This adds the QIODevice::Unbuffered flag to our rfcomm connections to
bypass the buffering layer in QIODevice.
This fixes so firmware upgrades work against the OSTC Sport.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
waitForReadyRead and waitForBytesWritten not overridden in
QBluetoothSocket and the default implementation in QIODevice are just
no-ops that always return false.
This removes those calls to lessen the confusion for anyone who looks at
the code.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This shouldn't happen, but in case there is a logic error higher up in the
code somewhere, this will prevent it from happening, period.
If the code asks for a new dive site with a specific uuid, simply return
the existing dive site with that uuid.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Normally the auto-created sites that are used when downloading GPS data
from the webservice that aren't associated with a dive will immediately
discarded. But if such a site is used for a dive but then the user changes
their mind and uses another site for that dive (for example because they
had a previous dive at that spot and decided they liked that one better
after all), then the dive site can stick around as an orphan.
Having orphan dive site may make sense for named dive sites that for some
reason are no longer used but may be used again in the future, but having
an auto-named orphaned dive site in the data file seems silly. So let's
remove them.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
In verbose mode we wanted to tell the user if they had a reference to a
non-existant dive site that we removed for consistency. Except the
condition was wrong and the warning was printed for every VALID dive site
reference. Oops.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
These messages are extremely useful to understand what Subsurface is
doing, but they are a bit too wordy for normal use.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When you have openend cloud storage, the Save and Save to cloud storage
are basically the same thing... so we need to show the progress bar in
that case, too.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Under certain conditions on my Mac the build process will happily find libssh2
but then at link time can't figure out where it is. Making the path to the
library explicit seems to fix that issue.
This commit also adds the code to dump all visible variables at the end of the
cmake file (commented out). I have looked this up online, added it to address
an issue, and then removed it quite a few times... I figure I might as well
keep it there for the next time I need it.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Counterintuitively, setting the minimum duration to 200ms actually makes
the dialog show up while waiting for the first progress signal. I had
thought that setting it to 0 would make it show up right away, but with a
value of 0 it waits for the first progress notification and with slow
internet connections that can take quite a while (and with some git
operations no progress notification will be sent out the whole time).
So this should make the situation with the progress bar a little better.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
pressure
At the beginning of the dive, let the cylinder cool down if the diver
starts off at the surface. And at the end of the dive, there may be
surface pressures where the diver has already turned off the air supply
(especially for computers like the Uemis Zurich that end up saving quite
a bit of samples after the dive has ended).
So just don't use surface samples to populate the beginning/ending
cylinder pressure data.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The VPM-B benchmark results are all based on nominal/zero conservatism, so we
should make sure we use zero conservatism in the tests.
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Nominal (aka 0) conservatism for VPM-B is generally considered aggressive. We
don't want to be aggressive, especially as some users might assume the default
is the correct value. Our default Buhlmann gradient factors are 30/75. The
most similar VPM-B conservatism level is probably +3, at least for dives in the
40-50 metre, 50-70 minute range.
Signed-off-by: Rick Walsh <rickmwalsh@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Since the location is only true with a valid uuid, set the
uuid on it, it will search for a valid name and set there.
this fixes a few inconsistencies handling the locations.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This makes the globe smoother while moving around same gps
dive sites.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Now mouse and keyboard navigation over the list of current
dive sites will update the marble globe position.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When the mouse moves over the dive site list, the globe
should show the current one under the mouse.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Each time we searched for an icon on the qt resource
we had to open, inflate, create, store, delete the icon.
now we search for it only once, use as cache and make
the world a better place.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Display an error message if no new dives are there for download.
[Dirk Hohndel: small adjustment to the message]
Signed-off-by: Guido Lerch <guido.lerch@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Clean up the text on the above issue. Many thanks, Claudiu.
Signed-off-by: Willem Ferguson <willemferguson@zoology.up.ac.za>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
In some cases the rename of the cache directory would fail in my testing.
Based on code that Lubomir provided, this tries a Windows specific
implementation of folder rename if the QDir based one fails for some
reason - but obviously only if we are running on Windows.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Apparently Windows doesn't like it if the proxy string is set to the empty
string.
Also give some better debugging output when run in verbose mode.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Update the user manual with information about the new Bluetooth
download mode feature.
Signed-off-by: Claudiu Olteanu <olteanu.claudiu@ymail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Firmware updates can only be done on a newly opened device.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Use model name to skip some configs which the Sport doesn't handle and
starts misbehaving on.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This looks at the OSTC3 hw descriptor and exposes that model info as a
read-only line edit, so you can see in clear text the name of the model
of computer.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
All the conflicting fields for suunto vyper configs are named _1, so
rename this one to be in the same style.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Put back the procedures for manually achieving a Bluetooth connection
into APPENDIX A.
Refer to the manual procedures from the text that described the new
built-in Subsurface Bluetooth interface.
Signed-off-by: Willem Ferguson <willemferguson@zoology.up.ac.za>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Let's use the same assumption for working pressure (where not defined)
as the origin uses.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
It looks odd if one item is missing an icon. Thus suggesting (yet
another) placeholder icon for the preferences dialog.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
1) Delete section on mobile application
2) Update Preferences and Menu options sections
No images are added or replaced.
Signed-off-by: Willem Ferguson <willemferguson@zoology.up.ac.za>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
If user selected a new dive site (a dive site that doesn't
exists yet, to be created) and clicked in edit it would edit
the *old* dive site. this is not optimal, but since it
removes a severe regression it will be like this (I won't
change it till 5.0) for a while, until I have time to
actually code something not messy for that.
The current dive location management is *much* nicer than
the old one, code-wise. No more magic numbers, magic codepaths,
magic constants, qt black magic. It's straigth-forward code
simple to follow, simple to edit.
This was something I was actually postponing for 5.0 but I
managed to get a boost of subsurface time at work.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When the message to 'This dive site is being edited' was
being show while the popup to choose the dive site, the
line edit was being covered. now it correctly moves to
the correct place.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Unused dead code / hack for the old QCompleter
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
If you select a dive site with a different uuid than your
current dive.dive_site_uuid, you should get a different
pallete to state clearly that something changed.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
