And add a fuction to parse tables that are not "relational", meaning
tables which are directly refered from Dives table.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
We can allocate fixed size arrays for smartrak tables as its size can
be known in advance. Simply reading table->num_rows is too dangereous
as smartrak tables have "holes" commonly. This is, they can look like:
Idx | Txt
1 | blablabla
2 | blebleble
4 | blobloblo
table->num_rows would give us 3, but we need to allocate 4 to get an
array like:
|0|blablabla |1|blebleble |2| |3|blobloblo
as the idea is to use the table index to reference the array data.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
Previously we built arrays for the tables each time we parsed a dive.
Now we simply build the lists once, and use them in each dive parsing.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
In most cases we can not foresee the maximum number of data of a given
type. It can be quite low or really big (a concerned diver can store
thousands of different fishes in Fish table).
Moving from arrays, where size has to be preset, to linked lists seems
the more logical option.
Here we set a (very limited) data structure, just an index and a text
fields following the format of most SmartTrak tables. Some special
table, like Buddy, needs a bit of processing before placing the data in
the list.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
We set the build type from the building script. Setting this in
CMakeLists.txt make the script setting useless and always build Release
type.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
In a previous commit, the get_gasmix_* functions were changed to
return by value. For consistency, also pass gasmix by value.
Note that on common 64-bit platforms struct gasmix is the size
of a pointer [2 * 32 bit vs. 64 bit] and therefore uses the
same space on the stack. On 32-bit platforms, the stack use
is probably doubled, but in return a dereference is avoided.
Supporting arbitrary gas-mixes (H2, Ar, ...) will be such an
invasive change that going back to pointers is probably the
least of our worries.
This commit is a step in const-ifying input parameters (passing
by value is the ultimate way of signaling that the input parameter
will not be changed [unless there are references to said parameter]).
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
The lastUsedDir() functions of MainWindow and Smrtk2ssrfcWindow don't
use any member-objects and are only used in their respective translation
units. Therefore, remove them from the class and made of static linkage.
The lastUsedImageDir() function was declared as a slog, which makes
no sense. Make it a normal static function (though one might argue
why it is assiociated with the DiveListView class in the first place).
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Replace constructs of the kind
s.toUtf8().data(),
s.toUtf8().constData(),
s.toLocal8Bit().data(),
s.toLocal8Bit.constData() or
qUtf8Printable(s)
by
qPrintable(s).
This is concise, consistent and - in principle - more performant than
the .data() versions.
Sadly, owing to a suboptimal implementation, qPrintable(s) currently
is a pessimization compared to s.toUtf8().data(). A fix is scheduled for
new Qt versions: https://codereview.qt-project.org/#/c/221331/
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Mostly replace "return (expression);" by "return expression;" and one
case of "function((parameter))" by "function(parameter)".
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
In commits eccd4b993 to 8f81a22e7 global error buffer and get_error_string()
func, were moved to a call back function.
This patch makes smtk2ssrf suport those changes and build again.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
There are ca. 50 constructs of the kind
same_string(s, "")
to test for empty or null strings. Replace them by the new helper
function empty_string().
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Don't change into a directory just to see if it exists.
Remove unnecessary braces of one of the changed if statements.
Signed-off-by: Berthold Stoeger <bstoeger@mail.tuwien.ac.at>
Correct spelling and typos in file filters.
Unify and translate file filter names.
Don't pass a file filter to a directory open dialog - not needed.
Signed-off-by: Stefan Fuchs <sfuchs@gmx.de>
6963b52d introduced a cmake option, COMMANDLINE to enable building a
pure command line version of smtk2ssrf importer, but then the
#define COMMANDLINE=1 forces building CLI mode.
This patch allows building GUI or CLI versions depending on selection of
the COMMANDLINE option.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
As Lubomir pointed out in his patch for datatrak.c, the format option %m
for sscanf doesn't work in mingw/windows. Fortunately it's unnecessary
as dates are dropped and we just get times.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
For different reasons some used functions aren't portable or simply are
not included in mingw. This includes index, rindex, strptime, and
timegm.
A workaround for this is needed, if we want to build for windows using
mingw based mxe environment. This patch does:
- drops index and rindex in favor of strchr and strrchr
- substitute strptime with a sscanf parsing
- emulate timegm with a private func smtk_timegm()
- remove definitions needed by strptime
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
Needed to crossbuild to windows.
