1 #!/usr/bin/env python3 1 #!/usr/bin/env python3
2 # SPDX-License-Identifier: GPL-2.0 2 # SPDX-License-Identifier: GPL-2.0
3 # Copyright Thomas Gleixner <tglx@linutronix.de 3 # Copyright Thomas Gleixner <tglx@linutronix.de>
4 4
5 from argparse import ArgumentParser 5 from argparse import ArgumentParser
6 from ply import lex, yacc 6 from ply import lex, yacc
7 import locale 7 import locale
8 import traceback 8 import traceback
9 import fnmatch <<
10 import sys 9 import sys
11 import git 10 import git
12 import re 11 import re
13 import os 12 import os
14 13
15 class ParserException(Exception): 14 class ParserException(Exception):
16 def __init__(self, tok, txt): 15 def __init__(self, tok, txt):
17 self.tok = tok 16 self.tok = tok
18 self.txt = txt 17 self.txt = txt
19 18
20 class SPDXException(Exception): 19 class SPDXException(Exception):
21 def __init__(self, el, txt): 20 def __init__(self, el, txt):
22 self.el = el 21 self.el = el
23 self.txt = txt 22 self.txt = txt
24 23
25 class SPDXdata(object): 24 class SPDXdata(object):
26 def __init__(self): 25 def __init__(self):
27 self.license_files = 0 26 self.license_files = 0
28 self.exception_files = 0 27 self.exception_files = 0
29 self.licenses = [ ] 28 self.licenses = [ ]
30 self.exceptions = { } 29 self.exceptions = { }
31 30
32 class dirinfo(object): <<
33 def __init__(self): <<
34 self.missing = 0 <<
35 self.total = 0 <<
36 self.files = [] <<
37 <<
38 def update(self, fname, basedir, miss): <<
39 self.total += 1 <<
40 self.missing += miss <<
41 if miss: <<
42 fname = './' + fname <<
43 bdir = os.path.dirname(fname) <<
44 if bdir == basedir.rstrip('/'): <<
45 self.files.append(fname) <<
46 <<
47 # Read the spdx data from the LICENSES directo 31 # Read the spdx data from the LICENSES directory
48 def read_spdxdata(repo): 32 def read_spdxdata(repo):
49 33
50 # The subdirectories of LICENSES in the ke 34 # The subdirectories of LICENSES in the kernel source
51 # Note: exceptions needs to be parsed as l 35 # Note: exceptions needs to be parsed as last directory.
52 license_dirs = [ "preferred", "dual", "dep 36 license_dirs = [ "preferred", "dual", "deprecated", "exceptions" ]
53 lictree = repo.head.commit.tree['LICENSES' 37 lictree = repo.head.commit.tree['LICENSES']
54 38
55 spdx = SPDXdata() 39 spdx = SPDXdata()
56 40
57 for d in license_dirs: 41 for d in license_dirs:
58 for el in lictree[d].traverse(): 42 for el in lictree[d].traverse():
59 if not os.path.isfile(el.path): 43 if not os.path.isfile(el.path):
60 continue 44 continue
61 45
62 exception = None 46 exception = None
63 for l in open(el.path, encoding="u 47 for l in open(el.path, encoding="utf-8").readlines():
64 if l.startswith('Valid-License 48 if l.startswith('Valid-License-Identifier:'):
65 lid = l.split(':')[1].stri 49 lid = l.split(':')[1].strip().upper()
66 if lid in spdx.licenses: 50 if lid in spdx.licenses:
67 raise SPDXException(el 51 raise SPDXException(el, 'Duplicate License Identifier: %s' %lid)
68 else: 52 else:
69 spdx.licenses.append(l 53 spdx.licenses.append(lid)
70 54
71 elif l.startswith('SPDX-Except 55 elif l.startswith('SPDX-Exception-Identifier:'):
72 exception = l.split(':')[1 56 exception = l.split(':')[1].strip().upper()
73 spdx.exceptions[exception] 57 spdx.exceptions[exception] = []
74 58
75 elif l.startswith('SPDX-Licens 59 elif l.startswith('SPDX-Licenses:'):
76 for lic in l.split(':')[1] 60 for lic in l.split(':')[1].upper().strip().replace(' ', '').replace('\t', '').split(','):
77 if not lic in spdx.lic 61 if not lic in spdx.licenses:
