The hardcoded widths with fixed values was just flat our broken and
created all kinds of strange artifacts.
But even with this somewhat more idiomatic approach this still isn't
perfect - on wider screens it doesn't shrink back to the minimum column
width. But that seems like a bit of a corner case. Overall this feels
like a huge improvement.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The number of possible sequences of events when the user is connecting a
USB device is rather surprising. These changes try to ensure that in
every case we get USB connection information we do in fact show the
correct (or best guess) connection.
See #2686
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Adding dives uses the number of the last dive to create a new
dive number. Ignore invalid dives.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The user had to restart the application or manually change the filter
if they changed the flag.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Mark invalid dives in the dive list by striking them out
and rendering them with a grey color. The color-change is
not sufficient, because the default model delegate ignores
color hints if the item is selected.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The location information model is updated automatically by the
divelist-model and the undo commands. Therefore remove the
QMLManager::updateSiteList() function. We do have to keep
the locationListChanged() signal though, because the list
of dive sites is not exported via model/view but rather via
a Q_PROPERTY. We really should change that.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
When the dive list is cleared or updated, the entries in the
divesite-model become stale and therefore the divesite-model
(with the actual name LocationInformationModel) also must be
updated. This was done manually in some parts of the code and
forgotten in others. Therefore, do it directly in the clear()
and reset() function of the dive list-model.
This might be a bit of a layering violation: why should one
model call into another if they are not in parent/child
relationship? However, this seems easier than introducing
a global "reset dives" function that coordinates the models.
Moreover, it does not appear 100% safe: if the clearing of
the divesite model causes accesses to the divelist-model,
they happen in the midst of a model reset and we had horrible
bugs with that kind of things. However, I don't think that
should happen.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The setData() function was used to edit the number of a dive.
However, that doesn't appear to be functional. Therefore, remove
the code. There is a context-menu entry "renumber dives" for that.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
In edit commands, the fulltext might have changed and therefore
we have to update the fulltext index.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The git parser loads into the global dive table, even if it
is called indirectly via parse_file(). However, parse_file()
may be given a different table. Fix this by extending the
git parser state.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
While I remember some of the thinking that went into doing things this way,
the more I read the code, the less it makes sense to me.
This is a rather drastic step, but in reasonably extensive testing it seems
to work in every case that I tried.
That's rather embarrassing, actually.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When editing a dive on mobile we might have to create a new
dive site. That site is added to the global dive site table
in the undo command. However, the code in QMLManager created
the dive site with create_dive_site*() functions, which already
adds it to the table. The undo command then added the dive
site again leading to a hang of the application.
To solve this problem, create new alloc_dive_site*()
functions that do the same as create_dive_site*()
but do not add it to the table.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
This seems to make more sense than to not select anything (which was the
previous behavior).
If the same connection is still available, select it again. Otherwise
pick the top connection (which is the right thing to do if the user has
just plugged in a USB device and refreshes the list -- that device will
show up as top entry).
Based on a suggestion by Christof Arnosti <charno@charno.ch>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This makes sense because USB devices will only show up when connected,
while BT/BLE will get listed once they are paired, even if not currently
connected. So the higher likelihood of being the device a user is looking
for is for any USB devices that show up, so those should go first.
Suggested-by: Christof Arnosti <charno@charno.ch>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The pop_cstring() function is used by the git parser to
duplicate a quoted string. On error, it returns an empty
string literal. Since the caller expects a copied string
and takes ownership of that string, it will ultimately
be freed.
Concrete example: a log with erroneous cylinder data was opened
getting such an empty string literal as description. On closing or
syncing with the cloud, the dive is freed, leading to a free
of the string literal -> crash.
Return a copy of the empty string instead.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
This needs more work and especially new screen shots, but for now this
covers at least some of the most important changes.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Instead of relying on the std::vector staying unchanged and not freeing
its members, instead keep a copy of the object in our DCDeviceData class.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
If the user tries to download from a device that he hasn't given the app
permission to read from, Android will pop up a dialogue asking for that
permission. With this after giving the permission we continue (well,
technically, restart) the download which is likely the expected behavior.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The usbRescan() code is smart enough not to duplicate entries that might
already be there. And with this the user doesn't have to manually tap
rescan if they didn't open Subsurface-mobile via the intent after a
device was plugged in.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This way even if a USB device wasn't added through an Android intent, we
still have a way to scan for it and select it. This is especially
important in case a user has a cable that we haven't seen yet (i.e. with
a VID/PID that we haven't added to Subsurface-mobile), but that
nevertheless works with the android usb serial drivers.
This also makes the flow a little more logical / consistent when
deciding which connections to show.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This finally allows us to download from not just the first device, but specifically
the device that the user picks.
Passing the object through a void pointer is not nice - but since this traverses
C code other solutions (like passing an index into the list) seemed even worse.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
If the user hadn't granted USB permissions, yet, we asynchronously get informed
once they did that. This ensures that the user gets taken back to the download
page once they approve.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This vastly simplifies our handling of devive information as we simply use
what is already in the descriptor. This way we do not duplicate information
about USB devices in the QMLManager.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