The cmake modules are just clones from those under
subsurface/cmake/Modules, tweaked to build smtk-import for windows.
As SmartTrak importer is building separately from Subsurface, a full
CMakeLists.txt file is necessary for the smtk-import directory.
It includes cross building to windows capability.
Curly braces are necessary in nested conditionals or results may not be
those expected. In this case, the "else" clause applied on second "if",
instead of first one.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
SmartTrak's bookmarks work in the same fashion Subsurface's do. The user
may set a bookmark while underwater or set it just on the divelog
software.
At this time we are parsing those set in the DC twice, as we get one
from libdivecomputer and another from smarttrak's database.
This patch just checks if we have a bookmark event downloaded by
libdivecomputer which has the same time that the one parsed from the
.slg file. If so, merge the names taking the one from smarttrak.
Text from smarttrak is preferred because the user may have entered some
interesting note there and libdivecomputer's name is just "bookmark".
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
It was assumed that every data field in a wreck table was filled or
zeroed. This assumption is actually false, so this patch adds testing
for the existence of strings before working with them.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
Until now, we did the cylinder import based on its initial pressure (a
tank without pressure is an unused tank). Based in this assumption, we
just dropped those tanks whose initial press was 0, losing user
introduced tank definitions and getting some duplicities due to one
cylinder being numbered (e.g.) 2 by libdivecomputer and 3 by SmartTrak.
The new workflow is: get every single tank reported by SmartTrak (giving
preference to libdivecomputer parsed data), then clean the cylinder
table reverse order, dropping tanks without description and init or end
pressures, and checkig them against the previous cylinder to do a merge,
if they look the same, and try to avoid duplicities.
The new logic assumes a heavier workload for the benefit of lower data
loss (e.g. a user may get his/her tanks descriptions despite he/she
hasn't recorded their pressures because forgot the values or had an issue
with the gas transmitter).
Suggested-by: Alessandro Volpi <volpial@gmail.com>
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
In the past subsurface managed up to 8 tanks, but now it manages up to
20. SmartTrak manages 10 (3 in older no trimix versions) so there is no
more need to drop the last tanks.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
The cylinder info is built one by one. This way, instead of passing dive
and tank number parameters, just passing a pointer to the tank been
worked seems preferable.
It also introduces lrint() in the function to round the doubles values
obtained from libmdb for tank size and workingpressure.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
Framed in a full rework of the cylinders info import management, this
funcs will be used in next patches.
Functions merge_cylinder_type(), merge_cylinder_mix() and
merge_cylinder_info(), have been shamelessly copied from dive.c as they
have been removed by 162767aa dated 2017-03-11 just when I was going to
use them.
Macros MERGE_MAX() and MERGE_MIN() have just been copied from dive.c,
but they could be moved to dive.h so they could be used elsewhere if
needed.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
As commented on mailing list. Most numerical values from libmdb are
doubles obtained via strtod(), so, rounding them instead of just
truncating seems the correct way.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
User can manually add bookmarks on the profile display of SmartTrak.
These are stored in a table with column names which made me think of
cartesian spatial coordinates. In the end the X coordinate is the time.
This makes possible to build subsurface events in those moments of the
dive. The other interesting data is just the text entered by the user.
Suggested-by: Alessandro Volpi <volpial@gmail.com>
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
Fixes some memory leaks, most of them excesive use of copy_string() in
places where it was unnecesary.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
hdr_buffer pointer was not always allocated (just on dc dives), but was
always freed.
This should cause a crash while processing a manually added dive (or a dive
whose dc model couldn't be stablished), but not always does, crashes only
if the pointed memory hasn't been previously allocated. In most cases causes
memory corruption, which goes easily unnoticed as it is correctly freed in
next parse run.
This means, obviously, I have some work TODO fixing leaks with valgrind.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
There were 1 or 2 hour differences between real dive time and the
imported time because of the time zones and energy saving in some locales.
Using timegm() ensures us UTC times instead of localized times.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
Can't remember what I was thinking when wrote that crappy thing. A
simple sscanf call will do the job, and a sanity check, off course.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
Libmdbº:xturns localized strings while parsing the data bases. This is
bad for time calculations as we may end with different strings formats
(e.g. en_US vs almost the rest of the world). Solution is simple: set a
fixed locale and parse only this format.
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
