Several Android targets (e.g. ANDROID86) fail to build because the required glm header lib is missing.

In src/Form/DataField/Prefix.cpp, PrefixDataField::GetAsDisplayString() it is possible that 's' is null.
const TCHAR *s = DataFieldString::GetAsDisplayString();
if (StringIsEmpty(s))

Issue Summary:

OLC is giving the oddly-phrased-english-error; "Used Aircraft has engine, but ENL logging deactivated" For any .icg files I enter from a self launch glider. There's an old post on a forum to put in a ticket... so here I am.

"ENL Logging" requires detecting engine noise and flags the corresponding .igc b-record

Per the IGC Spec...

"I-Record" needs offset and "ENL" appended and b-record needs a 3-digit decimal added at the offset specified in the I-Record.

200 - typical glider flight
400 - engine on
999 - maximum

The IGC says you need to measure noise level.

Expected Behavior:
OLC will accept .IGCs of a self-launch glider without "errors" as submitted from an xcsoar icg.

It seems that between 6412cdf and 439cd89 the CI builds started failing.

From the build log:

output/UNIX/dbg/driver.a(KRT2.o): In function `KRT2Device::Send(unsigned char const*, unsigned int, OperationEnvironment&)':
/root/project/src/Device/Driver/KRT2.cpp:181: undefined reference to `KRT2Device::CMD_TIMEOUT'

/cc @MaxKellermann

Running current master / 4c6e754 on UNIX, downloading the file "Germany.cup" fails.

Steps to reproduce

Start XCSoar, choose any mode (e.g. SIM), open the FileManager, Add "Germany.cup"

Result

Germany.cup appears in the file list only for a really short time and then disappears again. Additionally the phrase Out of memory appears on stderr.

Debug steps

Backtrace when stepping through the code:

(gdb) c
Continuing.
cannot open /home/kerel//.xcsoar/lua/init.lua: No such file or directory
[New Thread 0x7fffe980e700 (LWP 1479)]
[New Thread 0x7fffe3fff700 (LWP 1480)]
[Thread 0x7fffe3fff700 (LWP 1480) exited]
[New Thread 0x7fffe3fff700 (LWP 1496)]
[Thread 0x7fffe3fff700 (LWP 1496) exited]
[Switching to Thread 0x7fffe980e700 (LWP 1479)]

Thread 6 "DownloadMgr" hit Breakpoint 1, write () at ../sysdeps/unix/syscall-template.S:84
84	in ../sysdeps/unix/syscall-template.S
(gdb) bt
#0  write () at ../sysdeps/unix/syscall-template.S:84
#1  0x00007ffff5263817 in _IO_new_file_write (f=0x7ffff558b520 <_IO_2_1_stderr_>, data=0x7fffe980b680, n=14) at fileops.c:1271
#2  0x00007ffff5262b62 in new_do_write (fp=fp@entry=0x7ffff558b520 <_IO_2_1_stderr_>, data=data@entry=0x7fffe980b680 "Out of memory\n", 
    to_do=to_do@entry=14) at fileops.c:526
#3  0x00007ffff5263f6e in _IO_new_file_xsputn (f=0x7ffff558b520 <_IO_2_1_stderr_>, data=<optimized out>, n=14) at fileops.c:1350
#4  0x00007ffff523acc5 in buffered_vfprintf (s=0x7ffff558b520 <_IO_2_1_stderr_>, format=<optimized out>, args=<optimized out>)
    at vfprintf.c:2346
#5  0x00007ffff5237f25 in _IO_vfprintf_internal (s=0x7ffff558b520 <_IO_2_1_stderr_>, format=0x5555559fb97a "%s\n", 
    ap=ap@entry=0x7fffe980dc28) at vfprintf.c:1293
#6  0x00007ffff5240187 in __fprintf (stream=<optimized out>, format=<optimized out>) at fprintf.c:32
#7  0x0000555555680acc in LogString (p=0x7fffe4022078 "Out of memory") at src/LogFile.cpp:83
#8  0x0000555555680ccb in LogError (exception=...) at src/LogFile.cpp:152
#9  0x0000555555996045 in DownloadManagerThread::ProcessQueue (this=0x555555f21080, session=...) at src/Net/HTTP/DownloadManager.cpp:293
#10 0x000055555599505a in DownloadManagerThread::Tick (this=0x555555f21080) at src/Net/HTTP/DownloadManager.cpp:322
#11 0x000055555599e17e in StandbyThread::Run (this=0x555555f21080) at src/Thread/StandbyThread.cpp:115
#12 0x000055555599c8ae in Thread::ThreadProc (p=0x555555f21080) at src/Thread/Thread.cpp:166
#13 0x00007ffff603c494 in start_thread (arg=0x7fffe980e700) at pthread_create.c:333
#14 0x00007ffff52d9acf in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:97

