This is fairly straight forward. What I dislike is the check for the magic
number of "14 indeces". I'm sure there's a better way to tell if more than
one dive is selected...
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This data structure was quite fragile and made 'undo' when editing
rather hard to implement. So instead I decided to turn this into a
QMultiMap which seemed like the ideal data structure for it.
This map holds all the dive computer related data indexed by the model. As
QMultiMap it allows multiple entries per key (model string) and
disambiguates between them with the deviceId.
This commit turned out much larger than I wanted. But I didn't manage to
find a clean way to break it up and make the pieces make sense.
So this brings back the Ok / Cancel button for the dive computer edit
dialog. And it makes those two buttons actually do the right thing (which
is what started this whole process). For this to work we simply copy the
map to a working copy and do all edits on that one - and then copy that
over the 'real' map when we accept the changes.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
After the 3.1 release it is time to shift the focus on the Qt effort - and
the best way to do this is to merge the changes in the Qt branch into
master.
Linus was extremely nice and did a merge for me. I decided to do my own
merge instead (which by accident actually based on a different version of
the Qt branch) and then used his merge to double check what I was doing.
I resolved a few things differently but overall what we did was very much
the same (and I say this with pride since Linus is a professional git
merger)
Here's his merge commit message:
This is a rough and tumble merge of the Qt branch into 'master',
trying to sort out the conflicts as best as I could.
There were two major kinds of conflicts:
- the Makefile changes, in particular the split of the single
Makefile into Rules.mk and Configure.mk, along with the obvious Qt
build changes themselves.
Those changes conflicted with some of the updates done in mainline
wrt "release" targets and some helper macros ($(NAME) etc).
Resolved by largely taking the Qt branch versions, and then editing
in the most obvious parts of the Makefile updates from mainline.
NOTE! The script/get_version shell script was made to just fail
silently on not finding a git repository, which avoided having to
take some particularly ugly Makefile changes.
- Various random updates in mainline to support things like dive tags.
The conflicts were mainly to the gtk GUI parts, which obviously
looked different afterwards. I fixed things up to look like the
newer code, but since the gtk files themselves are actually dead in
the Qt branch, this is largely irrelevant.
NOTE! This does *NOT* introduce the equivalent Qt functionality.
The fields are there in the code now, but there's no Qt UI for the
whole dive tag stuff etc.
This seems to compile for me (although I have to force
"QMAKE=qmake-qt4" on f19), and results in a Linux binary that seems to
work, but it is otherwise largely untested.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Conditional inclusion of libzip, xslt and osm-gps-map just
makes testing more cumbersome, since testers might lack
Subsurface features without knowing.
Signed-off-by: Henrik Brautaset Aronsen <subsurface@henrik.synth.no>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Doing this on Arch Linux with gcc 4.8.0 helped find one real bug.
The rest are simply changes to make static functions externally visible
(as they are kept around to eventually become helpers used by Qt) which
for now avoids the warnings.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
- rip all Gtk code from qt-gui.cpp
- don't compile Gtk specific files
- don't link against Gtk libraries
- don't compile modules we don't use at all (yet)
- use #if USE_GTK_UI on the remaining files to disable Gtk related parts
- disable the non-functional Cochran support while I'm at it
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Functionality is unchanged, except we now have a nice process_dives
function that deals with all the logic and that gets called from
report_dives from the Gtk code.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
For dives where divecomputers haven't provided us with a cns, we
calculate our cns accumulated during that dive based on a simple table.
We also check if we did a dive in the prior 12 ours and grab the cns
from it and calculate how much of that still affects us.
[Dirk Hohndel: a couple of small changes: remove unnecessary check of cns
values in the samples of the first dive computer, changed
the way we determine the 'previous dive' and used the end
time of that previous dive for the decay calculation]
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The Qt ui will need to read the dive_table to populate widgets with
dives. Gtk functionality in init_ui is required to parse the dives.
Split init_ui to allow parsing to proceed and complete before Qt ui
mainwindow constructor is called.
Play with qDebug()'s printf style (Thiago!)
Signed-off-by: Amit Chaudhuri <amit.k.chaudhuri@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is the downside of having to track our own selection state. We now
have a class of dives that potentially isn't known to Gtk as we aren't
adding them to the TreeView. So we need to make sure that their selection
state is consistent by deselecting them (either when they are marked as
invalid OR when we change preferences to no longer show invalid dives).
And because Gtk sends out another set of selection events when clearing
the TreeView (not sure why), we also need to make sure that their
selection state isn't reset to selected by mistake when the divelist is
recreated.
A bit of a pain, but it seems to work now and be consistent.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
For dives where divecomputers haven't provided us with a cns, we
calculate our cns accumulated during that dive based on a simple table.
We also check if we did a dive in the prior 12 ours and grab the cns
from it and calculate how much of that still affects us.
