Sinon: Should we implement automated releases?

[mroderick]

1656 was caused by me overlooking a manual step when preparing the release.

This has likely happened before, and will likely happen again.

I think we should consider implementing automated releases of some sort.

Ideas

• https://github.com/semantic-release/semantic-release

[lucasfcosta]

Hello @mroderick,

As I want to step back-in now that I've got time to contribute again, here's my 2 cents on this matter.

We've been using semantic-release in chai for quite a while, which has been very helpful especially considering the number of different packages that we maintain which form the main library.

That might not be the case here, but manual releases are still error-prone and time-consuming. If I recall correctly, Chai once broke AirBnB's internal build because of a breaking change released under a non-major bump.

Since in this repo also contains the documentation for sinon I don't think keeping the docs in sync would be a problem.

I'd be happy to have a try at this if everyone thinks it's worth it.

[mroderick]

Hello @mroderick,

As I want to step back-in now that I've got time to contribute again, here's my 2 cents on this matter.

We've been using semantic-release in chai for quite a while, which has been very helpful especially considering the number of different packages that we maintain which form the main library.

That might not be the case here, but manual releases are still error-prone and time-consuming. If I recall correctly, Chai once broke AirBnB's internal build because of a breaking change released under a non-major bump.

Since in this repo also contains the documentation for sinon I don't think keeping the docs in sync would be a problem.

I'd be happy to have a try at this if everyone thinks it's worth it.

That would be great! If it would be easier, feel free to try it out on one of the smaller libraries like @sinonjs/samsam, etc.