def testAPKManifest(self):
from androguard.core.bytecodes.apk import APK
a = APK("examples/android/TestsAndroguard/bin/TestActivity.apk",
testzip=True)
self.assertEqual(a.get_app_name(), "TestsAndroguardApplication")
self.assertEqual(a.get_app_icon(), "res/drawable-hdpi/icon.png")
self.assertEqual(a.get_app_icon(max_dpi=120),
"res/drawable-ldpi/icon.png")
self.assertEqual(a.get_app_icon(max_dpi=160),
"res/drawable-mdpi/icon.png")
self.assertEqual(a.get_app_icon(max_dpi=240),
"res/drawable-hdpi/icon.png")
self.assertIsNone(a.get_app_icon(max_dpi=1))
self.assertEqual(a.get_main_activity(),
"tests.androguard.TestActivity")
self.assertEqual(a.get_package(), "tests.androguard")
self.assertEqual(a.get_androidversion_code(), '1')
self.assertEqual(a.get_androidversion_name(), "1.0")
self.assertEqual(a.get_min_sdk_version(), "9")
self.assertEqual(a.get_target_sdk_version(), "16")
self.assertIsNone(a.get_max_sdk_version())
self.assertEqual(a.get_permissions(), [])
self.assertEqual(a.get_declared_permissions(), [])
self.assertTrue(a.is_valid_APK())
def get_apk_info(self):
apk = APK(self.apk_file)
app_icon_file = apk.get_app_icon()
app_icon_data = apk.get_file(app_icon_file)
size = (256, 256)
buffered = BytesIO()
im = Image.open(BytesIO(app_icon_data))
im = im.resize(size, Image.ANTIALIAS)
im.save(buffered, "PNG")
app_icon_b64 = "data:image/png;base64," + base64.b64encode(
buffered.getvalue()).decode('utf-8')
self.package_name = apk.get_package()
self.app_name = apk.get_app_name()
self.report_saver.package_name = self.package_name
self.report_saver.app_name = self.app_name
self.report_saver.version = apk.get_androidversion_code()
self.report_saver.app_icon = app_icon_b64
permission_parser = PermissionParser(mode='groups')
permission_values = permission_parser.transform(
apk.get_permissions()).flatten().tolist()
permission_labels = permission_parser.labels()
self.report_saver.permissions_actual = {
permission_labels[i]: bool(v)
for i, v in enumerate(permission_values)
}
def parse_apk(apk_path):
apk = APK(apk_path)
package = apk.get_package()
version_code = apk.get_androidversion_code()
version_name = apk.get_androidversion_name()
tmp_icon = save_icon(apk, package)
return {"version_code": version_code,
"version_name": version_name,
"tmp_icon": tmp_icon,
"package": package}
def get_apk_information(_file):
'''
get apk information by androguard
params:
_file: apk file path
return:
1. apkname: apk name
2. apkcode: apk version code, eg: 222, 333
3. apkiconpath: get icon path
'''
apkinformation = APK(_file)
apkname = apkinformation.get_app_name()
apkcode = apkinformation.get_androidversion_code()
apkiconpath = apkinformation.get_app_icon()
# packagename = apkinformation.get_package()
return apkname, apkcode, apkiconpath
def analyse_local_apks(self, list_of_apks, download_folder):
"""
Analyse apks in the list list_of_apks
to check for updates and download updates
in the download_folder folder.