So, the error is logged by this statement:
https://github.com/XCSoar/XCSoar/blob/master/src/Net/HTTP/DownloadManager.cpp#L293
Cause by an exception raised by this statement:

(preliminary) Explanation

Currently Germany.cup is the only waypoint file using HTTPS. The libcurl error code CURLE_OUT_OF_MEMORY is translated by curl_easy_strerror to the above mentioned error message.

A similar error was reported 2008 on the ML curl-library: http://curl-library.cool.haxx.narkive.com/RwCtIULD/https-request-with-curl-library-gives-out-of-memory-error

It would be great if there could be a third vario needle indicator available for average climb over the whole actual thermal. may be a 90degree right rotatet "T" to indicate the average climb for the whole thermal (additional to the 30second average). This allows you to see extreamly fast if the 30s average is getting below the overall climb rate.
Additionally the color can indicate the tendency. Green for increasing overall climb rate and red for decreasing.

Just wondering why ios was dropped as a supported target, it seems lk8000 has added a build to testflight and it appears to work well. Just wondering why XCSoar hasn't done something similar.

Hello,
I've installed XCSoar to Kobo Clara HD. Almost everything works but when I go to "Network" and tap to "WIFI ON"; nothing happens and "Wifi" button stays inactive.
XCSoar version: 6.8.11

Currently with xcsoar-testing from play store, the plane icon does not point in the same direction to the track. Below a video of the problem:
https://youtu.be/btRwD5vzxsk

This does not happen with 6.8.10 on the same platform.

CXX output/ANDROID/dbg/src/Device/Driver.o
CXX output/ANDROID/dbg/src/Dialogs/Settings/Panels/AudioConfigPanel.o
CXX output/ANDROID/dbg/src/Monitor/MatTaskMonitor.o
CXX output/ANDROID/dbg/src/Monitor/AllMonitors.o
CXX output/ANDROID/dbg/src/MapWindow/OverlayBitmap.o
CXX output/ANDROID/dbg/src/Dialogs/Weather/MapOverlayWidget.o
CXX output/ANDROID/dbg/src/Hardware/Battery.o
CXX output/ANDROID/dbg/src/Dialogs/Device/ScanBluetoothLeDialog.o
CXX output/ANDROID/dbg/src/Java/Global.o
CXX output/ANDROID/dbg/src/Java/Object.o
CXX output/ANDROID/dbg/src/Java/String.o
CXX output/ANDROID/dbg/src/Java/Exception.o
CXX output/ANDROID/dbg/src/Java/File.o
src/Dialogs/Settings/Panels/AudioConfigPanel.cpp:44:66: error: function 'AudioConfigPanel' could be declared
with attribute 'noreturn' [-Werror,-Wmissing-noreturn]
AudioConfigPanel() : RowFormWidget(UIGlobals::GetDialogLook()) {
^
1 error generated.
CXX output/ANDROID/dbg/src/Java/InputStream.o
build/compile.mk:111: recipe for target 'output/ANDROID/dbg/src/Dialogs/Settings/Panels/AudioConfigPanel.o' failed
make: *** [output/ANDROID/dbg/src/Dialogs/Settings/Panels/AudioConfigPanel.o] Error 1
make: *** Waiting for unfinished jobs....

