Is it possible to inject shell/python commands from a configuration file? - python

Say you have a some meta data for a custom file format that your python app reads. Something like a csv with variables that can change as the file is manipulated:
var1,data1
var2,data2
var3,data3
So if the user can manipulate this meta data, do you have to worry about someone crafting a malformed meta data file that will allow some arbitrary code execution? The only thing I can imagine if you you made the poor choice to make var1 be a shell command that you execute with os.sys(data1) in your own code somewhere. Also, if this were C then you would have to worry about buffers being blown, but I don't think you have to worry about that with python. If your reading in that data as a string is it possible to somehow escape the string "\n os.sys('rm -r /'), this SQL like example totally wont work, but is there similar that is possible?

If you are doing what you say there (plain text, just reading and parsing a simple format), you will be safe. As you indicate, Python is generally safe from the more mundane memory corruption errors that C developers can create if they are not careful. The SQL injection scenario you note is not a concern when simply reading in files in python.
However, if you are concerned about security, which it seems you are (interjection: good for you! A good programmer should be lazy and paranoid), here are some things to consider:
Validate all input. Make sure that each piece of data you read is of the expected size, type, range, etc. Error early, and don't propagate tainted variables elsewhere in your code.
Do you know the expected names of the vars, or at least their format? Make sure the validate that it is the kind of thing you expect before you use it. If it should be just letters, confirm that with a regex or similar.
Do you know the expected range or format of the data? If you're expecting a number, make sure it's a number before you use it. If it's supposed to be a short string, verify the length; you get the idea.
What if you get characters or bytes you don't expect? What if someone throws unicode at you?
If any of these are paths, make sure you canonicalize and know that the path points to an acceptable location before you read or write.
Some specific things not to do:
os.system(attackerControlledString)
eval(attackerControlledString)
__import__(attackerControlledString)
pickle/unpickle attacker controlled content (here's why)
Also, rather than rolling your own config file format, consider ConfigParser or something like JSON. A well understood format (and libraries) helps you get a leg up on proper validation.
OWASP would be my normal go-to for providing a "further reading" link, but their Input Validation page needs help. In lieu, this looks like a reasonably pragmatic read: "Secure Programmer: Validating Input". A slightly dated but more python specific one is "Dealing with User Input in Python"

Depends entirely on the way the file is processed, but generally this should be safe. In Python, you have to put in some effort if you want to treat text as code and execute it.

Related

How to dynamically store and execute equations and functions

During my current project, I have been receiving data from a set of long-range sensors, which are sending data as a series of bytes. Generally, due to having multiple types of sensors, the bytes structures and data contained are different, hence the need to make the functionality more dynamic as to avoid having to hard-code every single setup in the future (which is not practical).
The server will be using Django, which I believe is irrelevant to the issue at hand but I have mentioned just in case it might have something that can be used.
The bytes data I am receiving looks like this:
b'B\x10Vu\x87%\x00x\r\x0f\x04\x01\x00\x00\x00\x00\x1e\x00\x00\x00\x00ad;l'
And my current process looks like this:
Take the first bytes to get the deviceID (deviceID = val[0:6].hex())
Look up the format to be used in the struct.unpack() (here: >BBHBBhBHhHL after removing the first bytes for the id.
Now, the issue is the next step. Many of the datas I have have different forms of per-processing that needs to be done. F.e. some values need to be ran with a join statement (e.g. ".".join(str(values[2]) ) while others need some simple mathematical changes (-113 + 2 * values[4]) and finally, others needs a simple logic check (values[7]==0x80) to return a boolean value.
My question is, what's the best way to code those methods? I would really like to avoid hardcoding them, but it almost seems like the best idea. another idea I saw was to store the functionalities as a string and execute them such as seen here, but I've been reading that its a very bad idea, and that it also slows down execution. The last idea I had was to hardcode some general functions only and use something similar to here, but this doesn't solve the issue of having to hard-code every new sensor-type, which is not realistic in a live-installation. Are there any better methods to achieve the same thing?
I have also looked at here, with the idea that some functionality can be somehow optimized as an equation, but I didn't see that a possibility for every occurrence, especially when any string manipulation is needed at all.
Additionally, is there a possibility of using some maths to apply some basic string manipulation? I can hard-code one string manipulation maybe, but to be honest this whole thing has been bugging me...
Finally, I am considering if I go with the function storing as string then executing, is there a way to set some "security" to avoid any malicious exploitation? Since such a method is... awful insecure to say the least.
However, after almost a week total of searching I am so far unable to find a better solution than storing functions as a string and running eval on them, despite not liking that option. If anyone finds a better option before then, I would be extremely grateful to any tips or ideas.
Appendum: Minimum code that can be used to show-case and test different methods:
import struct
def decode(input):
val = bytearray(input)
deviceID = val[0:6].hex()
del(val[0:6])
print(deviceID)
values = list(struct.unpack('>BBHBBhBHhHL', val))
print(values)
# Now what?
decode(b'B\x10Vu\x87%\x00x\r\x0f\x04\x01\x00\x00\x00\x00\x1e\x00\x00\x00\x00ad;l')

json data for/in python unittests

I'm writing an ORM-like library for Mongo. I've written some models, and want to make sure that they and the machinery that supports them are correct, so I'd like to write unit tests for them. I figured the best way to do that would be to simply write out some test data as JSON, then pass them to my models and see if the valid data is considered valid and the invalid data as invalid.
My question is where to put that data: it seems like a lot of non-TestCase stuff to put in test_models.py, but adding a separate test_data_for_models.json contributes its own headaches (versioning, etc.)
Which is the more recommended/idiomatic? Bear in mind I don't really need to generate any data--I'll likely just adapt data that's already in the db--I'm just unsure where to put it.

