def domain(self, domain):
normalizedDomain = DomainFactory.normalizeDomain(domain)
self.__domain = normalizedDomain
def execute(self, field):
"""Executes the field edition following the specified messages.
Children of the specified field will be replaced with new fields.
:param field: the format definition that will be user
:type field: :class:`netzob.Model.Vocabulary.AbstractField.AbstractField`
:raise Exception: if something bad happens
"""
if field is None:
raise TypeError("The field cannot be None")
fieldValues = [
TypeConverter.convert(data, Raw, HexaString)
for data in field.getValues(encoded=False)
]
if len(fieldValues) == 0:
raise Exception("No value found in the field.")
# Retrieve longuest field value
maxLengthFieldValue = len(max(fieldValues, key=len))
# definies the step following specified unitsize
stepUnitsize = self.__computeStepForUnitsize()
# Vertical identification of variation
indexedValues = []
for i in range(0, maxLengthFieldValue, stepUnitsize):
currentIndexValue = []
for fieldValue in fieldValues:
if i < len(fieldValue):
currentIndexValue.append(
fieldValue[i:min(len(fieldValue), i + stepUnitsize)])
else:
currentIndexValue.append(b'')
indexedValues.append(currentIndexValue)
# If requested, merges the adjacent static fields
if self.mergeAdjacentStaticFields:
result = []
staticSequences = []
for values in indexedValues:
if len(set(values)) == 1:
# static
staticSequences.append(values[0])
else:
# dynamic
if len(staticSequences) > 0:
result.append([b''.join(staticSequences)])
staticSequences = []
result.append(values)
if len(staticSequences) > 0:
result.append([b''.join(staticSequences)])
indexedValues = result
# If requested, merges the adjacent dynamic fields
if self.mergeAdjacentDynamicFields:
result = []
dynamicSequences = []
for values in indexedValues:
if len(set(values)) > 1:
# dynamic
dynamicSequences.append(values)
else:
# static
if len(dynamicSequences) > 0:
dynValues = zip(*dynamicSequences)
tmp_result = []
for d in dynValues:
tmp_result.append(b''.join(
[x if x is not None else b'' for x in d]))
result.append(tmp_result)
dynamicSequences = []
result.append(values)
if len(dynamicSequences) > 0:
dynValues = zip(*dynamicSequences)
tmp_result = []
for d in dynValues:
tmp_result.append(b''.join(
[x if x is not None else b'' for x in d]))
result.append(tmp_result)
indexedValues = result
# Create a field for each entry
newFields = []
for (i, val) in enumerate(indexedValues):
fName = "Field-{0}".format(i)
fDomain = DomainFactory.normalizeDomain([
Raw(TypeConverter.convert(v, HexaString, BitArray))
for v in set(val)
])
newFields.append(Field(domain=fDomain, name=fName))
# attach encoding functions
for newField in newFields:
newField.encodingFunctions = list(field.encodingFunctions.values())
field.fields = newFields
def execute(self, field):
"""Executes the field edition following the specified messages.
Children of the specified field will be replaced with new fields.
:param field: the format definition that will be user
:type field: :class:`netzob.Model.Vocabulary.AbstractField.AbstractField`
:raise Exception: if something bad happens
"""
if field is None:
raise TypeError("The field cannot be None")
fieldValues = [
TypeConverter.convert(data, Raw, HexaString)
for data in field.getValues(encoded=False)
]
if len(fieldValues) == 0:
raise Exception("No value found in the field.")
