In general this patch enables building of subsurface without being
forced to use the official Qt binary packages. This is particularly helpful
when having to debug Qt internals or having to deal with custom patches
on top of the official Qt releases.
The architecture dependent file path layout is only employed by official
Qt binary packages. They are the result of a reordering at package
generation time. If Qt was build for a single architecture, the standard
layout does not add the architecture specific top level patch for the resulting
binaries.
Signed-off-by: Alex Blasche <alexander.blasche@qt.io>
It seems that the Qt team deviated from their previous practice to keep
the Qt/x.y directory structure the same for all minor releases - so now
it is indeed Qt/5.9.1
Oh well.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This way in the en_US locale we no longer get shown the odd "dive(s)"
and instead get correct singular and plural forms.
Most of the patch is just a reindentation as it removes the if clause
that used to do the special case of NOT loading a translation for the
en_US case.
Right now we start with a trivial en_US translation file. My guess is
that this will be overwritten once we do the next round of "new strings,
new translations".
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
I don't quite understand why this isn't correctly substituted to lrand48()
by the header file, but patching it in the source is easy enough.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Previous script would fail on git folders that pull from more than one
repo. That's not a general case, but pretty usual for developers
pulling/pushing to github.
BTW in this cases, a lind git pull doesn't ensure we are pulling from
the right repo.
When run without flag, default to pull --rebase.
At the same time introduce a tiny func to return a message and bail out
after some command failures.
Needed to crossbuild to windows.
The cmake modules are just clones from those under
subsurface/cmake/Modules, tweaked to build smtk-import for windows.
The used cmake toolkit for building the Android Subsurface-mobile app
(qt-android-cmake) recently moved away from compiling with Ant in favor
of Gradle. The most recent Android SDK will not support Ant any more.
This calls for the addition of the Android SDK BUILDTOOLS_REVISION define
to the cmake of Subsurface-mobile. Without this, the build will fail.
The value has to be set to an existing directory in
.../android-sdk/build-tools/
Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
This should help us with better iOS apps as it allows Apple to run
llvm against our code to improve performance.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Our qmake based build for iOS still requires the dummy.qml to trigger
inclusion of the correct plugins in the iOS app.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Give a hint how to enable build of debug Qt5 DLLs when building MXE.
Take care about the "d"/"xxxd.dll" suffix for DLLs.
Copy libastro.dll from marble to correct loation as well (nevertheless we don't use it)
Signed-off-by: Stefan Fuchs <sfuchs@gmx.de>
Just an update of comment. The stange issue with Qt5.7.1 is
still present in Qt5.8. Extend the comment accordingly.
Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
Qt plugins and some other dependencies cannot be copied
by CMake install targets. They need to be manually deployed
to staging_tests directory too.
Signed-off-by: Jeremie Guichard <djebrest@gmail.com>
Because `[ "$foo" != "" ] is equivalent to `[ "$foo" ]'
in all POSIX shells.
Signed-off-by: Cristian Ionescu-Idbohrn <cristian.ionescu-idbohrn@axis.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Don't quote if you don't have to. Spend those cpu cycles on doing
something more useful, instead.
Signed-off-by: Cristian Ionescu-Idbohrn <cristian.ionescu-idbohrn@axis.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The `which' command is a fork and possible not standard in various
distributions, or builtin in certain (odd)? shells, like `zsh'.
Signed-off-by: Cristian Ionescu-Idbohrn <cristian.ionescu-idbohrn@axis.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This reworks build.sh for proper argument parsing and variable quoting.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
In the Qt installer, there is a MaintenanceTool which can upgrade your
install, so don't install in a "versioned" directory, just install in a
plain Qt-directory.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Shellcheck wanted quotes around "$USE_X" - but that makes the script
fail if you run it without the '-x' argument.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Move it to packaging/android where it belongs
Use direct URLs to download Android components
Make sure required packages are installed (only tested on Ubuntu)
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Updates mainly in the comments of the MXE build script:
- File system layout: different proposal
- Hint to not use MXE current version from git
- MXE build JOBS setting to (very) safe value of 1
- export CXXFLAGS=-std=c++11 added for marble build
android-mobile nowadays hardcoded in CMakeLists.txt, so workaround it
here.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This one's weird. We actually don't access the Photo Library. But
maybe it's the access to the local files (in order to store the
dive data) that causes this?
