| Age | Commit message (Collapse) | Author |
|---|
|
In commit 304526850c91 ("Don't deselect all dives on all selection
"change" events") the handling of "selected_dive" is incorrect. We ended
up with non-sensical values for the selected dive, including dives that
Gtk didn't think were selected.
This commit tries to be smart about what to do when the dive that we
currently consider selected is unselected (we have this weird notion of
many dives being selected, but one of them is shown in the profile and
that is the "selected_dive"). As long as there are others selected, we
pick one of them (first walking to earlier dives and if there are none
that are selected, looking for a later dive) as the new selected dive.
This appears to give us a rather intuitive behavior when playing with
multiple selected dives.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
gtk sends the selection change events all the time, for pretty much any
"divelist changed - so selection changed". The expansion of a trip, the
switch to a new model, yadda yadda. But we actually want selections to
be sticky across these events, so we can't just forget all of our old
selection state and repopulate it.
So we re-introduce the "am I allowed to change this row" callback, which
we used to use to create a list of every actual selection that was
changed. But instead of remembering the list (and having the stale
entries issue with that remembered list that caused problems), we now
just use that as a "that *particular* selection cleared" event.
So this callback works as the "which part of the visible, currently
selected state got cleared" notifier, and handles unselection.
Then, when the selection is over, we use the new model of "let's just
traverse the list of things gtk thinks are selected" and use that to
handle new selections in the visible state that gtk actually tracks
well. So that logic handles the new selections.
This way, dives that aren't visible to gtk don't ever get modified: gtk
won't ask about them being selected or not, and gtk won't track them in
its selection logic, so with this model their state never changes for
us.
gtk selections are annoying. They are simple for the case gtk knows
about (ie they are *visually* selected in the GUI), but since we very
much want to track selection across events that change the visual state,
we need to have this insane "impedance match".
Reported-by: Dirk Hohdnel <dirk@hohndel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
The dive selection rewrite didn't set the selected dive index, breaking
the cursor key logic.
Reported-by: Dirk Hohndel <dirk@hohndel.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
We used to generate a list of possibly changed selections using the gtk
tree selection "selection function".
But that's actually meant to just tell gtk whether an entry can be
selected or not, and our list of possibly changed entries ended up being
stale if the selection change was due to a list entry removal, for
example.
So rip out the old model entirely, and instead just walk the whole
selection that gtk gives us on a selection "change" event. We throw all
our old selections away when this happens, and just rebuild it all.
This should fix the occasional internal gtklib-quartz assertion that
Henrik is seeing. And it actually simplifies the code too.
Reported-by: Henrik Brautaset Aronsen <subsurface@henrik.synth.no>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
Add a sample at time 0 to allow for a pO2 from the start of the dive.
Remember the last pO2 so it doesn't have to be repeated (and the right
thing happens for the planned part of the dive).
This still doesn't allow us to change the setpoint at a certain depth
(which would be analogous to being able to switch to a certain gas at a
certain depth in OC plans), but with this commit it's already usable.
This commit also fixes a couple of small bugs in commit b8ee3de870fa
("Dive planning for closed circuit rebreather") where a pO2 of 1.1 was
hardcoded in one place, throwing off all plan calculations and integer
math was used to calculate a floating point value (leading to most pO2
values actually used being 1.0).
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
This misses a single issue to be used as a base for further discussion:
The CC setpoint is used for the next segment, not the one specified for. I
also have in mind to modify the existing code to use setpoints specified
in mbar and plain integer instead of float values.
Signed-off-by: Jan Schubert <Jan.Schubert@GMX.li>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
I have some concerns about the way this is implemented - especially the
use of gtk_grab_add to make the map widget work has me worried. But it
seems to work and survived some test cases that I threw at it.
The GtkButton with the Pixmap looks a little off on my screen, but this
way it was easy to implement. Feel free to come up with a better design.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
There are paths through this function that reach the comparison at the end
of it without trip_a and/or trip_b being initialized.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
divelist.c:copy_tree_node():
pass the pointer "icon" to gtk_tree_store_set()
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
The 'preexisting' value is used for downloading dives: we want to add
new dives but, but then compare those new dives against the
preexisting ones before we start sorting things and possibly merging
them.
However, the value was only updated sporadically, resulting in it
having stale information in it. Which would cause problems
particularly if you deleted dives, so that the preexisting value would
point past the actual existing values!