78 raise SPDXExceptio 62 raise SPDXException(None, 'Exception %s missing license %s' %(exception, lic))
79 spdx.exceptions[except 63 spdx.exceptions[exception].append(lic)
80 64
81 elif l.startswith("License-Tex 65 elif l.startswith("License-Text:"):
82 if exception: 66 if exception:
83 if not len(spdx.except 67 if not len(spdx.exceptions[exception]):
84 raise SPDXExceptio 68 raise SPDXException(el, 'Exception %s is missing SPDX-Licenses' %exception)
85 spdx.exception_files + 69 spdx.exception_files += 1
86 else: 70 else:
87 spdx.license_files += 71 spdx.license_files += 1
88 break 72 break
89 return spdx 73 return spdx
90 74
91 class id_parser(object): 75 class id_parser(object):
92 76
93 reserved = [ 'AND', 'OR', 'WITH' ] 77 reserved = [ 'AND', 'OR', 'WITH' ]
94 tokens = [ 'LPAR', 'RPAR', 'ID', 'EXC' ] + 78 tokens = [ 'LPAR', 'RPAR', 'ID', 'EXC' ] + reserved
95 79
96 precedence = ( ('nonassoc', 'AND', 'OR'), 80 precedence = ( ('nonassoc', 'AND', 'OR'), )
97 81
98 t_ignore = ' \t' 82 t_ignore = ' \t'
99 83
100 def __init__(self, spdx): 84 def __init__(self, spdx):
101 self.spdx = spdx 85 self.spdx = spdx
102 self.lasttok = None 86 self.lasttok = None
103 self.lastid = None 87 self.lastid = None
104 self.lexer = lex.lex(module = self, re 88 self.lexer = lex.lex(module = self, reflags = re.UNICODE)
105 # Initialize the parser. No debug file 89 # Initialize the parser. No debug file and no parser rules stored on disk
106 # The rules are small enough to be gen 90 # The rules are small enough to be generated on the fly
107 self.parser = yacc.yacc(module = self, 91 self.parser = yacc.yacc(module = self, write_tables = False, debug = False)
108 self.lines_checked = 0 92 self.lines_checked = 0
109 self.checked = 0 93 self.checked = 0
110 self.excluded = 0 <<
111 self.spdx_valid = 0 94 self.spdx_valid = 0
112 self.spdx_errors = 0 95 self.spdx_errors = 0
113 self.spdx_dirs = {} <<
114 self.dirdepth = -1 <<
115 self.basedir = '.' <<
116 self.curline = 0 96 self.curline = 0
117 self.deepest = 0 97 self.deepest = 0
118 98
119 def set_dirinfo(self, basedir, dirdepth): <<
120 if dirdepth >= 0: <<
121 self.basedir = basedir <<
122 bdir = basedir.lstrip('./').rstrip <<
123 if bdir != '': <<
124 parts = bdir.split('/') <<
125 else: <<
126 parts = [] <<
127 self.dirdepth = dirdepth + len(par <<
128 <<
129 # Validate License and Exception IDs 99 # Validate License and Exception IDs
130 def validate(self, tok): 100 def validate(self, tok):
131 id = tok.value.upper() 101 id = tok.value.upper()
132 if tok.type == 'ID': 102 if tok.type == 'ID':
133 if not id in self.spdx.licenses: 103 if not id in self.spdx.licenses:
134 raise ParserException(tok, 'In 104 raise ParserException(tok, 'Invalid License ID')
135 self.lastid = id 105 self.lastid = id
136 elif tok.type == 'EXC': 106 elif tok.type == 'EXC':
137 if id not in self.spdx.exceptions: 107 if id not in self.spdx.exceptions:
138 raise ParserException(tok, 'In 108 raise ParserException(tok, 'Invalid Exception ID')
139 if self.lastid not in self.spdx.ex 109 if self.lastid not in self.spdx.exceptions[id]:
140 raise ParserException(tok, 'Ex 110 raise ParserException(tok, 'Exception not valid for license %s' %self.lastid)
141 self.lastid = None 111 self.lastid = None
142 elif tok.type != 'WITH': 112 elif tok.type != 'WITH':
143 self.lastid = None 113 self.lastid = None
144 114
145 # Lexer functions 115 # Lexer functions
146 def t_RPAR(self, tok): 116 def t_RPAR(self, tok):
147 r'\)' 117 r'\)'
148 self.lasttok = tok.type 118 self.lasttok = tok.type
149 return tok 119 return tok
150 120
