Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

What was breaking in 0.16 and 0.17 ? #348

Closed
lrnv opened this issue Feb 25, 2024 · 7 comments
Closed

What was breaking in 0.16 and 0.17 ? #348

lrnv opened this issue Feb 25, 2024 · 7 comments

Comments

@lrnv
Copy link

lrnv commented Feb 25, 2024

Hey,

I have trouble understanding why the PR merged in 0.16 and 0.17 were breaking ?

  • If they were not breaking, these releases should have been 0.15.5 and 0.15.6
  • If they broke something, I'd like to know what to check that my package test suite (depending on TaylorSeries.jl) covers the change correctly.

Could you clarify the matter for me please ?

@lbenet
Copy link
Member

lbenet commented Feb 25, 2024

Thanks for reaching out. #347 introduced a new method zero! which is widely used now, and #333 a bunch of methods for Taylor1{TaylorN{T}} mixtures, aimed to improve performance; you are right that they are not (too) breaking. Also, #346 fixed a Taylor1 constructor bug for mixtures. All that, and still being in major version 0, I think are sufficient for tagging a new minor version. Sorry if that caused troubles in your packages...

@lrnv
Copy link
Author

lrnv commented Feb 25, 2024

There is no troubles in my packages, it is just that i wanted to ensure that the usage I had of TaylorSeries.jl did not broke in a way that i would not test for.

So if you tell me that nothing was really breaking, this simplifies the matter for me and i'll merge the compat bounds. Thanks a lot for the explanation !

@lbenet
Copy link
Member

lbenet commented Feb 25, 2024

I suggest to have a look on #346, because it fixed a constructor method. Aside from that, I think everything should be fine.

EDIT: Corrected the PR reference

@lrnv
Copy link
Author

lrnv commented Feb 25, 2024

You meant #346 I think.
I have checked out and this constructor change does not affect my usage. This is Ok for me, you may close this (or keep it open for further reference? as you want.)

@PerezHz
Copy link
Contributor

PerezHz commented Feb 25, 2024

Maybe worth mentioning: changes introduced by version 0.17 were breaking enough for TaylorIntegration, where we make heavy use of mutating functions and operations on "mixture"-types (e.g., Taylor1{TaylorN{Float64}}), and #347 in particular broke some tests where such features are used. But if you're not working with such mixtures, then probably these changes were not breaking for your packages.

@lbenet
Copy link
Member

lbenet commented Feb 29, 2024

If it is ok with you @lrnv, I'll close this issue...

@lrnv
Copy link
Author

lrnv commented Feb 29, 2024

Yes please close. Ty for the details !

@lbenet lbenet closed this as completed Feb 29, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants