Welcome!
This is the community forum for my apps Pythonista and Editorial.
For individual support questions, you can also send an email. If you have a very short question or just want to say hello — I'm @olemoritz on Twitter.
stash failed to install scipy, pillow, imreg_dft etc.
-
For my project, I need to use scipy, imreg_dft, imageio, numpy, etc.
But when using "pip install scipy"
Package scipy is blocklisted and marked fatal. Failing install.
Reason: This package uses C-Code, which can not be compiled by StaSh.
Error: Package 'scipy' blocklisted. Reason: This package uses C-Code, which can not be compiled by StaSh.But when using "pip install pillow"
Extracting archive file ...
Archive extracted.
Running setup file ...
TypeError("unsupported operand type(s) for +: 'OmniClass' and 'list'",)
Failed to run setup.py
Fall back to directory guessing ...
Error: Cannot locate packages. Manual installationWhat am I going to do with this?
-
In Pythonista, you cannot pip install a new version of a library over an old version that is already builtin.
Pythonista already has an old version of Pillow builtin...
>>> import PIL >>> PIL.VERSION '1.1.7' >>> PIL.PILLOW_VERSION '2.9.0'
but the current Pillow version is 8.3.2.
Pythonista already has an old version of Numpy builtin...
>>> import numpy >>> numpy.__version__ '1.8.0'
but the current Numpy version is 1.21.2.
Scipy contains a lot of C code (and perhaps Fortran!) so, that is not gonna work with a simple pip install.
You probably want https://pyto.app rather than Pythonista because it includes more of this stuff out-of-the-box and more current versions. It is open source so you can build your own special flavors.
-
The error message seems pretty clear:
But when using "pip install scipy"
Package scipy is blocklisted and marked fatal. Failing install.
Reason: This package uses C-Code, which can not be compiled by StaSh.C code cannot be compiled or installed on iOS. This is not a pythonista thing, it is not possible to run unsigned executables on iOS.
Numpy comes preinstalled (an old but functional version), so does pillow. Scipy cannot be installed. Pandas cannot be installed. Only pure python modules may be installed.
-
It seems that Pyto works with varies module/plugins.
But I like Pythonista better. Does it mean I have to wait for developer's solution? -
There has not been a Pythonista update in quite a while (Feb 26, 2020) so it might be a long wait.