Running latest git clone on linux.
When connecting Condor2 over TCP 4353 all works well and it connects. However after approximately 30sec of moving (the glider) it crashes.
The output is:
xcsoar: src/IGC/IGCWriter.cpp:107: void IGCWriter::WriteHeader(const BrokenDateTime&, const TCHAR*, const TCHAR*, const TCHAR*, const TCHAR*, const char*, const TCHAR*, bool): Assertion `date_time.IsPlausible()' failed.
Aborted (core dumped)

Add sound out via USB so a USB audio dongle can be used

Multiple reports in the forum, that Android 8 breaks current stable on startup.

Currently investigating...

There are some issues when trying to compile the v6.8.x branch for the Kobo. This is on an Ubuntu 18.04 system, using the arm-linux-gnueabihf-gcc-5 in version 5.5.0, with the debug options active.

• The glibc 2.23 that is used seems to be bugged, see https://sourceware.org/git/gitweb.cgi?p=glibc.git;a=commitdiff;h=388b4f1a02f3a801965028bbfcd48d905638b797 for details. A solution is to patch the glibc before compiling in the way mentioned there. I patched a version and uploaded it to github, see nico-krais@3bb96ff

Error message example:

/tmp/ccWYEnuM.s: Assembler messages:
/tmp/ccWYEnuM.s: Error: `loc1@GLIBC_2.4' can't be versioned to common symbol 'loc1'
/tmp/ccWYEnuM.s: Error: `loc2@GLIBC_2.4' can't be versioned to common symbol 'loc2'
/tmp/ccWYEnuM.s: Error: `locs@GLIBC_2.4' can't be versioned to common symbol 'locs'
../o-iterator.mk:9: recipe for target '/home/xxx/codes/XCSoar/output/KOBO/lib/arm-linux-gnueabihf/build/glibc-2.23/misc/regexp.os' failed

• An include is missing in src/Event/Poll/Linux/Input.cpp, meaning compilation will abort. Fix see nico-krais@d178970.

Error message example:

src/Event/Poll/Linux/Input.cpp: In member function ‘void LinuxInputDevice::Read()’:
src/Event/Poll/Linux/Input.cpp:133:9: error: ‘errno’ was not declared in this scope
     if (errno != EAGAIN && errno != EINTR)
         ^
src/Event/Poll/Linux/Input.cpp:133:18: error: ‘EAGAIN’ was not declared in this scope
     if (errno != EAGAIN && errno != EINTR)
                  ^
src/Event/Poll/Linux/Input.cpp:133:37: error: ‘EINTR’ was not declared in this scope
     if (errno != EAGAIN && errno != EINTR)
                                     ^
build/compile.mk:114: recipe for target 'output/KOBO/dbg/src/Event/Poll/Linux/Input.o' failed

Hi!

I recently decided that I might want to look further into contributing to XCSoar.
I feel like the build system is kind of difficult to grasp, especially for non-Linux devs, so I was thinking about providing a CMake alternative for building with conan as a dependency manager.
My desired outcome would be that it is possible to build XCSoar natively on all desktop operating systems (disclaimer: I unfortunately have very little experience with Android and iOS development).

Did anyone look into this before and would this be a desired change?

Feedback is very welcome.

Currently the ICAO code of the airfield. (Field 2 in a cup file) is not visible to xcsoar.

A great way would be to

• display it in the waypoint details
• make it searchable in the waypoint search
• display it in the waypoint results lists along with the airfield name like: Zurich (LSZH)

In 2018 the German Goverment updated the DWD law and mandated that certain DWD products are provided under an open data license.
The license under which the data is published can be found here.

The data itself is accessible via a a geoServer in various formats (GeoTiff, OpenLayer, ....) here.

This enhancement proposed to enable XCSoar to access this data and provide it on request as screen overlay.

Source file src/Screen/X11/TopWindow.cpp fails to build due to missing header.

https://github.com/xcsoar/xcsoar/releases

Hi, Since I wanted to dedicate myself to my "pull request" again and want to test the usability of my patch, I have cloned pure from the master repository xcsoar and compiled. Unfortunately
With the current master version no working xcsoar for Linux (Debian) can be produced. The presentation is not OK.
commit b45fe17

In version 7.0_preview13-Android the pc_met access is not working any more.

