
88 self.lexer = lex.lex(module = self, reflags = re.UNICODE)
89 # Initialize the parser. No debug file and no parser rules stored on disk
90 # The rules are small enough to be generated on the fly
91 self.parser = yacc.yacc(module = self, write_tables = False, debug = False)
92 self.lines_checked = 0
93 self.checked = 0
94 self.spdx_valid = 0
95 self.spdx_errors = 0
96 self.curline = 0
97 self.deepest = 0
98
99 # Validate License and Exception IDs
100 def validate(self, tok):
101 id = tok.value.upper()
102 if tok.type == 'ID':
103 if not id in self.spdx.licenses:
104 raise ParserException(tok, 'Invalid License ID')
105 self.lastid = id
106 elif tok.type == 'EXC':
107 if id not in self.spdx.exceptions:
108 raise ParserException(tok, 'Invalid Exception ID')
109 if self.lastid not in self.spdx.exceptions[id]:
110 raise ParserException(tok, 'Exception not valid for license %s' %self.lastid)
111 self.lastid = None
112 elif tok.type != 'WITH':
113 self.lastid = None
114
115 # Lexer functions
116 def t_RPAR(self, tok):
117 r'\)'
118 self.lasttok = tok.type
119 return tok
120
121 def t_LPAR(self, tok):
122 r'\('
123 self.lasttok = tok.type
124 return tok
125
126 def t_ID(self, tok):
127 r'[A-Za-z.0-9\-+]+'
128
129 if self.lasttok == 'EXC':
130 print(tok)
131 raise ParserException(tok, 'Missing parentheses')
132
133 tok.value = tok.value.strip()
134 val = tok.value.upper()
135
136 if val in self.reserved:
137 tok.type = val
138 elif self.lasttok == 'WITH':
139 tok.type = 'EXC'
140
141 self.lasttok = tok.type
142 self.validate(tok)
143 return tok
144
145 def t_error(self, tok):
146 raise ParserException(tok, 'Invalid token')
147
148 def p_expr(self, p):
149 '''expr : ID
150 | ID WITH EXC
151 | expr AND expr
152 | expr OR expr
153 | LPAR expr RPAR'''
154 pass
155
156 def p_error(self, p):
157 if not p:
158 raise ParserException(None, 'Unfinished license expression')
159 else:
160 raise ParserException(p, 'Syntax error')
161
162 def parse(self, expr):
163 self.lasttok = None
164 self.lastid = None
165 self.parser.parse(expr, lexer = self.lexer)
166
167 def parse_lines(self, fd, maxlines, fname):
168 self.checked += 1
169 self.curline = 0
170 try:
171 for line in fd:
172 line = line.decode(locale.getpreferredencoding(False), errors='ignore')
173 self.curline += 1
174 if self.curline > maxlines:
175 break
176 self.lines_checked += 1
177 if line.find("SPDX-License-Identifier:") < 0:
178 continue
179 expr = line.split(':')[1].strip()
180 # Remove trailing comment closure
181 if line.strip().endswith('*/'):
182 expr = expr.rstrip('*/').strip()
183 # Special case for SH magic boot code files
184 if line.startswith('LIST \"'):
185 expr = expr.rstrip('\"').strip()
186 self.parse(expr)
187 self.spdx_valid += 1
188 #
189 # Should we check for more SPDX ids in the same file and
190 # complain if there are any?
191 #
192 break
193
194 except ParserException as pe:
195 if pe.tok:
196 col = line.find(expr) + pe.tok.lexpos
197 tok = pe.tok.value
198 sys.stdout.write('%s: %d:%d %s: %s\n' %(fname, self.curline, col, pe.txt, tok))
199 else:
200 sys.stdout.write('%s: %d:0 %s\n' %(fname, self.curline, col, pe.txt))
201 self.spdx_errors += 1
202
203def scan_git_tree(tree):
204 for el in tree.traverse():
205 # Exclude stuff which would make pointless noise
206 # FIXME: Put this somewhere more sensible
207 if el.path.startswith("LICENSES"):
208 continue
209 if el.path.find("license-rules.rst") >= 0:
210 continue
211 if not os.path.isfile(el.path):
212 continue
213 with open(el.path, 'rb') as fd:
214 parser.parse_lines(fd, args.maxlines, el.path)
215
216def scan_git_subtree(tree, path):
217 for p in path.strip('/').split('/'):
218 tree = tree[p]
219 scan_git_tree(tree)
220
221if __name__ == '__main__':
222
223 ap = ArgumentParser(description='SPDX expression checker')
224 ap.add_argument('path', nargs='*', help='Check path or file. If not given full git tree scan. For stdin use "-"')
225 ap.add_argument('-m', '--maxlines', type=int, default=15,
226 help='Maximum number of lines to scan in a file. Default 15')
227 ap.add_argument('-v', '--verbose', action='store_true', help='Verbose statistics output')
228 args = ap.parse_args()
229
230 # Sanity check path arguments
231 if '-' in args.path and len(args.path) > 1:
232 sys.stderr.write('stdin input "-" must be the only path argument\n')
233 sys.exit(1)
234
235 try:
236 # Use git to get the valid license expressions
237 repo = git.Repo(os.getcwd())
238 assert not repo.bare
239
240 # Initialize SPDX data
241 spdx = read_spdxdata(repo)
242
243 # Initilize the parser
244 parser = id_parser(spdx)
245
246 except SPDXException as se:
247 if se.el:
248 sys.stderr.write('%s: %s\n' %(se.el.path, se.txt))
249 else:
250 sys.stderr.write('%s\n' %se.txt)
251 sys.exit(1)
252
253 except Exception as ex:
254 sys.stderr.write('FAIL: %s\n' %ex)
255 sys.stderr.write('%s\n' %traceback.format_exc())
256 sys.exit(1)
257
258 try:
259 if len(args.path) and args.path[0] == '-':
260 stdin = os.fdopen(sys.stdin.fileno(), 'rb')
261 parser.parse_lines(stdin, args.maxlines, '-')
262 else:
263 if args.path:
264 for p in args.path:
265 if os.path.isfile(p):
266 parser.parse_lines(open(p, 'rb'), args.maxlines, p)
267 elif os.path.isdir(p):
268 scan_git_subtree(repo.head.reference.commit.tree, p)
269 else:
270 sys.stderr.write('path %s does not exist\n' %p)
271 sys.exit(1)
272 else:
273 # Full git tree scan
274 scan_git_tree(repo.head.commit.tree)
275
276 if args.verbose:
277 sys.stderr.write('\n')
278 sys.stderr.write('License files: %12d\n' %spdx.license_files)
279 sys.stderr.write('Exception files: %12d\n' %spdx.exception_files)
280 sys.stderr.write('License IDs %12d\n' %len(spdx.licenses))
281 sys.stderr.write('Exception IDs %12d\n' %len(spdx.exceptions))
282 sys.stderr.write('\n')
283 sys.stderr.write('Files checked: %12d\n' %parser.checked)
284 sys.stderr.write('Lines checked: %12d\n' %parser.lines_checked)
285 sys.stderr.write('Files with SPDX: %12d\n' %parser.spdx_valid)
286 sys.stderr.write('Files with errors: %12d\n' %parser.spdx_errors)
287
288 sys.exit(0)
289
290 except Exception as ex:
291 sys.stderr.write('FAIL: %s\n' %ex)
292 sys.stderr.write('%s\n' %traceback.format_exc())
293 sys.exit(1)