Will Furnass

The problem for me is that shebang lines aren't updated in various scripts in the cloned virtualenv. Following on from my example above: ``` will@banana ~ $ fgrep -RI testenv1...

Think the problem stems from there being symlinks from `.virtualenvs/testenv1/local/{bin,include,lib}` to the `.virtualenvs/testenv1/{bin,include,lib}` that aren't updated when `testenv1` is cloned, and `/home/will/.virtualenvs/testenv2/local/lib/python2.7/site-packages` coming before `/home/will/.virtualenvs/testenv2/lib/python2.7/site-packages` in `sys.path` once `testenv2` is...

NB folks still on EL7 can get Python 3.8 from the `rh-python3.8` RedHat Software Collection ([more info](https://access.redhat.com/documentation/en-us/red_hat_software_collections/3/html/3.8_release_notes/chap-rhscl)).

Hmm, odd that Travis CI jobs now only fail when Python=3.7 and COVERAGE="yes". NB those tests result in multiple occurrences of: ``` ... if self._parameterized: > self.indepvararray.resize(new_len) E ValueError: cannot...

i.e. the use of Sphinx' autodoc

At the same time, someone commented that they liked that the example package contained very little real code as it helped them focus on the package/repo structure. However, I do...

A possible workaround is to disable the post-install script for the `openssh-client` Ubuntu .deb package e.g. ```bash export DEBIAN_FRONTEND=noninteractive apt-get update apt-get install -y `apt-cache depends openssh-client | awk '/Depends:/{print$2}'`...

Also, I can confirm that the error msg shown in the original post can be seen/reproduced if one is trying to build an image using just `fakeroot` on CentOS 7...

@mrocklin some folks at the University of Sheffield use the [ruffus](http://www.ruffus.org.uk/) computational pipeline manager on our HPC clusters, which uses these DRMAA bindings behind the scenes to submit jobs to...

> Repos will be public How will you guard against malicious code execution?