def testReadPlistCanParseXMLPlistAtOffset(self):
for _ in range(5):
self.min_xml.seek(0, os.SEEK_SET)
rand_int = random.randint(0, 2048)
padding = "A" * rand_int
padded_xml_plist = padding + self.min_xml.read()
fd = StringIO.StringIO(padded_xml_plist)
fd.seek(rand_int, os.SEEK_SET)
binplist.readPlist(fd)
def testReadPlistCanParseXMLPlistAtOffset(self):
for _ in range(5):
self.min_xml.seek(0, os.SEEK_SET)
rand_int = random.randint(0, 2048)
padding = "A" * rand_int
padded_xml_plist = padding + self.min_xml.read()
fd = StringIO.StringIO(padded_xml_plist)
fd.seek(rand_int, os.SEEK_SET)
binplist.readPlist(fd)
def testReadPlist(self):
blank_file = StringIO.StringIO()
self.assertRaises(binplist.FormatError, binplist.readPlist, blank_file)
bplist15 = StringIO.StringIO("bplist15")
self.assertRaises(binplist.FormatError, binplist.readPlist, bplist15)
# Incomplete XML does not parse (expat error from plistlib is caught)
xml_file = StringIO.StringIO("<xml")
self.assertRaises(binplist.FormatError, binplist.readPlist, xml_file)
# Check that a minimal XML plist does parse.
# We're not testing full XML parsing per se as that's plistlib's tests job.
binplist.readPlist(self.min_xml)
def testReadPlist(self):
blank_file = StringIO.StringIO()
self.assertRaises(binplist.FormatError, binplist.readPlist, blank_file)
bplist15 = StringIO.StringIO("bplist15")
self.assertRaises(binplist.FormatError, binplist.readPlist, bplist15)
# Incomplete XML does not parse (expat error from plistlib is caught)
xml_file = StringIO.StringIO("<xml")
self.assertRaises(binplist.FormatError, binplist.readPlist, xml_file)
# Check that a minimal XML plist does parse.
# We're not testing full XML parsing per se as that's plistlib's tests job.
binplist.readPlist(self.min_xml)
def Run(self, args):
self.context = args.context
self.filter_query = args.query
with vfs.VFSOpen(args.pathspec, progress_callback=self.Progress) as fd:
data = fd.Read(self.MAX_PLIST_SIZE)
plist = binplist.readPlist(cStringIO.StringIO(data))
# Create the query parser
parser = plist_lib.PlistFilterParser(self.filter_query).Parse()
filter_imp = plist_lib.PlistFilterImplementation
matcher = parser.Compile(filter_imp)
if self.context:
# Obtain the values for the context using the value expander
value_expander = filter_imp.FILTERS["ValueExpander"]
iterator = value_expander().Expand(plist, self.context)
else:
# If we didn't get a context, the context is the whole plist
iterator = [plist]
reply = protodict.RDFValueArray()
for item in iterator:
# As we're setting the context manually, we need to account for types
if isinstance(item, types.ListType):
for sub_item in item:
partial_plist = plist_lib.PlistValueToPlainValue(sub_item)
if matcher.Matches(partial_plist):
reply.Append(sub_item)
else:
partial_plist = plist_lib.PlistValueToPlainValue(item)
if matcher.Matches(partial_plist):
reply.Append(partial_plist)
self.SendReply(reply)
def Run(self, args):
self.context = args.context
self.filter_query = args.query
with vfs.VFSOpen(args.pathspec, progress_callback=self.Progress) as fd:
data = fd.Read(self.MAX_PLIST_SIZE)
plist = binplist.readPlist(cStringIO.StringIO(data))
# Create the query parser
parser = plist_lib.PlistFilterParser(self.filter_query).Parse()
filter_imp = plist_lib.PlistFilterImplementation
matcher = parser.Compile(filter_imp)
if self.context:
# Obtain the values for the context using the value expander
value_expander = filter_imp.FILTERS["ValueExpander"]
iterator = value_expander().Expand(plist, self.context)
else:
# If we didn't get a context, the context is the whole plist
iterator = [plist]
reply = rdfvalue.RDFValueArray()
for item in iterator:
# As we're setting the context manually, we need to account for types
if isinstance(item, types.ListType):
for sub_item in item:
partial_plist = plist_lib.PlistValueToPlainValue(sub_item)
if matcher.Matches(partial_plist):
reply.Append(sub_item)
else:
partial_plist = plist_lib.PlistValueToPlainValue(item)
if matcher.Matches(partial_plist):
reply.Append(partial_plist)
self.SendReply(reply)
def GetTopLevel(self, file_object, file_name=u''):
"""Returns the deserialized content of a plist as a dictionary object.