So just update it unconditionally in dive_list_update_dives(), which
anything that changes the dive list is supposed to call in order to
display the changes anyway.
Also, just for safety, when removing a dive, put NULL in the last dive
table location. Nobody should ever access past the end anyway (this
is enforced by 'get_dive()') but there are places that access the dive
list table directly, and the libdivecomputer download was one of
those. No reason to leave stale dive pointers possibly around for
uses like that.
Reported-by: Henrik Brautaset Aronsen <subsurface@henrik.synth.no>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
This moves the fields 'duration', 'surfacetime', 'maxdepth',
'meandepth', 'airtemp', 'watertemp', 'salinity' and 'surface_pressure'
to the per-divecomputer data structure. They are filled in by the dive
computer, and normally not edited.
NOTE! All actual *use* of this data was then changed from dive->field to
dive->dc.field programmatically with a shell-script and sed, and the
result then edited for details. So while the XML save and restore code
has been updated, all the displaying etc will currently always just show
the first dive computer entry.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
In icon_click_cb() we need to check if a correct GtkTreePath is found
(using gtk_tree_view_get_path_at_pos()) before requesting a GtkTreeIter
for it.
Without this patch a bug is reproducible, where the user may click
outside of the GtkTreeView entries, but still in the GtkTreeView -
e.g. when only one entry is available.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
This makes the code use the "dive_has_location()" function rather than
check the longitude and latitude directly.
It also uses "for_each_dive()" rather than open-coding it.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
Silly oversight.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
This replaces the really lame "italics text" from commit abe810ca1a29
("Mark locations that have GPS location data attached") with a marginally
less lame GPS icon.There's a reason why I am not making a living as
graphics artist. But I think this is a huge step forward from what we had
before...
The satellite.svg file is very loosely based on a different icon that I
found as public domain here http://www.clker.com/clipart-30400.html.
From that I created the PNG and then that was converted into the
GdkPixdata via gdk-pixbuf-csource; a rule for that was added to
the Makefile but commented out as I don't know if this tool will always be
available in the path. Having this icon included in the sources avoids
locating yet another icon file.
Better icons are certainly welcome!
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
This is rather lame - we simply turn the location text into italics for those
dives where we have GPS location data. Underlining might be more natural, but
Gtk plays games with the underline attribute if the mouse hovers over text.
Ideally I would have prefered a little GPS logo next to the location text - but
I couldn't figure out how to do that without writing my own cell renderer which
seemed total overkill.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
While we are waiting for an autotools generated Makefile, this should allow
people to build that don't have osm-gps-map.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
This adds the "Show in map" menu entry to the divelist only if we
actually have a location to show.
Of course, having some way to visually see whether we have a GPS
location even before we show the menu would probably be good. Maybe a
marker in the "location" string or something. But in the meanwhile, at
least we don't have that menu entry if we have nothing to show.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
This adds a "Show in map" entry in the dive list context menu. It will
zoom to the dive location if it exists, otherwise the full map will be
displayed.
I've also switched map tiles from OpenStreetMap to Google Maps just to
show off that we can.
Signed-off-by: Henrik Brautaset Aronsen <subsurface@henrik.synth.no>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
This patch centralizes the definition for surface pressure, oxygen in
air, (re)defines all such values as plain integers and adapts calculations.
It eliminates 11 (!) occurrences of definitions for surface pressure and
also a few for oxygen in air.
It also rewrites the calculation for EAD, END and EADD using the new
definitons, harmonizing it for OC and CC and fixes a bug for EADD OC
calculation.
And finally it removes the unneeded variable entry_ead in gtk-gui.c.
Jan
Signed-off-by: Jan Schubert <Jan.Schubert@GMX.li>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
Just put the code to do so in a function and reuse that function elsewhere
as well.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
Previously we would simply show the first dive in the divelist - which
worked fine in the default sort by trip setting and assuming that there
are no dives from the future in the divelist.
With this commit we actually find the correct dive in the divelist and
select it instead. If you sort by depth you will see the dive move around
in the divelist, but it will stay selected and visible in the profile.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
The loop would actually get entered for dive 0 and try to compare things
with dive -1. Which of course fails.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
Previously the code in init_decompression() would mindlessly walk back the
dive_table until it found a gap of at least 48h and take all those dives
into consideration when calculating tissue saturation. This goes horribly
wrong if you load dives from two divers into the same data file.
