Hi,
Did you know the R function install.packages() or its cousin
devtools::install_github()?
The devtools::install_github() is a function provided by the RStudio
environment. RStudio in this context would be akin to PyMOL, a software for
final users in general.
install.packages() is provided by R itself.
Both are capable to install custom libraries directly into the user's
environment.
If PyMOL had some function like that any developer could share its plugins
without the burden that is right now.
Would be a neat function and not very tricky either.
Maybe install_package(CONDA, "scikit-learn") or install_package(PIP,
"requests"). And then run these after import pymol, but before the
remaining imports.
Or maybe some special variable like __REQUIREMENTS__ = {"conda": "lxml"}
would be possible also. Of course would not be possible to run the file
before evaluating the variable, the ast module would work nice for this.
I never saw nothing like that for any software except R, RStudio or Golang,
but would be a great addition for PyMOL, and a very easy one. Just import
the conda and pip API and fetch the packages. But in my opinion it is very
clever.
The commercial version of PyMOL is already shipped with its own version of
Python (and conda I guess). But would not be that messy for user local pip
packages, after all is for this kind of reason to encourage user local
packages instead of system wide.
What do you think?
Best wishes,
Have a great weekend,
Pedro Lacerda
_______________________________________________
PyMOL-users mailing list
Archives: http://www.mail-archive.com/pymol-users@lists.sourceforge.net
Unsubscribe:
https://sourceforge.net/projects/pymol/lists/pymol-users/unsubscribe