Verify directory write privileges

I am writing a program in Python 2.7 that retrieves remote files and dumps them in a directory that can be specified by the user. Currently, in order to verify that the program can in fact write to that directory, I do the following (assuming that dumpdir is a string containing the name of the directory to check):
try:
os.mkdir(dumpdir+'/.mwcrawler')
os.rmdir(dumpdir+'/.mwcrawler')
except:
logging.error('Could not open %s for writing, using default', dumpdir)
But this feels even more hackish than my usual code. What's the correct way to go about this? Maybe some sort of assertion on privileges?
In general, it's better to ask for forgiveness than permission—you have to handle errors in writing each file anyway, so why check in advance?
But, when you have a user interface—even a command-line interface, where you may read a prefs file long before you get to any writing—it's often much more convenient to the user to return errors as soon as possible. As long as that's the only reason you're doing this check, there's nothing wrong with it.
However, there are many little ways you could improve the way you do the check.
First, you should almost never just use except: without specifying anything. Besides the fact that this catches different things in different versions of Python (and therefore also confuses human readers used to other versions), it means you have no way of distinguishing between not writable, a bad Unicode character, or even a typo in the code.
Plus, your error message says "not readable" if it can't write, which is pretty odd.
Second, unless you're sure nobody will ever have a file named .mwcrawler (e.g., because you refuse to transfer files starting with '.' or something), using any fixed name is just asking for trouble. A better solution is to use, e.g., tempfile.mkdtemp.
Also, you should avoid using string manipulation for paths if you want to be portable. That's what os.path (and higher-level utilities) are for—so you don't have to learn or think about Windows, zOS, etc.
Putting it all together:
try:
d = tempfile.mkdtemp(prefix='.mwcrawler', dir=dumpdir)
except Exception as e:
logging.error('Could not open %s for reading (%s), using default', dumpdir, e)
else:
os.rmdir(d)
This link describes the usage of os.access, a method specifically created for your needs.
It also explains a better way of approaching rights checking.
As also rightfully mentioned in comments, os.access will have issues in a few specific cases, so just to be totally sure, "hit-n-run" approach is actually better, try writing, catch exception, see what happened - go from there.

lists or dicts over zeromq in python

What is the correct/best way to send objects like lists or dicts over zeromq in python?
What if we use a PUB/SUB pattern, where the first part of the string would be used as a filter?
I am aware that there are multipart messages, but they where originally meant for a different purpose. Further you can not subscribe all messages, which have a certain string as the first element.
Manual serialization
You turn the data into a string, concatenate or else, do your stuff. It's fast and doesn't take much space but requires work and maintenance, and it's not flexible.
If another language wants to read the data, you need to code it again. No DRY.
Ok for very small data, but really the amount of work is usually not worth it unless you are looking for speed and memory effiency and that you can measure that your implementation is significantly better.
Pickle
Slow, but you can serialize complex objects, and even callable. It's powerfull, and it's so easy it's a no brainer.
On the other side it's possible to end up with something you can't pickle and break your code. Plus you can't share the data with any lib written in an other language.
Eventually, the format is not human readable (hard do debug) and quite verbose.
Very nice to share objects and tasks, not so nice for messages.
json
Reasonably fast, easy to implement with simple to averagely complex data structures. It's flexible, human readible and data can be shared accross languages easily.
For complex data, you'll have to write a bit of code.
Unless you have a very specific need, this is probably the best balance between features and complexity. Espacially since the last implementation in the Python lib is in C and speed is ok.
xml
Verbose, hard to create and a pain to maintain unless you got some heavy lib that that does all the job for you. Slow.
Unless it's a requirement, I would avoid it.
In the end
Now as usual, speed and space efficiency is relative, and you must first answer the questions:
what efficiency do I need ?
what am I ready to pay (money, time, energy) for that ?
what solution fits in my current system ?
It's all what matters.
That wonderful moment of philosophy passed, use JSON.
JSON:
# Client
socket.send(json.dumps(message))
# Server
message = json.loads(socket.recv())
More info:
JSON encoder and decoder
hwserver.py
hwclient.py
In zeroMQ, a message is simple a binary blob. You can put anything in it that you want. When you have an object that has multiple parts, you need to first serialize it into something that can be deserialized on the other end. The simplest way to do this is to use obj.repr() which produces a string that you can execute at the other end to recreate the object. But that is not the best way.
First of all, you should try to use a language independent format because sooner or later you will need to interact with applications written in other languages. A JSON object is a good choice for this because it is a single string that can be decoded by many languages. However, a JSON object might not be the most efficient representation if you are sending lots of messages across the network. Instead you might want to consider a format like MSGPACK or Protobufs.
If you need a topic identiffier for PUB_SUB, then simply tack it onto the beginning. Either use a fixed length topic, or place a delimiter between the topic and the real message.
Encode as JSON before sending, and decode as JSON after receiving.
Also check out MessagePack
http://msgpack.org/
"It's like JSON. but fast and small"
In case you are interested in seeing examples, I released a small package called pyRpc that shows you how to do a simple python RPC setup where you expose services between different apps. It uses the python zeromq built-in method for sending and receiving python objects (which I believe is simply cPickle)
http://pypi.python.org/pypi/pyRpc/0.1
https://github.com/justinfx/pyRpc
While my examples use the pyobj version of the send and receive calls, you can see there are other versions available that you can use, like send_json, send_unicode... Unless you need some specific type of serialization, you can easily just use the convenience send/receive functions that handle the serialization/deserialization on both ends for you.
http://zeromq.github.com/pyzmq/api/generated/zmq.core.socket.html
json is probably the fastest, and if you need even faster than what is included in zeromq, you could manually use cjson. If your focus is speed then this is a good option. But if you know you will be communicating only with other python services, then the benefit of cPickle is a native python serialization format that gives you a lot of control. You can easily define your classes to serialize the way you want, and end up with native python objects in the end, as opposed to basic values. Im sure you could also write your own object hook for json if you wanted.
There are a few questions in that question but in terms of best / correct way to send objects / dics obviously it depends. For a lot of situations JSON is simple and familiar to most. To get it to work I had to use send_string and recv_string e.g.
# client.py
socket.send_string(json.dumps({'data': ['a', 'b', 'c']}))
# server.py
result = json.loads(socket.recv_string())
Discussion in docs https://pyzmq.readthedocs.io/en/latest/unicode.html