• Log give no important informations.
• The credentials are all correct.

Seems dwd changed their web structure on flugwetter.de. In Image.cpp in src/Weather/PCMet the urls are not working by copying them manually in the browser.

• xcsoar.log
• Screenshot_XCSoar

P.S.: Metar and Taf download is also not working. I fixed the error by setting the ftp to https. See #183

Any help would be greatly appreciated! Cheers :)

In our group we're doing a new kind of task that requires passing "goal line" before the end of the flight. For example see "One-line OLC+ Flight" on www.teamxc.us/how-we-fly (2nd graphic). In XCSoar we could easily use the Racing task type if a "Semicircle" option were added. It would be oriented perpendicular to the line leading to it. It looks like a finish line, but it isn't the end of the flight

Alternatively, a "Generic" turnpoint sector would work great: arbitrary angle width of sector, arbitrary orientation to line leading to the sector, and arbitrary width.

Thanks, this would add great flexibility to create new kinds of tasks and competitions.

So, I'm investigating a SEGV in v6.8.11. The symptoms are that, after a few hours of running, 'listener' in class Port (Device/Port/Port.?pp) becomes invalid and Port::StateChanged() fails. However, my hypothesis is that the problem is not here: I suspect that listener is being over written by a memory corruption else where - probably outside the thread in which Port::StateChanged() is running.

I have tried the usual - running under gdb with a hardware watch on listener, running under valgrind and some lint type stuff. The problem is that the SEGV does not occur when run under gdb or valgrind - even when gdb is "clean" (no break points or watches).

So, the question is, is there another way to approach this problem that anybody can think of?

This issue replicates Ticket 3844 on github.

This issue replicates Ticket 1687 here on github.

tietz@tietz-EX58-UD5:~/XCSoar$ make TARGET=ANDROID
./build/thirdparty.py ./output/ANDROID ANDROID arm-linux-androideabi arm-linux-androideabi "-march=armv7-a -mfloat-abi=softfp -mfpu=vfpv3-d16 -fpic -funwind-tables -target armv7a-none-linux-androideabi -gcc-toolchain /home/tietz/opt/android-ndk-r18-beta2/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64 -integrated-as" "-I./output/ANDROID/include -DHAVE_POSIX -DHAVE_STDINT_H -DHAVE_UNISTD_H -DHAVE_VASPRINTF --sysroot=/home/tietz/opt/android-ndk-r18-beta2/sysroot -isystem /home/tietz/opt/android-ndk-r18-beta2/sysroot/usr/include/arm-linux-androideabi -DANDROID -D__ANDROID_API__=21 -isystem ./output/ANDROID/lib/arm-linux-androideabi/root/include -DEYE_CANDY -DTESTING -DBIONIC -DLIBCPP_NO_IOSTREAM " "-Wl,--no-undefined -L/home/tietz/opt/android-ndk-r18-beta2/platforms/android-21/arch-arm/usr/lib -B/home/tietz/opt/android-ndk-r18-beta2/platforms/android-21/arch-arm/usr/lib -Wl,--fix-cortex-a8 -Wl,--no-warn-mismatch" /home/tietz/opt/android-ndk-r18-beta2/toolchains/llvm/prebuilt/linux-x86_64/bin/clang /home/tietz/opt/android-ndk-r18-beta2/toolchains/llvm/prebuilt/linux-x86_64/bin/clang++ /home/tietz/opt/android-ndk-r18-beta2/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/bin/arm-linux-androideabi-ar "rcs" /home/tietz/opt/android-ndk-r18-beta2/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/bin/arm-linux-androideabi-ranlib /home/tietz/opt/android-ndk-r18-beta2/toolchains/arm-linux-androideabi-4.9/prebuilt/linux-x86_64/bin/arm-linux-androideabi-strip
Applying patch /home/tietz/XCSoar/lib/proj/patches/no_programs.diff
patching file src/Makefile.am
Hunk #1 succeeded at 1 with fuzz 2.