[Dirk Hohndel: a couple of small changes: remove unnecessary check of cns
values in the samples of the first dive computer, changed
the way we determine the 'previous dive' and used the end
time of that previous dive for the decay calculation]
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is simplistic & brute force: any function that touches Gtk related
data structures is moved to divelist-gtk.c, everything else stays in
divelist.c.
Header files have been adjusted so that this still compiles and appears to
work. More thought is needed to truly abstract this out, but this seems to
be a good point to commit this change.
Signed-off-by: Henrik Brautaset Aronsen <subsurface@henrik.synth.no>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is simplistic & brute force: any function that touches Gtk related
data structures is moved to divelist-gtk.c, everything else stays in
divelist.c.
Header files have been adjusted so that this still compiles and appears to
work. More thought is needed to truly abstract this out, but this seems to
be a good point to commit this change.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This patch implements exporting dives from Subsurface to UDDF format.
Events and cylinder info are the most remarkable things still missing
from the export.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
close() should be included from unistd.h, instead of fcntl.h for
better portability.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Gas with index 0 is assumed to be the first gas only if there is no
gaschage event in the first 30 seconds of the dive.
[Dirk Hohndel: minor formatting change in a logical expression]
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Here is a patch that restricts the gases listed in the divelist to those
that are actually used. I seem to have the indentation now under control
but I am not sure about the logic:
1) First gas (with index 0) is always used.
2) If there is a gas switch event, the new gas is also used (determined by
walking the list of dive computers and then the list of events).
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The zip_int64_t type appears to be missing with mingw so it breaks my cross
builds - and this is clearly equivalent.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This has no user interface and hardcodes a testing username / password.
But it can successfully create a DLD file (thanks to Miika and Lubomir)
and then uses libsoup to upload that to the server.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
patch contains:
- replacement for open_memstream()
- storage of the temporary zip file in the OS temporary directory
- replacement usage of mktemp() with g_mkstemp()
patch based on work by Miika Turkia.
[Dirk Hohndel: as suggested by Miika I commented out the deletion of the
DLD file]
Acked-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This generates a .DLD file of selected dives to be uploaded to
divelogs.de. The actual upload functionality along with sensible user
interface is still to be implemented. However, the resulting file from
this patch is tested to work (as far as I can tell) using upload API of
divelogs.de.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The current data file is not the same as the file to which individual
dives were saved.
Reported-by: Jan Schubert <Jan.Schubert@GMX.li>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
A couple of these could clearly cause a crash just like the one fixed by
commit 00865f5a1e1a ("equipment.c: Fix potential buffer overflow in
size_data_funct()").
One would append user input to fixed length buffer without checking.
We were hardcoding the (correct) max path length in macos.c - replaced by
the actual OS constant.
But the vast majority are just extremely generous guesses how long
localized strings could possibly be.
Yes, this commit is likely leaning towards overkill. But we have now been
bitten by buffer overflow crashes twice that were caused by localization,
so I tried to go through all of the code and identify every possible
buffer that could be affected by this.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
So far we only allowed setting the max depth.
This changes the layout of the entry widget and makes our helper function
create_date_time_widget return the hbox in which it positions the time
entry. I plan to reuse this later when allowing to edit the duration as
well as the start date and time under certain circumstances.
This is a small part of a feature request; see #75
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This creates a helper function called "gas_volume()" that takes the
cylinder and a particular pressure, and returns the estimated volume of
the gas at surface pressure, including proper approximation of the
incompressibility of gas.
It very much is an approximation, but it's closer to reality than
assuming a pure ideal gas. See for example compressibility at
http://en.wikipedia.org/wiki/Compressibility_factor
Suggested-by: Jukka Lind <jukka.lind@iki.fi>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The code was written to get the SAC rate correct, but we probably do
want to have the duration and mean depth of the dive always be shown for
the non-surface-time.
So move the code from the sac-rate calculation to the generic dive fixup
part. This makes the dive list and statistics all show the duration as
the under-water duration, which is not necessarily the same as
"difference between beginning and end of dive".
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This assumes that you are not breathing your cylinders while at the
surface, which may or may not be correct, but is usually the right
thing. Regardless, we're better off giving a conservative (higher) SAC
rate estimate for a diver that breathes his cylinder at the surface too
than giving an artificially low one because the diver ended up using his
snorkel and we didn't take that into account.
NOTE! This basically calculates a better duration and average depth than
the ones we end up showing in the dive list. Maybe we should actually
show this "no-surface-time" duration and average depth instead of the
ones we do show?
That's a separate question, though.
Added a test-case for the surface case to the sac-test.xml dives.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
We even documented that we did SAC in bar*l/min, but the "S" in SAC
stands for "Surface". So we should normalize SAC rate to surface
pressure, not one bar.
It's a tiny 1% difference, and doesn't actually matter in practice, but
it's noticeable when you want to explicitly test for SAC-rate by
creating a test-dive that averages exactly 10m. Suddenly you don't get
the round numbers you expect.