151 def t_LPAR(self, tok): 121 def t_LPAR(self, tok):
152 r'\(' 122 r'\('
153 self.lasttok = tok.type 123 self.lasttok = tok.type
154 return tok 124 return tok
155 125
156 def t_ID(self, tok): 126 def t_ID(self, tok):
157 r'[A-Za-z.0-9\-+]+' 127 r'[A-Za-z.0-9\-+]+'
158 128
159 if self.lasttok == 'EXC': 129 if self.lasttok == 'EXC':
160 print(tok) 130 print(tok)
161 raise ParserException(tok, 'Missin 131 raise ParserException(tok, 'Missing parentheses')
162 132
163 tok.value = tok.value.strip() 133 tok.value = tok.value.strip()
164 val = tok.value.upper() 134 val = tok.value.upper()
165 135
166 if val in self.reserved: 136 if val in self.reserved:
167 tok.type = val 137 tok.type = val
168 elif self.lasttok == 'WITH': 138 elif self.lasttok == 'WITH':
169 tok.type = 'EXC' 139 tok.type = 'EXC'
170 140
171 self.lasttok = tok.type 141 self.lasttok = tok.type
172 self.validate(tok) 142 self.validate(tok)
173 return tok 143 return tok
174 144
175 def t_error(self, tok): 145 def t_error(self, tok):
176 raise ParserException(tok, 'Invalid to 146 raise ParserException(tok, 'Invalid token')
177 147
178 def p_expr(self, p): 148 def p_expr(self, p):
179 '''expr : ID 149 '''expr : ID
180 | ID WITH EXC 150 | ID WITH EXC
181 | expr AND expr 151 | expr AND expr
182 | expr OR expr 152 | expr OR expr
183 | LPAR expr RPAR''' 153 | LPAR expr RPAR'''
184 pass 154 pass
185 155
186 def p_error(self, p): 156 def p_error(self, p):
187 if not p: 157 if not p:
188 raise ParserException(None, 'Unfin 158 raise ParserException(None, 'Unfinished license expression')
189 else: 159 else:
190 raise ParserException(p, 'Syntax e 160 raise ParserException(p, 'Syntax error')
191 161
192 def parse(self, expr): 162 def parse(self, expr):
193 self.lasttok = None 163 self.lasttok = None
194 self.lastid = None 164 self.lastid = None
195 self.parser.parse(expr, lexer = self.l 165 self.parser.parse(expr, lexer = self.lexer)
196 166
197 def parse_lines(self, fd, maxlines, fname) 167 def parse_lines(self, fd, maxlines, fname):
198 self.checked += 1 168 self.checked += 1
199 self.curline = 0 169 self.curline = 0
200 fail = 1 <<
201 try: 170 try:
202 for line in fd: 171 for line in fd:
203 line = line.decode(locale.getp 172 line = line.decode(locale.getpreferredencoding(False), errors='ignore')
204 self.curline += 1 173 self.curline += 1
205 if self.curline > maxlines: 174 if self.curline > maxlines:
206 break 175 break
207 self.lines_checked += 1 176 self.lines_checked += 1
208 if line.find("SPDX-License-Ide 177 if line.find("SPDX-License-Identifier:") < 0:
209 continue 178 continue
210 expr = line.split(':')[1].stri 179 expr = line.split(':')[1].strip()
211 # Remove trailing comment clos 180 # Remove trailing comment closure
212 if line.strip().endswith('*/') 181 if line.strip().endswith('*/'):
213 expr = expr.rstrip('*/').s 182 expr = expr.rstrip('*/').strip()
214 # Remove trailing xml comment 183 # Remove trailing xml comment closure
215 if line.strip().endswith('-->' 184 if line.strip().endswith('-->'):
216 expr = expr.rstrip('-->'). 185 expr = expr.rstrip('-->').strip()
217 # Special case for SH magic bo 186 # Special case for SH magic boot code files
218 if line.startswith('LIST \"'): 187 if line.startswith('LIST \"'):
219 expr = expr.rstrip('\"').s 188 expr = expr.rstrip('\"').strip()
220 self.parse(expr) 189 self.parse(expr)
221 self.spdx_valid += 1 190 self.spdx_valid += 1
222 # 191 #
223 # Should we check for more SPD 192 # Should we check for more SPDX ids in the same file and
224 # complain if there are any? 193 # complain if there are any?
