Some subsurface notes from a week of diving

Dirk Hohndel dirk at
Sun Mar 16 13:57:00 PDT 2014

On Mar 16, 2014, at 1:45 PM, Anton Lundin <glance at> wrote:

> I just came home from a nice week of diving in the Red Sea, and here is
> some sort of braindump about the issues i found in subsurface during
> this week.
> Crashes when trying to add a Gas Change event after selecting another dc
> with the left arrow.
> #0  0x000000000042c048 in add_event (dc=0x0, time=1124, type=25, flags=0, value=25, name=0x52eda8 "gaschange") at dive.c:41
> #1  0x000000000043d5ff in add_gas_switch_event (dive=0x6554b60, dc=0x0, seconds=1124, idx=0) at parse-xml.c:780
> #2  0x00000000004e8ddf in ProfileWidget2::changeGas (this=0xd52930) at qt-ui/profile/profilewidget2.cpp:777
> #3  0x0000000000504068 in ProfileWidget2::qt_static_metacall (_o=0xd52930, _c=QMetaObject::InvokeMetaMethod, _id=3, _a=0x7fffffffc880) at .moc/moc_profilewidget2.cpp:62
> The real problem is that dataModel->dcNr ends up going negative when
> flipping divecomputers with the left arrow key, and then dc becomes

Oops. We used to have code that fixes that. Wonder where that got lost. We basically implemented a somewhat simpleminded wrap-around feature… should be trivial to add back.

> The equipment boxes doesn't save if you don't hit enter before save.

Yeah, keyboard entry in the equipment boxes is still buggered. We went a few rounds on this and it’s not quite as broken as it used to be, but it’s still not the way I want it.

> When switching to oxygen(EAN100), the EAD and END values in the info box
> goes -2147483m. Unsigned zero as signed?

No idea. But a cute effect. Should also be easy to track down.

Dear GSoC students… the first and third problem here would be a great opportunity to show your understanding of our sources, our process, and your eagerness to contribute.

Let’s point out once again that GSoC is first and foremost about you learning to successfully participate in an open source projects - so students who show interest, enthusiasm and willingness to help out are most likely to get chosen.

I have no problem accepting zero students or one (I have seen really good work from one of your). But I’m also willing to accept four or five if we have enough promising looking candidates.


More information about the subsurface mailing list