search

anaconda / nb_conda

Conda environment and package access extension from within Jupyter
BSD 3-Clause "New" or "Revised" License
140 stars 32 forks source link

Add conda 4.4+ support (conda not on path errors) #75

Open tjd2002 opened 6 years ago

tjd2002 commented 6 years ago

Uses the $CONDA_EXE environment variable to find conda, rather than assuming it's on the path. Inspired by (and fixes the same problems as: https://github.com/Anaconda-Platform/nb_conda_kernels/pull/79 )

Also adds Travis testing for new- and old-style installs (borrowing from: https://github.com/Anaconda-Platform/nb_conda_kernels/pull/82 and https://github.com/Anaconda-Platform/nb_conda_kernels/pull/83)

Discussion of 4.4+ support across all nb extensions at: https://github.com/Anaconda-Platform/anaconda-nb-extensions/issues/168

tjd2002 commented 6 years ago

@mcg1969 @benbovy could you take a look?

tjd2002 commented 6 years ago

(This also incorporates #74 to get tests passing again)

coveralls commented 6 years ago

Coverage Status

Coverage decreased (-0.2%) to 45.477% when pulling 65cda30fbf96c206e5eb4ebbb882042001e97c43 on tjd2002:conda44-fix into c27196f15528b92c808cff1123a7579ba046d180 on Anaconda-Platform:master.

coveralls commented 6 years ago

Coverage Status

Coverage decreased (-0.2%) to 45.477% when pulling b1f24988c789bb12b900d345ff22e5dc284df04a on tjd2002:conda44-fix into c27196f15528b92c808cff1123a7579ba046d180 on Anaconda-Platform:master.

tjd2002 commented 6 years ago

Added @voidworker's fix from #66, which handles new behavior of conda info --json since 4.4

elgalu commented 6 years ago

Hi @tjd2002 how is your fork doing?

Given this project doesn't seem maintained I was thinking about switching to your branch but maybe you faced new issues and no longer using nb_conda ?

BharatKambalur commented 5 years ago

Has there been any progress merging this in?

tjd2002 commented 5 years ago

I would like to see this merged, but I don't use the conda tab anymore. (I think that Anaconda Navigator is the recommended way to manage conda envs if you need a GUI).

Note that you can still use conda kernels with Jupyter by installing the nb_conda_kernels package.

tjd2002 commented 5 years ago

@mcg1969 Any interest in pushing this forward?

tjd2002 commented 5 years ago

This repo seems to be quite soundly dead but there's an actively-developed fork at : https://github.com/fcollonval/jupyter_conda (Thanks @fcollonval)