"""
list_apks_to_update = []
package_bunch = []
version_codes = []
unavail_items = []
UNAVAIL = "This app is not available in the Play Store"
for filename in list_of_apks:
filepath = os.path.join(download_folder, filename)
logger.info("Analyzing %s", filepath)
apk = APK(filepath)
packagename = apk.package
package_bunch.append(packagename)
version_codes.append(util.vcode(apk.get_androidversion_code()))
# BulkDetails requires only one HTTP request
# Get APK info from store
details = self.api.bulkDetails(package_bunch)
for detail, packagename, filename, apk_version_code in zip(
details, package_bunch, list_of_apks, version_codes):
# this app is not in the play store
if not detail:
unavail_items.append(((packagename, filename), UNAVAIL))
continue
store_version_code = detail['versionCode']
# Compare
if apk_version_code < store_version_code:
# Add to the download list
list_apks_to_update.append([
packagename, filename, apk_version_code, store_version_code
])
if self.logging_enable:
self.write_logfiles(None, None,
[item[0][0] for item in unavail_items])
self.print_failed(unavail_items)
return list_apks_to_update
def testAPKManifest(self):
from androguard.core.bytecodes.apk import APK
a = APK("examples/android/TestsAndroguard/bin/TestActivity.apk", testzip=True)
self.assertEqual(a.get_app_name(), "TestsAndroguardApplication")
self.assertEqual(a.get_app_icon(), "res/drawable-hdpi/icon.png")
self.assertEqual(a.get_app_icon(max_dpi=120), "res/drawable-ldpi/icon.png")
self.assertEqual(a.get_app_icon(max_dpi=160), "res/drawable-mdpi/icon.png")
self.assertEqual(a.get_app_icon(max_dpi=240), "res/drawable-hdpi/icon.png")
self.assertIsNone(a.get_app_icon(max_dpi=1))
self.assertEqual(a.get_main_activity(), "tests.androguard.TestActivity")
self.assertEqual(a.get_package(), "tests.androguard")
self.assertEqual(a.get_androidversion_code(), '1')
self.assertEqual(a.get_androidversion_name(), "1.0")
self.assertEqual(a.get_min_sdk_version(), "9")
self.assertEqual(a.get_target_sdk_version(), "16")
self.assertIsNone(a.get_max_sdk_version())
self.assertEqual(a.get_permissions(), [])
self.assertEqual(a.get_declared_permissions(), [])
self.assertTrue(a.is_valid_APK())
def get_apk_info(self):
apk = APK(self.apk_file)
app_icon_file = apk.get_app_icon()
app_icon_data = apk.get_file(app_icon_file)
size = (256, 256)
buffered = BytesIO()
im = Image.open(BytesIO(app_icon_data))
im = im.resize(size, Image.ANTIALIAS)
im.save(buffered, "PNG")
app_icon_b64 = "data:image/png;base64," + base64.b64encode(
buffered.getvalue()).decode('utf-8')
self.package_name = apk.get_package()
self.app_name = apk.get_app_name()
self.report_saver.package_name = self.package_name
self.report_saver.app_name = self.app_name
self.report_saver.version = apk.get_androidversion_code()
self.report_saver.app_icon = app_icon_b64
def get_package_name(apk_path):
"""
Extract the package name of an APK
"""
a = APK(apk_path)
return a.get_package(), a.get_androidversion_code()
def extract_features(file_path):
result = {}
try:
a = APK(file_path)
d = DalvikVMFormat(a.get_dex())
dx = Analysis(d)
vm = dvm.DalvikVMFormat(a.get_dex())
#vmx = analysis.uVMAnalysis(vm)
d.set_vmanalysis(dx)
d.set_decompiler(DecompilerDAD(d, dx))
except:
return None
result['android_version_code'] = a.get_androidversion_code()
result['android_version_name'] = a.get_androidversion_name()
result['max_sdk'] = a.get_max_sdk_version()
result['min_sdk'] = a.get_min_sdk_version()
#result['libraries'] = a.get_libraries()
result['filename'] = a.get_filename()
result['target_sdk'] = a.get_target_sdk_version()
result['md5'] = hashlib.md5(a.get_raw()).hexdigest()
result['sha256'] = hashlib.sha256(a.get_raw()).hexdigest()
result['permissions'] = a.get_permissions()
result['activities'] = a.get_activities()
result['providers'] = a.get_providers()
result['services'] = a.get_services()
result['strings'] = d.get_strings()
result['class_names'] = [c.get_name() for c in d.get_classes()]
result['method_names'] = [m.get_name() for m in d.get_methods()]
result['field_names'] = [f.get_name() for f in d.get_fields()]
#result['is_native_code'] = 1 if analysis.is_native_code(dx) else 0
result['is_obfuscation'] = 1 if analysis.is_ascii_obfuscation(d) else 0
#result['is_crypto_code'] = 1 if analysis.is_crypto_code(dx) else 0
'''result['is_dyn_code'] = 1 if analysis.is_dyn_code(dx) else 0
result['is_reflection_code'] = 1 if analysis.is_reflection_code(vmx) else 0'''
