So I have been debugging this bug for the last 10 hours. The problem is that I wrote a program that get's data from a usb scope and processes it, but after I frooze it using pyinstaller it would crash without any error. I made test program and by adding print statements throughout my code I traced down the bug to this one line:
y = signal.convolve(prev_values,filter_b,mode ='valid')
I tried replacing this line with:
y = np.convolve(prev_values,filter_b,mode ='valid')
but that also crashes. I tried replacing the line with
y = np.ones(len(prev_values)-len(filter_b)+1)
and everything is fine. The command I use to freeze my app is:
pyinstaller --log-level=DEBUG --add-binary "./dlls/*;./" --add-binary "./interface.ui;./" --add-binary "./settings.yaml;./" "app.py"
The problem now is that I have now idea where this goes wrong. Is it python having some memory issues or something? Or could both numpy and scipy have a bug that causes this? Or is it something with pyinstaller? Any tips are welcome. Also if you know a good alternative to do a convolution for numpy and signal.
this np.convolve function was also essencial for my app and also was giving me some problems while using pyinstaller. What happened is that my numpy library was up to date while my pyinstaller wasn't, so i think all the hooks, that called the aproppriate dlls and etc to make everything work, were wrong. I've tried to manually fix it adding all the missing stuff manually, but in the end updating pyinstaller made everything work without any further issue.
No python expert here, but I hope updating everything helps you.
Related
Even after installing dash - getting module import error.
Pycharm does not throw any compilation errors on the line as well.
How could this be resolved?
I tried everything as suggested in comments and stuff known to me but could not resolve this - have done this a thousand times earlier but don't know what went wrong this time.
The only solution the=at worked was to create a new project, copy files over manually and then create a fresh interpreter and VE.
So I am trying to write some python code that will do two things, that seem to be mutually exclusive on my machine. My PC's host operating system is windows and I run Kali-Linux in WSL2 when I need to test my code on Linux. My code's main function creates two separate multiprocessing.Process objects, assigning a different thread, starting them both one after the other and then calling for them both to be joined. The plan is to allow each to run a simple server application simultaneously on different ports. This does not work when running python3 in PowerShell, as it seems to require access to os.fork() which doesn't work in said environment. When I found this out I pivoted to running in WSL2 which worked fantastically, for a time. After a while of experimenting with some ideas I decided to take some of my code and spin it off into its own file, which I placed in its own 'Libs' folder. WSL2 however, was unable to import this new file, instead giving me the exception ModuleNotFoundError: No module named 'NetStuff'. I originally had added:
sys.path.append('./Libs')
as has worked for me in the past, however when I found that WSL2 was unable to find my module, I printed out sys.path and it revealed that rather than appending my $current_working_directory/Libs like I intended, I was just appending the literal string, which wasn't useful. I then decided to try:
sys.path.append(str(pathlib.Path().resolve()) + '/Libs')
which at the bare minimum shows up as I would expect in sys.path. This, however still didn't work, python was unable to find my module and would unceremoniously crash every time. This led me to try something else, I ran my code in python3 under PowerShell again, which had no issue importing my module, it did still crash due to lacking os.fork() but the import gave no issues. Confused and annoyed I opened my code in IDLE 3.9 which, for some inexplicable reason, was able to import the file, and seemingly use os.fork(). The only major issue with running in IDLE is that it is seemingly incapable of understanding ascii colour escape characters. Given that the goal is to run my code in bash, and ideally also PowerShell, I am not satisfied with this as a solution. I returned to trying to fix the issue in WSL2 by adding my module to /home/Noah/bin, and appending this directory to sys.path, but this has still not so much as given me a new symptom.
I am utterly at a loss at this point. none of the fixes I know off hand are working, and neither are the new ones I've found online. I can't tell if I'm just missing something fundamental about python or if I'm running into a bug, if it's the latter, i can't seem to find other people with the same issue. As a result of my confusion and frustration I am appealing to you, kind users of stackoverflow.
