models.h / .cpp was getting too big (around 2.5k lines), and each change
triggered a rebuild in tons of parts of Subsurface, this is the first
commit trying to make the models code sane by removing them all of the
models.h/cpp file and also clearning code in the process.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is an attempt to help share code between the desktop version of
Subsurface and the mobile version.
More code will be moved around and the models will be split in a way that
will help recompile times and also creation of different interfaces for
different form-factors.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Just as we would expect, the libgit2 developers of course once again broke
their API. In order to compile against current master we need to remap
those APIs once again.
Simply call cmake with -DUSE_LIBGIT23_API
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This doesn't work on QtCreator yet - but it's the documented way in CMake
to display the project hierarchy in some IDEs (the one that I know that it
works with is Visual Studio). Even knowing that this doesn't work with
QtCreator, it's a valid change because someone can fix creator in the
future.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The SimpleWidgets file was getting too big, and location information will
also need a new model - a good way to do not mix everything is to put
things in a new file.
[Dirk Hohndel: added missing include of stdint.h]
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
MACOSX_BUNDLE and WIN32 give hints to cmake for different
install behaviors. for instance, trying to copy frameworks
and other stuff to the correct places.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Only build the tests when running them. The downside of the way this is
implemented is that build errors for the tests actually show up as test
errors and are only seen in the test log - but this seemed worth not
having every build include the tests.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
- don't put anything on the first line after the opening "("
- fix indentation of the closing "("
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
remoevs the extra white sapce in the generated
ssrf-version.h.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
These two new options allow usage of pkg-config to retrieve
libdivecomputer and libgit2.
For some reason the variables *_LIBRARIES have to be set
to blank strings in the process, otherwise "-l" is passed to
the linker.
LIBGIT2_FROM_PKGCONFIG is a rename from PREFER_GIT_FROMSOURCE
for consistency. The IF() check is also fixed.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
If NO_MARBLE is set to ON, the MARBLE package is not seeked.
If set to OFF the package is seeked, but if the find fails
the gcc definition -DNO_MARBLE is set and marble is disabled.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
NO_TESTS: disable the tests
NO_DOCS: disable the docs
Both are set to OFF by default.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
We shouldn't need this. Ubuntu is once again broken. What else is new.
But since this shouldn't hurt any of the sane Linux version, I'll just add
it for any Linux flavor.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is being taken care by Qt5::LinguistTool
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
CMake still confuses me. When testing if a variable was set it sometimes needs
to want "DEFINED", but sometimes I need to compare it to an empty string. Let's
do both (this seems to fix the problem I ran into) - but this is still weird.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
And use this to look for libusb.
If it's there then libdivecomputer likely was linked against it. If it isn't
then we don't need it, either.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
By adding a new CMakeLists.txt file and configuring
the translations inside of it's folder, the buildsystem
will compile everything on that folder.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
And remove the old code that did the same thing in
a very complex way.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Currently we search for a static library and if we don't
find it, we search for a shared library. If one of them
is found, we return success but if none of them is found,
error. So - A static library for libgit and libdivecomputer
is preferred over the shared one.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
We will use the OPTIONS variable to decide to find the
.a or the .so, so there's no need to do checks here.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
- add an option to toogle between libgit2 from package vs built from source
- add two options that will be used in the future: use static/dynamic lib
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
We are already looking for the default directory.
*if* cmake doesn't find Marble automatically, the user needs
to specify -DMARBLE_INCLUDE_DIR and -DMARBLE_LIBRARIES
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The pkg_config_library macro is a helper macro and was generating noise
on the main CMakeLists.txt file, so I moved it to another file.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is the correct way to deal with CMake Modules (or at least, the
default way). All our custom CMakeFiles will be here.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Qt5 Supports Components, which means that we can list the
components we need to find, instead of specifying one for
each line.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
While I don't need this when building on Fedora 20, OBS builds for
openSUSE appear to fail without this. Strange.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is still not something I'm 100% sure I got right. But at least now it
correctly rebuilds the .qm files...
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Don't try to install the TARGET into the Subsurface.app or we get a neat
infinite recursion. As a bundle cmake does all these things automagically.
Right now the version is hard coded which is ugly but since it's determined in
a separate cmake file it seemed non-obvious how to do this right...
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The cmake keywords still confuse me at times.
A custom target is always recreated. How silly. But the correct (I think)
implementation is actually easier...
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
I guess I should add the code to move that app to /Applications but I never use
it this way... not sure how many people really install from source on a Mac,
anyway.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When building on a Mac for some reason I was getting an empty translation file
name (and therefore and entry that was the whole directory and not a
translation file. This is a bit clumsy but it fixes the problem.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Silly cmake doesn't allow a target to depend on the install target. This
ugly hack appears to be the recommended workaround :-(
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The previous attempt to install them was completely bogus as it only
worked when the source directory had previously used for an in-source
build using qmake.