result['is_database'] = 1 if d.get_regex_strings(DB_REGEX) else 0
arr = []
s = a.get_elements("action", "name")
for i in s:
arr.append(i)
result['intents'] = arr
s_list = []
s_list.extend(result['class_names'])
s_list.extend(result['method_names'])
s_list.extend(result['field_names'])
result['entropy_rate'] = entropy_rate(s_list)
result['feature_vectors'] = {}
# Search for the presence of api calls in a given apk
result['feature_vectors']['api_calls'] = []
for call in API_CALLS:
status = 1 if dx.get_method(call) else 0
result['feature_vectors']['api_calls'].append(status)
# Search for the presence of permissions in a given apk
result['feature_vectors']['permissions'] = []
for permission in PERMISSIONS:
status = 1 if permission in result['permissions'] else 0
result['feature_vectors']['permissions'].append(status)
#Search for the presence of intents in a given apk
result['feature_vectors']['intents'] = []
n = len(INTENTS)
m = len(result['intents'])
for i in range(n):
stri = INTENTS[i]
flg = False
for j in range(m):
if stri in result['intents'][j]:
flg = True
break
if flg:
status = 1
else:
status = 0
result['feature_vectors']['intents'].append(status)
#Check for special strings in code
result['feature_vectors']['special_strings'] = []
for word in SPECIAL_STRINGS:
status = 1 if d.get_regex_strings(word) else 0
result['feature_vectors']['special_strings'].append(status)
return result
class XAPK:
def __init__(self, folder):
self.folder = Path(folder)
for x in self.folder.glob('*.apk'):
self.apk_src = Path(x)
break
for x in self.folder.glob('*.obb'):
self.obb_src = Path(x)
break
self.apk = APK(self.apk_src)
self.manifest = self.make_manifest()
self.icon = self.apk.get_file(self.apk.get_app_icon())
def make_manifest(self):
apk_size = self.apk_src.stat().st_size
if self.obb_src:
obb_size = self.obb_src.stat().st_size
else:
obb_size = 0
total_size = apk_size + obb_size
filename = self.apk.get_filename()
manifest = {}
manifest['xapk_version'] = 1
manifest['package_name'] = self.apk.get_package()
manifest['name'] = self.apk.get_app_name()
# manifest['locales_name'] = {} # TODO
manifest['version_code'] = self.apk.get_androidversion_code()
manifest['version_name'] = self.apk.get_androidversion_name()
manifest['min_sdk_version'] = self.apk.get_min_sdk_version()
manifest['target_sdk_version'] = self.apk.get_target_sdk_version()
manifest['permissions'] = self.apk.get_declared_permissions()
manifest['total_size'] = total_size
manifest['expansions'] = []
if obb_size:
main_obb = {}
main_obb[
'file'] = 'Android/obb/{package_name}/main.{version_code}.{package_name}.obb'.format(
**manifest)
main_obb['install_location'] = 'EXTERNAL_STORAGE'
main_obb[
'install_path'] = 'Android/obb/{package_name}/main.{version_code}.{package_name}.obb'.format(
**manifest)
manifest['expansions'].push(main_obb)
return manifest
def save(self):
self.name = '{package_name}_v{version_name}.xapk'.format(
**self.manifest)
zip_path = self.folder.joinpath(self.name)
zip_dir = tempfile.mkdtemp()
try:
print('copying apk to temp directory...')
apk_name = '{package_name}.apk'.format(**self.manifest)
apk_src = self.apk_src.resolve()
apk_dest = PurePath(zip_dir).joinpath(apk_name)
shutil.copy2(apk_src, apk_dest)
print('apk: OK')
if self.manifest.get('expansions'):
print('copying obb to temp directory...')
obb_name = self.manifest['expansions'][0]['install_path']
obb_src = self.obb_src.resolve()
obb_dest = PurePath(zip_dir).joinpath(obb_name)
os.makedirs(Path(obb_dest).parent, exist_ok=True)
shutil.copy2(obb_src, obb_dest)
print('obb: OK')
else:
print('no obb found')
print('creating icon in temp directory...')
icon = self.icon
icon_dest = PurePath(zip_dir).joinpath('icon.png')
with open(icon_dest, 'wb') as iconfile:
iconfile.write(icon)
print('icon: OK')
print('creating manifest in temp directory...')
manifest_dest = PurePath(zip_dir).joinpath('manifest.json')
with open(manifest_dest, 'w') as manifestfile:
s = json.dumps(self.manifest, separators=(':', ','))
manifestfile.write(s)
print('manifest: OK')
print('creating xapk archive...')
with zipfile.ZipFile(zip_path,
'w',
compression=zipfile.ZIP_DEFLATED) as zfd:
for root, dirs, files in os.walk(zip_dir):
for f in files:
filename = os.path.join(root, f)
zfd.write(filename, os.path.relpath(filename, zip_dir))
print('xapk: OK')
finally:
print('cleaning up temp directory...')
