Uncompyle6 translates Python bytecode back into equivalent Python source code. It accepts bytecodes from Python version 1.3 to version 3.8, spanning over 24 years of Python releases. We include Dropbox’s Python 2.5 bytecode and some PyPy bytecode.

A native Python cross-version decompiler and fragment decompiler. The successor to decompyle, uncompyle, and uncompyle2.

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. Almost all 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 all of them are no longer actively maintained.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 or so, 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.

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 are serious about testing, and use 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.

Also Read – VulnWhisperer : Create Actionable Data From Your Vulnerability Scans


The code here can be run on Python versions 2.6 or later, PyPy 3-2.4, or PyPy-5.0.1. Python versions 2.4-2.7 are supported in the python-2.4 branch. 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 the above-mentioned PyPy versions.


This uses setup.py, so it follows the standard Python routine:

pip install -e . # set up to run from source tree
# Or if you want to install instead
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.

Running Tests

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



$ 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 another python decompiler like pycdc . Since they 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.

Known Bugs/Restrictions

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 strongest in Python 2 for 2.7 and drops off as you get further away from that. Support is also probably pretty good for python 2.3-2.4 since a lot of the goodness of early the version of the decompiler from that era has been preserved (and Python compilation in that era was minimal)

There is some work to do on the lower end Python versions which is more difficult for us to handle since 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 and 3.7 there have been major changes to the MAKE_FUNCTION and CALL_FUNCTION instructions.

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 bytcode. 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 obfuscated code. For that try: PJOrion Deobfuscator to unscramble the bytecode to get valid bytecode before trying this tool. 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. For situations like this, you might want to consider a decompilation service like Crazy Compilers. Handling pathologically long lists of expressions or statements is slow.