fundamental design issues with trips and dives

[Dirk Hohndel]

> On Nov 7, 2019, at 10:03 AM, Linus Torvalds <torvalds at linux-foundation.org> wrote:
> 
> On Thu, Nov 7, 2019 at 9:46 AM Dirk Hohndel <dirk at hohndel.org> wrote:
>> 
>> I like the thinking here. Radically simple. It does, however, breaks a lot of assumptions in our code, but I'm willing to deal with that in favor of something that's logical and consistent.
> 
> Hmm. It shouldn't break any assumptions in any of the original code,
> because that was very much the original design of a trip. A trip is
> just a collection of dives.

It breaks a LOT of assumptions in the mobile code I should have said.

> 
> Of course, the fact that this all used to work fine doesn't mean that
> it still does.
> 
> [ Goes off and tests ]
> 
> Sure it does.
> 
> At least on desktop. You can happily do this:

Yes. I haven't looked at that code for the desktop in a while.
On mobile this does very odd and non-sensical things. Which I'll need to fix as part of that N+1-st rewrite.

> 
> So the whole "trip is just a collection of dives independently of
> dates" is important, and has actually been the design all along. You
> seem to just not have been aware of that  ;)

I was vaguely aware of it. I was also aware of the fact that the mobile code made these assumptions. So I wanted to put it out there to see what people thought.

/D