shutil.rmtree(zip_dir)
print('cleanup: OK')
def extract_features(file_path):
result = {}
try:
a = APK(file_path)
d = DalvikVMFormat(a.get_dex())
dx = Analysis(d)
vm = dvm.DalvikVMFormat(a.get_dex())
vmx = analysis.Analysis(vm)
d.set_vmanalysis(dx)
d.set_decompiler(DecompilerDAD(d, dx))
except Exception as e:
print e
return None
result['android_version_code'] = a.get_androidversion_code()
result['android_version_name'] = a.get_androidversion_name()
result['max_sdk'] = a.get_max_sdk_version()
result['min_sdk'] = a.get_min_sdk_version()
result['libraries'] = a.get_libraries()
result['filename'] = a.get_filename()
result['target_sdk'] = a.get_target_sdk_version()
result['md5'] = hashlib.md5(a.get_raw()).hexdigest()
result['sha256'] = hashlib.sha256(a.get_raw()).hexdigest()
result['permissions'] = a.get_permissions()
result['activities'] = a.get_activities()
result['providers'] = a.get_providers()
result['services'] = a.get_services()
result['strings'] = d.get_strings()
result['class_names'] = [c.get_name() for c in d.get_classes()]
result['method_names'] = [m.get_name() for m in d.get_methods()]
result['field_names'] = [f.get_name() for f in d.get_fields()]
# result['is_native_code'] = 1 if analysis.is_native_code(dx) else 0
result['is_obfuscation'] = 1 if analysis.is_ascii_obfuscation(d) else 0
# result['is_crypto_code'] = 1 if analysis.is_crypto_code(dx) else 0
# result['is_dyn_code'] = 1 if analysis.is_dyn_code(dx) else 0
# result['is_reflection_code'] = 1 if analysis.is_reflection_code(vmx) else 0
result['is_database'] = 1 if d.get_regex_strings(DB_REGEX) else 0
s_list = []
s_list.extend(result['class_names'])
s_list.extend(result['method_names'])
s_list.extend(result['field_names'])
result['entropy_rate'] = entropy_rate(s_list)
result['feature_vectors'] = {}
result['feature_vectors']['api_calls'] = []
for call in API_CALLS:
status = 1 if dx.get_method_by_name(".", call, ".") else 0
result['feature_vectors']['api_calls'].append(status)
result['feature_vectors']['permissions'] = []
for permission in PERMISSIONS:
status = 1 if permission in result['permissions'] else 0
result['feature_vectors']['permissions'].append(status)
result['feature_vectors']['special_strings'] = []
for word in SPECIAL_STRINGS:
status = 1 if d.get_regex_strings(word) else 0
result['feature_vectors']['special_strings'].append(status)
result['feature_vectors']['others'] = [
# result['is_reflection_code'],
# result['is_crypto_code'],
# result['is_native_code'],
result['is_obfuscation'],
result['is_database'],
# result['is_dyn_code']
]
return result
class StaticAnalysis:
def __init__(self, apk_path=None):
self.apk = None
self.apk_path = apk_path
self.signatures = None
self.compiled_tracker_signature = None
self.classes = None
self.app_details = None
if apk_path is not None:
self.load_apk()
def _compile_signatures(self):
"""
Compiles the regex associated to each signature, in order to speed up
the trackers detection.
:return: A compiled list of signatures.
"""
self.compiled_tracker_signature = []
try:
self.compiled_tracker_signature = [
re.compile(track.code_signature) for track in self.signatures
]
except TypeError:
print("self.signatures is not iterable")
def load_trackers_signatures(self):
"""
Load trackers signatures from the official Exodus database.
:return: a dictionary containing signatures.
"""
self.signatures = []
exodus_url = "https://reports.exodus-privacy.eu.org/api/trackers"
r = requests.get(exodus_url)
data = r.json()
for e in data['trackers']:
self.signatures.append(
namedtuple(
'tracker',
data['trackers'][e].keys())(*data['trackers'][e].values()))
self._compile_signatures()
logging.debug('{} trackers signatures loaded'.format(
len(self.signatures)))
def load_apk(self):
"""
Load the APK file.
"""
if self.apk is None:
self.apk = APK(self.apk_path)
def get_embedded_classes(self):
"""
Get the list of Java classes embedded into all DEX files.