225 # 194 #
226 fail = 0 <<
227 break 195 break
228 196
229 except ParserException as pe: 197 except ParserException as pe:
230 if pe.tok: 198 if pe.tok:
231 col = line.find(expr) + pe.tok 199 col = line.find(expr) + pe.tok.lexpos
232 tok = pe.tok.value 200 tok = pe.tok.value
233 sys.stdout.write('%s: %d:%d %s 201 sys.stdout.write('%s: %d:%d %s: %s\n' %(fname, self.curline, col, pe.txt, tok))
234 else: 202 else:
235 sys.stdout.write('%s: %d:0 %s\ 203 sys.stdout.write('%s: %d:0 %s\n' %(fname, self.curline, pe.txt))
236 self.spdx_errors += 1 204 self.spdx_errors += 1
237 205
238 if fname == '-': !! 206 def scan_git_tree(tree):
239 return <<
240 <<
241 base = os.path.dirname(fname) <<
242 if self.dirdepth > 0: <<
243 parts = base.split('/') <<
244 i = 0 <<
245 base = '.' <<
246 while i < self.dirdepth and i < le <<
247 base += '/' + parts[i] <<
248 i += 1 <<
249 elif self.dirdepth == 0: <<
250 base = self.basedir <<
251 else: <<
252 base = './' + base.rstrip('/') <<
253 base += '/' <<
254 <<
255 di = self.spdx_dirs.get(base, dirinfo( <<
256 di.update(fname, base, fail) <<
257 self.spdx_dirs[base] = di <<
258 <<
259 class pattern(object): <<
260 def __init__(self, line): <<
261 self.pattern = line <<
262 self.match = self.match_file <<
263 if line == '.*': <<
264 self.match = self.match_dot <<
265 elif line.endswith('/'): <<
266 self.pattern = line[:-1] <<
267 self.match = self.match_dir <<
268 elif line.startswith('/'): <<
269 self.pattern = line[1:] <<
270 self.match = self.match_fn <<
271 <<
272 def match_dot(self, fpath): <<
273 return os.path.basename(fpath).startsw <<
274 <<
275 def match_file(self, fpath): <<
276 return os.path.basename(fpath) == self <<
277 <<
278 def match_fn(self, fpath): <<
279 return fnmatch.fnmatchcase(fpath, self <<
280 <<
281 def match_dir(self, fpath): <<
282 if self.match_fn(os.path.dirname(fpath <<
283 return True <<
284 return fpath.startswith(self.pattern) <<
285 <<
286 def exclude_file(fpath): <<
287 for rule in exclude_rules: <<
288 if rule.match(fpath): <<
289 return True <<
290 return False <<
291 <<
292 def scan_git_tree(tree, basedir, dirdepth): <<
293 parser.set_dirinfo(basedir, dirdepth) <<
294 for el in tree.traverse(): 207 for el in tree.traverse():
295 if not os.path.isfile(el.path): !! 208 # Exclude stuff which would make pointless noise
>> 209 # FIXME: Put this somewhere more sensible
>> 210 if el.path.startswith("LICENSES"):
296 continue 211 continue
297 if exclude_file(el.path): !! 212 if el.path.find("license-rules.rst") >= 0:
298 parser.excluded += 1 !! 213 continue
>> 214 if not os.path.isfile(el.path):
299 continue 215 continue
300 with open(el.path, 'rb') as fd: 216 with open(el.path, 'rb') as fd:
301 parser.parse_lines(fd, args.maxlin 217 parser.parse_lines(fd, args.maxlines, el.path)
302 218
303 def scan_git_subtree(tree, path, dirdepth): !! 219 def scan_git_subtree(tree, path):
304 for p in path.strip('/').split('/'): 220 for p in path.strip('/').split('/'):
305 tree = tree[p] 221 tree = tree[p]
306 scan_git_tree(tree, path.strip('/'), dirde !! 222 scan_git_tree(tree)
307 <<
308 def read_exclude_file(fname): <<
309 rules = [] <<
310 if not fname: <<
311 return rules <<
