Contents
A native Python cross-version decompiler and fragment decompiler. The successor to decompyle, uncompyle, and uncompyle2.
uncompyle6 translates Python bytecode back into equivalent Python source code. It accepts bytecodes from Python version 1.0 to version 3.8, spanning over 24 years of Python releases. We include Dropbox's Python 2.5 bytecode and some PyPy bytecodes.
Ok, I'll say it: this software is amazing. It is more than your normal hacky decompiler. Using compiler technology, the program creates a parse tree of the program from the instructions; nodes at the upper levels that look a little like what might come from a Python AST. So we can really classify and understand what's going on in sections of Python bytecode.
Building on this, another thing that makes this different from other CPython bytecode decompilers is the ability to deparse just fragments of source code and give source-code information around a given bytecode offset.
I use the tree fragments to deparse fragments of code at run time inside my trepan debuggers. For that, bytecode offsets are recorded and associated with fragments of the source code. This purpose, although compatible with the original intention, is yet a little bit different. See this for more information.
Python fragment deparsing given an instruction offset is useful in showing stack traces and can be encorporated into any program that wants to show a location in more detail than just a line number at runtime. This code can be also used when source-code information does not exist and there is just bytecode. Again, my debuggers make use of this.
There were (and still are) a number of decompyle, uncompyle, uncompyle2, uncompyle3 forks around. Many of them come basically from the same code base, and (almost?) all of them are no longer actively maintained. One was really good at decompiling Python 1.5-2.3, another really good at Python 2.7, but that only. Another handles Python 3.2 only; another patched that and handled only 3.3. You get the idea. This code pulls all of these forks together and moves forward. There is some serious refactoring and cleanup in this code base over those old forks. Even more experimental refactoring is going on in decompyle3.
This demonstrably does the best in decompiling Python across all Python versions. And even when there is another project that only provides decompilation for subset of Python versions, we generally do demonstrably better for those as well.
How can we tell? By taking Python bytecode that comes distributed with that version of Python and decompiling these. Among those that successfully decompile, we can then make sure the resulting programs are syntactically correct by running the Python interpreter for that bytecode version. Finally, in cases where the program has a test for itself, we can run the check on the decompiled code.
We use an automated processes to find bugs. In the issue trackers for other decompilers, you will find a number of bugs we've found along the way. Very few to none of them are fixed in the other decompilers.
The code in the git repository can be run from Python 2.4 to the latest Python version, with the exception of Python 3.0 through 3.2. Volunteers are welcome to address these deficiencies if there a desire to do so.
The way it does this though is by segregating consecutive Python versions into git branches:
- master
- Python 3.6 and up (uses type annotations)
- python-3.3-to-3.5
- Python 3.3 through 3.5 (Generic Python 3)
- python-2.4
- Python 2.4 through 2.7 (Generic Python 2)
PyPy 3-2.4 and later works as well.
The bytecode files it can read have been tested on Python bytecodes from versions 1.4, 2.1-2.7, and 3.0-3.8 and later PyPy versions.
You can install from PyPI using the name uncompyle6:
pip install uncompyle6
To install from source code, this project uses setup.py, so it follows the standard Python routine:
$ pip install -e . # set up to run from source tree
or:
$ python setup.py install # may need sudo
A GNU Makefile is also provided so make install (possibly as root or
sudo) will do the steps above.
make check
A GNU makefile has been added to smooth over setting running the right command, and running tests from fastest to slowest.
If you have remake installed, you can see the list of all tasks
including tests via remake --tasks
Run
$ uncompyle6 *compiled-python-file-pyc-or-pyo*
For usage help:
$ uncompyle6 -h
In older versions of Python it was possible to verify bytecode by decompiling bytecode, and then compiling using the Python interpreter for that bytecode version. Having done this, the bytecode produced could be compared with the original bytecode. However as Python's code generation got better, this no longer was feasible.
If you want Python syntax verification of the correctness of the
decompilation process, add the --syntax-verify option. However since
Python syntax changes, you should use this option if the bytecode is
the right bytecode for the Python interpreter that will be checking
the syntax.
You can also cross compare the results with either another version of uncompyle6 since there are are sometimes regressions in decompiling specific bytecode as the overall quality improves.
For Python 3.7 and 3.8, the code in decompyle3 is generally better.
Or try specific another python decompiler like uncompyle2, unpyc37, or pycdc. Since the later two work differently, bugs here often aren't in that, and vice versa.
There is an interesting class of these programs that is readily available give stronger verification: those programs that when run test themselves. Our test suite includes these.
And Python comes with another a set of programs like this: its test
suite for the standard library. We have some code in test/stdlib to
facilitate this kind of checking too.
The biggest known and possibly fixable (but hard) problem has to do with handling control flow. (Python has probably the most diverse and screwy set of compound statements I've ever seen; there are "else" clauses on loops and try blocks that I suspect many programmers don't know about.)
All of the Python decompilers that I have looked at have problems decompiling Python's control flow. In some cases we can detect an erroneous decompilation and report that.