:return: array of Java classes names as string
"""
if self.classes is not None:
return self.classes
class_regex = re.compile(r'classes.*\.dex')
with TemporaryDirectory() as tmp_dir:
with zipfile.ZipFile(self.apk_path, "r") as apk_zip:
class_infos = (info for info in apk_zip.infolist()
if class_regex.search(info.filename))
for info in class_infos:
apk_zip.extract(info, tmp_dir)
dexdump = which('dexdump')
cmd = '{} {}/classes*.dex | perl -n -e\'/[A-Z]+((?:\w+\/)+\w+)/ && print "$1\n"\'|sort|uniq'.format(
dexdump, tmp_dir)
try:
self.classes = subprocess.check_output(
cmd,
stderr=subprocess.STDOUT,
shell=True,
universal_newlines=True).splitlines()
logging.debug('{} classes found in {}'.format(
len(self.classes), self.apk_path))
return self.classes
except subprocess.CalledProcessError:
logging.error('Unable to decode {}'.format(self.apk_path))
raise Exception('Unable to decode the APK')
def detect_trackers_in_list(self, class_list):
"""
Detect embedded trackers in the provided classes list.
:return: list of embedded trackers
"""
if self.signatures is None:
self.load_trackers_signatures()
def _detect_tracker(sig, tracker, class_list):
for clazz in class_list:
if sig.search(clazz):
return tracker
return None
results = []
args = [(self.compiled_tracker_signature[index], tracker, class_list)
for (index, tracker) in enumerate(self.signatures)
if len(tracker.code_signature) > 3]
for res in itertools.starmap(_detect_tracker, args):
if res:
results.append(res)
trackers = [t for t in results if t is not None]
logging.debug('{} trackers detected in {}'.format(
len(trackers), self.apk_path))
return trackers
def detect_trackers(self, class_list_file=None):
"""
Detect embedded trackers.
:return: list of embedded trackers
"""
if self.signatures is None:
self.load_trackers_signatures()
if class_list_file is None:
return self.detect_trackers_in_list(self.get_embedded_classes())
else:
with open(class_list_file, 'r') as classes_file:
classes = classes_file.readlines()
return self.detect_trackers_in_list(classes)
def get_version(self):
"""
Get the application version name
:return: version name
"""
self.load_apk()
return self.apk.get_androidversion_name()
def get_version_code(self):
"""
Get the application version code
:return: version code
"""
self.load_apk()
return self.apk.get_androidversion_code()
def get_permissions(self):
"""
Get application permissions
:return: application permissions list
"""
self.load_apk()
return self.apk.get_permissions()
def get_app_name(self):
"""
Get application name
:return: application name
"""
self.load_apk()
return self.apk.get_app_name()
def get_package(self):
"""
Get application package
:return: application package
"""
self.load_apk()
return self.apk.get_package()
def get_libraries(self):
"""
Get application libraries
:return: application libraries list
"""
self.load_apk()
return self.apk.get_libraries()
def get_icon_path(self):
"""
Get the icon path in the ZIP archive
:return: icon path in the ZIP archive
"""
self.load_apk()
return self.apk.get_app_icon()
def get_application_details(self):
"""
Get the application details like creator, number of downloads, etc.
:param handle: application handle
:return: application details dictionary
"""
self.load_apk()
if self.app_details is not None:
return self.app_details
details = get_details_from_gplaycli(self.get_package())
if details is not None:
self.app_details = details
return details
def _get_icon_from_details(self, path):
"""
Get icon from applications details dictionary
:param path: path where to write the icon file
:return: icon path
:raises Exception: if unable to find icon
"""
details = self.get_application_details()
if details is not None:
for i in details.get('images'):
if i.get('imageType') == 4:
f = requests.get(i.get('url'))
with open(path, mode='wb') as fp:
fp.write(f.content)
if os.path.isfile(path) and os.path.getsize(path) > 0:
return path
raise Exception('Unable to download the icon from details')
@staticmethod
def _render_drawable_to_png(self, bxml, path):
ap = axml.AXMLPrinter(bxml)
print(ap.get_buff())
def save_icon(self, path):
"""
Extract the icon from the ZIP archive and save it at the given path
:param path: destination path of the icon
