Add multi-year domain purchases

[umurkontaci]

This PR enables users to purchase domains for more than 1 year. As this is a big change, we're adding a feature flag to enable it when we're ready.

Screenshot

Testing

You'll need to test this with D341-code.

• Add a domain to the cart
• Go to checkout
• Change the year from the select box
• Buy the domain
• Make sure the domain is registered for the duration you have picked
• Make sure you are charged the correct amount
• If you opted for a privacy protection, make sure you are charged for it correctly as well.
• Cancel the domain you've just registered

[dmsnell]

@umurkontaci looks like we have some merge conflicts - does this need a rebase?

[umurkontaci]

@dmsnell Yup, thanks for the ping. Rebased now

[dmsnell]

It seems more natural to me to use the word quantity to specify how many of a specific product item we want in the cart. Maybe we can have some others chime in. volume sounds a bit odd to me and maybe misleading (though maybe I don't understand this well enough!)

[umurkontaci]

volume is the name that's being used in the WP codebase, hence the usage.

[dmsnell]

thanks for rebasing @umurkontaci - it looks like there are now a few anachronisms after the update and I pointed out one or two, but we should probably make sure the style matches the surrounding code if the old stuff had been updated since this PR was born.

[klimeryk]

You've used spaces here - it will look weird for anyone that has configured tab to be expanded to something else than 2 spaces :)

[umurkontaci]

@dmsnell @klimeryk I've updated and rebased the code. Do you mind taking a look at it?

We're shelving the option to buy multi-year domains for now. But I would rather keep this code merged rather than having to keep track of everything to rebase.

[dmsnell]

@umurkontaci I'm a bit confused - why would we be merging this code if we aren't enabling the feature? Wouldn't it make more sense to leave the code out of the repository until we need it?

[umurkontaci]

No it wouldn't. If we are not merging this code, by the time we are ready to launch, everything in this code would be irrelevant / stale and I have rewrite the whole thing all over again.

[dmsnell]

If we are not merging this code, by the time we are ready to launch, everything in this code would be irrelevant / stale and I have rewrite the whole thing all over again.

That's kind of my point. If we merge this in, it seems like it will sit as dead code if and until we decide to allow multi-year purchases, and by the time that happens, this will all be irrelevant / stale and we will have to redo it all anyway, so why not just let this branch go (or bookmark it) and defer the development until we need it - otherwise we'll just be adding an unnecessary maintenance burden, unless I misunderstand you

[umurkontaci]

It will still be less work to do though. We would have to test the stuff to be working again and make small adjustments, but this piece will be included in code refactoring processes. Refactoring a bunch of functions is easier than coming with an old code and adjusting the whole to the 10 times refactored code.

It would be unnecessary maintenance burden if we were adding something we didn't know we'd need it or not.

This PR has been thoroughly tested in here and in pre-oss. I don't want to throw everyone's work away.

[dmsnell]

@umurkontaci no hard feeling if you end up merging this in, but if I understand you correctly I can't vote to do so - seems like a bad idea to me. I understand not wanting to throw everyone's work away, but that's not a compelling argument to me to put in dead code and I don't think that's the right way to look at this branch. in any case, this code will live on in git

feel free to ping other people for review. I appreciate your work on this.

[stephanethomas]

I agree with @dmsnell, we shouldn't merge this pull request. This will add extra maintenance and introduce additional complexity for something that is not used - without mentioning that it will mislead developers who don't know this part of the code and who will wonder what it does and how it fits in the whole application.