# Retrieve longuest field value
maxLengthFieldValue = len(max(fieldValues, key=len))
# definies the step following specified unitsize
stepUnitsize = self.__computeStepForUnitsize()
# Vertical identification of variation
indexedValues = []
for i in range(0, maxLengthFieldValue, stepUnitsize):
currentIndexValue = []
for fieldValue in fieldValues:
if i < len(fieldValue):
currentIndexValue.append(
fieldValue[i:min(len(fieldValue), i + stepUnitsize)])
else:
currentIndexValue.append(b'')
indexedValues.append(currentIndexValue)
# If requested, merges the adjacent static fields
if self.mergeAdjacentStaticFields:
result = []
staticSequences = []
for values in indexedValues:
if len(set(values)) == 1:
# static
staticSequences.append(values[0])
else:
# dynamic
if len(staticSequences) > 0:
result.append([b''.join(staticSequences)])
staticSequences = []
result.append(values)
if len(staticSequences) > 0:
result.append([b''.join(staticSequences)])
indexedValues = result
# If requested, merges the adjacent dynamic fields
if self.mergeAdjacentDynamicFields:
result = []
dynamicSequences = []
for values in indexedValues:
if len(set(values)) > 1:
# dynamic
dynamicSequences.append(values)
else:
# static
if len(dynamicSequences) > 0:
dynValues = zip(*dynamicSequences)
tmp_result = []
for d in dynValues:
tmp_result.append(b''.join(
[x if x is not None else b'' for x in d]))
result.append(tmp_result)
dynamicSequences = []
result.append(values)
if len(dynamicSequences) > 0:
dynValues = zip(*dynamicSequences)
tmp_result = []
for d in dynValues:
tmp_result.append(
b''.join([x if x is not None else b'' for x in d]))
result.append(tmp_result)
indexedValues = result
# Create a field for each entry
newFields = []
for (i, val) in enumerate(indexedValues):
fName = "Field-{0}".format(i)
fDomain = DomainFactory.normalizeDomain([
Raw(TypeConverter.convert(v, HexaString, BitArray))
for v in set(val)
])
newFields.append(Field(domain=fDomain, name=fName))
# attach encoding functions
for newField in newFields:
newField.encodingFunctions = list(field.encodingFunctions.values())
field.fields = newFields
def split(field, delimiter):
"""Split a field (or symbol) with a specific delimiter. The
delimiter can be passed either as an ASCII, a Raw, an
HexaString, or any objects that inherit from AbstractType.
>>> from netzob.all import *
>>> samples = [b"aaaaff000000ff10", b"bbff110010ff00000011", b"ccccccccfffe1f000000ff12"]
>>> messages = [RawMessage(data=sample) for sample in samples]
>>> symbol = Symbol(messages=messages[:3])
>>> Format.splitDelimiter(symbol, ASCII("ff"))
>>> print(symbol)
Field-0 | Field-sep-6666 | Field-2 | Field-sep-6666 | Field-4
---------- | -------------- | ------------ | -------------- | ----------
'aaaa' | 'ff' | '000000' | 'ff' | '10'
'bb' | 'ff' | '110010' | 'ff' | '00000011'
'cccccccc' | 'ff' | 'fe1f000000' | 'ff' | '12'
---------- | -------------- | ------------ | -------------- | ----------
>>> samples = [b"434d446964656e74696679230400000066726564", b"5245536964656e74696679230000000000000000", b"434d44696e666f2300000000", b"524553696e666f230000000004000000696e666f", b"434d4473746174732300000000", b"52455373746174732300000000050000007374617473", b"434d4461757468656e7469667923090000006d7950617373776421", b"52455361757468656e74696679230000000000000000", b"434d44656e6372797074230a00000031323334353674657374", b"524553656e637279707423000000000a00000073707176777436273136", b"434d4464656372797074230a00000073707176777436273136", b"5245536465637279707423000000000a00000031323334353674657374", b"434d446279652300000000", b"524553627965230000000000000000", b"434d446964656e746966792307000000526f626572746f", b"5245536964656e74696679230000000000000000", b"434d44696e666f2300000000", b"524553696e666f230000000004000000696e666f", b"434d4473746174732300000000", b"52455373746174732300000000050000007374617473", b"434d4461757468656e74696679230a000000615374726f6e67507764", b"52455361757468656e74696679230000000000000000", b"434d44656e63727970742306000000616263646566", b"524553656e6372797074230000000006000000232021262724", b"434d44646563727970742306000000232021262724", b"52455364656372797074230000000006000000616263646566", b"434d446279652300000000", b"524553627965230000000000000000"]