:return: destination path of the icon, None in case of error
"""
try:
icon = self.get_icon_path()
if icon is None:
raise Exception('Unable to get icon path')
with zipfile.ZipFile(self.apk_path) as z:
with open(path, 'wb') as f:
f.write(z.read(icon))
with Image.open(path) as _:
logging.info('Get icon from APK: success')
return path
except Exception:
logging.warning('Unable to get the icon from the APK')
return None
# TODO: Set this back once details download is working again
# logging.warning('Downloading icon from details')
# try:
# saved_path = self._get_icon_from_details(path)
# logging.debug('Icon downloaded from application details')
# return saved_path
# except Exception as e:
# logging.warning(e)
def get_icon_phash(self):
"""
Get the perceptual hash of the application icon
:return: the perceptual hash, empty string in case of error
"""
with NamedTemporaryFile() as ic:
path = self.save_icon(ic.name)
if path is None:
logging.error('Unable to save the icon')
return ''
return self.get_phash(ic.name)
@staticmethod
def get_phash(image_name):
"""
Get the perceptual hash of the given image
:param image_name: name of the image file
:return: the perceptual hash, empty string in case of error
"""
dhash.force_pil() # Force PIL
try:
image = Image.open(image_name).convert("RGBA")
row, col = dhash.dhash_row_col(image, size=PHASH_SIZE)
return row << (PHASH_SIZE * PHASH_SIZE) | col
except IOError as e:
logging.error(e)
return ''
@staticmethod
def get_icon_similarity(phash_origin, phash_candidate):
"""
Get icons similarity score [0,1.0]
:param phash_origin: original icon
:param phash_candidate: icon to be compared
:return: similarity score [0,1.0]
"""
diff = dhash.get_num_bits_different(phash_origin, phash_candidate)
return 1 - 1. * diff / (PHASH_SIZE * PHASH_SIZE * 2)
def get_application_universal_id(self):
parts = [self.get_package()]
for c in self.get_certificates():
parts.append(c.fingerprint.upper())
return sha1(' '.join(parts).encode('utf-8')).hexdigest().upper()
def get_certificates(self):
certificates = []
def _my_name_init(self, oid, value, _type=_SENTINEL):
if not isinstance(oid, ObjectIdentifier):
raise TypeError(
"oid argument must be an ObjectIdentifier instance.")
if not isinstance(value, six.text_type):
raise TypeError("value argument must be a text type.")
if len(value) == 0:
raise ValueError("Value cannot be an empty string")
if _type == _SENTINEL:
_type = _NAMEOID_DEFAULT_TYPE.get(oid, _ASN1Type.UTF8String)
if not isinstance(_type, _ASN1Type):
raise TypeError("_type must be from the _ASN1Type enum")
self._oid = oid
self._value = value
self._type = _type
NameAttribute.__init__ = _my_name_init
signs = self.apk.get_signature_names()
for s in signs:
c = self.apk.get_certificate(s)
cert = Certificate(c)
certificates.append(cert)
return certificates
def get_apk_size(self):
"""
Get the APK file size in bytes
:return: APK file size
"""
return os.path.getsize(self.apk_path)
def get_sha256(self):
"""
Get the sha256sum of the APK file
:return: hex sha256sum
"""
BLOCKSIZE = 65536
hasher = sha256()
with open(self.apk_path, 'rb') as apk:
buf = apk.read(BLOCKSIZE)
while len(buf) > 0:
hasher.update(buf)
buf = apk.read(BLOCKSIZE)
return hasher.hexdigest()
def save_embedded_classes_in_file(self, file_path):
"""
Save list of embedded classes in file.
:param file_path: file to write
"""
with open(file_path, 'w+') as f:
f.write('\n'.join(self.get_embedded_classes()))
def print_apk_infos(self):
"""
Print APK information
"""
permissions = self.get_permissions()
libraries = self.get_libraries()
certificates = self.get_certificates()
print("=== Information")
print('- APK path: {}'.format(self.apk_path))
print('- APK sum: {}'.format(self.get_sha256()))
print('- App version: {}'.format(self.get_version()))
print('- App version code: {}'.format(self.get_version_code()))
print('- App UID: {}'.format(self.get_application_universal_id()))
print('- App name: {}'.format(self.get_app_name()))
print('- App package: {}'.format(self.get_package()))
print('- App permissions: {}'.format(len(permissions)))
for perm in permissions:
print(' - {}'.format(perm))
print('- App libraries:')
for lib in libraries:
print(' - {}'.format(lib))