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This still isn't quite straight forward, but at least now the README matches
the process that I use again.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
We don't use ssh-based git in Subsurface-mobile, so there's no reason to
link against it.
This should hopefully fix the current issues with the Android APK on some
devices.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
It appears that instead of statically linking against ssl/crypto/ssh2, you
instead have to dynamically link against it and then bundle the library in
the APK. The documentation is not 100% clear and I don't have an Android
Nougat device to test this with, so for now this is an attempt.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Just link it directly into Subsurface-mobile. That's what we already do
with the qmake file for iOS, now the cmake based builds do the same. This
should remove a lot of issues.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Run all scripts with -e so they exit as soon as something breaks. That
way the build stops at the first error, not some other error.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Build kirkigami plugin out of source and make sure that we use the same QT
version for the plugin and the app.
Signed-off-by: Joakim Bygdell <j.bygdell@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Instead of building a library that we link against, let's just use the .pri
file and include Kirigami in the primary build.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Now kirigami needs to be built with a C++ plugin.
In cases of mobile operating systems such as iOS (and in a lesser measuse,
Android) having a proper plugin loaded at runtime may be difficult, so
statically link it together with all of its qml files compiled as a
qresource inside the static library.
Signed-off-by: Marco Martin <notmart@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This creates the possibility to pass configuration, where the ndk and
sdk is installed, to the build.sh script via environment variables.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Since c78e4f we build the mobile and desktop versions with different
package id's.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
With Qt 5.7, they started to require c++11 support, and in 5.6.1 some
nullptr's showed up in QtAndroidExtras/qandroidfunctions.h, so now we
need to compile our c++ code with c++11 support in our compiler.
As Thiago pointed out, this effectively "downgrades" GCC 6 from c++14
support to c++11.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Simply have the Qt link in packagin/ios point to whatever Qt version
you want to build against and the script picks the right one.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The iOS build process is rather stupid - it scans all .qml files
in the root directory of the project to determine which QML dependencies
are required.
This is why we had the weird leftover fake QML project in our sources.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Having build number as mandatory parameter seems unnecessary, thus
setting default value (used for andoird:versionCode) to 0.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Running subsurface/packaging/android/build.sh after deleting
subsurface-mobile-build-arm directory fails for me due to missing
ssrf-version.h file. Just ignore if it doesn't exist.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Having subsurface-core as a directory name really messes with
autocomplete and is obviously redundant. Simmilarly, qt-mobile caused an
autocomplete conflict and also was inconsistent with the desktop-widget
name for the directory containing the "other" UI.
And while cleaning up the resulting change in the path name for include
files, I decided to clean up those even more to make them consistent
overall.
This could have been handled in more commits, but since this requires a
make clean before the build, it seemed more sensible to do it all in one.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Can't believe I didn't notice them earlier... I must have stared at these
strings countless times.
Reported-by: Scott Ireland <scott@sdj.ca>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This reverts commit 78a8137963.
These keys don't enable access, they require the device to have GPS support.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This key is needed so that the iOS app asks the user for permission to use the
GPS position information.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
A lot of this is still black magic, but at least this now documents what I
understand about the process.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The name subsurface-ios was used in many places and that was just not helpful
to fight against. This should work much better.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
And make sure the version displayed for Android includes both that version
and the build version, which is our regular canonical 4 part version
number - so this release will be something like "1.0.0 (4.5.2.1047)"
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This now can create all the support libraries in armv7, but that isn't
sufficient for QtCreator which wants fat libraries with both armv7 and arm64 in
them.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This reverts commit 7fac2a38b2.
Oops, that should have been dropped in the rebase before pushing the
changes. I did this instead in commit 99d1d8876e ("iOS build: create
ssrf-version.h by hand")
It's unclear why the build fails if we don't add the sample app files as
well.
[Dirk Hohndel: refactored the patches]
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This now only builds all the prerequisits but not the actual Subsurface
binaries - that will be done with qmake (oh the irony) in a later commit.