Args:
file_object: A file-like object to parse.
file_name: The name of the file-like object.
Returns:
A dictionary object representing the contents of the plist.
Raises:
UnableToParseFile: when the file cannot be parsed.
"""
# Note that binplist.readPlist does not seek to offset 0.
try:
top_level_object = binplist.readPlist(file_object)
except binplist.FormatError as exception:
if not isinstance(exception, py2to3.BYTES_TYPE):
error_string = str(exception).decode(u'utf8',
errors=u'replace')
else:
error_string = exception
raise errors.UnableToParseFile(
u'File is not a plist file: {0:s}'.format(error_string))
except (AttributeError, LookupError, ValueError,
binascii.Error) as exception:
raise errors.UnableToParseFile(
u'Unable to parse XML file, reason: {0:s}'.format(exception))
except OverflowError as exception:
raise errors.UnableToParseFile(
u'Unable to parse: {0:s} with error: {1:s}'.format(
file_name, exception))
if not top_level_object:
raise errors.UnableToParseFile(
u'File is not a plist: missing top level object')
# Since we are using readPlist from binplist now instead of manually
# opening the binary plist file we loose this option. Keep it commented
# out for now but this needs to be tested a bit more.
# TODO: Re-evaluate if we can delete this or still require it.
#if bpl.is_corrupt:
# logging.warning(
# u'Corruption detected in binary plist: {0:s}'.format(file_name))
return top_level_object
def GetTopLevel(self, file_object, file_name=u''):
"""Returns the deserialized content of a plist as a dictionary object.
Args:
file_object: A file-like object to parse.
file_name: The name of the file-like object.
Returns:
A dictionary object representing the contents of the plist.
Raises:
UnableToParseFile: when the file cannot be parsed.
"""
# Note that binplist.readPlist does not seek to offset 0.
try:
top_level_object = binplist.readPlist(file_object)
except binplist.FormatError as exception:
if not isinstance(exception, py2to3.BYTES_TYPE):
error_string = str(exception).decode(u'utf8', errors=u'replace')
else:
error_string = exception
raise errors.UnableToParseFile(
u'File is not a plist file: {0:s}'.format(error_string))
except (
AttributeError, LookupError, ValueError, binascii.Error) as exception:
raise errors.UnableToParseFile(
u'Unable to parse XML file, reason: {0:s}'.format(exception))
except OverflowError as exception:
raise errors.UnableToParseFile(
u'Unable to parse: {0:s} with error: {1:s}'.format(
file_name, exception))
if not top_level_object:
raise errors.UnableToParseFile(
u'File is not a plist: missing top level object')
# Since we are using readPlist from binplist now instead of manually
# opening the binary plist file we loose this option. Keep it commented
# out for now but this needs to be tested a bit more.
# TODO: Re-evaluate if we can delete this or still require it.
#if bpl.is_corrupt:
# logging.warning(
# u'Corruption detected in binary plist: {0:s}'.format(file_name))
return top_level_object
def Parse(self, cmd, args, stdout, stderr, return_val, time_taken,
knowledge_base):
"""Parse the system profiler output. We get it in the form of a plist."""
