def parseCerts(self, signature, offset):
prev = self._f.tell()
true_offset = signature.getOffset() + offset
self._f.seek(true_offset)
magic = getInt(self._f)
if magic != dictionary.signatures['BLOBWRAPPER']:
data = {
'offset': true_offset,
'magic': hex(magic),
'expected': hex(dictionary.signatures['BLOBWRAPPER'])
}
a = Abnormality(title='BAD MAGIC - BLOBWRAPPER', data=data)
self.addAbnormality(a)
self._f.seek(prev)
return
size = getInt(self._f) - 8
# out = open('cms', 'wb')
# out.write(self._f.read(size))
# out.close()
# exit(0)
if size > 0:
signed_data = cms.CMS(self._f.read(size), format='DER')
for cert in signed_data.certs:
serial = cert.serial
subject = {
'country': self.getCertNameData(cert.subject,
oid.Oid('C')),
'org': self.getCertNameData(cert.subject, oid.Oid('O')),
'org_unit': self.getCertNameData(cert.subject,
oid.Oid('OU')),
'common_name': self.getCertNameData(cert.subject,
oid.Oid('CN'))
}
issuer = {
'country': self.getCertNameData(cert.issuer, oid.Oid('C')),
'org': self.getCertNameData(cert.issuer, oid.Oid('O')),
'org_unit': self.getCertNameData(cert.issuer,
oid.Oid('OU')),
'common_name': self.getCertNameData(cert.issuer,
oid.Oid('CN'))
}
ca = cert.check_ca()
cert = Certificate(serial=serial, subject=subject,
issuer=issuer, ca=ca)
signature.addCert(cert)
else:
data = {
'offset': true_offset,
'size': size
}
a = Abnormality(title='NON-POSITIVE CMS SIZE', data=data)
self.addAbnormality(a)
self._f.seek(prev)
def parse_codedirectory(self, signature, offset):
prev = self.f.tell()
true_offset = signature.offset + offset
self.f.seek(true_offset)
magic = get_int(self.f)
if magic != dictionary.signatures['CODEDIRECTORY']:
data = {
'offset': true_offset,
'magic': hex(magic),
'expected': hex(dictionary.signatures['CODEDIRECTORY'])
}
a = Abnormality(title='BAD MAGIC - CODEDIRECTORY', data=data)
self.add_abnormality(a)
self.f.seek(prev)
return
# Skip size
self.f.read(4)
version = get_int(self.f)
# Not sure how to parse flags yet...
flags = get_int(self.f)
hash_offset = get_int(self.f)
ident_offset = get_int(self.f)
n_special_slots = get_int(self.f)
n_code_slots = get_int(self.f)
code_limit = get_int(self.f)
hash_size = int(self.f.read(1).encode('hex'), 16)
hash_type = dictionary.hashes[int(self.f.read(1).encode('hex'), 16)]
if version >= 0x20200:
platform = int(self.f.read(1).encode('hex'), 16)
else:
# Skip spare1
self.f.read(1)
page_size = int(round(exp(int(self.f.read(1).encode('hex'),
16) * log(2))))
# Skip spare2
self.f.read(4)
if version >= 0x20100:
scatter_offset = get_int(self.f)
if version >= 0x20200:
team_id_offset = get_int(self.f)
self.f.seek(true_offset + team_id_offset)
team_id = readstring(self.f)
self.f.seek(true_offset + ident_offset)
identity = readstring(self.f)
codedirectory = CodeDirectory(version=version, flags=flags,
hash_offset=hash_offset,
n_special_slots=n_special_slots,
n_code_slots=n_code_slots,
code_limit=code_limit,
hash_size=hash_size, hash_type=hash_type,
page_size=page_size, identity=identity)
if version >= 0x20100:
codedirectory.scatter_offset = scatter_offset
if version >= 0x20200:
codedirectory.platform = platform
codedirectory.team_id_offset = team_id_offset
codedirectory.team_id = team_id
self.f.seek(true_offset + hash_offset - n_special_slots * hash_size)
count = n_special_slots + n_code_slots
while count > 0:
hash = self.f.read(hash_size).encode('hex')
codedirectory.add_hash(hash)
count -= 1
signature.codedirectory = codedirectory
self.f.seek(prev)
