I made a simple, one-file script that I would like to share with the end-user. I found that Pyinstaller does the job so I refactor my project structure regarding.
I have a project dir. Inside that, I have the package that has an empty __init__.py and a __main__.py with the actual script with a few include to libraries like opencv-python, numpy, etc. Outside of the project, I have a setup.py and an entry-point script that imports the main function from the __main__.py and calls it.
Then I have created the executable against this entry-point with PyInstaller in --onefile mode. When I use the created executable on my machine it does the job perfectly but when I send it to the end-user it ends up with an error (see below). I am not quite sure whats this error means but I saw paths in that to my dev environment which should be not there on other machines. It looks like it missing the dependencies but I thought that PyInstaller bundles these dependencies into the executable.
What am I missing here? I have read many regarding questions here in StackOverflow but I couldn't find a solution.
I developed this script using PyCharm on the latest macOS within a venv created by PyCharm. The venv folder is in the project dir - I don't know that could be a problem.
Traceback (most recent call last):
File "PyInstaller/loader/rthooks/pyi_rth_pkgres.py", line 11, in <module>
File "/Users/hordon/Documents/DEV/projects/scan_detect/venv/lib/python3.7/site-packages/PyInstaller/loader/pyimod03_importers.py", line 627, in exec_module
File "setuptools-40.8.0-py3.7.egg/pkg_resources/__init__.py", line 33, in <module>
File "/Users/hordon/Documents/DEV/projects/scan_detect/venv/lib/python3.7/site-packages/PyInstaller/loader/pyimod03_importers.py", line 627, in exec_module
File "platform.py", line 116, in <module>
File "/Users/hordon/Documents/DEV/projects/scan_detect/venv/lib/python3.7/site-packages/PyInstaller/loader/pyimod03_importers.py", line 627, in exec_module
File "subprocess.py", line 153, in <module>
ImportError: dlopen(/var/folders/wk/cwx1b16j50z5_yt1ynq82hr00000gn/T/_MEI7eUUkV/select.cpython-37m-darwin.so, 2): Symbol not found: ____chkstk_darwin
Referenced from: /var/folders/wk/cwx1b16j50z5_yt1ynq82hr00000gn/T/_MEI7eUUkV/select.cpython-37m-darwin.so (which was built for Mac OS X 10.15)
Expected in: /usr/lib/libSystem.B.dylib
in /var/folders/wk/cwx1b16j50z5_yt1ynq82hr00000gn/T/_MEI7eUUkV/select.cpython-37m-darwin.so
[15314] Failed to execute script pyi_rth_pkgres
i have the same problem ，and sadly it's a problem of mac os. So, if you want to support different version of mac os, you need to build your app on the oldest version of mac os.
"In Mac OS X, components from one version of the OS are usually compatible with later versions, but they may not work with earlier versions.
The only way to be certain your app supports an older version of Mac OS X is to run PyInstaller in the oldest version of the OS you need to support."
https://pyinstaller.readthedocs.io/en/stable/usage.html#making-mac-os-x-apps-forward-compatible
Related
On PC1, I have a program in which package A B C is employed, and I'd
like to pack the program by Pyinstaller together with package A B
(excluding C for some reason), obtaining "main.exe".
Assuming that there's another machine PC2 installed with python and
package C, I hope to run this "main.exe" (without package C
embedded) on PC2, during which package C installed on PC2 is called.
Is this realizable? and How?
I pack the program on PC1 by:pyinstaller main.py --exclude-module C and as expected, failing to execute main.exe on PC2 with ModuleNotFoundError: No module named 'C', since package C installed on PC2 can not be recognized by main.exe.
To be precise, Package C mentioned refers to "geatpy" http://geatpy.com, Package A, B refer to Numpy, Pandas, xlwt, etc. Bundling all of these into one bundle and running it, some errors occur:
Traceback (most recent call last):
File "main.py", line 9, in <module>
from optimization import *
File "PyInstaller\loader\pyimod03_importers.py", line 540, in exec_module
File "optimization.py", line 20, in <module>
import geatpy as ea
File "PyInstaller\loader\pyimod03_importers.py", line 540, in exec_module
File "geatpy\__init__.py", line 16, in <module>
ModuleNotFoundError: No module named 'awGA'
[48140] Failed to execute script main
After enquiring with the developer of this module, he suggests excluding this module in the bundle, and install via pip cmd instead, because pyinstaller may be not well compatible with pyd/dll file. I am quite new to coding, so maybe I can't get his point or there may be some misunderstanding.
