Uncompyle2 issue: array indices must be integers - python

I have an issue when I am trying to decompile a .pyc file.
The traceback is the following:
Traceback (most recent call last):
File "my.py", line 4, in <module>
uncompyle2.uncompyle_file("/home/user/Downloads/asd.pyc", fileobj)
File "/home/user/Desktop/uncompyle2/uncompyle2/__init__.py", line 130, in uncompyle_file
uncompyle(version, co, outstream, showasm, showast, deob)
File "/home/user/Desktop/uncompyle2/uncompyle2/__init__.py", line 93, in uncompyle
tokens, customize = scanner.disassemble(co, deob=deob)
File "/home/user/Desktop/uncompyle2/uncompyle2/Scanner.py", line 214, in disassemble
cf = self.find_jump_targets(code)
File "/home/user/Desktop/uncompyle2/uncompyle2/Scanner.py", line 926, in find_jump_targets
self.detect_structure(i, op)
File "/home/user/Desktop/uncompyle2/uncompyle2/Scanner.py", line 737, in detect_structure
if int(self.code[jmp]) == RETURN_VALUE:
**TypeError: array indices must be integers**
Any ideas about this ?
I'm using Python 2.7.6 on a Ubuntu machine.
The command that I'm running to have this work is the following:
uncompyle2 asd.pyc
//EDIT: As far as I can tell, this happens only on a specific file(asd.py). It works on other files.
Any workaround ?

The .pyc file that you're trying to decompile is probably obfuscated. It's not uncompyle2s job to also deofuscate the file.
Try something else like pyc2py. Maybe it will work.

Related

Using mysql connector with pyinstaller --onefile causes issue at runtime

I am trying to use pyinstaller to create a .exe for someone else to run. The program uses mysql to ping a sql database and return information from it. When I run the program in PyCharm and with pyinstaller --onedir, everything works fine. However, when I bundle the program with --onefile, I get this error and traceback when running:
Traceback (most recent call last):
File "main.py", line 266, in <module>
File "main.py", line 88, in main
File "main.py", line 108, in grabData
File "sql.py", line 12, in SQLconnect
File "mysql\connector\pooling.py", line 286, in connect
File "mysql\connector\connection_cext.py", line 101, in __init__
File "mysql\connector\abstracts.py", line 1095, in connect
File "mysql\connector\connection_cext.py", line 199, in _open_connection
TypeError: argument 6 must be str, not None
For reference, here is the relevant line in the _open_connection function call in connection_cext.py:
self._cmysql = _mysql_connector.MySQL(
buffered=self._buffered,
raw=self._raw,
charset_name=charset_name,
connection_timeout=(self._connection_timeout or 0),
use_unicode=self._use_unicode,
auth_plugin=self._auth_plugin,
plugin_dir=self._plugin_dir,
)
The __init__ of this file has this code snippet:
self._plugin_dir = os.path.join(
os.path.dirname(os.path.abspath(_mysql_connector.__file__)),
"mysql",
"vendor",
"plugin",
)
I believe this plugin directory is the reason my code is failing, but I do not know how to bypass this or set it so that it references the right file at runtime. I know pyinstaller creates a temp file at runtime, but I don't know how to get the module to use it as the plugin directory.
Any help would be greatly appreciated! I really really wanna keep it in --onefile mode for simplicity for the user, so any solutions that maintain that would be ideal. Thanks!

FMUException: Failed to setup the experiment

I have a fmu which is created in GT-Suite and am trying to work with it in python.
I have followed jmodelica tutorials
from pyfmi import load_fmu
model = load_fmu('myFMU.fmu')
res = model.simulate(final_time=10)
My fmu gets loaded but when I try to run model.simulate step it throws an error
Traceback (most recent call last):
File "<ipython-input-3-4812da4bb52b>", line 1, in <module>
res = model.simulate(final_time=10)
File "src\pyfmi\fmi.pyx", line 6981, in pyfmi.fmi.FMUModelCS2.simulate
File "src\pyfmi\fmi.pyx", line 304, in pyfmi.fmi.ModelBase._exec_simulate_algorithm
File "src\pyfmi\fmi.pyx", line 298, in pyfmi.fmi.ModelBase._exec_simulate_algorithm
File "C:\Users\chinn\Anaconda3\envs\test_env\lib\site-packages\pyfmi\fmi_algorithm_drivers.py", line 761, in __init__
self.model.setup_experiment(start_time=start_time, stop_time_defined=self.options["stop_time_defined"], stop_time=final_time)
File "src\pyfmi\fmi.pyx", line 4292, in pyfmi.fmi.FMUModelBase2.setup_experiment
FMUException: Failed to setup the experiment.
I have tried running it in multiple environments in my pc but am getting the same error. Googled a lot but couldn't find anything. Can some one help me with resolving this issue?
The fmu is probably not exported with the correct license setting.

