Unable to complete onefile creation

Hi all,

Following Kay Hayen’s advice on Gitter, here’s a full report of the issue I found.

• Nuitka version:

Python: 3.9.2 (default, Feb 28 2021, 17:03:44)
Executable: /usr/bin/python3
OS: Linux (Windows WSL/Debian)
Arch: x86_64

• How did you install Nuitka and Python: Python was downloaded from the original site, Nuitka was installed using sudo apt-get -y install nuitka. Additional installation includes sudo apt-get -y install clang.

• The specific PyPI names and versions:

beautifulsoup4==4.9.3
chardet==4.0.0
click==8.1.3
dlib==19.24.0
face-recognition==1.3.0
face-recognition-models==0.3.0
html5lib==1.1
imageio==2.19.3
joblib==1.1.0
lxml==4.6.3
natsort==8.1.0
networkx==2.8.4
Nuitka==0.6.11.3
numpy==1.22.4
packaging==21.3
pbr==5.9.0
Pillow==9.1.1
Pygments==2.7.1
pyparsing==3.0.9
PyWavelets==1.3.0
PyYAML==5.3.1
scikit-image==0.19.3
scikit-learn==1.1.1
scipy==1.8.1
SCons==4.0.1
six==1.16.0
soupsieve==2.2.1
testresources==2.0.1
threadpoolctl==3.1.0
tifffile==2022.5.4
torch==1.7.1+cpu
typing_extensions==4.2.0
webencodings==0.5.1

• Compiling hello.py also fails:

Nuitka:INFO: Starting Python compilation.
Nuitka:INFO: Completed Python level compilation and optimization.
Nuitka:INFO: Generating source code for C backend compiler.
Nuitka:INFO: Running data composer tool for optimal constant value handling.
Nuitka:INFO: Running C level backend compilation via Scons.
Nuitka-Scons:INFO: Backend C compiler: gcc (gcc).
Nuitka-Scons:INFO: Compiled 12 C files using ccache.
Nuitka-Scons:INFO: Cached C files (using ccache) with result 'disabled': 1
Nuitka-Scons:INFO: Cached C files (using ccache) with result 'cache hit': 9
Nuitka-Scons:INFO: Cached C files (using ccache) with result 'cache miss': 2
Nuitka-Postprocessing:INFO: Creating single file from dist folder, this may take a while.
FATAL: Error, expected output file hello.bin not created by AppImage.

• Also supply a Short, Self Contained, Correct, Example: stripped actual code

import cv2
import numpy as np
import torch
import os
import time
import skimage

import configparser

#------------------------------------------------------
#                     Read parameters
# Set this depending on your camera type:
# - True = Raspberry Pi 2.x camera module
# - False = USB webcam or other USB video input (like an HDMI capture device)
#------------------------------------------------------
config = configparser.ConfigParser()
config.read(initFilename)

USING_RPI_CAMERA_MODULE = (config.get("init", "CAMERA_INPUT") == "RPI")
FLIP_IMAGE_MODE = int(config.get("init", "FLIP_IMAGE_MODE"))
MEASURE_TIME = (config.get("init", "MEASURE_TIME") == "True")
SIMILARITY_THRESHOLD = float(config.et("crypto", "SIMILARITY_THRESHOLD"))

##################### Training Parametres #############
OUTPUT_PATH = config.get("init", "OUTPUT_PATH")
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")

IMAGE_SIZE = int(config.get("crypto", "IMAGE_SIZE"))
CRYPTO_BLOCK_SIZE = int(config.get("crypto", "CRYPTO_BLOCK_SIZE"))
USE_CRYPTO_RANDOM_IMAGE = (config.get("crypto", "USE_CRYPTO_RANDOM_IMAGE") == "True")
MAX_GRAYSCALE = float(config.get("crypto", "MAX_GRAYSCALE"))

BACKBONE_WEIGHT_NAME = OUTPUT_PATH + "/" + config.get("results", "BACKBONE_WEIGHT_NAME")
REFERENCE_NAME = OUTPUT_PATH + "/" + config.get("crypto", "REFERENCE_NAME")

####################  Encryption parameters ###########
BLOCK_ORDER_REFERENCE = OUTPUT_PATH + "/" + config.get("crypto", "BLOCK_ORDER_REFERENCE")
with open(BLOCK_ORDER_REFERENCE, 'rb') as f:
    shuffle_block_order = np.load(f)
LIST_BLOCK_REFERENCE = OUTPUT_PATH + "/" + config.get("crypto", "LIST_BLOCK_REFERENCE")
with open(LIST_BLOCK_REFERENCE, 'rb') as f:
    list_block_keys = np.load(f)
CLASS_NAME_REFERENCE = OUTPUT_PATH + "/" + config.get("crypto", "CLASS_NAME_REFERENCE")
with open(CLASS_NAME_REFERENCE, 'rb') as f:
    class_names = np.load(f)
LABEL_REFERENCE = OUTPUT_PATH + "/" + config.get("crypto", "LABEL_REFERENCE")
with open(LABEL_REFERENCE, 'rb') as f:
    labelSet = np.load(f)

referenceSet = torch.load(REFERENCE_NAME).to(device)

