It looks like libxml2 has some internal limitations by default that
causes parse failures in some situations. Avoid them with
XML_PARSE_HUGE.
Without this, you get errors like
test.xml:349250: parser error : internal error: Huge input lookup
όμουν τουλάχιστον αλλά +2kg και ενδεχομένως +4
^
when something in the xml file grows too large.
I don't know libxml2 internals, so I have no idea what exactly goes
wrong, but the docs say:
XML_PARSE_HUGE = 524288 : relax any hardcoded limit from the parser
and that makes us successfully parse the Greek file from Kostas.
Reported-by: Kostas Katsioulis <kostaskatsioulis@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Move the ARRAY_SIZE macro into a header file and use it to determine the
number of cloud servers that we need to check.
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
If we can't reach the cloud server in the URL (which might come from the
settings or be passed in by the user), we try the alternative server(s).
If we end up changing servers, we need to update the remote that we have
already parsed from the URL.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
If we can't reach our preferred server, try using a different one.
The diff makes more sense when ignoring white space.
With this we check the connection to the cloud server much earlier and
in case of failure to connect try a different cloud_base_url.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
With the new names for the cloud server we'd get different local cache
directory names depending on which server gets used. In order to avoid
that, normalize the name before generating the hash that determines the
local directory name.
Additionally, the old code had an extra '/' in the URL, due to the way
the URL was assembled. Again, to match the existing hash for people
upgrading from older Subsurface versions, add that to our normalized
name as well.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The backend infrastructure will soon be able to support more than one
cloud server which automagically stay in sync with each other.
One critical requirement for that to work is that once a session was
started with one of the servers, the complete session happens with that
server - we must not switch from server to server while doing a git
transaction. To make sure that's the case, we aren't trying to use DNS
tricks to make this load balancing scheme work, but instead try to
determine at program start which server is the best one to use.
Right now this is super simplistic. Two servers, one in the US, one in
Europe. By default we use the European server (most of our users appear
to be in Europe), but if we can figure out that the client is actually
in the Americas, use the US server. We might improve that heuristic over
time, but as a first attempt it seems not entirely bogus.
The way this is implemented is a simple combination of two free
webservices that together appear to give us a very reliable estimate
which continent the user is located on.
api.ipify.org gives us our external IP address
ip-api.com gives us the continent that IP address is on
If any of this fails or takes too long to respond, we simply ignore it
since either server will work. One oddity is that if we decide to change
servers we only change the settings that are stored on disk, not the
runtime preferences. This goes back to the comment above that we have to
avoid changing servers in mid sync.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
We know the preference is never empty, so stop testing for this. But
don't maintain two different preferences with basically the same
content. Instead add the '/git' suffix where needed and keep this all in
one place.
Simplify the extraction of the branch name from the cloud URL.
Also a typo fix and a new comment.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This was a hack for a very early SSL certificate that was rejected on
some platforms. We haven't used that one in ages, so let's just remove
the whole hack - but always show in the console output when there was an
SSL error.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The dive selection was initialized during data-reset. However,
this emitted a signal before all data-reset routines were run.
Ultimately, this led to access-after-free in the statistics code.
Instead, move the select_newest_visible_dive() signal from the
divelist-model to the process_loaded_dives() function. There
is no point in initializing the selection if the dive data
is cleared after all.
This change broke closing of the log, because the UI-selection
was not reset. Therefore, when clearing the data, clear the
selection before proceeding with clearing.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
In 9bfc6d252, testing of the planner was changed to use the
planner_ds parameter instead of a global variable.
Unfortunately, two conditionals were inverted, leading to
an erroneous ceiling calculation when in the planner.
Restore the proper conditions. Moreover, instead of testing
the planner_ds parameter, use the already existing in_planner
flag, which is derived from said parameter.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Both the calculations for CNS and OTU did not take
into account the pO2 drop when using a PSCR. Furthermore,
there was some unit confusion due to not using internal
units.
