On 02/07/2018 03:17 PM, Grant Edwards wrote:
On 2018-02-07, Rob Gaddi <rgaddi@highlandtechnology.invalid> wrote:

When I'm working on a module, the trick is to write a setup.py (using
setuptools) from the very get-go.  Before I write a single line of code,
I've got a setup.py and the directory framework.

Then you install the package using pip -e (or in practice --user -e).
That's the missing piece.  That way you can import your module from the
interpreter, because it's now on the path, but its physical location is
right there where you left it, complete with your VCS metadata and etc.

How do you work on a package that must remain installed and usable the
whole time you're working on it?

IOW, only specific test apps or apps run in a specific directory
should get the "in-progress" foo module when they do an "import foo".


Achievable with a virtualenv, but now the process is even that much more complicated.

--
Rob Gaddi, Highland Technology -- www.highlandtechnology.com
Email address domain is currently out of order.  See above to fix.
--
https://mail.python.org/mailman/listinfo/python-list

Reply via email to