On 06-05-14 18:13, Linus Torvalds wrote:
On Tue, May 6, 2014 at 6:21 AM, Dirk Hohndel dirk@hohndel.org wrote:
Maybe you are right - we need to rip out the libdivecomputer event model and create our own and simply populate it from the data libdivecomputer provides us. That way we can deal with some of the other annoyances that we have inherited from libdivecomputer as well
I'm the last to think that the libdivecomputer event model is good, but I think translating it to some saner internal format is likely even worse. Because then you have the "new version of libdivecomputer, new event number, now we need to add more translation, and in the meantime we lose data".
How is that different from what we have today? If libdivecomputer gets some new event today, then subsurface will not be able to use it, until some code is added to handle it properly. How else will it know the interpretation for the new value? Without the correct interpretation, it's just some meaningless number.
I don't see much difference between adding an extra translation or some extra code to handle the new event directly.
(The exception here is when you just store the new number as-is, and then an more recent subsurface version, which does know about the new event, will be able to interpret it correctly afterwards. But, correct me if I'm wrong, I don't think subsurface does that right now. Subsurface stores the event name in the xml and not the event number, if I remember correctly?)
Jef