def parse_imports_and_strings(self, macho):
prev = self.f.tell()
true_offset = macho.offset + macho.strtab.offset
if macho.has_flag('TWOLEVEL'):
for i in macho.symtab.gen_syms():
if i.is_imp():
self.f.seek(true_offset + i.index)
if ((self.f.tell() > (true_offset +
macho.strtab.size)) or
(self.f.tell() > self.file.size)):
data = {
'offset': self.f.tell(),
'strtab_offset': true_offset,
'strtab_size': macho.strtab.size,
'file_size': self.file.size
}
a = Abnormality(title='BAD STRING INDEX', data=data)
self.add_abnormality(a)
continue
func = readstring(self.f)
if i.dylib == 0:
dylib = 'SELF_LIBRARY'
elif i.dylib <= len(macho.dylibs):
dylib = macho.dylibs[i.dylib - 1]
elif i.dylib == 254:
dylib = 'DYNAMIC_LOOKUP'
elif i.dylib == 255:
dylib = 'EXECUTABLE'
else:
data = {
'dylib': i.dylib,
'dylib_len': len(macho.dylibs)
}
a = Abnormality(title='DYLIB OUT OF RANGE', data=data)
self.add_abnormality(a)
dylib = str(i.dylib) + ' (OUT OF RANGE)'
imp = FunctionImport(func=func, dylib=dylib)
macho.add_import(imp)
else:
self.f.seek(true_offset + i.index)
if ((self.f.tell() > (true_offset +
macho.strtab.size)) or
(self.f.tell() > self.file.size)):
data = {
'offset': self.f.tell(),
'strtab_offset': true_offset,
'strtab_size': macho.strtab.size,
'file_size': self.file.size
}
a = Abnormality(title='BAD STRING INDEX', data=data)
self.add_abnormality(a)
continue
string = readstring(self.f)
if string != '':
macho.strtab.add_string(string)
else:
for i in macho.symtab.gen_syms():
if i.is_imp():
self.f.seek(true_offset + i.index)
if self.f.tell() > (true_offset +
macho.strtab.size):
data = {
'offset': self.f.tell(),
'strtab_offset': true_offset,
'strtab_size': macho.strtab.size
}
a = Abnormality(title='BAD STRING INDEX', data=data)
self.add_abnormality(a)
continue
func = readstring(self.f)
imp = FunctionImport(func=func)
macho.add_import(imp)
else:
self.f.seek(true_offset + i.index)
string = readstring(self.f)
if string != '':
macho.strtab.add_string(string)
self.f.seek(prev)
def parseMachO(self, macho):
self._f.seek(macho.getOffset())
# skip magic
self._f.read(4)
cputype = getInt(self._f)
# print 'cputype: ' + str(cputype)
# print 'offset: ' + str(self._f.tell())
subtype = getInt(self._f)
filetype = getInt(self._f)
nlcs = getInt(self._f)
slcs = getInt(self._f)
flags = getInt(self._f)
if macho.is64Bit():
# skip padding
self._f.read(4)
if macho.isLittle():
cputype = little(cputype, 'I')
subtype = little(subtype, 'I')
filetype = little(filetype, 'I')
nlcs = little(nlcs, 'I')
slcs = little(slcs, 'I')
flags = little(flags, 'I')
try:
cpu = dictionary.cputypes[cputype][-2]
except:
cpu = cputype
data = {
'offset': macho.getOffset() + 4,
'cputype': cputype
}
a = Abnormality(title='UNKNOWN CPUTYPE', data=data)
self.addAbnormality(a)
try:
subtype = dictionary.cputypes[cputype][subtype]
except:
data = {
'offset': macho.getOffset() + 8,
'cputype': cputype,
'subtype': subtype
}
a = Abnormality(title='UNKNOWN SUBTYPE', data=data)
self.addAbnormality(a)
try:
filetype = dictionary.filetypes[filetype]
except:
data = {
'offset': macho.getOffset() + 12,
'filetype': filetype
}
a = Abnormality(title='UNKNOWN FILETYPE', data=data)
self.addAbnormality(a)
flags = self.listMachOFlags(flags)
macho.setCPUType(cpu)
macho.setSubType(subtype)
macho.setFileType(filetype)
macho.setNLCs(nlcs)
macho.setSLCs(slcs)
macho.setFlags(flags)
lc = LoadCommander(f=self._f, macho=macho, file_size=self._file.getSize())
lc.parseLCs()
self._abnormalities += lc.getAbnormalities()