With this commit things will still turn out correctly, as long as the
dives are in separate trips in for each of the divers. So with this I can
load both Linus' and my divelog and things stay sane even on our shared
dive trips.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
.. and add the usual logic to not save the default values.
This also simplifies the initial system-specific setup of both of these:
since we have defaults for all the preferences that get set up at
startup, we can just initialize those defaults to the system-specific
fonts then and there.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
.. and rename the badly named 'output_units/input_units' variables.
We used to have this confusing thing where we had two different units
(input vs output) that *look* like they are mirror images, but in fact
"output_units" was the user units, and "input_units" are the XML parsing
units.
So this renames them to be clearer. "output_units" is now just "units"
(it's the units a user would ever see), and "input_units" is now
"xml_parsing_units" and set by the XML file parsers to reflect the units
of the parsed file.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
We have several places where we interpolate the depth based on two
samples and the time between them. Some of them use floating point, some
of them don't, some of them meant to do it but didn't.
Just use a common helper function for it. I seriously doubt the floating
point here really matters, since doing it in integers is not going to
overflow unless we're interpolating between two samples that are hours
apart at hundreds of meters of depth, but hey, it gives that rounding to
the nearest millimeter. Which I'm sure matters.
Anyway, we can probably just get rid of the rounding and the floating
point math, but it won't really hurt either, so at least do it
consistently.
The interpolation could be for other things than just depth, but we
probably don't have anything else we'd want to interpolate. But make the
function naming generic just in case.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
o) Instead of using gradient factors as means of comparison, I now use
pressure (as in: maximal ambient pressure).
o) tissue_tolerance_calc() now computes the maximal ambient pressure now
respecting gradient factors. For this, it needs to know about the
surface pressure (as refernce for GF_high), thus gets *dive as an
argument. It is called from add_segment() which this also needs *dive
as an additional argument.
o) This implies deco_allowed_depth is now mainly a ambient-pressure to
depth conversion with decorations to avoid negative depth (i.e. no deco
obliation), implementation of quantization (!smooth => multiples of 3m)
and explicit setting of last deco depth (e.g. 6m for O2 deco).
o) gf_low_pressure_this_dive (slight change of name), the max depth in
pressure units is updated in add_segment. I set the minimal value in
buehlmann_config to the equivalent of 20m as otherwise good values of
GF_low add a lot of deco to shallow dives which do not need deep stops
in the first place.
o) The bogus loop is gone as well as actual_gradient_limit() and
gradient_factor_calculation() and large parts of deco_allowed_depth()
although I did not delete the code but put it in comments.
o) The meat is in the formula in lines 147-154 of deco.c. Here is the
rationale:
Without gradient factors, the M-value (i.e the maximal tissue pressure)
at a given depth is given by ambient_pressure / buehlmann_b + a.
According to "Clearing Up The Confusion About "Deep Stops" by Erik C.
Baker (as found via google) the effect of the gradient factors is no
replace this by a reduced affine relation (i.e. another line) such that
at the surface the difference between M-value and ambient pressure is
reduced by a factor GF_high and at the maximal depth by a factor
GF_low.
That is, we are looking for parameters alpha and beta such that
alpha surface + beta = surface + gf_high * (surface/b + a - surface)
and
alpha max_p + beta = max_p + gf_low * (max_p/b + a - max_p)
This can be solved for alpha and beta and then inverted to obtain the
max ambient pressure given tissue loadings. The result is the above
mentioned formula.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
We do want to compare "loose" dives too, but we need to be a bit
careful, and always use the trip date as the primary sort key for any
dives that are not in the same trip.
Reported-and-tested-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
As the user enters data into the entry fields, that data is validated and
as soon as there is enough data we start constructing a dive profile,
including the final ascent to the surface, including required deco stops,
etc.
This commit still has some serious issues.
- when data is input that doesn't validate, we just print a warning to
stdout - instead we need to change the backgroundcolor of the input
field or something.
- when we switch to the last dive in order to show the profile we don't
actually search for the last dive - we just show the first one in the
tree. This works for the default sort order but is of course wrong
otherwise
I'm sure there are many other bugs, but I want to push it out where it is
right now for others to be able to take a look.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
This comes with absolutely no gui - so the plan literally needs to be
compiled into Subsurface. Not exactly a feature, but this allowed me to
focus on the planning part instead of spending time on tedious UI work.