>>> messages = [RawMessage(data=TypeConverter.convert(sample, HexaString, Raw)) for sample in samples]
>>> symbol = Symbol(messages=messages)
>>> symbol.encodingFunctions.add(TypeEncodingFunction(ASCII)) # Change visualization to hexastring
>>> Format.splitDelimiter(symbol, ASCII("#"))
>>> print(symbol)
Field-0 | Field-sep-23 | Field-2 | Field-sep-23 | Field-4
--------------- | ------------ | -------------------- | ------------ | -------
'CMDidentify' | '#' | '....fred' | '' | ''
'RESidentify' | '#' | '........' | '' | ''
'CMDinfo' | '#' | '....' | '' | ''
'RESinfo' | '#' | '........info' | '' | ''
'CMDstats' | '#' | '....' | '' | ''
'RESstats' | '#' | '........stats' | '' | ''
'CMDauthentify' | '#' | '....myPasswd!' | '' | ''
'RESauthentify' | '#' | '........' | '' | ''
'CMDencrypt' | '#' | '....123456test' | '' | ''
'RESencrypt' | '#' | "........spqvwt6'16" | '' | ''
'CMDdecrypt' | '#' | "....spqvwt6'16" | '' | ''
'RESdecrypt' | '#' | '........123456test' | '' | ''
'CMDbye' | '#' | '....' | '' | ''
'RESbye' | '#' | '........' | '' | ''
'CMDidentify' | '#' | '....Roberto' | '' | ''
'RESidentify' | '#' | '........' | '' | ''
'CMDinfo' | '#' | '....' | '' | ''
'RESinfo' | '#' | '........info' | '' | ''
'CMDstats' | '#' | '....' | '' | ''
'RESstats' | '#' | '........stats' | '' | ''
'CMDauthentify' | '#' | '....aStrongPwd' | '' | ''
'RESauthentify' | '#' | '........' | '' | ''
'CMDencrypt' | '#' | '....abcdef' | '' | ''
'RESencrypt' | '#' | '........' | '#' | " !&'$"
'CMDdecrypt' | '#' | '....' | '#' | " !&'$"
'RESdecrypt' | '#' | '........abcdef' | '' | ''
'CMDbye' | '#' | '....' | '' | ''
'RESbye' | '#' | '........' | '' | ''
--------------- | ------------ | -------------------- | ------------ | -------
>>> print(symbol.fields[0]._str_debug())
Field-0
|-- Alt
|-- Data (Raw=b'CMDidentify' ((0, 88)))
|-- Data (Raw=b'RESidentify' ((0, 88)))
|-- Data (Raw=b'CMDinfo' ((0, 56)))
|-- Data (Raw=b'RESinfo' ((0, 56)))
|-- Data (Raw=b'CMDstats' ((0, 64)))
|-- Data (Raw=b'RESstats' ((0, 64)))
|-- Data (Raw=b'CMDauthentify' ((0, 104)))
|-- Data (Raw=b'RESauthentify' ((0, 104)))
|-- Data (Raw=b'CMDencrypt' ((0, 80)))
|-- Data (Raw=b'RESencrypt' ((0, 80)))
|-- Data (Raw=b'CMDdecrypt' ((0, 80)))
|-- Data (Raw=b'RESdecrypt' ((0, 80)))
|-- Data (Raw=b'CMDbye' ((0, 48)))
|-- Data (Raw=b'RESbye' ((0, 48)))
Below is another example of the FieldSplitDelimiter usage: it splits fields based on a Raw string.
>>> from netzob.all import *
>>> samples = [b"\\x01\\x02\\x03\\xff\\x04\\x05\\xff\\x06\\x07", b"\\x01\\x02\\xff\\x03\\x04\\x05\\x06\\xff\\x07", b"\\x01\\xff\\x02\\x03\\x04\\x05\\x06"]
>>> messages = [RawMessage(data=sample) for sample in samples]
>>> symbol = Symbol(messages=messages)
>>> Format.splitDelimiter(symbol, Raw(b"\\xff"))
>>> print(symbol)
Field-0 | Field-sep-ff | Field-2 | Field-sep-ff | Field-4
-------------- | ------------ | ---------------------- | ------------ | ----------
'\\x01\\x02\\x03' | b'\\xff' | '\\x04\\x05' | b'\\xff' | '\\x06\\x07'
'\\x01\\x02' | b'\\xff' | '\\x03\\x04\\x05\\x06' | b'\\xff' | '\\x07'
'\\x01' | b'\\xff' | '\\x02\\x03\\x04\\x05\\x06' | '' | ''
-------------- | ------------ | ---------------------- | ------------ | ----------
:param field : the field to consider when spliting
:type: :class:`netzob.Model.Vocabulary.AbstractField.AbstractField`
:param delimiter : the delimiter used to split messages of the field
:type: :class:`netzob.Model.Types.AbstractType.AbstractType`
"""
if delimiter is None:
raise TypeError("Delimiter cannot be None.")