print('- Certificates: {}'.format(len(certificates)))
for cert in certificates:
print(' - {}'.format(cert))
def print_embedded_trackers(self):
"""
Print detected trackers
"""
trackers = self.detect_trackers()
print('=== Found trackers: {}'.format(len(trackers)))
for t in trackers:
print(' - {}'.format(t.name))
def analyze(path):
try:
start = process_time()
hashfunctions = dict(md5=hashlib.md5,
sha1=hashlib.sha1,
sha256=hashlib.sha256,
sha512=hashlib.sha512)
a = APK(path)
certs = set(
a.get_certificates_der_v3() + a.get_certificates_der_v2() +
[a.get_certificate_der(x) for x in a.get_signature_names()])
for cert in certs:
x509_cert = x509.Certificate.load(cert)
issuer = {
'commonName': None,
'organizationName': None,
'organizationalUnitName': None,
'countryName': None,
'stateOrProvinceName': None,
'localityName': None
}
subject = {
'commonName': None,
'organizationName': None,
'organizationalUnitName': None,
'countryName': None,
'stateOrProvinceName': None,
'localityName': None
}
strIssuer = get_certificate_name_string(x509_cert.issuer,
short=False)
strSubject = get_certificate_name_string(x509_cert.subject,
short=False)
arrIssuer = strIssuer.split(',')
for i in arrIssuer:
if i.lstrip().split('=')[0] == 'commonName':
issuer['commonName'] = i.lstrip().split('=')[1]
elif i.lstrip().split('=')[0] == 'organizationName':
issuer['organizationName'] = i.lstrip().split('=')[1]
elif i.lstrip().split('=')[0] == 'organizationalUnitName':
issuer['organizationalUnitName'] = i.lstrip().split('=')[1]
elif i.lstrip().split('=')[0] == 'countryName':
issuer['countryName'] = i.lstrip().split('=')[1]
elif i.lstrip().split('=')[0] == 'stateOrProvinceName':
issuer['stateOrProvinceName'] = i.lstrip().split('=')[1]
elif i.lstrip().split('=')[0] == 'localityName':
issuer['localityName'] = i.lstrip().split('=')[1]
arrSubject = strSubject.split(',')
for i in arrSubject:
if i.lstrip().split('=')[0] == 'commonName':
subject['commonName'] = i.lstrip().split('=')[1]
elif i.lstrip().split('=')[0] == 'organizationName':
subject['organizationName'] = i.lstrip().split('=')[1]
elif i.lstrip().split('=')[0] == 'organizationalUnitName':
subject['organizationalUnitName'] = i.lstrip().split(
'=')[1]
elif i.lstrip().split('=')[0] == 'countryName':
subject['countryName'] = i.lstrip().split('=')[1]
elif i.lstrip().split('=')[0] == 'stateOrProvinceName':
subject['stateOrProvinceName'] = i.lstrip().split('=')[1]
elif i.lstrip().split('=')[0] == 'localityName':
subject['localityName'] = i.lstrip().split('=')[1]
for k, v in hashfunctions.items():
if k == 'md5':
md5 = v(cert).hexdigest()
elif k == 'sha1':
sha1 = v(cert).hexdigest()
elif k == 'sha256':
sha256 = v(cert).hexdigest()
elif k == 'sha512':
sha512 = v(cert).hexdigest()
md5 = md5
appName = a.get_app_name()
fileSize = os.stat(a.get_filename()).st_size
sha1 = sha1
sha256 = sha256
sha512 = sha512
timestamp = time.time()
dateTime = datetime.fromtimestamp(timestamp)
timeOfSubmit = dateTime.strftime("%Y-%m-%d %H:%M:%S")
package = a.get_package()
androidversionCode = a.get_androidversion_code()
androidversionName = a.get_androidversion_name()
minSDKVersion = a.get_min_sdk_version()
maxSDKVersion = a.get_max_sdk_version()
targetSDKVersion = a.get_target_sdk_version()
mainActivity = a.get_main_activity()
attributes = {
'validFrom':
x509_cert['tbs_certificate']['validity']
['not_before'].native.strftime("%Y-%m-%d %H:%M:%S"),
'validTo':
x509_cert['tbs_certificate']['validity']
['not_after'].native.strftime("%Y-%m-%d %H:%M:%S"),
'serialNumber':
hex(x509_cert.serial_number),
'hashAlgorithm':
x509_cert.hash_algo,
'signatureAlgorithm':
x509_cert.signature_algo
}
certificateAttributes = json.dumps(attributes)
certificateIssuer = json.dumps(issuer)
certificateSubject = json.dumps(subject)
declaredPermissions = json.dumps(a.get_declared_permissions())
requestedPermissions = json.dumps(a.get_permissions())
activities = json.dumps(a.get_activities())
services = json.dumps(a.get_services())
receivers = json.dumps(a.get_receivers())
providers = json.dumps(a.get_providers())
stop = process_time()
analysisTime = stop - start
connect = mysql.connect()
cursor = connect.cursor()