A new menu "Planner" with entry "Test Planner" calls into the hard-coded
function in planner.c. There a simple dive plan can be constructed with
calls to plan_add_segment(&diveplan, duration, depth at the end, fO2, pO2)
Calling plan(&diveplan) does the deco calculations and creates deco stops
that keep us below the ceiling (with the GFlow/high values currently
configured). The stop levels used are defined at the top of planner.c in
the stoplevels array - there is no need to do the traditional multiples of
3m or anything like that.
The dive including the ascents and deco stops all the way to the surface
is completed and then added as simulated dive to the end of the divelist
(I guess we could automatically select it later) and can be viewed.
This is crude but shows the direction we can go with this. Envision a nice
UI that allows you to simply enter the segments and pick the desired
stops.
What is missing is the ability to give the algorithm additional gases that
it can use during the deco phase - right now it simply keeps using the
last gas used in the diveplan.
All that said, there are clear bugs here - and sadly they seem to be in
the deco calculations, as with the example given the ceiling that is
calculated makes no sense. When displayed in smooth mode it has very
strange jumps up and down that I wouldn't expect. For example with GF
35/75 (the default) the deco ceiling when looking at the simulated dive
jumps from 16m back up to 13m around 14:10 into the dive. That seems very
odd.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
It's still a mess (and the symbols aren't used consistently), but it's a
tiny bit more logical...
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
The old implementation was broken in several ways.
For one thing the GF values are percentages, so they should normally be
0 < GF < 1 (well, some crazy people like to go above that).
With this most of the Bühlmann config constants were wrong.
Furthermore, after we adjust the pressure tolerance based on the gradient
factors, we need to convert this back into a depth (instead of passing
back the unmodified depth - oops).
Finally, this commit adds closed circuit support to the deco calculations.
Major progress and much more useful at this stage.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
Bringing in the first attempts to do our own deco calculations
|
|
There were some minor problems when moving the selection
cursor around:
1) If the selection was larger than 1, it was possible
for the selection to get "stuck" in the middle of the list.
This patch approaches this by always calling
gtk_tree_selection_unselect_all() before
gtk_tree_selection_select_iter(), or simply always making
sure we have one selected iterator when navigating with the keys.
2) When there was a single top level dive before the first trip
it wasn't possible to navigate trough the child dives of said
trip in both directions.
The patch attempts to fix this by having the hunks/checks:
if (idx < 0) {
(idx is of a trip) performed regardless of other conditions.
*** Note: testing was done by importing all test*.xml
dives with auto-group on.
[Dirk Hohndel: adjusted the patch to also fix on_key_press to only grab
the key if no modifier key is pressed; otherwise this
breaks shift-cursor-keys for selecting multiple dives.]
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
This also initializes the N2 tissue saturations to correct numbers
(setting them to zero was clearly silly).
With this commit we walk back in the dive_table until we find a surface
intervall that's longer than 48h. Or a dive that comes after the last one
we looked at; that would indicate that this is a divelist that contains
dives from multiple divers or dives that for other reasons are not
ordered. In a sane environment one would assume that the dives that need
to be taken into account when doing deco calculations are organized as one
trip in the XML file and so this logic should work.
One major downside of the current implementation is that we recalculate
everything whenever the plot_info is recreated - which happens quite
frequently, for example when resizing the window or even when we go into
loup mode. While this isn't all that compute intensive, this is an utter
waste and we should at least cache the saturation inherited from previous
dives (and clear that number when the selected dive changes). We don't
want to cache all of it as the recreation of the plot_info may be
triggered by the user changing equipment (and most importantly, gasmix)
information. In that case the deco data for this dive does indeed have to
be recreated. But without changing the current dive the saturation after
the last surface intervall should stay the same.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
For dives with more than 4 cylinders, the frame got very crowded and we
needed a magnifier to see the numbers.
If we used more than four tanks, let's put the info in another frame, if not, print
the OTUs, the maxcns and the weight sytem in the new frame.
There is still room for two more short data.
Changed naming of nitrox and trimix mixes.
Changed cylinder description.