if field is None:
raise TypeError("Field cannot be None.")
if len(field.messages) < 1:
raise ValueError("The associated symbol does not contain any message.")
# Find message substrings after applying delimiter
splittedMessages = []
for cell in field.getValues(encoded=False, styled=False):
splittedMessage = cell.split(delimiter.value.tobytes())
splittedMessages.append(splittedMessage)
import itertools
# Inverse the array, so that columns contains observed values for each field
splittedMessages = list(itertools.zip_longest(*splittedMessages))
# If the delimiter does not create splitted fields
if len(splittedMessages) <= 1:
return
# Else, we add (2*len(splittedMessages)-1) fields
newFields = []
iField = -1
for i in range(len(splittedMessages)):
iField += 1
fieldDomain = list()
# temporary set that hosts all the observed values to prevent useless duplicate ones
observedValues = set()
has_inserted_empty_value = False
isEmptyField = True # To avoid adding an empty field
for v in splittedMessages[i]:
if v != "" and v is not None:
isEmptyField = False
if v not in observedValues:
fieldDomain.append(Raw(v))
observedValues.add(v)
else:
if not has_inserted_empty_value:
fieldDomain.append(Raw(nbBytes=0))
has_inserted_empty_value = True
if not isEmptyField:
newField = Field(domain=DomainFactory.normalizeDomain(fieldDomain), name="Field-"+str(iField))
newField.encodingFunctions = list(field.encodingFunctions.values())
newFields.append(newField)
iField += 1
str_delimiter = TypeConverter.convert(delimiter.value, BitArray, HexaString).decode('utf-8')
fieldName = "Field-sep-{}".format(str_delimiter)
newFields.append(Field(domain=Alt([delimiter, Raw(nbBytes=0)]), name=fieldName))
newFields.pop()
# Reset the field
from netzob.Inference.Vocabulary.Format import Format
Format.resetFormat(field)
# Create a field for each entry
field.fields = newFields
def split(field, delimiter):
"""Split a field (or symbol) with a specific delimiter. The
delimiter can be passed either as an ASCII, a Raw, an
HexaString, or any objects that inherit from AbstractType.
>>> from netzob.all import *
>>> samples = [b"aaaaff000000ff10", b"bbff110010ff00000011", b"ccccccccfffe1f000000ff12"]
>>> messages = [RawMessage(data=sample) for sample in samples]
>>> symbol = Symbol(messages=messages[:3])
>>> Format.splitDelimiter(symbol, ASCII("ff"))
>>> print(symbol)
Field-0 | Field-sep-6666 | Field-2 | Field-sep-6666 | Field-4
---------- | -------------- | ------------ | -------------- | ----------
'aaaa' | 'ff' | '000000' | 'ff' | '10'
'bb' | 'ff' | '110010' | 'ff' | '00000011'
'cccccccc' | 'ff' | 'fe1f000000' | 'ff' | '12'
---------- | -------------- | ------------ | -------------- | ----------
>>> samples = [b"434d446964656e74696679230400000066726564", b"5245536964656e74696679230000000000000000", b"434d44696e666f2300000000", b"524553696e666f230000000004000000696e666f", b"434d4473746174732300000000", b"52455373746174732300000000050000007374617473", b"434d4461757468656e7469667923090000006d7950617373776421", b"52455361757468656e74696679230000000000000000", b"434d44656e6372797074230a00000031323334353674657374", b"524553656e637279707423000000000a00000073707176777436273136", b"434d4464656372797074230a00000