def get_scancode_file(file_dict):
'''Given a file dictionary from the scancode results, return a FileData
object with the results'''
# scancode records paths from the target directory onwards
# which in tern's case is tern.utils.constants.untar_dir
# removing that portion of the file path
fspath = file_dict['path'].replace(
constants.untar_dir + os.path.sep, '')
fd = FileData(
file_dict['name'], fspath, file_dict['date'], file_dict['file_type'])
fd.short_file_type = get_file_type(file_dict)
fd.add_checksums({'sha1': file_dict['sha1'], 'md5': file_dict['md5']})
if file_dict['licenses']:
fd.licenses = [li['short_name'] for li in file_dict['licenses']]
fd.license_expressions = file_dict['license_expressions']
if file_dict['copyrights']:
fd.copyrights = [c['value'] for c in file_dict['copyrights']]
if file_dict['urls']:
fd.urls = [u['url'] for u in file_dict['urls']]
fd.packages = file_dict['packages']
fd.authors = [a['value'] for a in file_dict['authors']]
if file_dict['scan_errors']:
# for each scan error make a notice
for err in file_dict['scan_errors']:
fd.origins.add_notice_to_origins(
'File: ' + fd.path, Notice(err, 'error'))
return fd
def testInstance(self):
file1 = FileData('file1', 'path/to/file1')
self.assertEqual(file1.name, 'file1')
self.assertEqual(file1.path, 'path/to/file1')
self.assertFalse(file1.checksum_type)
self.assertFalse(file1.checksum)
self.assertFalse(file1.date)
self.assertFalse(file1.version_control)
self.assertFalse(file1.version)
self.assertFalse(file1.file_type)
self.assertFalse(file1.licenses)
self.assertFalse(file1.license_expressions)
self.assertFalse(file1.copyrights)
self.assertFalse(file1.authors)
self.assertFalse(file1.packages)
self.assertFalse(file1.urls)
with self.assertRaises(ValueError):
file2 = FileData('file2', 'path/to/file2', '12355')
file2 = FileData('file2', 'path/to/file2', '2020-01-01', 'binary')
self.assertEqual(file2.date, '2020-01-01')
self.assertEqual(file2.file_type, 'binary')
file2.licenses = ['MIT', 'GPL']
file2.license_expressions = ['GPLv2 or MIT', 'MIT and GPLv2']
file2.copyrights = ['copyrights']
file2.authors = ['author1', 'author2']
file2.packages = ['package1', 'package2']
self.assertEqual(file2.licenses, ['MIT', 'GPL'])
self.assertEqual(file2.license_expressions,
['GPLv2 or MIT', 'MIT and GPLv2'])
self.assertEqual(file2.copyrights, ['copyrights'])
self.assertEqual(file2.authors, ['author1', 'author2'])
self.assertEqual(file2.packages, ['package1', 'package2'])
def testMerge(self):
file1 = FileData('switch_root', 'sbin/switch_root')
file1.set_checksum('sha256', '123abc456def')
file1.extattrs = '-rwxr-xr-x|1000|1000|14408|1'
file2 = FileData('switch_root', 'sbin/switch_root')
file2.add_checksums({
'SHA1': '12345abcde',
'MD5': '1ff38cc592c4c5d0c8e3ca38be8f1eb1'
})
file2.set_checksum('sha256', '123abc456def')
file2.extattrs = '-rwxr-xr-x|1000|1000|14408|1'
file2.date = '2012-02-02'
file2.file_type = 'binary'
file2.short_file_type = 'BINARY'
file2.licenses = ['MIT', 'GPL']
file2.license_expressions = ['MIT or GPL']
file2.copyrights = ['copyrights']
file2.authors = ['author1', 'author2']
file2.packages = ['package1', 'package2']
file2.urls = ['url1', 'url2']
file2.origins.add_notice_to_origins(
'scanning', Notice('something happened', 'error'))
file3 = FileData('switch_root', 'sbin/switch_root')
file3.set_checksum('sha1', '456def123abc')
file4 = FileData('e2image', 'sbin/e2image')
self.assertFalse(file1.merge(file4))
self.assertTrue(file1.merge(file3))
self.assertEqual(file1.checksum, '123abc456def')
self.assertEqual(file1.extattrs, '-rwxr-xr-x|1000|1000|14408|1')
self.assertFalse(file1.merge('astring'))
self.assertTrue(file1.merge(file2))
self.assertEqual(file1.checksums, {
'sha1': '12345abcde',
'md5': '1ff38cc592c4c5d0c8e3ca38be8f1eb1'
})
self.assertEqual(file1.date, '2012-02-02')
self.assertEqual(file1.file_type, 'binary')
self.assertEqual(file1.short_file_type, 'BINARY')
self.assertEqual(file1.licenses, ['MIT', 'GPL'])
self.assertEqual(file1.license_expressions, ['MIT or GPL'])
self.assertEqual(file1.copyrights, ['copyrights'])
self.assertEqual(file1.authors, ['author1', 'author2'])
self.assertEqual(file1.packages, ['package1', 'package2'])
self.assertEqual(file1.urls, ['url1', 'url2'])
self.assertEqual(len(file1.origins.origins), 1)
self.assertEqual(file1.origins.origins[0].origin_str, 'scanning')
self.assertEqual(len(file1.origins.origins[0].notices), 1)
self.assertEqual(file1.origins.origins[0].notices[0].message,
'something happened')
def collect_layer_data(layer_obj):
'''Use scancode to collect data from a layer filesystem. This function will
create a FileData object for every file found. After scanning, it will
return a list of FileData objects.
'''
files = []
# run scancode against a directory
command = 'scancode -ilpcu --quiet --json -'
full_cmd = get_filesystem_command(layer_obj, command)
origin_layer = 'Layer: ' + layer_obj.fs_hash[:10]
result, error = rootfs.shell_command(True, full_cmd)
if not result:
logger.error("No scancode results for this layer: %s", str(error))
layer_obj.origins.add_notice_to_origins(origin_layer,
Notice(str(error), 'error'))
else:
# make FileData objects for each result
data = json.loads(result)
for f in data['files']:
if f['type'] == 'file':
# scancode records paths from the target directory onwards
# which in tern's case is tern.utils.constants.untar_dir
# removing that portion of the file path
fspath = f['path'].replace(constants.untar_dir + os.path.sep,
'')
fd = FileData(f['name'], fspath, f['date'], f['file_type'])
if f['licenses']:
fd.licenses = [l['short_name'] for l in f['licenses']]
fd.license_expressions = f['license_expressions']
if f['copyrights']:
fd.copyrights = [c['value'] for c in f['copyrights']]
if f['urls']:
fd.urls = [u['url'] for u in f['urls']]
fd.packages = f['packages']
fd.authors = [a['value'] for a in f['authors']]
if f['scan_errors']:
# for each scan error make a notice
for err in f['scan_errors']:
fd.origins.add_notice_to_origins(
'File: ' + fd.path, Notice(err, 'error'))
files.append(fd)
return files