_ = stderr, time_taken, args, knowledge_base # Unused
self.CheckReturn(cmd, return_val)
serial_number = []
hardware_list = []
plist = binplist.readPlist(cStringIO.StringIO(stdout))
if len(plist) > 1:
raise parsers.ParseError("SPHardwareDataType plist has too many items.")
hardware_list = plist[0]["_items"][0]
serial_number = hardware_list["serial_number"]
yield rdf_client.HardwareInfo(serial_number=serial_number)
def Parse(self, cmd, args, stdout, stderr, return_val, time_taken,
knowledge_base):
"""Parse the system profiler output. We get it in the form of a plist."""
_ = stderr, time_taken, args, knowledge_base # Unused
self.CheckReturn(cmd, return_val)
serial_number = []
hardware_list = []
plist = binplist.readPlist(cStringIO.StringIO(stdout))
if len(plist) > 1:
raise parsers.ParseError("SPHardwareDataType plist has too many items.")
hardware_list = plist[0]["_items"][0]
serial_number = hardware_list["serial_number"]
yield rdfvalue.HardwareInfo(serial_number=serial_number)
def Read(self, file_object):
"""Reads a plist from a file-like object.
Args:
file_object: the file-like object.
Raises:
IOError: if the plist file-like object cannot be read.
"""
try:
file_object.seek(0, os.SEEK_SET)
# Note that binplist.readPlist does not seek to offset 0.
self.root_key = binplist.readPlist(file_object)
except (AttributeError, binascii.Error, binplist.FormatError,
LookupError, OverflowError, ValueError) as exception:
raise IOError(exception)
def Read(self, file_object):
"""Reads a plist from a file-like object.
Args:
file_object: the file-like object.
Raises:
IOError: if the plist file-like object cannot be read.
"""
try:
file_object.seek(0, os.SEEK_SET)
# Note that binplist.readPlist does not seek to offset 0.
self.root_key = binplist.readPlist(file_object)
except (
AttributeError, binascii.Error, binplist.FormatError, LookupError,
OverflowError, ValueError) as exception:
raise IOError(exception)
def Parse(self, statentry, file_object, knowledge_base):
"""Parse the Plist file."""
plist = binplist.readPlist(file_object)
if not isinstance(plist, list):
raise parser.ParseError(
"InstallHistory plist is a '%s', expecting a list" %
type(plist))
for sw in plist:
yield rdf_client.SoftwarePackage(
name=sw.get("displayName"),
version=sw.get("displayVersion"),
description=",".join(sw.get("packageIdentifiers")),
# TODO(hanuszczak): make installed_on an RDFDatetime
installed_on=_DateToEpoch(sw.get("date")),
install_state=rdf_client.SoftwarePackage.InstallState.INSTALLED
)
def GetTopLevel(self, file_object, file_name=u''):
"""Returns the deserialized content of a plist as a dictionary object.
Args:
file_object: A file-like object to parse.
file_name: The name of the file-like object.
Returns:
A dictionary object representing the contents of the plist.
"""
# Note that binplist.readPlist does not seek to offset 0.
try:
top_level_object = binplist.readPlist(file_object)
except binplist.FormatError as exception:
raise errors.UnableToParseFile(
u'[{0:s}] File is not a plist file: {1:s}'.format(
self.NAME, utils.GetUnicodeString(exception)))
except (
LookupError, binascii.Error, ValueError, AttributeError) as exception:
raise errors.UnableToParseFile(
u'[{0:s}] Unable to parse XML file, reason: {1:s}'.format(
self.NAME, exception))
except OverflowError as exception:
raise errors.UnableToParseFile(
u'[{0:s}] Unable to parse: {1:s} with error: {2:s}'.format(
self.NAME, file_name, exception))
if not top_level_object:
raise errors.UnableToParseFile(
u'[{0:s}] File is not a plist: {1:s}'.format(
self.NAME, utils.GetUnicodeString(file_name)))
# Since we are using readPlist from binplist now instead of manually
# opening the binary plist file we loose this option. Keep it commented
# out for now but this needs to be tested a bit more.
# TODO: Re-evaluate if we can delete this or still require it.
#if bpl.is_corrupt:
# logging.warning(
# u'[{0:s}] corruption detected in binary plist: {1:s}'.format(
# self.NAME, file_name))
return top_level_object
def GetTopLevel(self, file_object, file_name=u''):
"""Returns the deserialized content of a plist as a dictionary object.