Oops.
With this change we now create the translations in the build directory and
install them from there.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
- don't reference .so files, reference libraries
- handle libdivecomputer just like the other libraries
- add the ability to link libdivecomputer statically
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
With no V2 question shown
- parsing fails when a V2 file is loaded
- parsing succeeds when a V3 file is loaded
- import of CSV file succeeds
With V2 question shown
- parsing succeeds when a V2 file is loaded
Finally compare the output of reading in the various files with reference
output included in the sources.
My guess is that this test might be a bit fragile, but hey, it's a start.
(reminder: the tests only get built when using cmake)
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Sequentially parses a file, expected to be a Datatrak/WLog divelog, and
converts the dive info into Subsurface's dive structure.
As my first DC, back in 90s, was an Aladin Air X, the obvious choice of log
software was DTrak (Win version). After using it for some time we moved to WLog
(shareware software more user friendly than Dtrak, printing capable, and still
better, it runs under wine, which, as linux user, was definitive for me). Then,
some years later, my last Aladin died and I moved to an OSTC, forcing me to
look for a software that support this DC.
I found JDivelog which was capable of import Dtrak logs and used it for some
time until discovered Subsurface existence and devoted to it.
The fact was that importing Dtrak dives in JDivelog and then re-importing them
in Subsurface caused a significant data loss (mainly in the profile events and
alarms) and weird location of some other info in the dive notes (mostly tag
items in the original Dtrak software). This situation can't actually be solved
with tools like divelogs.de which causes similar if no greater data loss.
Although this won't be a core feature for Subsurface, I expect it can be useful
for some other divers as has been for me.
Comments and issues:
Datatrak/Wlog files include a lot of diving data which are not directly
supported in Subsurface, in these cases we choose mostly to use "tags".
The lack of some important info in Datatrak archives (e.g. tank's initial
pressure) forces us to do some arbitrary assumptions (e.g. initial pressure =
200 bar).
There might be archives coming directly from old DOS days, as first versions
of Datatrak run on that OS; they were coded CP437 or CP850, while dive logs
coming from Win versions seems to be coded CP1252. Finally, Wlog seems to use a
mixed confusing style. Program directly converts some of the old encoded chars
to iso8859 but is expected there be some issues with non alphabetic chars, e.g.
"ª".
There are two text fields: "Other activities" and "Dive notes", both limited to
256 char size. We have merged them in Subsurface's "Dive Notes" although the
first one could be "tagged", but we're unsure that the user had filled it in
a tag friendly way.
WLog adds some information to the dive and lets the user to write more than
256 chars notes. This is achieved, while keeping compatibility with DTrak
divelogs, by adding a complementary file named equally as the .log file and
with .add extension where all this info is stored. We have, still, not worked
with this complementary files.
This work is based on the paper referenced in butracker #194 which has some
errors (e.g. beginning of log and beginning of dive are changed) and a lot of
bytes of unknown meaning. Example.log shows, at least, one more byte than those
referred in the paper for the O2 Aladin computer, this could be a byte referred
to the use of SCR but the lack of an OC dive with O2 computer makes impossible
for us to compare.
The only way we have figured out to distinguish a priori between SCR and non
SCR dives with O2 computers is that the dives are tagged with a "rebreather"
tag. Obviously this is not a very trusty way of doing things. In SCR dives,
the O2% in mix means, probably, the maximum O2% in the circuit, not the O2%
of the EAN mix in the tanks, which would be unknown in this case.
The list of DCs related in bug #194 paper seems incomplete, we have added
one or two from WLog and discarded those which are known to exist but whose
model is unknown, grouping them under the imaginative name of "unknown". The
list can easily be increased in the future if we ever know the models
identifiers.
BTW, in Example.log, 0x00 identifier is used for some DC dives and from my own
divelogs is inferred that 0x00 is used for manually entered dives, this could
easily be an error in Example.log coming from a preproduction DC model.
Example.log which is shipped in datatrak package is included in dives
directory for testing pourposes.
[Dirk Hohndel: some small cleanups, merged with latest master, support
divesites, remove the pointless memset() before free() calls
add to cmake build]
Signed-off-by: Salvador Cuñat <salvador.cunat@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Fixed the icon problem Dirk found.
We really should choose between qmake and cmake. I wouldn't care about
cmake if qmake was building the UTs...
From 8eeea28a523fd6ef588d81b82ab904d4512b3d7a Mon Sep 17 00:00:00 2001
From: Patrick Valsecchi <patrick@thus.ch>
Date: Tue, 24 Feb 2015 09:06:37 +0100
Subject: [PATCH 3/3] Cmake build now contains icons
Signed-off-by: Patrick Valsecchi <patrick@thus.ch>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Non parallel build was broken because qthelper.cpp was depending
indirectly on a QT generated file and no cmake dependency was enforcing
the file to be generated before. I've removed the not needed #include
that was introducing this dependency.