073707176777436273136", b"5245536465637279707423000000000a00000031323334353674657374", b"434d446279652300000000", b"524553627965230000000000000000", b"434d446964656e746966792307000000526f626572746f", b"5245536964656e74696679230000000000000000", b"434d44696e666f2300000000", b"524553696e666f230000000004000000696e666f", b"434d4473746174732300000000", b"52455373746174732300000000050000007374617473", b"434d4461757468656e74696679230a000000615374726f6e67507764", b"52455361757468656e74696679230000000000000000", b"434d44656e63727970742306000000616263646566", b"524553656e6372797074230000000006000000232021262724", b"434d44646563727970742306000000232021262724", b"52455364656372797074230000000006000000616263646566", b"434d446279652300000000", b"524553627965230000000000000000"]
>>> messages = [RawMessage(data=TypeConverter.convert(sample, HexaString, Raw)) for sample in samples]
>>> symbol = Symbol(messages=messages)
>>> symbol.encodingFunctions.add(TypeEncodingFunction(ASCII)) # Change visualization to hexastring
>>> Format.splitDelimiter(symbol, ASCII("#"))
>>> print(symbol)
Field-0 | Field-sep-23 | Field-2 | Field-sep-23 | Field-4
--------------- | ------------ | -------------------- | ------------ | -------
'CMDidentify' | '#' | '....fred' | '' | ''
'RESidentify' | '#' | '........' | '' | ''
'CMDinfo' | '#' | '....' | '' | ''
'RESinfo' | '#' | '........info' | '' | ''
'CMDstats' | '#' | '....' | '' | ''
'RESstats' | '#' | '........stats' | '' | ''
'CMDauthentify' | '#' | '....myPasswd!' | '' | ''
'RESauthentify' | '#' | '........' | '' | ''
'CMDencrypt' | '#' | '....123456test' | '' | ''
'RESencrypt' | '#' | "........spqvwt6'16" | '' | ''
'CMDdecrypt' | '#' | "....spqvwt6'16" | '' | ''
'RESdecrypt' | '#' | '........123456test' | '' | ''
'CMDbye' | '#' | '....' | '' | ''
'RESbye' | '#' | '........' | '' | ''
'CMDidentify' | '#' | '....Roberto' | '' | ''
'RESidentify' | '#' | '........' | '' | ''
'CMDinfo' | '#' | '....' | '' | ''
'RESinfo' | '#' | '........info' | '' | ''
'CMDstats' | '#' | '....' | '' | ''
'RESstats' | '#' | '........stats' | '' | ''
'CMDauthentify' | '#' | '....aStrongPwd' | '' | ''
'RESauthentify' | '#' | '........' | '' | ''
'CMDencrypt' | '#' | '....abcdef' | '' | ''
'RESencrypt' | '#' | '........' | '#' | " !&'$"
'CMDdecrypt' | '#' | '....' | '#' | " !&'$"
'RESdecrypt' | '#' | '........abcdef' | '' | ''
'CMDbye' | '#' | '....' | '' | ''
'RESbye' | '#' | '........' | '' | ''
--------------- | ------------ | -------------------- | ------------ | -------
>>> print(symbol.fields[0]._str_debug())
Field-0
|-- Alt
|-- Data (Raw=b'CMDidentify' ((0, 88)))
|-- Data (Raw=b'RESidentify' ((0, 88)))
|-- Data (Raw=b'CMDinfo' ((0, 56)))
|-- Data (Raw=b'RESinfo' ((0, 56)))
|-- Data (Raw=b'CMDstats' ((0, 64)))
|-- Data (Raw=b'RESstats' ((0, 64)))
|-- Data (Raw=b'CMDauthentify' ((0, 104)))
|-- Data (Raw=b'RESauthentify' ((0, 104)))
|-- Data (Raw=b'CMDencrypt' ((0, 80)))
|-- Data (Raw=b'RESencrypt' ((0, 80)))
|-- Data (Raw=b'CMDdecrypt' ((0, 80)))
|-- Data (Raw=b'RESdecrypt' ((0, 80)))
|-- Data (Raw=b'CMDbye' ((0, 48)))
|-- Data (Raw=b'RESbye' ((0, 48)))
Below is another example of the FieldSplitDelimiter usage: it splits fields based on a Raw string.