Args:
file_object: A file-like object to parse.
file_name: The name of the file-like object.
Returns:
A dictionary object representing the contents of the plist.
"""
# Note that binplist.readPlist does not seek to offset 0.
try:
top_level_object = binplist.readPlist(file_object)
except binplist.FormatError as exception:
raise errors.UnableToParseFile(
u'[{0:s}] File is not a plist file: {1:s}'.format(
self.NAME, utils.GetUnicodeString(exception)))
except (LookupError, binascii.Error, ValueError,
AttributeError) as exception:
raise errors.UnableToParseFile(
u'[{0:s}] Unable to parse XML file, reason: {1:s}'.format(
self.NAME, exception))
except OverflowError as exception:
raise errors.UnableToParseFile(
u'[{0:s}] Unable to parse: {1:s} with error: {2:s}'.format(
self.NAME, file_name, exception))
if not top_level_object:
raise errors.UnableToParseFile(
u'[{0:s}] File is not a plist: {1:s}'.format(
self.NAME, utils.GetUnicodeString(file_name)))
# Since we are using readPlist from binplist now instead of manually
# opening the binary plist file we loose this option. Keep it commented
# out for now but this needs to be tested a bit more.
# TODO: Re-evaluate if we can delete this or still require it.
#if bpl.is_corrupt:
# logging.warning(
# u'[{0:s}] corruption detected in binary plist: {1:s}'.format(
# self.NAME, file_name))
return top_level_object
def Parse(self, cmd, args, stdout, stderr, return_val, time_taken,
knowledge_base):
"""Parse the system profiler output. We get it in the form of a plist."""
_ = stderr, time_taken, args, knowledge_base # Unused
self.CheckReturn(cmd, return_val)
plist = binplist.readPlist(io.BytesIO(stdout))
if len(plist) > 1:
raise parser.ParseError(
"SPHardwareDataType plist has too many items.")
hardware_list = plist[0]["_items"][0]
serial_number = getattr(hardware_list, "serial_number", None)
system_product_name = getattr(hardware_list, "machine_model", None)
bios_version = getattr(hardware_list, "boot_rom_version", None)
yield rdf_client.HardwareInfo(serial_number=serial_number,
bios_version=bios_version,
system_product_name=system_product_name)
def Parse(self, cmd, args, stdout, stderr, return_val, time_taken,
knowledge_base):
"""Parse the system profiler output. We get it in the form of a plist."""
_ = stderr, time_taken, args, knowledge_base # Unused
self.CheckReturn(cmd, return_val)
plist = binplist.readPlist(io.BytesIO(stdout))
if len(plist) > 1:
raise parser.ParseError("SPHardwareDataType plist has too many items.")
hardware_list = plist[0]["_items"][0]
serial_number = getattr(hardware_list, "serial_number", None)
system_product_name = getattr(hardware_list, "machine_model", None)
bios_version = getattr(hardware_list, "boot_rom_version", None)
yield rdf_client.HardwareInfo(
serial_number=serial_number,
bios_version=bios_version,
system_product_name=system_product_name)
def Parse(self, statentry, file_object, knowledge_base):
"""Parse the Plist file."""
