search

FasterXML / jackson-module-kotlin

Module that adds support for serialization/deserialization of Kotlin (http://kotlinlang.org) classes and data classes.
Apache License 2.0
1.12k stars 175 forks source link

Added note about #670 #672

Closed k163377 closed 1 year ago

k163377 commented 1 year ago

Added note regarding disruptive changes reported in #670.

@cowtowncoder Is this an appropriate way to write annotations in such cases? Please let me know if it should be reflected outside of the release notes.

cowtowncoder commented 1 year ago

@k163377 Yes, this looks good to me.

In addition, it'd be good to add something on 2.15 Wiki release notes here:

https://github.com/FasterXML/jackson/wiki/Jackson-Release-2.15#changes-compatibility

(in whichever section makes most sense)

I try to keep these updated based on findings, things users report; so even if not initially included on actual release it should have complete set of information we have.

k163377 commented 1 year ago

@cowtowncoder Thank you very much. I probably do not have edit permission to your page, so could you please grant it to me?

cowtowncoder commented 1 year ago

@cowtowncoder Thank you very much. I probably do not have edit permission to your page, so could you please grant it to me?

I sent you an invite, that should give access to this wiki and jackson-future-ideas repo wiki too.

k163377 commented 1 year ago

@cowtowncoder Thank you, I have added it. https://github.com/FasterXML/jackson/wiki/Jackson-Release-2.15#kotlin

pjfanning commented 1 year ago

Can I suggest this instead:

jackson-module-kotlin changes the serialization result of getter-like functions starting with 'is'. For example, a function defined as fun isValid(): Boolean, which was previously output with the name valid, is now output with the name isValid (#670).

The term 'regression' implies that we will change the behaviour back to what it was before - that we are accepting it as a bug that needs to be fixed. Documenting it in the release notes then implies the opposite - that we are keeping the change. So, I think we should not use the word 'regression' here.

k163377 commented 1 year ago

@pjfanning Thank you for pointing this out.

I am very sorry, but would you be able to help me with PR? I have to rely on machine translation to read and write English, and I am not confident that I can express the nuances of such a case well.

pjfanning commented 1 year ago

@k163377 I modified the wiki. If anyone feels like my changes are inaccurate, feel free to further edit them.

k163377 commented 1 year ago

@pjfanning Thank you very much. It looks very good.

I have submitted a PR regarding the release note.

676

cowtowncoder commented 1 year ago

Thank you @k163377 and @pjfanning ! Good point on semantics of regression: using optimal term helps avoid misunderstanding by users.