[thirdparty/linux.git] / scripts / spdxcheck.py

#!/usr/bin/env python
# SPDX-License-Identifier: GPL-2.0
# Copyright Thomas Gleixner <tglx@linutronix.de>

from argparse import ArgumentParser
from ply import lex, yacc
import locale
import traceback
import sys
import git
import re
import os

class ParserException(Exception):
    def __init__(self, tok, txt):
        self.tok = tok
        self.txt = txt

class SPDXException(Exception):
    def __init__(self, el, txt):
        self.el = el
        self.txt = txt

class SPDXdata(object):
    def __init__(self):
        self.license_files = 0
        self.exception_files = 0
        self.licenses = [ ]
        self.exceptions = { }

# Read the spdx data from the LICENSES directory
def read_spdxdata(repo):

    # The subdirectories of LICENSES in the kernel source
    # Note: exceptions needs to be parsed as last directory.
    license_dirs = [ "preferred", "dual", "deprecated", "exceptions" ]
    lictree = repo.head.commit.tree['LICENSES']

    spdx = SPDXdata()

    for d in license_dirs:
        for el in lictree[d].traverse():
            if not os.path.isfile(el.path):
                continue

            exception = None
            for l in open(el.path).readlines():
                if l.startswith('Valid-License-Identifier:'):
                    lid = l.split(':')[1].strip().upper()
                    if lid in spdx.licenses:
                        raise SPDXException(el, 'Duplicate License Identifier: %s' %lid)
                    else:
                        spdx.licenses.append(lid)

                elif l.startswith('SPDX-Exception-Identifier:'):
                    exception = l.split(':')[1].strip().upper()
                    spdx.exceptions[exception] = []

                elif l.startswith('SPDX-Licenses:'):
                    for lic in l.split(':')[1].upper().strip().replace(' ', '').replace('\t', '').split(','):
                        if not lic in spdx.licenses:
                            raise SPDXException(None, 'Exception %s missing license %s' %(exception, lic))
                        spdx.exceptions[exception].append(lic)

                elif l.startswith("License-Text:"):
                    if exception:
                        if not len(spdx.exceptions[exception]):
                            raise SPDXException(el, 'Exception %s is missing SPDX-Licenses' %exception)
                        spdx.exception_files += 1
                    else:
                        spdx.license_files += 1
                    break
    return spdx

class id_parser(object):

    reserved = [ 'AND', 'OR', 'WITH' ]
    tokens = [ 'LPAR', 'RPAR', 'ID', 'EXC' ] + reserved

    precedence = ( ('nonassoc', 'AND', 'OR'), )

    t_ignore = ' \t'

    def __init__(self, spdx):
        self.spdx = spdx
        self.lasttok = None
        self.lastid = None
        self.lexer = lex.lex(module = self, reflags = re.UNICODE)
        # Initialize the parser. No debug file and no parser rules stored on disk
        # The rules are small enough to be generated on the fly
        self.parser = yacc.yacc(module = self, write_tables = False, debug = False)
        self.lines_checked = 0
        self.checked = 0
        self.spdx_valid = 0
        self.spdx_errors = 0
        self.curline = 0
        self.deepest = 0

    # Validate License and Exception IDs
    def validate(self, tok):
        id = tok.value.upper()
        if tok.type == 'ID':
            if not id in self.spdx.licenses:
                raise ParserException(tok, 'Invalid License ID')
            self.lastid = id
        elif tok.type == 'EXC':
            if id not in self.spdx.exceptions:
                raise ParserException(tok, 'Invalid Exception ID')
            if self.lastid not in self.spdx.exceptions[id]:
                raise ParserException(tok, 'Exception not valid for license %s' %self.lastid)
            self.lastid = None
        elif tok.type != 'WITH':
            self.lastid = None