312 with open(fname) as fd: <<
313 for line in fd: <<
314 line = line.strip() <<
315 if line.startswith('#'): <<
316 continue <<
317 if not len(line): <<
318 continue <<
319 rules.append(pattern(line)) <<
320 return rules <<
321 223
322 if __name__ == '__main__': 224 if __name__ == '__main__':
323 225
324 ap = ArgumentParser(description='SPDX expr 226 ap = ArgumentParser(description='SPDX expression checker')
325 ap.add_argument('path', nargs='*', help='C 227 ap.add_argument('path', nargs='*', help='Check path or file. If not given full git tree scan. For stdin use "-"')
326 ap.add_argument('-d', '--dirs', action='st <<
327 help='Show [sub]directory <<
328 ap.add_argument('-D', '--depth', type=int, <<
329 help='Directory depth for <<
330 ap.add_argument('-e', '--exclude', <<
331 help='File containing file <<
332 ap.add_argument('-f', '--files', action='s <<
333 help='Show files without S <<
334 ap.add_argument('-m', '--maxlines', type=i 228 ap.add_argument('-m', '--maxlines', type=int, default=15,
335 help='Maximum number of li 229 help='Maximum number of lines to scan in a file. Default 15')
336 ap.add_argument('-v', '--verbose', action= 230 ap.add_argument('-v', '--verbose', action='store_true', help='Verbose statistics output')
337 args = ap.parse_args() 231 args = ap.parse_args()
338 232
339 # Sanity check path arguments 233 # Sanity check path arguments
340 if '-' in args.path and len(args.path) > 1 234 if '-' in args.path and len(args.path) > 1:
341 sys.stderr.write('stdin input "-" must 235 sys.stderr.write('stdin input "-" must be the only path argument\n')
342 sys.exit(1) 236 sys.exit(1)
343 237
344 try: 238 try:
345 # Use git to get the valid license exp 239 # Use git to get the valid license expressions
346 repo = git.Repo(os.getcwd()) 240 repo = git.Repo(os.getcwd())
347 assert not repo.bare 241 assert not repo.bare
348 242
349 # Initialize SPDX data 243 # Initialize SPDX data
350 spdx = read_spdxdata(repo) 244 spdx = read_spdxdata(repo)
351 245
352 # Initialize the parser 246 # Initialize the parser
353 parser = id_parser(spdx) 247 parser = id_parser(spdx)
354 248
355 except SPDXException as se: 249 except SPDXException as se:
356 if se.el: 250 if se.el:
357 sys.stderr.write('%s: %s\n' %(se.e 251 sys.stderr.write('%s: %s\n' %(se.el.path, se.txt))
358 else: 252 else:
359 sys.stderr.write('%s\n' %se.txt) 253 sys.stderr.write('%s\n' %se.txt)
360 sys.exit(1) 254 sys.exit(1)
361 255
362 except Exception as ex: 256 except Exception as ex:
363 sys.stderr.write('FAIL: %s\n' %ex) 257 sys.stderr.write('FAIL: %s\n' %ex)
364 sys.stderr.write('%s\n' %traceback.for 258 sys.stderr.write('%s\n' %traceback.format_exc())
365 sys.exit(1) 259 sys.exit(1)
366 260
367 try: 261 try:
368 fname = args.exclude <<
369 if not fname: <<
370 fname = os.path.join(os.path.dirna <<
371 exclude_rules = read_exclude_file(fnam <<
372 except Exception as ex: <<
373 sys.stderr.write('FAIL: Reading exclud <<
374 sys.exit(1) <<
375 <<
376 try: <<
377 if len(args.path) and args.path[0] == 262 if len(args.path) and args.path[0] == '-':
378 stdin = os.fdopen(sys.stdin.fileno 263 stdin = os.fdopen(sys.stdin.fileno(), 'rb')