The following is the snippet that is causing me problems in WSL2:
path0 = ('/home/Noah/bin')
path1 = (str(pathlib.Path().resolve()) + '/Libs')
sys.path.append(path0)
sys.path.append(path1)
print(sys.path)
import NetStuff
The following is output of print(sys.path) in WSL2:
['/mnt/c/Users/Noah/VSCodeRepos/Python/BlackPack', '/usr/lib/python39.zip', '/usr/lib/python3.9', '/usr/lib/python3.9/lib-dynload', '/home/noah/.local/lib/python3.9/site-packages', '/usr/local/lib/python3.9/dist-packages', '/usr/lib/python3/dist-packages', '/home/Noah/bin', '/mnt/c/Users/Noah/VSCodeRepos/Python/BlackPack/Libs']
The following is the error being thrown by WSL2:
Traceback (most recent call last):
File "/mnt/c/Users/Noah/VSCodeRepos/Python/BlackPack/BlackPackServer.py", line 21, in <module>
import NetStuff
ModuleNotFoundError: No module named 'NetStuff'
I am specifically hoping to fix the issue with WSL2 at the moment as I am fairly certain that getting the code to run on PowerShell is merely going to require rewriting my code so that it doesn't rely on os.fork(). Thank you for reading my problem, and if I left out any information that you would like to see just tell me and I'll add it in an edit!
Edit: I instantly realized that I should specify that my host machine is running windows 10.
I really need help dealing with this issue.
I run my script on PyScripter and everything works perfectly, but when I try to bundle all the scripts into .exe using PyInstaller I got an error as shown in the picture below.
I also try to run from the Anaconda prompt in my virtual environment, there is no issue at all.
How can I fix this issue??
FYI, I also provide my conda information.
and all the packages that I used.
Plus, a snippet from my code (I don't know whether it's relevant or not)
The figure below should be the output that I look for
Thank you for your help, it's been many days and I still couldn't fix this issue.
I'm not sure if this gonna solve your problem, but you can try my method.
I directly change the code from fiona's __init__ file.
import fiona._loading
with fiona._loading.add_gdal_dll_directories():
to
import _loading
with _loading.add_gdal_dll_directories():
Try to use cx_freeze instead of pyinstaller, it worked for me.
I am building a GUI tkinter python3 application and attempting to compile it with py2app. For some reason when I try to launch the .app bundle in the dist folder it gives me this error:
A main script could not be located in the Resources folder
I was wondering why it is doing this, as it is rather frustrating, and I can not find anything about it anywhere. I copied my .py file into the resources folder (Networking.py). Previous to this error I also found an error in the Info.plist. In the key where it states the runtime executable, I found it was trying to get python2.7, which I have updated and am no longer using. I changed it to my current version, which the path for looks like this:
/Library/Frameworks/Python.framework/Versions/3.6/Python
It may be worth noting that it had a strange path previously, which did not look like a proper path to me. It was #executable_path/../Frameworks/Python.framework/Versions/2.7/PythonI removed this completely... Was this wrong? I have no idea about anything about XML, which is what it seemed to be...
Also when compiling this happened:
error: [Errno 1] Operation not permitted: '/Users/Ember/dist/Networking.app/Contents/MacOS/Networking'
Any help would be highly appreciated! Thanks!
EDIT
I actually figured out: a bit of a stupid mistake, but since I'm using python 3.x I have to type in python3 before doing it.
In your "setup.py" file that you used to create the application, did you remember to list all the modules used in your code. For example, if you used the OS and Glob modules, then you would add this to your setup.py next to "OPTIONS":
OPTIONS = {'argv_emulation': True, 'includes':['glob', 'os']}
Basically, anything that you import into your module, you should include in the setup.py. Let me know if that works.
I actually figured out: a bit of a stupid mistake, but since I'm using python 3.x I have to type in python3 before doing it.
I had the same problem.
python setup.py py2app
I tried
python3 setup.py py2app
and it worked just fine. Hope this helps.
I have created a desktop gui application which i want to package. I was directed to use pyinstaller to package it so i did. I have a mainwindow.py file where i import a resource file (converted to python code) as well as other imports going on in there. However when i run the output file,it gives me an error. I suspected it was due to my resource file, from the error so i tried pyinstaller on one of the modules and it worked perfectly. however i keep getting an error with the MainWindow.py file. I have the image of the error here for your consideration. How do i go around this error?
I have managed to solve the problem. After reading this answer, I tried pyrcc4 with the '-py3' flag and it worked. I think without the py3 flag,the resource file is converted into a python 2.x code. And I was trying to interpret py2 code with py3 interpreter. That's my understanding of the problem tho. But in case you have a similar problem,try using the -'py3' flag for python3 interpreter.Good Luck