Using pipista
-
last edited bySo im hearing a lot about pipista and its ability to install other modules. I want to install nltk or mathplotlib with pipista. Van someone tell me where to get pipista, and how to put it onto pythonista and how to run it to get libraries from the pypi library
-
last edited byOk so i got it to work for numpy, but when the code snippet executes it says saved to: numpy-1.8.0.tar.gz where can i find that and do i have to extract it or anything?
-
last edited byAll i see when i run this code snippet is a folder called pypi-modules that contains one file called xmlrpclib
-
last edited byUse Shellista to gunzip and the untar numpy-1.8.tar.gz.
-
last edited byInstalling numpy this way won't work, you can only add pure Python modules yourself, numpy is mostly written in C (but will be included with the next update).
-
last edited byIs it possible such that the update will allow us to add non-python modules ourself or is that an extremely involved process?
-
last edited byAnd also how exactly does one use shellista to gunzip or untar a tar.gz file. Can someone show me an example code snippet
-
last edited bySimilar to using pipista. Go to https://gist.github.com/pudquick/4139094 and copy the code for shellista. Paste it into a new script in pythonista. Call it 'shellista' (don't include the .py).
Here is where it is different. You run shellista by tapping the run triangle. It will switch over to the console/interpreter window and show a prompt.
You can then type gunzip('path-to-file.tar.gz') and press return. However I'm not quite clear on the path yet. Anybody?
-
last edited byhmmm. The first time I tried pipista it told me that it download the .gz file, but when I tried shellista it couldn't find the file. But I tried it again and it worked just great. So not sure what happened.
Anyways. Thank you for all this, it is fantastic.
-
last edited byFor future generations who find themselves here.. pipista is no longer supported. Active development is happening on stash https://github.com/ywangd/stash, which includes a pip command that is a little more reliable.