[Dirk Hohndel: refactored the patches]
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
No reasons not to upgrade to the latest OpenSSL lib. The currently used
1.0.1 branch is ending end of 2016, so a switch to 1.0.2 is useful
anyway.
Signed-off-by: Jan Mulder <jlmulder@xs4all.nl>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Since commit c496d5fa05 ("Add helper script to pull Plasma Mobile Components
and icons") we had three different spots where we retrieved the Plasma Mobile
Components. That's a wee bit of overkill. So instead have the other two scripts
just call this one.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Don't ever link against a shared libcrypt. One of the recent changes to
make things build on the various Linux build systems apparently broke the
Android build as it now adds an extra -lcrypt right after the correct
static link to libcrypt.a. Instead of fiddling even more with this and
re-breaking all the other builds I just hack around it here and remove any
calls to a simply -lcrypt as that won't work on Android.
This also passes through the remaining options on the command line to make
so we can do things like VERBOSE=1 or -j12
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
It compiles but the link stage fails because of a missing -LSystem
but its a baby step.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Building the iOS command line utility fails. But frankly, we don't need that,
anyway. I cannot figure out how to tell sqlite that all I want is the library,
so I'm working around that by first building the library, then pretending that
sqlite3 was indeed built in order to be able to run make install. Horrible,
ugly, stupid. But it seems to work.
Also cleaned up the whitespace.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The cross compile script kind of works right now, it's missing
something that I'm really not sure where or what it is.
currently sqlite will not build because:
error: gethostuuid is not defined in iOS
This bug was already opened on sqlite bugtracker for about a year, the
workaround is to pass -DDSQLITE_ENABLE_LOCKING_STYLE=0 to the compilation
flags, which I did but did not work for some reason.
Which is a good error - it shows us that we are actually trying to compile
for iOS.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The build.sh and readme files are the same as the Android ones
and I'll be changing them over time.
The configure-for-ios.sh script is a file that manages to set
everything, compilers frameworks and such, for iOS compilation.
I'll probably dissecate the configure-for-ios.sh file and put it
back on the build.sh, but not now.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
this was requiring libconfuse and we don't need it.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The MXE script still tried to run things using the qmake project
file and we removed that ages ago.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@intel.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Also fixes a capitalization error that prevented finding libssh2 in some
circumstances. And adds a missing include when building with libzip on Mac.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This way build.sh can build a Subsurface.app that the user can use via
open subsurface/build/Subsurface.app
after running build.sh.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
It still is rather specific to my system layout, but at least removes a
reference to my home directory path...
It also removes @rpath references from the executable. This should in theory
work, but it failed for me on one machine that I tested on and doing things
this way doesn't appear to cause problems.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Some of the flags needed to build for android was set in cmake. There
are many more that needs to be set correctly for things to work, so
having some in cmake and some in the Android build.sh is just confusing.
This removes the bits from cmake and moves everything into build.sh.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
These are mostly a convenience for me, they'd obviously have to be updated for
someone else trying to use them.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Since pkg-config is installed in different places on linux and mac,
let's ask where it is.
Signed-off-by: Joakim Bygdell <j.bygdell@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Until we can get libftdi to build on mac we are better of just diabeling it.
Signed-off-by: Joakim Bygdell <j.bygdell@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
git pull -u isn't known on my systems (recent Debian and Ubuntu), so
better not use this flag.
Signed-off-by: Sebastian Kügler <sebas@kde.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Add a new dialog/page which is shown right before
the final "uninstall" click.
The dialog may contains two checkboxes - for registry
entries and for the user path. These checkboxes will not be
created if the user has not run the application yet,
as no registry keys will be available.
Selecting the user directory checkbox shows a warning message box,
that the user should make sure no important files are present there.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Since we ship it all in one piece there is no reason to use a static
libdivecomputer (like there is on Linux). This allows us to give the user
a different libdivecomputer.dll for testing when tracking down a bug.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This seems to fix our issues with being able to create zip files on the
fly (needed for the divelogs.de access).
Fixes#955
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
While this should be fixed in cmake, for now just manuallt get
libssrfmarblewidget.dll and QtXml.dll in place
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Instructions how to use it are in the script.
