fix: add strict flag validation and version-requested semantics#55
Conversation
|
I saw this today but I haven't been able to get to it. I will try again early next week. Thanks for taking the time to think about this. |
|
Would you be ok getting on a video call to review this. Changes like this can break things and I want to walk through the code. The initial idea was to return any flags we didn't process for config so people could use this one package for everything flag related. You wanting to fail if something is not matching is a big change in behavior. The only behavior needs to be maintained. I rather review this with you one-on-one so we can make sure nothing breaks. |
|
Hi @ardan-bkennedy, Apologies for the delay; I’ve been away on holiday for the past two weeks. I would love to hop on a call to align on this. I am generally available after 9:00 PM UTC. Regarding the technical details, I’ve re-evaluated the impact on backward compatibility. Here is the breakdown:
Essentially, the compatibility risk is localized to the specific error type handling. Let me know if that timeframe works for a quick sync! |
Fixes #54