>>> from netzob.all import *
>>> samples = [b"\\x01\\x02\\x03\\xff\\x04\\x05\\xff\\x06\\x07", b"\\x01\\x02\\xff\\x03\\x04\\x05\\x06\\xff\\x07", b"\\x01\\xff\\x02\\x03\\x04\\x05\\x06"]
>>> messages = [RawMessage(data=sample) for sample in samples]
>>> symbol = Symbol(messages=messages)
>>> Format.splitDelimiter(symbol, Raw(b"\\xff"))
>>> print(symbol)
Field-0 | Field-sep-ff | Field-2 | Field-sep-ff | Field-4
-------------- | ------------ | ---------------------- | ------------ | ----------
'\\x01\\x02\\x03' | b'\\xff' | '\\x04\\x05' | b'\\xff' | '\\x06\\x07'
'\\x01\\x02' | b'\\xff' | '\\x03\\x04\\x05\\x06' | b'\\xff' | '\\x07'
'\\x01' | b'\\xff' | '\\x02\\x03\\x04\\x05\\x06' | '' | ''
-------------- | ------------ | ---------------------- | ------------ | ----------
:param field : the field to consider when spliting
:type: :class:`netzob.Model.Vocabulary.AbstractField.AbstractField`
:param delimiter : the delimiter used to split messages of the field
:type: :class:`netzob.Model.Vocabulary.Types.AbstractType.AbstractType`
"""
if delimiter is None:
raise TypeError("Delimiter cannot be None.")
if field is None:
raise TypeError("Field cannot be None.")
if len(field.messages) < 1:
raise ValueError(
"The associated symbol does not contain any message.")
# Find message substrings after applying delimiter
splittedMessages = []
for cell in field.getValues(encoded=False, styled=False):
splittedMessage = cell.split(delimiter.value.tobytes())
splittedMessages.append(splittedMessage)
import itertools
# Inverse the array, so that columns contains observed values for each field
splittedMessages = list(itertools.zip_longest(*splittedMessages))
# If the delimiter does not create splitted fields
if len(splittedMessages) <= 1:
return
# Else, we add (2*len(splittedMessages)-1) fields
newFields = []
iField = -1
for i in range(len(splittedMessages)):
iField += 1
fieldDomain = list()
# temporary set that hosts all the observed values to prevent useless duplicate ones
observedValues = set()
has_inserted_empty_value = False
isEmptyField = True # To avoid adding an empty field
for v in splittedMessages[i]:
if v != "" and v is not None:
isEmptyField = False
if v not in observedValues:
fieldDomain.append(Raw(v))
observedValues.add(v)
else:
if not has_inserted_empty_value:
fieldDomain.append(Raw(nbBytes=0))
has_inserted_empty_value = True
if not isEmptyField:
newField = Field(
domain=DomainFactory.normalizeDomain(fieldDomain),
name="Field-" + str(iField))
newField.encodingFunctions = list(
field.encodingFunctions.values())
newFields.append(newField)
iField += 1
str_delimiter = TypeConverter.convert(delimiter.value, BitArray,
HexaString).decode('utf-8')
fieldName = "Field-sep-{}".format(str_delimiter)
newFields.append(
Field(domain=Alt([delimiter, Raw(nbBytes=0)]), name=fieldName))
newFields.pop()
# Reset the field
from netzob.Inference.Vocabulary.Format import Format
Format.resetFormat(field)
# Create a field for each entry
field.fields = newFields
def children(self, children):
from netzob.Model.Vocabulary.Domain.DomainFactory import DomainFactory
self._children = []
for child in children:
normalizedChild = DomainFactory.normalizeDomain(child)
self._children.append(normalizedChild)
def children(self, children):
from netzob.Model.Vocabulary.Domain.DomainFactory import DomainFactory
self._children = []
for child in children:
normalizedChild = DomainFactory.normalizeDomain(child)
self._children.append(normalizedChild)