##########################################################
##################### Network's configuration ############
print("Loading model...")

def main_loop():

    print("Begin testing...")

    while True:

        if MEASURE_TIME:
            start_time =  time.time()

        # Hit 'q' on the keyboard to quit!
        if cv2.waitKey(1) & 0xFF == ord('q'):
            break

    # Release handle to the webcam
    cv2.destroyAllWindows()

if __name__ == "__main__":
    main_loop()

• Provide in your issue the Nuitka options used: python3 -m nuitka --onefile --python-flag=no_site --prefer-source-code --plugin-enable=numpy --plugin-enable=torch --linux-onefile-icon=ARCLAN1.ico ssce.py

• Note if this is a regression: n/a

Here’s the output when compiling real code:

 python3 -m nuitka --onefile --python-flag=no_site --prefer-source-code --plugin-enable=numpy --plugin-enable=torch --linux-onefile-icon=test.ico testapp.py
Nuitka:INFO: Starting Python compilation.
Nuitka:INFO: Completed Python level compilation and optimization.
Nuitka:INFO: Generating source code for C backend compiler.
Nuitka:INFO: Running data composer tool for optimal constant value handling.
Nuitka:INFO: Running C level backend compilation via Scons.
Nuitka-Scons:INFO: Backend C compiler: gcc (gcc).
Nuitka-Scons:INFO: Slow C compilation detected, used 60s so far, this might indicate scalability problems.
Nuitka-Scons:INFO: Running /usr/bin/ccache /usr/bin/gcc -o module.scipy.stats._continuous_distns.o -c -std=c11 -fvisibility=hidden -fwrapv -fpartial-inlining -ftrack-macro-expansion=0 -Wno-deprecated-declarations -fno-var-tracking -fcompare-debug-second -O3 -pipe -D_NUITKA_STANDALONE -D__NUITKA_NO_ASSERT__ -D_NUITKA_CONSTANTS_FROM_INCBIN -D_NUITKA_FROZEN=164 -D_NUITKA_EXE -I/usr/include/python3.9 -I. -I/usr/lib/python3/dist-packages/nuitka/build/include -I/usr/lib/python3/dist-packages/nuitka/build/static_src module.scipy.stats._continuous_distns.c took 96.71 seconds
Nuitka-Scons:INFO: Slow C compilation detected, used 60s so far, this might indicate scalability problems.
Nuitka-Scons:INFO: Running /usr/bin/ccache /usr/bin/gcc -o module.tifffile.tifffile.o -c -std=c11 -fvisibility=hidden -fwrapv -fpartial-inlining -ftrack-macro-expansion=0 -Wno-deprecated-declarations -fno-var-tracking -fcompare-debug-second -O3 -pipe -D_NUITKA_STANDALONE -D__NUITKA_NO_ASSERT__ -D_NUITKA_CONSTANTS_FROM_INCBIN -D_NUITKA_FROZEN=164 -D_NUITKA_EXE -I/usr/include/python3.9 -I. -I/usr/lib/python3/dist-packages/nuitka/build/include -I/usr/lib/python3/dist-packages/nuitka/build/static_src module.tifffile.tifffile.c took 193.58 seconds
Nuitka-Scons:INFO: Compiled 1030 C files using ccache.
Nuitka-Scons:INFO: Cached C files (using ccache) with result 'disabled': 1
Nuitka-Scons:INFO: Cached C files (using ccache) with result 'cache hit': 389
Nuitka-Scons:INFO: Cached C files (using ccache) with result 'cache miss': 640
Nuitka-Plugins:INFO: torch: Copying files from 'torch' installation:
Nuitka-Plugins:INFO: torch: Copied 1436 files.
Nuitka-Postprocessing:INFO: Creating single file from dist folder, this may take a while.
Nuitka-Postprocessing:INFO: Completed onefile creation.
Traceback (most recent call last):
  File "/usr/lib/python3/dist-packages/nuitka/__main__.py", line 209, in <module>
    main()
  File "/usr/lib/python3/dist-packages/nuitka/__main__.py", line 195, in main
    MainControl.main()
  File "/usr/lib/python3/dist-packages/nuitka/MainControl.py", line 773, in main
    packDistFolderToOnefile(dist_dir, binary_filename)
  File "/usr/lib/python3/dist-packages/nuitka/freezer/Onefile.py", line 51, in packDistFolderToOnefile
    packDistFolderToOnefileLinux(onefile_output_filename, dist_dir, binary_filename)
  File "/usr/lib/python3/dist-packages/nuitka/freezer/Onefile.py", line 161, in packDistFolderToOnefileLinux
    assert result == 0, result
AssertionError: 1

Thanks in advance for any help!

[jlgarnier]

Hi again,

I managed to update the package but the compilation now stumbles on modprobe: FATAL: Module fuse not found in directory /lib/modules/4.4.0-19041-Microsoft, while the whole process stops after downloading appimagetools with FATAL: Error, error exit from AppImage because fuse kernel module was not loaded: I’m afraid fuse cannot properly install in a WSL container.

Any feedback on this?

[kayhayen]

On develop right now, the --onefile-tempdir behavior is the default, ought to use that, it doesn’t use fuse to build anymore.