    # Lexer functions
    def t_RPAR(self, tok):
        r'\)'
        self.lasttok = tok.type
        return tok

    def t_LPAR(self, tok):
        r'\('
        self.lasttok = tok.type
        return tok

    def t_ID(self, tok):
        r'[A-Za-z.0-9\-+]+'

        if self.lasttok == 'EXC':
            print(tok)
            raise ParserException(tok, 'Missing parentheses')

        tok.value = tok.value.strip()
        val = tok.value.upper()

        if val in self.reserved:
            tok.type = val
        elif self.lasttok == 'WITH':
            tok.type = 'EXC'

        self.lasttok = tok.type
        self.validate(tok)
        return tok

    def t_error(self, tok):
        raise ParserException(tok, 'Invalid token')

    def p_expr(self, p):
        '''expr : ID
                | ID WITH EXC
                | expr AND expr
                | expr OR expr
                | LPAR expr RPAR'''
        pass

    def p_error(self, p):
        if not p:
            raise ParserException(None, 'Unfinished license expression')
        else:
            raise ParserException(p, 'Syntax error')

    def parse(self, expr):
        self.lasttok = None
        self.lastid = None
        self.parser.parse(expr, lexer = self.lexer)

    def parse_lines(self, fd, maxlines, fname):
        self.checked += 1
        self.curline = 0
        try:
            for line in fd:
                line = line.decode(locale.getpreferredencoding(False), errors='ignore')
                self.curline += 1
                if self.curline > maxlines:
                    break
                self.lines_checked += 1
                if line.find("SPDX-License-Identifier:") < 0:
                    continue
                expr = line.split(':')[1].strip()
                # Remove trailing comment closure
                if line.strip().endswith('*/'):
                    expr = expr.rstrip('*/').strip()
                # Special case for SH magic boot code files
                if line.startswith('LIST \"'):
                    expr = expr.rstrip('\"').strip()
                self.parse(expr)
                self.spdx_valid += 1
                #
                # Should we check for more SPDX ids in the same file and
                # complain if there are any?
                #
                break

        except ParserException as pe:
            if pe.tok:
                col = line.find(expr) + pe.tok.lexpos
                tok = pe.tok.value
                sys.stdout.write('%s: %d:%d %s: %s\n' %(fname, self.curline, col, pe.txt, tok))
            else:
                sys.stdout.write('%s: %d:0 %s\n' %(fname, self.curline, col, pe.txt))
            self.spdx_errors += 1

def scan_git_tree(tree):
    for el in tree.traverse():
        # Exclude stuff which would make pointless noise
        # FIXME: Put this somewhere more sensible
        if el.path.startswith("LICENSES"):
            continue
        if el.path.find("license-rules.rst") >= 0:
            continue
        if not os.path.isfile(el.path):
            continue
        with open(el.path, 'rb') as fd:
            parser.parse_lines(fd, args.maxlines, el.path)

def scan_git_subtree(tree, path):
    for p in path.strip('/').split('/'):
        tree = tree[p]
    scan_git_tree(tree)

if __name__ == '__main__':

    ap = ArgumentParser(description='SPDX expression checker')
    ap.add_argument('path', nargs='*', help='Check path or file. If not given full git tree scan. For stdin use "-"')
    ap.add_argument('-m', '--maxlines', type=int, default=15,
                    help='Maximum number of lines to scan in a file. Default 15')
    ap.add_argument('-v', '--verbose', action='store_true', help='Verbose statistics output')
    args = ap.parse_args()

    # Sanity check path arguments
    if '-' in args.path and len(args.path) > 1:
        sys.stderr.write('stdin input "-" must be the only path argument\n')
        sys.exit(1)

    try:
        # Use git to get the valid license expressions
        repo = git.Repo(os.getcwd())
        assert not repo.bare

        # Initialize SPDX data
        spdx = read_spdxdata(repo)

        # Initilize the parser
        parser = id_parser(spdx)

    except SPDXException as se:
        if se.el:
            sys.stderr.write('%s: %s\n' %(se.el.path, se.txt))
        else:
            sys.stderr.write('%s\n' %se.txt)
        sys.exit(1)

    except Exception as ex:
        sys.stderr.write('FAIL: %s\n' %ex)
        sys.stderr.write('%s\n' %traceback.format_exc())
        sys.exit(1)

    try:
        if len(args.path) and args.path[0] == '-':
            stdin = os.fdopen(sys.stdin.fileno(), 'rb')
            parser.parse_lines(stdin, args.maxlines, '-')
        else:
            if args.path:
                for p in args.path:
                    if os.path.isfile(p):
                        parser.parse_lines(open(p, 'rb'), args.maxlines, p)
                    elif os.path.isdir(p):
                        scan_git_subtree(repo.head.reference.commit.tree, p)
                    else:
                        sys.stderr.write('path %s does not exist\n' %p)
                        sys.exit(1)
            else:
                # Full git tree scan
                scan_git_tree(repo.head.commit.tree)