sql = "INSERT INTO tbl_apkinfo (md5, appName, fileSize, analysisTime, sha1, sha256, sha512, firstSubmission, lastSubmission, package, androidversionCode, androidversionName, minSDKVersion, maxSDKVersion, targetSDKVersion, mainActivity, certificateAttributes, certificateIssuer, certificateSubject, declaredPermissions, requestedPermissions, activities, services, providers, receivers) VALUES (%s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s)"
param = (md5, appName, fileSize, analysisTime, sha1, sha256, sha512,
timeOfSubmit, timeOfSubmit, package, androidversionCode,
androidversionName, minSDKVersion, maxSDKVersion,
targetSDKVersion, mainActivity, certificateAttributes,
certificateIssuer, certificateSubject, declaredPermissions,
requestedPermissions, activities, services, providers,
receivers)
cursor.execute(sql, param)
connect.commit()
connect.close()
androaxml_main(path,
os.path.join(app.config['OUTPUT_PATH'], md5 + '.xml'))
return True
except:
return False
def extract_features(file_path):
result = {}
try:
a = APK(file_path)
d = DalvikVMFormat(a.get_dex())
dx = VMAnalysis(d)
vm = dvm.DalvikVMFormat(a.get_dex())
vmx = analysis.uVMAnalysis(vm)
d.set_vmanalysis(dx)
d.set_decompiler(DecompilerDAD(d, dx))
except:
return None
result['android_version_code'] = a.get_androidversion_code()
result['android_version_name'] = a.get_androidversion_name()
result['max_sdk'] = a.get_max_sdk_version()
result['min_sdk'] = a.get_min_sdk_version()
result['libraries'] = a.get_libraries()
result['filename'] = a.get_filename()
result['target_sdk'] = a.get_target_sdk_version()
result['md5'] = hashlib.md5(a.get_raw()).hexdigest()
result['sha256'] = hashlib.sha256(a.get_raw()).hexdigest()
result['permissions'] = a.get_permissions()
result['activities'] = a.get_activities()
result['providers'] = a.get_providers()
result['services'] = a.get_services()
#result['strings'] = d.get_strings()
#result['class_names'] = [c.get_name() for c in d.get_classes()]
#result['method_names'] = [m.get_name() for m in d.get_methods()]
#result['field_names'] = [f.get_name() for f in d.get_fields()]
class_names = [c.get_name() for c in d.get_classes()]
method_names = [m.get_name() for m in d.get_methods()]
field_names = [ f.get_name() for f in d.get_fields()]
result['is_native_code'] = 1 if analysis.is_native_code(dx) else 0
result['is_obfuscation'] = 1 if analysis.is_ascii_obfuscation(d) else 0
result['is_crypto_code'] = 1 if analysis.is_crypto_code(dx) else 0
result['is_dyn_code'] = 1 if analysis.is_dyn_code(dx) else 0
result['is_reflection_code'] = 1 if analysis.is_reflection_code(vmx) else 0
result['is_database'] = 1 if d.get_regex_strings(DB_REGEX) else 0
s_list = []
#s_list.extend(result['class_names'])
#s_list.extend(result['method_names'])
#s_list.extend(result['field_names'])
s_list.extend(class_names)
s_list.extend(method_names)
s_list.extend(method_names)
result['entropy_rate'] = entropy_rate(s_list)
result['feature_vectors'] = {}
# Search for the presence of api calls in a given apk
result['feature_vectors']['api_calls'] = []
for call in API_CALLS:
status = 1 if dx.tainted_packages.search_methods(".", call, ".") else 0
result['feature_vectors']['api_calls'].append(status)
# Search for the presence of permissions in a given apk
result['feature_vectors']['permissions'] = []
for permission in PERMISSIONS:
status = 1 if permission in result['permissions'] else 0
result['feature_vectors']['permissions'].append(status)
result['feature_vectors']['special_strings'] = []
for word in SPECIAL_STRINGS:
status = 1 if d.get_regex_strings(word) else 0
result['feature_vectors']['special_strings'].append(status)
opt_seq = []
for m in d.get_methods():
for i in m.get_instructions():
opt_seq.append(i.get_name())
optngramlist = [tuple(opt_seq[i:i+NGRAM]) for i in xrange(len(opt_seq) - NGRAM)]
optngram = Counter(optngramlist)
optcodes = dict()
tmpCodes = dict(optngram)
#for k,v in optngram.iteritems():
# if v>=NGRAM_THRE:
#optcodes[str(k)] = v
# optcodes[str(k)] = 1
tmpCodes = sorted(tmpCodes.items(),key =lambda d:d[1],reverse=True)
for value in tmpCodes[:NGRAM_THRE]:
optcodes[str(value[0])] = 1
result['feature_vectors']['opt_codes'] = optcodes
return result
def get_androidversion_code(self):
return androguard_apk.get_androidversion_code(self)
