HACKY

Python hacking suite.

INSTALLATION

Tested on CPython 3.4

python setup.py install (preferred)

or

pip install hacky

EXAMPLES

• Class transplantation:

>>> import hacky
>>> hacky.set_class(100, type('MyInt', (int,), {'__repr__': lambda self: "hello from hacky"}))
hello from hacky
>>> 100
hello from hacky
>>> 

• Messing around with tp_flags:

>>> from types import FunctionType
>>> import hacky
>>> hacky.set_flags(
        FunctionType,
        hacky.get_flags(FunctionType) | (1 << 10)
    )
<class 'function'>
# you can now subclass function type
>>> class T(FunctionType): pass
>>>

• Make your class final:

>>> hacky.set_flags(int, hacky.get_flags(int) ^ (1 << 10))
<class 'int'>
>>> class T(int): pass
TypeError: type 'int' is not an acceptable base type

• Read process's memory. Lets examine memory around int object 100.

>>> [hacky.read_memory_in(id(100)+shift) for shift in range(20)]
[7, 0, 0, 0, 0, 0, 0, 0, 192, 228, 158, 0, 0, 0, 0, 0, 1, 0, 0, 0]
>>> import struct
>>> [int(item) for item in struct.pack('<I', id(int))]
[192, 228, 158, 0]

There are 7 links on this instance and also we get correct address of type int located in memory.

• Modify process's memory:

>>> hacky.write_memory_in(id(100)+24, 200)
>>> 100
200

• Inject attr at type`s dictionary

>>> object_dict = hacky.fetch_dict(object)
>>> object_dict["foo"] = "hello world"
>>> (100).foo
'hello world'
>>> "hello world".foo
'hello world'
>>> (lambda x: x).foo
'hello world'

• More stuff soon.