Finally, I found a solution by manually copy all the .pyd file in folder Python38\Lib\site-packages\geatpy into the pyinstaller-produced folder dist\main, and I do it the same way for several other missed .py modules warned during execution.
To generate the dist\main folder, pyinstaller -c main.py is applied instead of bundling it into one executable file.
As buran say, the missing import problem is due to the inability of pyinstaller to detect what is needed, especially for .pyd file.
I try some other arguments for pyinstaller but fails, like: --paths=.... --hidden-import=geatpy. Maybe that's the limitation of pyinstaller.
For the second suggestion of buran, I haven't tried it yet, but it's believed to be feasible.
I have been trying to deploy a Cloud Function with some private dependencies (pyodbc) as I couldn't get it working thru requirements.txt. Please note, I don't want to use Docker here. So all I have built here is below files,
1. main.py
2. process.py ( this one use pyodbc to connect to teradata)
3. libs (folder)
3.1 pyodbc-4.0.30.dist-info (package)
3.2 pyodbc (python extension module)
3.3 __init.py__ ( this is to make this folder as module)
4.requirements.txt
I also updated process.py file to import pyodbc module as below,
import libs.pyodbc
Please note: I used GCP docs to install the pyodbc package and put in libs using https://cloud.google.com/functions/docs/writing/specifying-dependencies-python
On top this, I am also requirements.txt to import as default.
But I am still getting module error as below.
Error message: Code in file main.py can't be loaded.
Did you list all required modules in requirements.txt?
Detailed stack trace: Traceback (most recent call last):
File "/env/local/lib/python3.7/site-packages/google/cloud/functions/worker.py", line 305, in check_or_load_user_function
_function_handler.load_user_function()
File "/env/local/lib/python3.7/site-packages/google/cloud/functions/worker.py", line 184, in load_user_function
spec.loader.exec_module(main)
File "<frozen importlib._bootstrap_external>", line 728, in exec_module
File "<frozen importlib._bootstrap>", line 219, in _call_with_frames_removed
File "/user_code/main.py", line 9, in <module>
from process import process
File "/user_code/process.py", line 6, in <module>
import libs.pyodbc
ModuleNotFoundError: No module named 'libs.pyodbc'
Any leads or help from here is really appreciated. All I am trying to achieve here is, Read CSV files from GCP bucket and process it thru dataframe which loads into teradata and generate output file back into another GCP bucket. I am trying to achieve all with Cloud Functions only. Thank you
The pyodbc project might be a bit of a special case here, because:
The project requires some platform-specific code;
They haven't published source distributions for their latest release (only built distributions).
Here's what I did to get this to work. Starting with an empty libs directory, first download the latest available source distribution:
$ pip download pyodbc --no-binary :all:
Make a directory for the module:
$ mkdir libs/pyodbc
Untar the source distribution into the module:
$ tar xf pyodbc-4.0.28.tar.gz -C libs/pyodbc
Then, in the function you can do:
import libs.pyodbc
Past day has basically been a continuous frustration. I'm trying to create my first pyQt5 application and freeze it to OSX app. Here short chronology of my efforts so far:
Noticed that PyQt5 install is only for Python3. Installed Python3 and pyQt via Homebrew.
Developed application. Everything works when launched from PyCharm.
Installed cx_freeze from source to Python3 since Pycharm's pip installer failed in task.
Trying to freeze the application with cx_freeze only to get syntax error. Resolved it with following advice applied to pyQt5: SyntaxError when using cx_freeze on PyQt app
Checked tutorial for cx_freeze: http://www.pythonschool.net/cxfreeze_mac/ and created a setup.py by the example:
application_title = "simple_PyQt4" #what you want to application to be called
main_python_file = "main.py" #the name of the python file you use to run the program
import sys
from cx_Freeze import setup, Executable
base = None
if sys.platform == "win32":
base = "Win32GUI"
includes = ["atexit","re"]
setup(
name = application_title,
version = "0.1",
description = "Sample cx_Freeze PyQt4 script",
options = {"build_exe" : {"includes" : includes }},
executables = [Executable(main_python_file, base = base)])
Run the cx_freeze with "bdist_mac" parameter. This time .app file was generated. However, it doesn't launch. Can't see any errors or information anywhere. Just doesn't launch at all when clicked.
At this point I'm getting really tired of all of this so please help. If you can, please help me either:
1. Install PyQt5 to Python2.7 with Homebrew (tried, couldn't figure it out) so that I can use packaking tools compatible with Python2.