Python support is pretty good for Python 2
On the lower end of Python versions, decompilation seems pretty good although we don't have any automated testing in place for Python's distributed tests. Also, we don't have a Python interpreter for versions 1.6, and 2.0.
In the Python 3 series, Python support is is strongest around 3.4 or
3.3 and drops off as you move further away from those versions. Python
3.0 is weird in that it in some ways resembles 2.6 more than it does
3.1 or 2.7. Python 3.6 changes things drastically by using word codes
rather than byte codes. As a result, the jump offset field in a jump
instruction argument has been reduced. This makes the EXTENDED_ARG
instructions are now more prevalent in jump instruction; previously
they had been rare. Perhaps to compensate for the additional
EXTENDED_ARG instructions, additional jump optimization has been
added. So in sum handling control flow by ad hoc means as is currently
done is worse.
Between Python 3.5, 3.6, 3.7 there have been major changes to the
MAKE_FUNCTION and CALL_FUNCTION instructions.
Python 3.8 removes SETUP_LOOP, SETUP_EXCEPT,
BREAK_LOOP, and CONTINUE_LOOP, instructions which may
make control-flow detection harder, lacking the more sophisticated
control-flow analysis that is planned. We'll see.
Currently not all Python magic numbers are supported. Specifically in some versions of Python, notably Python 3.6, the magic number has changes several times within a version.
We support only released versions, not candidate versions. Note however that the magic of a released version is usually the same as the last candidate version prior to release.
There are also customized Python interpreters, notably Dropbox, which use their own magic and encrypt bytecode. With the exception of the Dropbox's old Python 2.5 interpreter this kind of thing is not handled.
We also don't handle PJOrion or otherwise obfuscated code. For PJOrion try: PJOrion Deobfuscator to unscramble the bytecode to get valid bytecode before trying this tool; pydecipher might help with that.
This program can't decompile Microsoft Windows EXE files created by Py2EXE, although we can probably decompile the code after you extract the bytecode properly. Pydeinstaller may help with unpacking Pyinstaller bundlers.
Handling pathologically long lists of expressions or statements is slow. We don't handle Cython or MicroPython which don't use bytecode.
There are numerous bugs in decompilation. And that's true for every other CPython decompiler I have encountered, even the ones that claimed to be "perfect" on some particular version like 2.4.
As Python progresses decompilation also gets harder because the compilation is more sophisticated and the language itself is more sophisticated. I suspect that attempts there will be fewer ad-hoc attempts like unpyc37 (which is based on a 3.3 decompiler) simply because it is harder to do so. The good news, at least from my standpoint, is that I think I understand what's needed to address the problems in a more robust way. But right now until such time as project is better funded, I do not intend to make any serious effort to support Python versions 3.8 or 3.9, including bugs that might come in. I imagine at some point I may be interested in it.
You can easily find bugs by running the tests against the standard test suite that Python uses to check itself. At any given time, there are dozens of known problems that are pretty well isolated and that could be solved if one were to put in the time to do so. The problem is that there aren't that many people who have been working on bug fixing.
Some of the bugs in 3.7 and 3.8 are simply a matter of back-porting the fixes in decompyle3. Volunteers are welcome to do so.
You may run across a bug, that you want to report. Please do so after reading How to report a bug and follow the instructions when opening an issue.
Be aware that it might not get my attention for a while. If you sponsor or support the project in some way, I'll prioritize your issues above the queue of other things I might be doing instead.
- https://github.com/rocky/python-decompile3 : Much smaller and more modern code, focusing on 3.7 and 3.8. Changes in that will get migrated back here.
- https://code.google.com/archive/p/unpyc3/ : supports Python 3.2 only. The above projects use a different decompiling technique than what is used here. Currently unmaintained.
- https://github.com/figment/unpyc3/ : fork of above, but supports Python 3.3 only. Includes some fixes like supporting function annotations. Currently unmaintained.
- https://github.com/wibiti/uncompyle2 : supports Python 2.7 only, but does that fairly well. There are situations where
uncompyle6results are incorrect whileuncompyle2results are not, but more often uncompyle6 is correct when uncompyle2 is not. Becauseuncompyle6adheres to accuracy over idiomatic Python,uncompyle2can produce more natural-looking code when it is correct. Currentlyuncompyle2is lightly maintained. See its issue tracker for more details. - How to report a bug
- The HISTORY file.
- https://github.com/rocky/python-xdis : Cross Python version disassembler
- https://github.com/rocky/python-xasm : Cross Python version assembler
- https://github.com/rocky/python-uncompyle6/wiki : Wiki Documents which describe the code and aspects of it in more detail
- https://github.com/zrax/pycdc : The README for this C++ code says it aims to support all versions of Python. You can aim your slign shot for the moon too, but I doubt you are going to hit it. This code is best for Python versions around 2.7 and 3.3 when the code was initially developed. Accuracy for current versions of Python3 and early versions of Python is lacking. Without major effort, it is unlikely it can be made to support current Python 3. See its issue tracker for details. Currently lightly maintained.