[ Side note: 10m is not _exactly_ one extra atmosphere according to our
calculations, but it's darn close in sea water: the standard salinity
of 1.03 kg/l together with the standard acceleration of 9.81m/s^2
gives an additional pressure of 1.01 bar, which is within a fraction
of a percent of one ATM.
Of course, divers have likely chosen that value exactly for the math
to come out that way, since the true average salinity of seawater is
actually slightly lower ]
So here's a few test-dives, along with the SAC rate fixup to make them
look right.
(There's also a one-liner to dive.c that makes the duration come out
right if the last sample has a non-zero depth, and the previous sample
did not: one of my original test-dives did the "average 10m depth" by
starting at 0 and ending at 20m, and dive.c got a tiny bit confused
about that ;)
[ The rationale for me testing our SAC rate calculations in the first
place was that on snorkkeli.net user "Poltsi" reported that our SAC rate
calculations differ from the ones that Suunto DM4 reports. So I wanted
to verify that we did things right.
Note that Poltsi reported differences larger than the difference of
BAR/ATM, so this is not the cause. I'll continue to look at this. ]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Even if these exists for the heap lifespan, we can call:
g_object_unref(dive_list.treemodel);
g_object_unref(dive_list.listmodel);
in divelist.c:dive_list_destroy()
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
There were some small leaks before here, related to gtk_tree_iter_copy(),
but there is another one in select_next_dive():
nextiter = gtk_tree_iter_copy(iter);
This now requires a SJ near the epilog where we do the memory cleanup.
Lets call this similar label consistently "free_iter" between
select_prev_dive and select_next_dive.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Each time subsurface starts there is supposedly no sorting order
(or divelist column) specified by the UI, yet the actual column is '#'
(or dive number column), since its the *only* column which allows
trips to be visible. If the user selects a different sorting order
then he has no idea which column was the one who had the trips visible.
"Where did all those 'trip' things go?"
This can be a bit confusing...
Lets provide indication by calling gtk_tree_view_column_set_sort_indicator().
Also call gtk_tree_view_column_set_sort_order() to specify a descending
order.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When the user changes the dive list sorting order via clicking
on different column titles, using gtk_widget_grab_focus() gives
keyboard focus back to the list itself (not staying on the column titles),
which gives a hint that the list itself has focus index of 0 and is
reset each time the widget receives this type of "initial" focus.
Acked-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
With commit 800b0482f39e ("When switching sort order, scroll the dive
list to the current dive") we introduced an unwanted new behavior. When
changing sort columns we could lose all selections except for the main
"selected_dive". This was caused by the call to gtk_tree_view_set_cursor
on the selected_dive which unselected all other dives.
As a side-effect this also fixed another bug that was introduced by the
same change: shift-cursor-up and -down now works again to select multiple
dives with the keyboard.
But fixing that bug unearthed a different issue. Our code that restored
the selection state of the tree model oddly decided to mark a divetrip as
selected if its first child was selected. The bug above masked that by
immediately unselecting the trip again, but now that this was fixed the
problem was immediately obvious: we would start with both the first trip
and the first dive in that trip selected (well, since we are in reverse
order it's actually the chronologically last trip and last dive...).
Instead we now use the remembered state of the trip to determine whether
it should be expanded or selected.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Instead, just keep track of the expanded state of trips as we get the
gtk callbacks for the state changes (which we need to track anyway for
the selection logic), and automatically restore the state whenever we
re-create the divelist.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The 'remember_tree_state()' thing is meant to remember if a dive trip is
expanded or not, but it missed the "or not" part. IOW, it never cleared
the expanded flag, it only ever set it.
As a result, if you were doing multiple operations on the divelist tree
(testing all the recent gtk-model removal, for example) the dive trips
would end up expanding more and more, even if you collapsed things by
hand in between operations.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This does the moving of dives into the trip above without the complexity
of the gtk data structures, and instead just recreates the whole
divelist afterwards. As usual, this simplifies things a lot, and the
less gtk-specific code we have, the better.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
.. use our own data structures instead, and regenerate the gtk ones
after having successfully created the new trip.
This simplifies the code enormously, and also makes it much more
generic. You can now create a new trip from any arbitrary set of
selected dives (it used to be that the "multiple selected dives" case
worked, but only for some very specific special cases of selected dives).
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Commit bcf1f8c4fe ("Don't do "remove_from_trip" by walking the gtk
data structures") made find_trip_by_idx() only work for negative indexes
(positive indexes are dives), but when it removed the unnecessary test
for negativity, the statement inside it should have been kept as
unconditional, rather than removed with the test.
Oops.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
They are complicated and confusing. Just use our own data structures
and re-generate the gtk ones from them.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
They are complicated and confusing. Just use our own data structures
and re-generate the gtk ones from them.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