# Need to investigate whether the presence of a
# symbol/string table is expected and whether the
# abscence is indicative of shenanigans.
if macho.hasLC('SYMTAB'):
self.parseSyms(macho)
self.parseImportsAndStrings(macho)
if macho.hasLC('CODE_SIGNATURE'):
self.parseSig(macho)
if not macho.isArchive():
self._file.setContent(macho)
def parseImportsAndStrings(self, macho):
prev = self._f.tell()
true_offset = macho.getOffset() + macho.getStrTab().getOffset()
# blacklist = ('dyld_', '_OBJC_', '.objc_', '___stack_chk_')
if macho.hasFlag('TWOLEVEL'):
for i in macho.getSymTab().genSyms():
if i.isImp():
self._f.seek(true_offset + i.getIndex())
if ((self._f.tell() > (true_offset +
macho.getStrTab().getSize())) or
(self._f.tell() > self._file.getSize())):
data = {
'offset': self._f.tell(),
'strtab_offset': true_offset,
'strtab_size': macho.getStrTab().getSize(),
'file_size': self._file.getSize()
}
a = Abnormality(title='BAD STRING INDEX', data=data)
self.addAbnormality(a)
continue
func = readstring(self._f)
# if func.startswith(blacklist):
# continue
if i.getDylib() == 0:
dylib = 'SELF_LIBRARY'
elif i.getDylib() <= len(macho.getDylibs()):
dylib = macho.getDylibs()[i.getDylib() - 1]
elif i.getDylib() == 254:
dylib = 'DYNAMIC_LOOKUP'
elif i.getDylib() == 255:
dylib = 'EXECUTABLE'
else:
data = {
'dylib': i.getDylib(),
'dylib_len': len(macho.getDylibs())
}
a = Abnormality(title='DYLIB OUT OF RANGE', data=data)
self.addAbnormality(a)
dylib = str(i.getDylib()) + ' (OUT OF RANGE)'
imp = FunctionImport(func=func, dylib=dylib)
macho.addImport(imp)
else:
self._f.seek(true_offset + i.getIndex())
if ((self._f.tell() > (true_offset +
macho.getStrTab().getSize())) or
(self._f.tell() > self._file.getSize())):
data = {
'offset': self._f.tell(),
'strtab_offset': true_offset,
'strtab_size': macho.getStrTab().getSize(),
'file_size': self._file.getSize()
}
a = Abnormality(title='BAD STRING INDEX', data=data)
self.addAbnormality(a)
continue
string = readstring(self._f)
if string != '':
macho.getStrTab().addString(string)
else:
for i in macho.getSymTab().genSyms():
if i.isImp():
self._f.seek(true_offset + i.getIndex())
if self._f.tell() > (true_offset +
macho.getStrTab().getSize()):
data = {
'offset': self._f.tell(),
'strtab_offset': true_offset,
'strtab_size': macho.getStrTab().getSize()
}
a = Abnormality(title='BAD STRING INDEX', data=data)
self.addAbnormality(a)
continue
func = readstring(self._f)
imp = FunctionImport(func=func)
macho.addImport(imp)
else:
self._f.seek(true_offset + i.getIndex())
string = readstring(self._f)
if string != '':
macho.getStrTab().addString(string)
self._f.seek(prev)
def parseSyms(self, macho):
prev = self._f.tell()
true_offset = macho.getOffset() + macho.getSymTab().getOffset()
if macho.is64Bit():
symbol_size = 60
else:
symbol_size = 56
# print 'to:', true_offset
# print macho.getOffset(), macho.getSize()
if (true_offset < macho.getOffset() + macho.getSize() and
true_offset < self._file.getSize()):
self._f.seek(true_offset)
for i in range(macho.getSymTab().getNSyms()):
# print self._f.tell()
if ((self._f.tell() + symbol_size > macho.getOffset() +
macho.getSize()) or (self._f.tell() + symbol_size >
self._file.getSize())):
data = {
'offset': self._f.tell(),
'mach-o_size': macho.getSize(),
'mach-o_offset': macho.getOffset(),
'file_size': self._file.getSize()
}
a = Abnormality(title='REMAINING SYMBOLS OUT OF BOUNDS',
data=data)
self.addAbnormality(a)
self._f.seek(prev)
return
else:
index = getInt(self._f)
sym_type = int(self._f.read(1).encode('hex'), 16)
sect = int(self._f.read(1).encode('hex'), 16)
desc = int(self._f.read(2).encode('hex'), 16)
value = None
if macho.is64Bit():
if macho.isLittle():
value = little(getLL(self._f), 'Q')
else:
value = getLL(self._f)
else:
if macho.isLittle():
value = little(getInt(self._f), 'I')
else:
value = getInt(self._f)
if macho.isLittle():
index = little(index, 'I')
if sym_type >= 32:
if sym_type in dictionary.stabs:
stab = dictionary.stabs[sym_type]
else:
offset = self._f.tell() - symbol_size
data = {
'offset': offset,
'index': index,
'sym_type': sym_type,
'sect': sect,
'desc': desc,
'value': value
}
a = Abnormality(title='UNKNOWN STAB', data=data)
self.addAbnormality(a)
continue
sym = Symbol(index=index, stab=stab, sect=sect,
value=value)
macho.getSymTab().addSym(sym)
else:
pext = sym_type & 0x10
if sym_type & 0x0e in dictionary.n_types:
n_type = dictionary.n_types[sym_type & 0x0e]
else:
offset = self._f.tell() - symbol_size
data = {
'offset': offset,
'index': index,
'pext': pext,
'n_type': sym_type & 0x0e,
'sect': sect,
'desc': desc,
'value': value
}
a = Abnormality(title='UNKNOWN N_TYPE', data=data)
self.addAbnormality(a)
ext = sym_type & 0x01
if macho.isLittle():
dylib = desc & 0x0f
ref = (desc >> 8) & 0xff
else:
dylib = (desc >> 8) & 0xff
ref = desc & 0x0f
sym = Symbol(index=index, pext=pext, sym_type=n_type,
ext=ext, sect=sect, dylib=dylib, ref=ref,
value=value)
macho.getSymTab().addSym(sym)
# print self._f.tell()
# print sym.getIndex(), sym.getValue()
else:
data = {
'offset': true_offset,
'mach-o_size': macho.getSize(),
'mach-o_offset': macho.getOffset(),
'file_size': self._file.getSize()
}
a = Abnormality(title='SYMBOL TABLE OUT OF BOUNDS', data=data)
self.addAbnormality(a)
self._f.seek(prev)