Timeline for The applicability of functional core - imperative shell to a cli program which contains a wrapper around a binary
Current License: CC BY-SA 4.0
11 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 7 at 8:05 | audit | Suggested edits | |||
| Aug 7 at 10:03 | |||||
| Aug 5 at 21:21 | vote | accept | sentientbottleofwine | ||
| Aug 5 at 8:51 | answer | added | Doc Brown | timeline score: 5 | |
| Aug 4 at 17:03 | answer | added | Basilevs | timeline score: -1 | |
| Aug 4 at 16:56 | comment | added | Basilevs | @amon do you have some specific business in mind? Because most of my projects were "stateful" and extracting a functional core did nothing for tests - yes, we could test it, but at cost of exposed internals and imperative parts always require more testing anyway. | |
| Aug 3 at 19:44 | answer | added | candied_orange | timeline score: 0 | |
| Aug 3 at 14:23 | comment | added | amon | Functional Core is great for writing testable business logic. But if there is no logic to test, this doesn't help. Seems your code is mostly a thin wrapper around an external program. Bugs would mostly lie in that interaction, not in your logic. (There are techniques like sans-io that can be used to convert protocol implementations like for HTTP into an IO-agnostic testable structure, which might also work for a CLI wrapper, but is probably more effort than it's worth). | |
| Aug 3 at 9:33 | review | Close votes | |||
| Aug 8 at 3:06 | |||||
| Aug 3 at 5:23 | comment | added | Doc Brown | I don't know tmux, but from what I read about it, I guess it is highly stateful.A wrapper around a given stateful component does not look like a good candidate for the FC-IS approach for me.. | |
| Aug 2 at 21:02 | history | edited | sentientbottleofwine | CC BY-SA 4.0 |
grammar errors
|
| Aug 2 at 20:37 | history | asked | sentientbottleofwine | CC BY-SA 4.0 |