There are issues with the size of some translations.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
In commit 96db56f89c76 ("Allow overlapping (and disjoint) dive trips")
I allowed dives to be part of arbitrary dive trips regardless of date,
which meant that the divelist tree model code needed to find the right
parent for a dive as it was inserted.
That code stupidly assumed that the top level of the dive list tree
containted *only* trips, which is not at all the case. It happens to
be true if you group all your dives into divetrips (the common case
for autogroup=1, which real users do tend to have), but now that Dirk
made the autogrouping be a per-xml-file setting, it became much easier
to trigger the "mixed trips and non-trip dives" case, and that showed
the stupid bug with the test dives.
So instead of just blindly iterating to the 'n'th entry, search for
the actual entry that is the dive trip we want to associate a dive
with.
Reported-by: Lubomir Ivanov <neolit123@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
This patch makes the divelist behave more as you would expect it as you
scroll up and down through its entries.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
This commit steals the cursor up and down keys away from gtk so regardless
where gtk thinks the focus may be, we can still use the keys to change
between dives.
In the current UI design where all editing happens in separate windows
this works as expected, as we only grab the keys for the main window. If
we manage to re-enable in-place editing then we need to make sure that
this doesn't cause problems (as gtk uses up/down for the ability to change
drop down selections in combo boxes or values in spin buttons. So we must
make sure that we stop stealing these keys once we start editing something
(in which case simply switching to the next/prev dive wouldn't be a good
thing, anyway).
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
This clarifies/changes the meaning of our "cylinderindex" entry in our
samples. It has been rather confused, because different dive computers
have done things differently, and the naming really hasn't helped.
There are two totally different - and independent - cylinder "indexes":
- the pressure sensor index, which indicates which cylinder the sensor
data is from.
- the "active cylinder" index, which indicates which cylinder we actually
breathe from.
These two values really are totally independent, and have nothing
what-so-ever to do with each other. The sensor index may well be fixed:
many dive computers only support a single pressure sensor (whether
wireless or wired), and the sensor index is thus always zero.
Other dive computers may support multiple pressure sensors, and the gas
switch event may - or may not - indicate that the sensor changed too. A
dive computer might give the sensor data for *all* cylinders it can read,
regardless of which one is the one we're actively breathing. In fact, some
dive computers might give sensor data for not just *your* cylinder, but
your buddies.
This patch renames "cylinderindex" in the samples as "sensor", making it
quite clear that it's about which sensor index the pressure data in the
sample is about.
The way we figure out which is the currently active gas is with an
explicit has change event. If a computer (like the Uemis Zurich) joins the
two concepts together, then a sensor change should also create a gas
switch event. This patch also changes the Uemis importer to do that.
Finally, it should be noted that the plot info works totally separately
from the sample data, and is about what we actually *display*, not about
the sample pressures etc. In the plot info, the "cylinderindex" does in
fact mean the currently active cylinder, and while it is initially set to
match the sensor information from the samples, we then walk the gas change
events and fix it up - and if the active cylinder differs from the sensor
cylinder, we clear the sensor data.
[Dirk Hohndel: this conflicted with some of my recent changes - I think
I merged things correctly...]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
We used to have the rule that a dive trip has to have all dives in it in
sequential order, even though our XML file really is much more flexible,
and allows arbitrary nesting of dives within a dive trip.
Put another way, the old model had fairly inflexible rules:
- the dive array is sorted by time
- a dive trip is always a contiguous slice of this sorted array
which makes perfect sense when you think of the dive and trip list as a
physical activity by one person, but leads to various very subtle issues
in the general case when there are no guarantees that the user then uses
subsurface that way.
In particular, if you load the XML files of two divers that have
overlapping dive trips, the end result is incredibly messy, and does not
conform to the above model at all.
There's two ways to enforce such conformance:
- disallow that kind of behavior entirely.
This is actually hard. Our XML files aren't date-based, they are
based on XML nesting rules, and even a single XML file can have
nesting that violates the date ordering. With multiple XML files,
it's trivial to do in practice, and while we could just fail at
loading, the failure would have to be a hard failure that leaves the
user no way to use the data at all.
- try to "fix it up" by sorting, splitting, and combining dive trips
automatically.
Dirk had a patch to do this, but it really does destroy the actual
dive data: if you load both mine and Dirk's dive trips, you ended up
with a result that followed the above two technical rules, but that
didn't actually make any *sense*.