_ = knowledge_base
kwargs = {}
try:
kwargs["aff4path"] = file_object.urn
except AttributeError:
pass
direct_copy_items = [
"Label", "Disabled", "UserName", "GroupName", "Program",
"StandardInPath", "StandardOutPath", "StandardErrorPath",
"LimitLoadToSessionType", "EnableGlobbing", "EnableTransactions",
"OnDemand", "RunAtLoad", "RootDirectory", "WorkingDirectory", "Umask",
"TimeOut", "ExitTimeOut", "ThrottleInterval", "InitGroups",
"StartOnMount", "StartInterval", "Debug", "WaitForDebugger", "Nice",
"ProcessType", "AbandonProcessGroup", "LowPriorityIO", "LaunchOnlyOnce"
]
string_array_items = [
"LimitLoadToHosts", "LimitLoadFromHosts", "LimitLoadToSessionType",
"ProgramArguments", "WatchPaths", "QueueDirectories"
]
flag_only_items = ["SoftResourceLimits", "HardResourceLimits", "Sockets"]
plist = {}
try:
plist = binplist.readPlist(file_object)
except (binplist.FormatError, ValueError, IOError) as e:
plist["Label"] = "Could not parse plist: %s" % e
# These are items that can be directly copied
for key in direct_copy_items:
kwargs[key] = plist.get(key)
# These could be a string, they could be an array, we don't know and neither
# does Apple so we check.
for key in string_array_items:
elements = plist.get(key)
if isinstance(elements, basestring):
kwargs[key] = [elements]
else:
kwargs[key] = elements
# These are complex items that can appear in multiple data structures
# so we only flag on their existence
for key in flag_only_items:
if plist.get(key):
kwargs[key] = True
if plist.get("inetdCompatability") is not None:
kwargs["inetdCompatabilityWait"] = plist.get("inetdCompatability").get(
"Wait")
keepalive = plist.get("KeepAlive")
if isinstance(keepalive, bool) or keepalive is None:
kwargs["KeepAlive"] = keepalive
else:
keepalivedict = {}
keepalivedict["SuccessfulExit"] = keepalive.get("SuccessfulExit")
keepalivedict["NetworkState"] = keepalive.get("NetworkState")
pathstates = keepalive.get("PathState")
if pathstates is not None:
keepalivedict["PathState"] = []
for pathstate in pathstates:
keepalivedict["PathState"].append(
rdf_plist.PlistBoolDictEntry(
name=pathstate, value=pathstates[pathstate]))
otherjobs = keepalive.get("OtherJobEnabled")
if otherjobs is not None:
keepalivedict["OtherJobEnabled"] = []
for otherjob in otherjobs:
keepalivedict["OtherJobEnabled"].append(
rdf_plist.PlistBoolDictEntry(
name=otherjob, value=otherjobs[otherjob]))
kwargs["KeepAliveDict"] = rdf_plist.LaunchdKeepAlive(**keepalivedict)
envvars = plist.get("EnvironmentVariables")
if envvars is not None:
kwargs["EnvironmentVariables"] = []
for envvar in envvars:
kwargs["EnvironmentVariables"].append(
rdf_plist.PlistStringDictEntry(name=envvar, value=envvars[envvar]))
startcalendarinterval = plist.get("StartCalendarInterval")
if startcalendarinterval is not None:
if isinstance(startcalendarinterval, dict):
kwargs["StartCalendarInterval"] = [
rdf_plist.LaunchdStartCalendarIntervalEntry(
Minute=startcalendarinterval.get("Minute"),
Hour=startcalendarinterval.get("Hour"),
Day=startcalendarinterval.get("Day"),
Weekday=startcalendarinterval.get("Weekday"),
Month=startcalendarinterval.get("Month"))
]
else:
kwargs["StartCalendarInterval"] = []
for entry in startcalendarinterval:
kwargs["StartCalendarInterval"].append(
rdf_plist.LaunchdStartCalendarIntervalEntry(
Minute=entry.get("Minute"),
Hour=entry.get("Hour"),
Day=entry.get("Day"),
Weekday=entry.get("Weekday"),
Month=entry.get("Month")))
yield rdf_plist.LaunchdPlist(**kwargs)
def Parse(self, statentry, file_object, knowledge_base):
"""Parse the Plist file."""