2. Get the freezing process to work properly. If possible give detailed explanations, I'm new to these tools.
I'm also interested in hearing how people are supposed to deploy standalone applications with pyQt5 since the process seems anything other than straightforward.
EDIT: I ran the result "main" in folder as suggested by ThomasK. I got following error but have no idea what it means:
Traceback (most recent call last):
File "/usr/local/Cellar/python3/3.3.4/Frameworks/Python.framework/Versions/3.3/lib/python3.3/site-packages/cx_Freeze-4.3.2-py3.3-macosx-10.9-x86_64.egg/cx_Freeze/initscripts/Console3.py", line 27, in <module>
exec(code, m.__dict__)
File "main.py", line 8, in <module>
File "/usr/local/Cellar/python3/3.3.4/Frameworks/Python.framework/Versions/3.3/lib/python3.3/importlib/_bootstrap.py", line 1565, in _find_and_load
return _find_and_load_unlocked(name, import_)
File "/usr/local/Cellar/python3/3.3.4/Frameworks/Python.framework/Versions/3.3/lib/python3.3/importlib/_bootstrap.py", line 1512, in _find_and_load_unlocked
_call_with_frames_removed(import_, parent)
File "/usr/local/Cellar/python3/3.3.4/Frameworks/Python.framework/Versions/3.3/lib/python3.3/importlib/_bootstrap.py", line 313, in _call_with_frames_removed
return f(*args, **kwds)
File "/usr/local/Cellar/python3/3.3.4/Frameworks/Python.framework/Versions/3.3/lib/python3.3/importlib/_bootstrap.py", line 1565, in _find_and_load
return _find_and_load_unlocked(name, import_)
File "/usr/local/Cellar/python3/3.3.4/Frameworks/Python.framework/Versions/3.3/lib/python3.3/importlib/_bootstrap.py", line 1529, in _find_and_load_unlocked
raise exc
ImportError: No module named 'ui'
On main.py I have this import on line 8:
from ui.main import Ui_MainWindow
I have all the .ui files and their .py counterparts at folder named "ui" in my project directory.
EDIT: Output of the freezing process: http://pastebin.com/RR9pNGfR
EDIT: And my project schema: http://pastebin.com/HmsdNXEb
I don't know how far the original poster ever got in solving their problem, and the pastebins referenced are long since deleted so there's not a lot of supplementary material, but the latest version of those cx_freeze instructions at pythonschool, updated on 2014-08-27 (after this question), says:
Unfortunately the current version of cx_Freeze on Mac OS X does not play particularly nicely with Python versions that have been installed from Python.org or PyQt if you have installed it previously using our instructions. Therefore, to ensure that cx_Freeze installs successfully you will need to install Python and PyQt using a package manager called MacPorts.
So, that may have been the problem all along. Installing Python3 and PyQt "via Homebrew" sounds like it may have also resulted in a problematic configuration.
This has been asked a couple of times but the answers do not make any sense at least to me. I have Ubuntu 12.04 and had to upgrade Python from default (2.7.3?) to 2.7.5 to get libraries I can use to compile modwsgi.
I obtained 2.7.5 sources, configured with --enable-shared, made and installed. A few obsolete sun audio etc libraries could not be built but otherwise the installation went through. Except that it doesn't work:
visti#ubuntu:~$ python
Traceback (most recent call last):
File "/usr/local/lib/python2.7/site.py", line 548, in <module>
main()
File "/usr/local/lib/python2.7/site.py", line 530, in main
known_paths = addusersitepackages(known_paths)
File "/usr/local/lib/python2.7/site.py", line 266, in addusersitepackages
user_site = getusersitepackages()
File "/usr/local/lib/python2.7/site.py", line 241, in getusersitepackages
user_base = getuserbase() # this will also set USER_BASE
File "/usr/local/lib/python2.7/site.py", line 231, in getuserbase
USER_BASE = get_config_var('userbase')
File "/usr/local/lib/python2.7/sysconfig.py", line 516, in get_config_var
return get_config_vars().get(name)
File "/usr/local/lib/python2.7/sysconfig.py", line 449, in get_config_vars
import re
File "/usr/local/lib/python2.7/re.py", line 105, in <module>
import sre_compile
File "/usr/local/lib/python2.7/sre_compile.py", line 14, in <module>
import sre_parse
File "/usr/local/lib/python2.7/sre_parse.py", line 17, in <module>
from sre_constants import *
File "/usr/local/lib/python2.7/sre_constants.py", line 18, in <module>
from _sre import MAXREPEAT
ImportError: cannot import name MAXREPEAT
The suggested solution here ubuntu ImportError: cannot import name MAXREPEAT was to reconfigure "virtualenv", whatever it is. /usr/local/bin has an ancient script named virtualenv but it tries to run python as well and doesn't work.