As of today this doesn't create working binaries - this worked perfectly
fine back when MXE was still based on Qt5.4
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is trying to clean things up quite a bit and make it less specific to
my directory layout.
It still calls the signing script that tries to sign things with my keys,
so it won't work for others without tuning, but this should get you
closer.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This connects the serial_ftdi implementation to subsurface, and builds
libftdi1 for the android builds.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This makes libusb to log straight to the android logging system. This
makes sure we can reach the logs and debug things.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This idea was inspired by:
https://github.com/PointCloudLibrary/mobile/blob/master/3rdparty/android/patches/libusb.patch
The whole thing is re-written from scratch but the idea came from there,
and its a way simpler way of getting a system-opened fd to the right
place than patching every call in the stack to pass a fd down.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This Changes the NDK, SDK, JAVA_HOME and ant to the appropriate versions
and locations when the script runs on a Mac.
Signed-off-by: Joakim Bygdell <j.bygdell@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Now when we build openssl for https support, its easy to add libssh2 to
get ssh support to, so this does that.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
SSL2 and SSL3 are deprecated, and we can gladly leave them out of our
build.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The openssl build step polluted the variable space, and overrode the CC
env, thus breaking the build if you did build openssl in the same
session.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When retrying a clean build several libraries failed to build and I
finally tracked this down to the cross build tools not finding their
sysroot.
Also, on my main build server I have an older cmake version and one of the
tools claims to require cmake 3 but I see no actual incompatibility, so
I'm patching out that check.
Hackish? Yes. But it seems to work.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This teaches android build.sh and cmake about how to build the qml ui of
subsurface-mobile as a apk.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This introduces code to use qt-android-cmake to produce a working apk.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Due to bugs in libgit2, we force linking with libssh2 to add
git-over-ssh support. On android we currency don't want libssh2 due to
its dependency chain, so this makes the libssh2 force linking opt-out.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This teaches the android build script how to do out of tree builds.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This updates the android build script to something that uses CMake.
This can't produce a working APK yet, but it at least builds the shared
object which should be wrapped into the APK.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Also remove some of the code for building on Ubuntu 12.04.
Signed-off-by: Tomaz Canabrava <tomaz.canabrava@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The MinGW based script is still there for reference, but that's no
longer how I build the Windows binaries.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
With this we also give up on building daily builds on Ubuntu 12.04
The cmake based infrastructure very much assumes Qt5 at this point.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This script likely is useless for anyone but me, but I like having it in the
source tree in case others can benefit from seeing how the packages are built.
This now is based on out-of-tree cmake builds.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Update the documentation with dependencies for cross-building on Linux
to Windows for OpenSuse platform and correct some building instructions.
Moreover fix the windows building script to use the architectural specific
binary.
Signed-off-by: Claudiu Olteanu <olteanu.claudiu@ymail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Over one year ago i started with a crazy idea, "Wouldn't Subsurface on
android be nice?" when i read about Android support in Qt.
After playing around with it and doing some quite ugly hacks i got it to
build and run.
Now are all the patches upstream and this imports the quite crude build
script, for others to continue on.
This is a squash-import of what have have happened in
https://github.com/glance-/subsurface-android during 2014.
Signed-off-by: Anton Lundin <glance@acc.umu.se>
CC: Joseph W. Joshua <joejoshw@gmail.com>
CC: Venkatesh Shukla <venkatesh.shukla.eee11@iitbhu.ac.in>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Dang. I didn't pay attention that commit 2677f3ca79 ("LIBMARBLEDEVEL
points to an install dir, not a build dir") broke the way I build the
Linux binaries.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
We only use this for the Ubuntu 12.04 builds. The goal is to move away
from Qt4 support, so this is mainly an afterthought.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
We now have perfectly fine 32bit binaries with Qt5 so no more reason to
steer people towards 64bit binaries. Actually, I don't plan to make 64bit
binaries for the next release.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This script is based on the mxe package and builds everything from source
instead of using the mingw packages from Fedora as I did in the past.
I'm keeping the old script around for now, but eventually I should remove it as
this is the current way to create a working installer that supports both 32 and
64 bit Windows and is Qt5 based.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is simply here for people to look at. It will age immediately and it makes
no sense to try to keep it current here as it is maintained in OBS. But I think
it might be a useful starting point for others who want to package daily builds
of Subsurface.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This reverts commit 7a7ce2c5e0.