Now at patch /home/tietz/XCSoar/lib/proj/patches/no_programs.diff
libtoolize: putting auxiliary files in AC_CONFIG_AUX_DIR, '.'.
libtoolize: linking file './ltmain.sh'
libtoolize: putting macros in AC_CONFIG_MACRO_DIRS, 'm4'.
libtoolize: linking file 'm4/libtool.m4'
libtoolize: linking file 'm4/ltoptions.m4'
libtoolize: linking file 'm4/ltsugar.m4'
libtoolize: linking file 'm4/ltversion.m4'
libtoolize: linking file 'm4/lt~obsolete.m4'
Traceback (most recent call last):
File "./build/thirdparty.py", line 140, in
x.build(toolchain)
File "/home/tietz/XCSoar/build/python/build/autotools.py", line 96, in build
build = self.configure(toolchain)
File "/home/tietz/XCSoar/build/python/build/autotools.py", line 46, in configure
subprocess.check_call(['aclocal'], cwd=src)
File "/usr/lib/python3.5/subprocess.py", line 576, in check_call
retcode = call(*popenargs, **kwargs)
File "/usr/lib/python3.5/subprocess.py", line 557, in call
with Popen(*popenargs, **kwargs) as p:
File "/usr/lib/python3.5/subprocess.py", line 947, in init
restore_signals, start_new_session)
File "/usr/lib/python3.5/subprocess.py", line 1551, in _execute_child
raise child_exception_type(errno_num, err_msg)
FileNotFoundError: [Errno 2] No such file or directory: 'aclocal'
build/thirdparty.mk:45: recipe for target 'output/ANDROID/lib/arm-linux-androideabi/stamp' failed
make: *** [output/ANDROID/lib/arm-linux-androideabi/stamp] Error 1

Currently there is no description of the git work flow in the Developers' Manual. As a consequence it is not clear to the potential patch submitter where they should branch from.

The work flow used in the XCSoar project is relatively simple. Despite it needs to be documented.

Hello, since last update I have an issue with the files attached to landing places, they don't open anymore when I go under details. I went back to 6.8.11 and there is no issue. It seems to be related to the last version.

When selecting airspace/waypoint database to download sometimes I can see the options listed twice on one big list:

output/UNIX/dbg/src/Logger/LoggerFRecord.o: In function LoggerFRecord::Update(GPSState const&, double, bool):
/root/project/src/Logger/LoggerFRecord.cpp:71: undefined reference to LoggerFRecord::ACCELERATED_UPDATE_TIME
/root/project/src/Logger/LoggerFRecord.cpp:71: undefined reference to LoggerFRecord::DEFAULT_UPDATE_TIME

xcsoar: ./src/Engine/Task/Ordered/OrderedTask.hpp:361: const TaskProjection& OrderedTask::GetTaskProjection() const: Assertion `!IsEmpty()' failed.
in @e0418b0e40d8b98329c290dbeb0a8eafac8c696a

Connecting a BLE serial interface device to XCSoar works, and the device will automatically be reconnected while that device's information remains in the Android Bluetooth cache. If the Bluetooth system is restarted (e.g. turn Bluetooth off and on) or the phone is rebooted, the systems's cached information about the device is lost. Then when BluetoothHelper#connect() is called with the MAC address of the device, it is assumed to be a legacy (RFCOMM) device and a connection attempt is made using legacy Bluetooth calls. This fails.

A side effect of this is that the device name is also lost, so it is displayed with the MAC address only in the device list.

So two changes are needed:

1. record the device type (BLE vs RFCOMM)
2. cache the device name for display purposes

Item 1) could be achieved by adding a new device type in DeviceConfig::PortType and changing the interface to the Android code. This won't solve 2)

An alternative would be to encode the device type and name into the MAC address string. This would require a change to the way the mac address is stored in the Config object.

Another alternative is to maintain persistent state for the device in the Android specific code, for example using a SharedPreferences instance or a file in the XCSoar directory. This would map each MAC address to a device type and name. This would not require any change to the non-Android code. The SharedPreferences option is easy, but has the drawback that copying the XCSoar default data directory to a new device would not transfer that information. Creating a file in the XCSoar directory would require knowledge of where that is, and a cursory inspection did not reveal any existing mechanism for Android Java code to access that in XCSoar.

