jdg / MBProgressHUD

MBProgressHUD + Customizations
http://www.bukovinski.com/
MIT License
16.01k stars 3.56k forks source link

Created protocol to improve injectability of lib. #462

Closed Panajev closed 7 years ago

Panajev commented 7 years ago

Signed-off-by: Goffredo Marocchi goffredo.marocchi@gamesys.co.uk

Panajev commented 7 years ago

@jdg is this build going to TravisCI?

matej commented 7 years ago

One of the original design goals for MBProgressHUD was to just have it be one implementation file (and a header) for easy integration. While the code has grown a lot over time, I still see this as a positive attribute in this case, so I'd like to preserve that.

Panajev commented 7 years ago

Would it be a folder better for easy integration? Also while the easier integration has a big merit, wouldn't the improved separation in components be preferable? There are merits to breaking down to a single class per file.

Would you agree to merge it more if this were changed to go back to a single file solution?

On Mon, Mar 6, 2017 at 12:18 PM, Matej Bukovinski notifications@github.com wrote:

One of the original design goals for MBProgressHUD was to just have it be one implementation file (and a header) for easy integration. While the code has grown a lot over time, I still see this as a positive attribute in this case, so I'd like to preserve that.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/jdg/MBProgressHUD/pull/462#issuecomment-284381219, or mute the thread https://github.com/notifications/unsubscribe-auth/AAiXLzK3gGw-HRyJTnzdCqDDTZti5dObks5ri_msgaJpZM4MTbR0 .

Panajev commented 7 years ago

Hello @matej would this be close to what you are looking for? I still think that the protocol for ProgressHUD should be a separate header and we should have separate implementation files for the other subcomponents (perhaps all importing the same MBProgressHUD header), but then again with Cocoapod I am not sure it matters if this takes one or multiple files.

I do not want to create an incompatible fork or a fork that cannot stay up to date, so I am happy to refactor it like this if it meets your standards for this library better.

Panajev commented 7 years ago

Hello @jdg, would you need additional changes for the PR?

matej commented 7 years ago

Sorry for the late replay, @Panajev. I think this complicates the structure unnecessary - without much benefit. I would like to keep this a simple one off class.

You can of course keep using it as you desire on your fork, but I'd opt to keep the main repository as is.