pythonrunscript - Answer.AI dev chat #3

This dev chat is about the tool "pythonrunscript". This tool makes it easier to use python for scripting, by letting you define single-file python scripts which declare their dependencies in comments (using pip or conda). The tool will then run those scripts, auto-installing their dependencies in cached, isolated environments.

0m00s : Preface comment, summarizing python-run-script and changes since the video.
1m30s : Hellos and introductions
2m45s : The problem: finding an "adequate scripting language"
4m20s : Requirements for an adequate scripting language
11m22s : Why existing languages fail as an adequate scripting language.
13m20s : Sources of inspiration: swift-shell, and rust-script
17m50s : How python-run-script smoothes over the problems presented by python's packaging and distribution infrastructure, to make it an adequate scripting language
19m05s. : Discussion of when conda is needed and how to use it, versus using venv
24m00s. : Demonstration of python-run-script (in an early development version)
33m00s. : Demonstration for the case of no dependencies
34m30s. : Demonstration for the case of only pip dependencies
36m00s. : This tool's origin story, in Alexis's need to deploy software to non-developers as a CTO of a manufacturing startup
38m30s. : Feedback and suggestions from Jeremy and Johno.
41m15s. : Discussion of the syntax for dependency block, and the merits of version pinning.
47m00s. : Syntax discussion, on the best delimiter for dependency blocks.
48m20s. : Syntax discussion, on how to specify conda requirements in a standard way.
54m00s. : Final thoughts, on the joy of small projects.

[niazhimselfangels]

Thanks again everybody for putting these out- I listen through these and learn a dozen nuggets about things I never spent time thinking about. It's great to hear the running commentary about design considerations and how to think about implementations. Great stuff!

[felipe]

20:50
Very interesting conversation. For me at least, working with docker has been a more stable and straight forward(albeit slower) way of running a good, isolated python environment.
After I polluted a fedora's workstation environment I just switched everything to containers. Really recomend it.