379 parser.parse_lines(stdin, args.max 264 parser.parse_lines(stdin, args.maxlines, '-')
380 else: 265 else:
381 if args.path: 266 if args.path:
382 for p in args.path: 267 for p in args.path:
383 if os.path.isfile(p): 268 if os.path.isfile(p):
384 parser.parse_lines(ope 269 parser.parse_lines(open(p, 'rb'), args.maxlines, p)
385 elif os.path.isdir(p): 270 elif os.path.isdir(p):
386 scan_git_subtree(repo. !! 271 scan_git_subtree(repo.head.reference.commit.tree, p)
387 args. <<
388 else: 272 else:
389 sys.stderr.write('path 273 sys.stderr.write('path %s does not exist\n' %p)
390 sys.exit(1) 274 sys.exit(1)
391 else: 275 else:
392 # Full git tree scan 276 # Full git tree scan
393 scan_git_tree(repo.head.commit !! 277 scan_git_tree(repo.head.commit.tree)
394 <<
395 ndirs = len(parser.spdx_dirs) <<
396 dirsok = 0 <<
397 if ndirs: <<
398 for di in parser.spdx_dirs.val <<
399 if not di.missing: <<
400 dirsok += 1 <<
401 278
402 if args.verbose: 279 if args.verbose:
403 sys.stderr.write('\n') 280 sys.stderr.write('\n')
404 sys.stderr.write('License file 281 sys.stderr.write('License files: %12d\n' %spdx.license_files)
405 sys.stderr.write('Exception fi 282 sys.stderr.write('Exception files: %12d\n' %spdx.exception_files)
406 sys.stderr.write('License IDs 283 sys.stderr.write('License IDs %12d\n' %len(spdx.licenses))
407 sys.stderr.write('Exception ID 284 sys.stderr.write('Exception IDs %12d\n' %len(spdx.exceptions))
408 sys.stderr.write('\n') 285 sys.stderr.write('\n')
409 sys.stderr.write('Files exclud <<
410 sys.stderr.write('Files checke 286 sys.stderr.write('Files checked: %12d\n' %parser.checked)
411 sys.stderr.write('Lines checke 287 sys.stderr.write('Lines checked: %12d\n' %parser.lines_checked)
412 if parser.checked: !! 288 sys.stderr.write('Files with SPDX: %12d\n' %parser.spdx_valid)
413 pc = int(100 * parser.spdx <<
414 sys.stderr.write('Files wi <<
415 missing = parser.checked - <<
416 mpc = int(100 * missing / <<
417 sys.stderr.write('Files wi <<
418 sys.stderr.write('Files with e 289 sys.stderr.write('Files with errors: %12d\n' %parser.spdx_errors)
419 if ndirs: <<
420 sys.stderr.write('\n') <<
421 sys.stderr.write('Director <<
422 pc = int(100 * dirsok / nd <<
423 sys.stderr.write('Director <<
424 <<
425 if ndirs and ndirs != dirsok and a <<
426 if args.verbose: <<
427 sys.stderr.write('\n') <<
428 sys.stderr.write('Incomplete d <<
429 for f in sorted(parser.spdx_di <<
430 di = parser.spdx_dirs[f] <<
431 if di.missing: <<
432 valid = di.total - di. <<
433 pc = int(100 * valid / <<
434 sys.stderr.write(' <<
435 <<
436 if ndirs and ndirs != dirsok and a <<
437 if args.verbose or args.dirs: <<
438 sys.stderr.write('\n') <<
439 sys.stderr.write('Files withou <<
440 for f in sorted(parser.spdx_di <<
441 di = parser.spdx_dirs[f] <<
442 for f in sorted(di.files): <<
443 sys.stderr.write(' <<
444 290
445 sys.exit(0) 291 sys.exit(0)
446 292
447 except Exception as ex: 293 except Exception as ex:
448 sys.stderr.write('FAIL: %s\n' %ex) 294 sys.stderr.write('FAIL: %s\n' %ex)
449 sys.stderr.write('%s\n' %traceback.for 295 sys.stderr.write('%s\n' %traceback.format_exc())
450 sys.exit(1) 296 sys.exit(1)
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.