In cmake, {FOO STREQUAL ""} is TRUE only if FOO is defined and
empty. Fixed a couple of bad usages.
Made the required libraries actually required in cmake.
From ef5ab90f258c5754d3022a023c28050cbafed3d5 Mon Sep 17 00:00:00 2001
From: Patrick Valsecchi <patrick@thus.ch>
Date: Tue, 24 Feb 2015 08:38:49 +0100
Subject: [PATCH 1/2] Fixed cmake build.
Non parallel build was broken because qthelper.cpp was depending
indirectly on a QT generated file and no cmake dependency was enforcing
the file to be generated before. I've removed the not needed #include
that was introducing this dependency.
In cmake, {FOO STREQUAL ""} is TRUE only if FOO is defined and
empty. Fixed a couple of bad usages.
Made the required libraries actually required in cmake.
Signed-off-by: Patrick Valsecchi <patrick@thus.ch>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This doesn't actually change any code, but it moves the 'is_git_repo()'
function that is used by both loading and saving into a new git-access.c
file.
This is where I'll start doing remote repo syncing too. Knock wood.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Refactored the parsing logic to make it more solid (no more guessing) and
more flexible (support more formats).
Added a test for checking that.
Fixed a few warnings.
[Dirk Hohndel: some changes to coding style]
Signed-off-by: Patrick Valsecchi <patrick@thus.ch>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
My previous cmake patch was relying on ssrf-version.h to be generated
correctly by qmake, in the source directory. This one fixes the generation
of this file in the build directory.
Signed-off-by: Patrick Valsecchi <patrick@thus.ch>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
We already have a bunch of filters, let's pack them together
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
I fixed up the decode and finished the parse for Cochran EMC, Commander
and Gemini computers. I suspect that this code may only work with files
from certain versions of Cochran Analyst. It works with my own CAN files
and with the samples that came with Analyst v4.01v.
A seemingly arbitrary offset of 0x4914 is needed to access data.
The previous code uses 0x4a14 and 0x4b14. I suspect these are from
different version of Analyst.
[Dirk Hohndel: whitespace cleanup, add files to subsurface.pro, made sure
this compiles without the corresponding patch to
libdivecomputer (that isn't upstream, yet), cleaned up the
usage of structs, removed a few unused variables]
Signed-off-by: John Van Ostrand <john@vanostrand.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Ignoring the fact that cmake is a disgusting pile of crap, let's at least
track the files in our project for the people who want to use it.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The statistics widget is a beast, one of the parts that I dislike most
on the current subsurface implementation. This is the initial work to
change that to something amazing. This first commit adds the first bunch
of files that I think are needed, and the correct setup for the qmake and
cmake buildsystems.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Just added the missing files to the CMakesList.txt
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This commit breaks the loading of images that were done in the divelist
into smaller bits. A bit of code refactor was done in order to correct the
placement of a few methods.
ShiftTimesDialog::EpochFromExiv got moved to Exif::epoch dive_add_picture
is now used instead of add_event picture_load_exif_data got implemented
using the old listview code. dive_set_geodata_from_picture got
implemented using the old listview code.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This has a few classes: the model is the list of pictures for the current
dive, the delegate is how this pictures will be displayed on screen, the
widget is the collection of delegates, and the DivePictureThumbnailThread
is a worker-thread to generate the thumbnails so the UI will not freeze.
[Dirk Hohndel: added the new files to subsurface.pro]
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This changes are only to make CMake compile again after the
addition of the export dialog.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The DiveList classes were a partial mess (and some of it is still in a
messy state). The classes that deal with it where done in 'qtHelpers.h',
the extern global variable in dive.h, a few methods here and there. This
concentrates most - but not all - functions in their own file. The reason
for that is to make the new developer faster when looking for things: if
it's a divecomputer related method, it should be in a single file, not
scattered around.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The C code should be compilable without the need to compile the Gui part,
too. This is expecially good for unit testing as we can test all the
algorithms without a window appearing out of nowhere.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is a preferences setting, it should belong to the preferences
structure.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Use cmake .. -DLIBDCDEVEL="../your/library" to specify
where the libdc is, or leave it blank if you have it
installed on the system. you can change the location later
by running ccmake.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
While filtering for *.c / *.cpp is much smaller, it has the
disvantage of getting garbage on some computers as the devs
tends to create test files around.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is just a stub test case to show how a test case should be
implemented. every 'private slot' on the test classes will be
automatically executed when you run 'make test' on the terminal,
and a report will be generated with a failure / success.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This commit adds a CMake based buildsystem for the tests,
it separated the subsurface sources in small libraries to
make testing easyer ( subsurface core, subsurface ui,
subsurface profile, etc. )
There's no test yet, this however produces a compiled
and executable binary *on linux*. This file shouldn't
be needed on any other platform as this is not the main
build system, but the test build system.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