            if args.verbose:
                sys.stderr.write('\n')
                sys.stderr.write('License files:     %12d\n' %spdx.license_files)
                sys.stderr.write('Exception files:   %12d\n' %spdx.exception_files)
                sys.stderr.write('License IDs        %12d\n' %len(spdx.licenses))
                sys.stderr.write('Exception IDs      %12d\n' %len(spdx.exceptions))
                sys.stderr.write('\n')
                sys.stderr.write('Files checked:     %12d\n' %parser.checked)
                sys.stderr.write('Lines checked:     %12d\n' %parser.lines_checked)
                sys.stderr.write('Files with SPDX:   %12d\n' %parser.spdx_valid)
                sys.stderr.write('Files with errors: %12d\n' %parser.spdx_errors)

            sys.exit(0)

    except Exception as ex:
        sys.stderr.write('FAIL: %s\n' %ex)
        sys.stderr.write('%s\n' %traceback.format_exc())
        sys.exit(1)
Commit	Line	Data
5385a295 TG	1	#!/usr/bin/env python
	2	# SPDX-License-Identifier: GPL-2.0
	3	# Copyright Thomas Gleixner <tglx@linutronix.de>
	4
	5	from argparse import ArgumentParser
	6	from ply import lex, yacc
bed95c43	7	import locale
5385a295 TG	8	import traceback
	9	import sys
	10	import git
	11	import re
	12	import os
	13
	14	class ParserException(Exception):
	15	def __init__(self, tok, txt):
	16	self.tok = tok
	17	self.txt = txt
	18
	19	class SPDXException(Exception):
	20	def __init__(self, el, txt):
	21	self.el = el
	22	self.txt = txt
	23
	24	class 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
	32	def read_spdxdata(repo):
	33
	34	# The subdirectories of LICENSES in the kernel source
8d7a7abf VF	35	# Note: exceptions needs to be parsed as last directory.
8d7a7abf VF	36	license_dirs = [ "preferred", "dual", "deprecated", "exceptions" ]
fde5e903	37	lictree = repo.head.commit.tree['LICENSES']
5385a295 TG	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:
8d7a7abf	62	raise SPDXException(None, 'Exception %s missing license %s' %(exception, lic))
5385a295 TG	63	spdx.exceptions[exception].append(lic)
	64
	65	elif l.startswith("License-Text:"):
	66	if exception:
	67	if not len(spdx.exceptions[exception]):
8d7a7abf	68	raise SPDXException(el, 'Exception %s is missing SPDX-Licenses' %exception)
5385a295 TG	69	spdx.exception_files += 1
	70	else:
	71	spdx.license_files += 1
	72	break
	73	return spdx
	74
	75	class 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':
bed95c43	107	if id not in self.spdx.exceptions:
5385a295 TG	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:
3a6ab5c7	172	line = line.decode(locale.getpreferredencoding(False), errors='ignore')
5385a295 TG	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
959b4968 TG	179	expr = line.split(':')[1].strip()
959b4968 TG	180	# Remove trailing comment closure
a5f4cb42	181	if line.strip().endswith('*/'):
959b4968 TG	182	expr = expr.rstrip('*/').strip()
	183	# Special case for SH magic boot code files
	184	if line.startswith('LIST \"'):
	185	expr = expr.rstrip('\"').strip()
5385a295 TG	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
	203	def 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
5385a295 TG	211	if not os.path.isfile(el.path):
5385a295 TG	212	continue
bed95c43 JC	213	with open(el.path, 'rb') as fd:
bed95c43 JC	214	parser.parse_lines(fd, args.maxlines, el.path)
5385a295 TG	215
	216	def scan_git_subtree(tree, path):
	217	for p in path.strip('/').split('/'):
	218	tree = tree[p]
	219	scan_git_tree(tree)
	220
	221	if __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] == '-':
3a6ab5c7 TR	260	stdin = os.fdopen(sys.stdin.fileno(), 'rb')
3a6ab5c7 TR	261	parser.parse_lines(stdin, args.maxlines, '-')
5385a295 TG	262	else:
	263	if args.path:
	264	for p in args.path:
	265	if os.path.isfile(p):
3a6ab5c7	266	parser.parse_lines(open(p, 'rb'), args.maxlines, p)
5385a295 TG	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)