Someone suggested the problem might be related to using old Python modules as MAXREPEAT appeared in 2.7.4. To counter this I removed everything related to Python with rm -rf (/usr/bin/python*, /usr/lib/python*, /usr/local/lib/python*, /usr/local/bin/python*, did make clean, make, make install to 2.7.5.
Still the same result. Now I don't have a working version of Python anymore. Is there any way to fix this or should I try reinstalling Ubuntu Python packages if they have 2.7.5 available?
Hannu
Just battled the same problem. Ended up deleting the entire /usr/local/lib/python2.7 install path just to get back to the previous working version.
Also be sure to remove /usr/local/lib/python2.7-config, then reinstall python. Worked for me anyway.
This usually happens when you mix and match libpython.so versions with python interpreter versions. If nuking your install paths didnt help, check these environment variables.
$PYTHONHOME $PYTHONINCLUDE $PYTHONLIB $PYTHONLIBSTATIC $PYTHONPATH
Make sure what they point to were all deleted as well. And then reinstall.
You can also try and install python to a home directory, by configuring the -prefix variable to point to a home dir.
make configure -prefix=$HOME/mypython`
Then make sure you set those above variables to the new home directory python resides in.
The way I fixed this was by searching my system for other instances of the sre_constants.py file. Then see if the location of those files is referenced in your user or system $PATH. If so, remove that location configuration from your $PATH. Rerun python to see if that fixed it.
In my case, it seemed that Python is referencing the wrong location which does not does not support MAXREPEAT.
I have a python application primarily dependent on matplotlib, tkinter, numpy and scipy I have been distributing for multiple operating systems. I build these on top of the python.org version of python 2.6. For mac, I have had good success in the past with py2app and pyinstaller to build standalone versions of my application. Recently, however, I have included functions that require patsy, pandas and igraph and have had to move to the mac distributed version of python 2.6 in order to get igraph to work. All works fine in building a py2app version of my software on the Mac for my OS (10.6) but when I run on 10.8 I encountered the following error:
Traceback (most recent call last):
File "UI.pyc", line 603 createHeatMap
File "clustering.pyc", line 31, in <module>
File "mpl_toolkits/__init__.pyc", line 2, in <module>
File "/System/Library/Frameworks/Python.framework/Versions/2.6/Extras/lib/python/pkg_resources.py", line 698, in <module>
File "/System/Library/Frameworks/Python.framework/Versions/2.6/Extras/lib/python/pkg_resources.py", line 701, in Environment
File "/System/Library/Frameworks/Python.framework/Versions/2.6/Extras/lib/python/pkg_resources.py", line 96, in get_supported_platform
File "/System/Library/Frameworks/Python.framework/Versions/2.6/Extras/lib/python/pkg_resources.py", line 222, in get_build_platform
File "/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/distutils/util.py", line 98, in get_platform
File "/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/distutils/sysconfig.py", line 537, in get_config_vars
File "/System/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/distutils/sysconfig.py", line 405, in _init_posix
DistutilsPlatformError: invalid Python installation: unable to open /user/include/python2.6/pyconfig.h (No such file or directory)
This is fixed by saving the file pyconfig.h to the designated location on the destination machine (not present in 10.8 but present in 10.6), which allows my application to work fine. In my py2app pacakge directory, under "Contents/Resources" there is a sym-link to the directory "include" which points to: /System/Library/Frameworks/Python.framework/Versions/2.6/include/python2.6/pyconfig.h. If I include the pyconfig.h file as an addition resource in my py2app setup file, it also includes it in the package, but it still is not recognized.
Can anyone suggest a good work around for this that doesn't require me to use the python.org version of python (I've had horrible luck getting igraph to work with this).
Thanks...
This is a known issue in py2app, see https://bitbucket.org/ronaldoussoren/py2app/issue/36/py2app-fails-to-build-a-working-executable. The pyconfig.h file is only present on machines where the developer tools are installed, which is why I haven't noticed the problem myself.
The easiest workaround is probably to replace the pyconfig.h symlink by a copy of the file after creating the app bundle.