Reported-by: arosl
Signed-off-by: Robert C. Helling <helling@atdotde.de>
When parsing "event 123" (?) a picture is added, without
initializing the picture structure. Thus, a picture with a
random gps location is added.
Use the "empty_picture" initializer to avoid that. Fixes a
Coverity warning.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
We used to round the ceilings for the individual tissues with
%.1f but the maximal (and thus effective) ceiling only with
%.0f. This makes no sense or be rounded up (to the conservative
side).
This commit shows also the maximal ceiling with higher accuracy.
Reported-by: Peter Hübner
Signed-off-by: Robert C. Helling <helling@atdotde.de>
When doing OC bailout from a CCR dive, there could still
be pO2 sensor readings but those are not valid.
This fixes a problem noticed by Justin Ashworth.
Signed-off-by: Robert C. Helling <helling@atdotde.de>
Even when diving a CCR, the pO2 cannot exceed ambient
pressure. This only makes a difference at shallow depths.
Fix this in the calculation of OTUs and CNS.
This affects some tests that now have slightly different CNS and OTU values.
Suggested-by: Justin Ashworth
Signed-off-by: Robert C. Helling <helling@atdotde.de>
We had a user request to allow for setpoint changes
at certain depths for CCR deco.
You can now enter a cylinder with name like
"SP 1.4" ('S' and 'P' and ' ' and a float) with
a switch depth and that cylinder is interpreted as
a depth dependent setpoint switch.
This user interface is a hack. But I believe that such
setpoint changes are similar enough to gas switches during
deco and should thus be handled in a simiar manner.
I would be happy to hear ideas how this could be made
less easter eggish.
Suggested-by: Justin Ashworth
Signed-off-by: Robert C. Helling <helling@atdotde.de>
The application state is a desktop-only thing. The mobile UI
also has its application state, but that is something completely
different.
The last remaining user of the application state was to flag
whether the planner is active. Since this has all been
unglobalized, the ApplicationState structure can be moved
from core to the desktop UI. And there it can be made local
to the MainWindow class.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The TemplateLayout prints different dives depending on
whether the planner is active. Instead of accessing a
global variable, pass the status down from the MainWindow.
That's all quite convoluted, since there are multiple
layers involved.
On the positive side, the in_planner() function has now
no users an can be removed.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
To remove reliance on global state, pass an "in_planner" argument
to decoMode(). Thus, calls to in_planner() can be removed.
This is a more-or-less automated change. Ultimately it would
probably be better to pass the current deco-mode to the affected
functions instead of calling decoMode() with an in_planner
parameter.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
To remove reliance on global state, pass an "in_planner" argument
to clear_deco(). Thus, calls to in_planner() can be removed.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
To remove reliance on global state, pass an "in_planner" argument
to vpmb_next_gradient(). Thus, calls to in_planner() can be removed.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
To remove reliance on global state, pass an "in_planner" argument
to add_segment(). Thus, calls to in_planner() can be removed.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The icons shown in the dive list were rendered for every single
access. Render them only once. This supposes that the
defaultIconMetrics structure does not change once the icons are
rendered!
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
There was the "application state", which decided what to show
in the "quadrants" and the "view state" which decided which
quadrant to show. These interacted in a hard-to-grasp way.
The "view state" is used to show the map or dive list in
full screen.
I simply couldn't get these two orthogonal states to interact
properly. Moreover the thing was buggy: If a quadrant was hidden,
the user could still show it, by dragging from the side of the
window, at least under KDE.
To solve these woes, merge the two states into a single
application state. If the widget of a quadrant is set to null,
don't show it. So the four "view states" are now "application
states" where three of the four quadrants are not shown.
This also changes the memory management of the widgets:
widgets that are not shown are now removed from the QSplitter
objects. This makes it possible that the same widget is
shown in *different* quadrants.
While writing this, I stumbled upon a Qt bug, which is known
since 2014:
https://forum.qt.io/topic/43176/qsplitter-sizes-return-0
When restoring the quadrant sizes there was a test whether
the quadrant size is 0. If that was the case, a default size
was set. This seems not to work if the widgets were recently
added. Since this test now always fails, make the quadrants
non-collapsible and thus guarantee that 0 is never saved as
a size.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Somehow three identical lines snuck into commit 0a4e37ee8b ("core/BT: simplify
detection of bluetooth names").