Mac: Opening large files creates `invalid argument` error

I am currently trying to open a large file on my 64 bit Mac (using 64 bit Miniconda, as this is the only version for mac). I use a framework 'sumatra' which is a library to track scientific computing work and possibly reproduce it. I get the following error. I was able to debug a little, and found out that it is mainly due to X_train.npy being too big (3.6GB when not unpacked).
Traceback (most recent call last):
File "/Users/davidal/miniconda3/envs/ml_project/bin/smt", line 31, in <module>
main(sys.argv[2:])
File "/Users/davidal/miniconda3/envs/ml_project/lib/python3.5/site-packages/sumatra/commands.py", line 372, in run
project.allow_command_line_parameters)
File "/Users/davidal/miniconda3/envs/ml_project/lib/python3.5/site-packages/sumatra/commands.py", line 76, in parse_arguments
parameters = build_parameters(arg)
File "/Users/davidal/miniconda3/envs/ml_project/lib/python3.5/site-packages/sumatra/parameters.py", line 586, in build_parameters
parameters = parameter_set_class(filename)
File "/Users/davidal/miniconda3/envs/ml_project/lib/python3.5/site-packages/parameters/__init__.py", line 387, in __init__
pstr = f.read()
OSError: [Errno 22] Invalid argument
Any experience resolving such issues? Where should I look further? Any ideas?

Python error trying to convert JaCoCo to Cobertura

I am trying to convert a JaCoCo coverage report to Cobertura format (since Shippable only supports Cobertura). This guy claims to have a tool to convert JaCoCo to Cobertura, however when running his script I get the following error:
Traceback (most recent call last):
File "cover2cover.py", line 151, in <module>
jacoco2cobertura(filename, source_root)
File "cover2cover.py", line 139, in jacoco2cobertura
convert_root(root, into, source_root)
File "cover2cover.py", line 127, in convert_root
packages.append(convert_package(package))
File "cover2cover.py", line 113, in convert_package
c_classes.append(convert_class(j_class, j_package))
File "cover2cover.py", line 100, in convert_class
c_methods.append(convert_method(j_method, j_method_lines))
File "cover2cover.py", line 85, in convert_method
convert_lines(j_lines, c_method)
File "cover2cover.py", line 33, in convert_lines
for jline in j_lines:
File "cover2cover.py", line 23, in method_lines
larger = list(int(jm.attrib['line']) for jm in jmethods if int(jm.attrib['line']) > start_line)
File "cover2cover.py", line 23, in <genexpr>
larger = list(int(jm.attrib['line']) for jm in jmethods if int(jm.attrib['line']) > start_line)
KeyError: 'line'
I know nothing about python, so any help would be appreciated.
I don't know python either, but I know that python 2 and python 3 have significant differences. Perhaps you ran into that?
I was able to run the script ok with this version:
$> python --version
Python 2.7.11
To ensure I got the script without any download or browser or line-ending type issues, I did clone the git repo:
$> git clone https://github.com/rix0rrr/cover2cover.git
Then the script ran first try on my jacoco XML file.

Getting TypeError: execve() arg 3 contains a non-string value when trying to run nodejs in sublime text

I am trying to run javascript with sublime text through nodejs plugin, and been getting the following message. I've found this question that seems to be relevant but I still couldn't figure it out. The script is encoded by UTF-8. Thank you.
Traceback (most recent call last):
File "/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/threading.py", line 532, in __bootstrap_inner
File "./lib/command_thread.py", line 41, in run
File "/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/subprocess.py", line 623, in __init__
File "/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/subprocess.py", line 1141, in _execute_child
TypeError: execve() arg 3 contains a non-string value
ps. I am running OS X (10.8).
I had the same problem and it was fixed by putting this in my Nodejs.sublime-settings file (accessible by hitting command+shift+p and selecting Nodejs::User File Settings):
{
"node_path": "/opt/local/lib/",
"node_command": "/opt/local/bin/node"
}
Replace the paths with something that makes sense on your system.

Categories