_ = knowledge_base
kwargs = {}
try:
kwargs["aff4path"] = file_object.urn
except AttributeError:
pass
direct_copy_items = [
"Label", "Disabled", "UserName", "GroupName", "Program",
"StandardInPath", "StandardOutPath", "StandardErrorPath",
"LimitLoadToSessionType", "EnableGlobbing", "EnableTransactions",
"OnDemand", "RunAtLoad", "RootDirectory", "WorkingDirectory",
"Umask", "TimeOut", "ExitTimeOut", "ThrottleInterval",
"InitGroups", "StartOnMount", "StartInterval", "Debug",
"WaitForDebugger", "Nice", "ProcessType", "AbandonProcessGroup",
"LowPriorityIO", "LaunchOnlyOnce"
]
string_array_items = [
"LimitLoadToHosts", "LimitLoadFromHosts", "LimitLoadToSessionType",
"ProgramArguments", "WatchPaths", "QueueDirectories"
]
flag_only_items = [
"SoftResourceLimits", "HardResourceLimits", "Sockets"
]
plist = {}
try:
plist = binplist.readPlist(file_object)
except (binplist.FormatError, ValueError, IOError) as e:
plist["Label"] = "Could not parse plist: %s" % e
# These are items that can be directly copied
for key in direct_copy_items:
kwargs[key] = plist.get(key)
# These could be a string, they could be an array, we don't know and neither
# does Apple so we check.
for key in string_array_items:
elements = plist.get(key)
if isinstance(elements, basestring):
kwargs[key] = [elements]
else:
kwargs[key] = elements
# These are complex items that can appear in multiple data structures
# so we only flag on their existence
for key in flag_only_items:
if plist.get(key):
kwargs[key] = True
if plist.get("inetdCompatability") is not None:
kwargs["inetdCompatabilityWait"] = plist.get(
"inetdCompatability").get("Wait")
keepalive = plist.get("KeepAlive")
if isinstance(keepalive, bool) or keepalive is None:
kwargs["KeepAlive"] = keepalive
else:
keepalivedict = {}
keepalivedict["SuccessfulExit"] = keepalive.get("SuccessfulExit")
keepalivedict["NetworkState"] = keepalive.get("NetworkState")
pathstates = keepalive.get("PathState")
if pathstates is not None:
keepalivedict["PathState"] = []
for pathstate in pathstates:
keepalivedict["PathState"].append(
rdf_plist.PlistBoolDictEntry(
name=pathstate, value=pathstates[pathstate]))
otherjobs = keepalive.get("OtherJobEnabled")
if otherjobs is not None:
keepalivedict["OtherJobEnabled"] = []
for otherjob in otherjobs:
keepalivedict["OtherJobEnabled"].append(
rdf_plist.PlistBoolDictEntry(
name=otherjob, value=otherjobs[otherjob]))
kwargs["KeepAliveDict"] = rdf_plist.LaunchdKeepAlive(
**keepalivedict)
envvars = plist.get("EnvironmentVariables")
if envvars is not None:
kwargs["EnvironmentVariables"] = []
for envvar in envvars:
kwargs["EnvironmentVariables"].append(
rdf_plist.PlistStringDictEntry(name=envvar,
value=envvars[envvar]))
startcalendarinterval = plist.get("StartCalendarInterval")
if startcalendarinterval is not None:
if isinstance(startcalendarinterval, dict):
kwargs["StartCalendarInterval"] = [
rdf_plist.LaunchdStartCalendarIntervalEntry(
Minute=startcalendarinterval.get("Minute"),
Hour=startcalendarinterval.get("Hour"),
Day=startcalendarinterval.get("Day"),
Weekday=startcalendarinterval.get("Weekday"),
Month=startcalendarinterval.get("Month"))
]
else:
kwargs["StartCalendarInterval"] = []
for entry in startcalendarinterval:
kwargs["StartCalendarInterval"].append(
rdf_plist.LaunchdStartCalendarIntervalEntry(
Minute=entry.get("Minute"),
Hour=entry.get("Hour"),
Day=entry.get("Day"),
Weekday=entry.get("Weekday"),
Month=entry.get("Month")))
yield rdf_plist.LaunchdPlist(**kwargs)