Is it ever polite to put code in a python configuration file?

One of my favorite features about python is that you can write configuration files in python that are very simple to read and understand. If you put a few boundaries on yourself, you can be pretty confident that non-pythonistas will know exactly what you mean and will be perfectly capable of reconfiguring your program.
My question is, what exactly are those boundaries? My own personal heuristic was
Avoid flow control. No functions, loops, or conditionals. Those wouldn't be in a text config file and people aren't expecting to have understand them. In general, it probably shouldn't matter the order in which your statements execute.
Stick to literal assignments. Methods and functions called on objects are harder to think through. Anything implicit is going to be a mess. If there's something complicated that has to happen with your parameters, change how they're interpreted.
Language keywords and error handling are right out.
I guess I ask this because I came across a situation with my Django config file where it seems to be useful to break these rules. I happen to like it, but I feel a little guilty. Basically, my project is deployed through svn checkouts to a couple different servers that won't all be configured the same (some will share a database, some won't, for example). So, I throw a hook at the end:
try:
from settings_overrides import *
LOCALIZED = True
except ImportError:
LOCALIZED = False
where settings_overrides is on the python path but outside the working copy. What do you think, either about this example, or about python config boundaries in general?
There is a Django wiki page, which addresses exactly the thing you're asking.
http://code.djangoproject.com/wiki/SplitSettings
Do not reinvent the wheel. Use configparser and INI files. Python files are to easy to break by someone, who doesn't know Python.
Your heuristics are good. Rules are made so that boundaries are set and only broken when it's obviously a vastly better solution than the alternate.
Still, I can't help but wonder that the site checking code should be in the parser, and an additional configuration item added that selects which option should be taken.
I don't think that in this case the alternative is so bad that breaking the rules makes sense...
-Adam
I think it's a pain vs pleasure argument.
It's not wrong to put code in a Python config file because it's all valid Python, but it does mean you could confuse a user who comes in to reconfigure an app. If you're that worried about it, rope it off with comments explaining roughly what it does and that the user shouldn't edit it, rather edit the settings_overrides.py file.
As for your example, that's nigh on essential for developers to test then deploy their apps. Definitely more pleasure than pain. But you should really do this instead:
LOCALIZED = False
try:
from settings_overrides import *
except ImportError:
pass
And in your settings_overrides.py file:
LOCALIZED = True
... If nothing but to make it clear what that file does.. What you're doing there splits overrides into two places.
As a general practice, see the other answers on the page; it all depends. Specifically for Django, however, I see nothing fundamentally wrong with writing code in the settings.py file... after all, the settings file IS code :-)
The Django docs on settings themselves say:
A settings file is just a Python module with module-level variables.
And give the example:
assign settings dynamically using normal Python syntax. For example:
MY_SETTING = [str(i) for i in range(30)]
Settings as code is also a security risk. You import your "config", but in reality you are executing whatever code is in that file. Put config in files that you parse first and you can reject nonsensical or malicious values, even if it is more work for you. I blogged about this in December 2008.

Categories