Shouldn't have pushed that one :-)
The fix was to modify the spec file, not the name of the directory and tar
file.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is still quite fragile and isn't enough for anyone to run it, but it
captures where I am in the automation process.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
* only build a static libdivecomputer
* only build the libgit2 library, not the executable
* don't echo all the symlinks when fake-installing libmarblewidget
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This now assumes that a running changelog is maintained in
src/debian.changelog, i.e., at the same level as the subsurface tree; the
organization now should look like this:
src/debian.changelog
src/subsurface # subsurface git checkout
src/subsurface/libdivecomputer # libdivecomputer git Subsurface-xx branch
src/subsurface/marble-source # marble git Subsurface-xx branch
src/subsurface/libgit2 # libgit2 git checkout
Instead of running dh_make to create all new debian build files, we add the
necessary files in our script.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Since we strip the .git data from the source tree (to conserve space and not
violate the packaging guidelines - or at least not violate THAT packaging
guideline) we need to create the correct revision before the tar file of sources
is packaged.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Marble can't be static, so instead we build a shared library but give it a
different name so it can be installed in parallel with the "real"
libmarblewidget.so.
Also make sure that the correct libusb is installed so that Atomics Aquatics
dive computers are supported.
Fixes#782
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Oops, I thought I had done that but that was flat out wrong.
Now the source upload shrinks from over 70MB to around 26MB. Much better.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
These files knowingly (one might say, intentionally) violate the spirit
and letter of the Debian / Ubuntu packaging rules. They are intended to be
able to create our own packages that include their own libdivecomputer,
libgit2 and (later) libmarble. Especially for daily builds this is WAY
easier than fighting with whatever may be the current version of these
packages in Ubuntu (especially since this allows us to use our private
libdivecomputer branch).
This assumes that the user runs the make-package.sh script from a
directory below which we have
subsurface/ <- Subsurface checked out git tree
subsurface/libdivecomputer <- desired libdivecomputer sources
subsurface/libgit2 <- desired libgit2 source
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
It makes more sense to do this on init and not have the user go through
any other screens in case this is the wrong binary.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Checking this in to make sure I don't end up creating broken installers
again. I doubt that this is useful for anyone but me - but then, I don't
think anyone but me creates Windows installers.
Background - when Fedora 20 updated the cross-built version of Qt for
Win64 something broke. Subsurfae installed with those DLLs will crash.
Replacing the older 5.3.1 DLLs fixes this for now, so I have a directory
with just those DLLs and simply replace them in the staging directory
before calling makensis.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
With these changes we link statically against libusb and libdivecomputer
but don't add the .a files to our installers.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This also makes sure that we package the Qt5 translations, not the Qt4
translations.
There was an odd issue that somehow a 32bit search path ended up being
used by win-dll which resulted in the wrong DLLs being packaged.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
I assume the theme directory should be deleted on uninstall the same way
e.g. Documentation directory is deleted.
Signed-off-by: Miika Turkia <miika.turkia@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
I was confused by the function name getSubsurfaceDataPath() - it does not
find paths relative to the "data" folder, if finds the path where we might
install folders like "data", "translations", or "theme".
"data" is for some reason where we install the "marbledata" files.
Therefore on both Mac and Windows we need to put the "theme" directory
next to the "data" directory, not below it.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Admittedly I believe I'm the only one using this script (and related .nsi
file), it still seems to make sense to keep it up to date in the
repository.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
When doing an out of tree build you don't want to stage the package with
the source but under your current directory. So let's make sure we
distinguish between source and target here... and instead of putting
things into packaging/windows they now end up in staging which is much
more consistent. And to make my life even easier, the installer .exe ends
up in the base dir in which you build the package.
Also, we link statically against libdivecomputer, so don't pack the dll.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is all mostly to make my life easier.