Reported-by: Henrik Brautaset Aronsen <subsurface@henrik.synth.no>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Instead of that super long if-else if chain, have something more
structured using a table for the common case of prefix based names.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
It would be so much nicer if we could just let libdivecomputer do this,
but the filter function there doesn't quite do things the way we need
them to be. Which is why we have our own function here.
This is a small attempt to rationalize the code that we have to make it
easier to maintain.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The goal is to enable a user experiencing crashes when applying GPS data
to their dive log to make all necessary data available to the
developers. Hopefully the clipboard is large enough to hold all the
data.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This only read accesses the dive and constructs a plot-info
structure. Make the dive parameter const.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
This function initializes decompression data from a dive.
The dive is not modified, therefore make it const.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
We need to do this before the preferences are loaded, or the system
default size is lost. Given that our other sizes are all relative to
this value, that would be a problem.
With this we can now ensure that we always have the right font size for
smaller, regular, and larger theme settings.
Also removes some obsolete commented out code.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Adds fields to the advanced preferences page to modify GFLow and GFHigh for
the Buhlmann decompression model for calculating ceilings. Updated preferences
code to set the Buhlmann parameters in core/deco.c when the GF prefs are
updated.
Signed-off-by: Doug Junkins <douglas.junkins@gmail.com>
First, the time zone adjustment was wrong - this as written could only
ever have worked in UTC or by pure chance.
Second, the order of alerting the UI of the availability of a GPS fix
was also incorrect creating a race between the UI and our data
structures.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When setting a CCR setpoint, the profile code(!) would turn
the dive into a CCR dive. Not only should the display layer
not alter dives, this also means that the action is not
undoable.
Move that to the appropriate undo command, where it makes
more sense, but obviously also makes things more complicated.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The make_first_dc() function clones a dive with a certain dive
computer moved to the front. This is used by the
MoveDiveComputerToFront undo command.
make_first_dc() calls invalidate_dive(). However, the undo
command does that by itself on every undo/redo. Thus,
remove the call in make_first_dc().
Aside from consistency, the goal is to move invalidate_dive()
to command/* so that we can be more aggressive about the whole
topic: Store only "const dive *" pointers and thus force any
writing access to explicitly invalidate the dive cache.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The DiveHandler shows a context menu where a cylinder can be
chosen. This indirectly accesses the global displayed_dive
variable.
Remove this in a step to make the profile reentrant.
The code was quite ominous: instead of simply generating the
list of cylinders, a global model was reset and then accessed
with Qt's cumbersome model/view API. All this trampling over
global state can be removed by simply making the function
that generates the list globally accessible.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Instead of accessing the global displayed_dive variable
in RulerItem, pass the dive. This is a step in making the
profile reentrant.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Instead of accessing the global displayed_dive variable,
pass the dive to the various profile items. This is a
step in making the profile code reentrant.
This removes the last user of the displayed_dc macro,
which can now be removed.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Don't access the global displayed_dive variable in an effort
to make the profile reentrant.
Note that this still accesses the global dc_number variable,
which will likely have to be removed.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The profile item that shows the ceilings adds a warning event
if the ceiling is violated. This is very unfortunate.
Improve this situation by adding the event up to the function
that calculates the ceiling. This is still not how it should
be - the display layer should not modify the dive that it
displays.
To make this clear, add a comment that details that this
is a contract between planner and display layer: The planner
uses a dive that can be trampled upon by the profile.
Still, this should be solved differently.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The in_planner() function is incompatible with a reentrant
profile, since it accesses a global variable. In
create_plot_info_new() it is essentially redundant, because
there is a planner_ds (ds = deco_state) parameter that
is used only when in the planner. Therefore use that as
the in_planner indicator: when non-null, the profile is
showing a planned dive.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