So this patch doesn't try to enforce the rules, and instead just changes
them to be more generic:
- the dive array is still sorted by dive time
- a dive trip is just an arbitrary collection of dives.
The relaxed rules means that mixing dives and dive trips for two people
is trivial, and we can easily handle any XML file. The dive trip is
defined by the XML nesting level, and is totally independent of any
date-based sorting.
It does require a few things:
- when we save our dive data, we have to do it hierarchically by dive
trip, not just by walking the dive array linearly.
- similarly, when we create the dive tree model, we can't just blindly
walk the array of dives one by one, we have to look up the correct
trip (parent)
- when we try to merge two dives that are adjacent (by date sorting),
we can't do it if they are in different trips.
but apart from that, nothing else really changes.
NOTE! Despite the new relaxed model, creating totally disjoing dive
trips is not all that easy (nor is there any *reason* for it to be
easty). Our GUI interfaces still are "add dive to trip above" etc, and
the automatic adding of dives to dive trips is obviously still based on
date.
So this does not really change the expected normal usage, the relaxed
data structure rules just mean that we don't need to worry about the odd
cases as much, because we can just let them be.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
We used to enable dive trips whenever we sorted by date, which can be
a bit annoying. Sometimes you really just want to sort all your dives
by date, without necessarily seeing the trip data.
So this changes the default sort to be the "dive number" table, and
then does *not* actually sort by the dive number, but instead enables
the trips, and then sorts the result by date. So the "dive number"
column - which used to be non-sortable - becomes semantically
equivalent to the old date column sorting.
And now sorting by date makes it act like sorting by depth or any
other attribute - we hide the dive trip tree, and just show the plain
list of dives (sorted by date, obviously).
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
This patch makes a couple of modifications:
1) divelist.c:delete_single_dive() now tries to free all memory associated
with a dive, such as the string values for divemaster, location, notes &
etc.
2) dive.c:merge_text(), now always makes a copy in memory for the returned
string - either combined or one of the two which are passed
to the function.
The reason for the above two changes is that when (say) importing the same
data over and over, technically a merge will occur for the contained dives,
but mapped pointers can go out of scope.
main.c:report_dives() calls try_to_merge() and if succeeds the two dives
that were merged are deleted from the table. when we delete a dive,
we now make sure all string data is cleared with it, but also in the actual merge
itself, which precedes, copies of the merged texts are made (with merge_text()),
so that the new, resulted dive has his own text allocations.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
Interesting crash. Importing a file gets us to a stage where we have a
trip tree note with a date that doesn't exist as trip date. That's clearly
bogus. And in import_files() we assume that all is still fine and try to
restore the old expanded / selected state for the various trips.
There is clearly a bigger issue here, this patch at least prevents the
actual crash from happening by making sure the pointer is non-NULL before
dereferencing it.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
From the GTK docs on gtk_tree_model_get_path():
"Returns a newly-created GtkTreePath referenced by iter.
This path should be freed with gtk_tree_path_free()."
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
This will normally happen for CCR dives with a set pO2. The old
calculations was only valid for OC dives.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
The idea behind editing timestamps had of course been the typical "oops, I
forgot to set my time correctly" which shifts a dive (or a few of them) by
a few hours but keeps the overall order of dives the same).
But reasonable people might argue that they can envision a scenario where
more dramatic changes are being made. And we need to deal with the impact
this has on dive trips.
Here we handle a couple of simple cases:
- this is the only dive in a trip; just update the trip
(this can still cause problems if the new time is in the middle of an
existing trip).
- this dives moves before the start of the trip it is in; let's remove it
from that trip (this response is a bit simplistic - but as I tried to
say, I don't expect this to be a common use case; and removing it at
least doesn't lead to entirely unexpected behavior).
- this dive moves past the end of this trip into the range of a different
trip (in this case we remove the dive from the current trip and allow it
to interrupt the trip it is moving into).
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
|
|
I foolishly changed visible_columns in both the (ill-named) cns branch and
master...
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Conflicts:
divelist.c
gtk-gui.c
profile.c
|
Dirk Hohndel
Linus Torvalds
Jan Schubert
Lubomir I. Ivanov
Henrik Brautaset Aronsen
Robert C. Helling
Salvador Cuñat