I'm not thrilled with the marble changes - as Linus pointed out before the
way we do these "LIBxxxDEVEL" changes is broken as it will still first
link against any library installed in the system. But since I have removed
any globally installed copies of these libraries this actually works for
me and it does help when experimenting with different build options for
the main libraries that we depend on.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This way I can have a different directory from where I build Windows
binary without interfering with my native build in the source directory.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The bundle signature is not a necessary property in any of the OS X
versions we support. And the current bundle version identifier is 6.0,
not 1.0.
Signed-off-by: Thiago Macieira <thiago@macieira.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is required to enable HiDPI support for the Retina displays. The
Info.plist that comes with Qt had this, but the one we supply with
Subsurface didn't.
Done-with: Jake Petroules <jake.petroules@petroules.com>
Signed-off-by: Thiago Macieira <thiago@macieira.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
On one of my machines codesign doesn't find my signing key by hash, but
does find it by name. Go figure.
Also on that same system (32bit Mac Mini with running Snow Leopard / 10.6)
gcc 4.2 doesn't support the -unused-result warning. So let's only turn
that on for more modern versions of gcc
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This means we no longer need to keep them on disk and worry about
installing / uninstalling them. They will always be kept in-memory
(compressed).
Signed-off-by: Thiago Macieira <thiago@macieira.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
We appear to be missing the correct dll. I'm out of time trying to track
this down, so I just switched Subsurface to access divelogs.de via http on
Windwos.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
These are only useful for me, but having them in git makes my life so much
easier...
Instead of using macdeployqt to create my DMG I use the tool that I used for
Subsurface 3. This allows for much prettier DMG content as well.
Fixes#329
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
They are loaded into a Qt resource and always accessed via it.
[Dirk Hohndel: had to hand edit / apply the changes to the .pri file]
Signed-off-by: Thiago Macieira <thiago@macieira.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Install the Documentation and include it in the installer.
Try and get all the directories and files removed in the uninstaller.
Where the heck does 'oldshare' come from?
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The executable shortcuts were lacking icons. This should
do the trick, by using the packaged subsurface.ico.
Perhaps it would be better if we hardcode the icon into
the executable as a resource.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
And increase our list of plugins to be deployed to include the GIF and
SVG image plugins, the SVG icon engine and the CJK text codecs.
The install rules now iterate over the plugin list and deploy the
plugins in the right path in the staging area. The plugins must also
be scanned for dependencies (Fedora's qjpeg4.dll depends on
libjpeg-62.dll, which neds to be copied to the staging area).
Finally, fix qt.conf needed to be fixed.
Signed-off-by: Thiago Macieira <thiago@macieira.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
That way, the NSIS rules also work for creating an installer for debug
builds. Which you'd do by running:
make -f Makefile.Debug installer
Signed-off-by: Thiago Macieira <thiago@macieira.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
My system has libzip-1.dll, but Dirk's is probably newer and has
libzip-2.dll
Signed-off-by: Thiago Macieira <thiago@macieira.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The "make install" step will copy all we depend on DLLs there.
Signed-off-by: Thiago Macieira <thiago@macieira.org>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
This is seriously flawed. makensis is run twice for some reason. I also
noticed that the data and xslt directories under packaging/windows aren't
created when running make install. Running
make -f Makefile.Release install_marbledir install_deploy
works, but obviously this should be taken care of by the dependency.
The installed binary under Windows is not finding its icon, the
translations are missing... lots of work left to do here.
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
Marble doesn't work, yet (Google Maps aren't loaded), but at least
Subsurface starts under Windows with the installer built.
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>
On installation, set the "DisplayVersion" registry value
to ${SUBSURFACE_VERSION}, so that a version is displayed
when browsing the list of installed programs.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
The NSIS script on installation will write a key to the registry
that will be shown to the user as a "Subsurface" entry (with icon)
in the list of installed programs that can be uninstalled
(e.g. in the Control Panel).
On uninstall, said registry key will be removed.
Signed-off-by: Lubomir I. Ivanov <neolit123@gmail.com>
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
commit 59294029f3d1 ("Capitalize package name and add capitalized tar-ball
prefix") had an unintended side effect: the cross build for Windows on
Linux no longer worked (as it set NAME=subsurface.exe).
Fixed this by introducing a TARGET variable that is derived from $(NAME).
Signed-off-by: Dirk Hohndel <dirk@hohndel.org>