My plan is to implement the last alternative using a SharedPreferences instance. Comments?

When wanting to change the infoboxes positions using the InfoBox Sets Config screens the paste function does not overwrite the selected infobox.

Test Case:
Navigate to

• Config -> System -> Look -> InfoBox Sets -> {any infobox set}
• Select an infobox
• press Copy
• select another infobox
• press Paste
• Press Yes when asked to overwrite

The destination infobox does not change.

Tested in:
XCS7 on Android & XCS6 on Android, Mac and Linux

Currently, Airspace warning dialog replace the bottom area (in this case Cross section). Ideally this shouldn't replace but show it on above or on top of the bottom area.

Before Airspace warning:

Airspace warning appear and Cross section disappear:

After acknowledge Airspace warning, Cross section is back:

I would be happy to get some code pointer so I can help contribute to this.

Using a device with an OLED panel, having a black terrain color ramp would greatly reduce battery consumption.

Hello,
I got this error:
tietz@tietz-EX58-UD5:~/XCSoar$ make TARGET=UNIX64
UNTAR boost_1_65_1.tar.bz2
/bin/sh: 1: quilt: not found
build/libboost.mk:18: recipe for target 'output/src/stamp-boost_1_65_1' failed
make: *** [output/src/stamp-boost_1_65_1] Error 127

My environment:
tietz@tietz-EX58-UD5:/etc$ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu 16.04.5 LTS
Release: 16.04
Codename: xenial
tietz@tietz-EX58-UD5:/etc$ uname -i
x86_64

I used this manual to install the development:
https://download.xcsoar.org/releases/6.8.11/XCSoar-developer-manual.pdf

This shows in short form the steps to provide and install the development environment:
sudo apt-get clean
sudo apt-get update
sudo apt-get upgrade

sudo apt-get install make
librsvg2-bin xsltproc
imagemagick gettext
git

sudo apt-get install make g++
zlib1g-dev
libsdl1.2-dev libfreetype6-dev
libpng-dev libjpeg-dev
libcurl4-openssl-dev
libxml-parser-perl
librsvg2-bin xsltproc
imagemagick gettext
libegl1-mesa-dev
libgl1-mesa-dev
fonts-dejavu

git clone git://github.com/XCSoar/XCSoar
cd XCSoar
git pull
git submodule init
git submodule update
make TARGET=UNIX64

Hi,

I've seen a problem on XCSoar when you want to import big shapefiles, the rendering is not fine and the map looks not good.

French electricity companies have published in open data all the shapefiles of the overground electricity lines (low voltage, high voltage and very high voltage) this month, so i wanted to integrate all the electricity lines into a XCSoar map file for improve pilot's security and share this after with all pilots; but i've not succeed and it was not as expected.

The integration in a .xcm file of shapefile has been easy, after the integration XCSoar have read normally the .xcm file and the maps appears in XCSoar, but all the electricuty lines are divided in little parts, so i can not see the correct route of the electricity lines.

I think it will be easyier to see the problem on my screenshot

All the shapefiles i've used are OK because they came from electricity companies.

I think XCSoar is only rendering on the map a small part of the lines and that is the issue.

I don't know if it is a problem, if it is normal is there a parameter to change in the source code for improve the rendering of shapefiles ?

$ make TARGET=ANDROID
cp: cannot stat 'output/ANDROID/dbg/build/res/xml/accessory_filter.xml': No such file or directory
build/android.mk:178: recipe for target 'output/ANDROID/dbg/build/res/xml/accessory_filter.xml' failed
make: *** [output/ANDROID/dbg/build/res/xml/accessory_filter.xml] Error 1

Since the move of the project to GitHub the text of the README file has become outdated. Make the required changes.

Hello guys,
Please, can you support GDL90 interface? We have a problem getting ADS-B traffic data on XCSoar.
Thanks

PepperJo/stratux#4

The Air Control Display (Manufacturer Air Avionics) can be controlled via RS-232 interface. The device is sending NMEA sentences. Depending on the connected hardware and licence purchased, altimeter, COM and transponder interaction are possible.

An XCSoar driver for the altimeter and COM interaction would be desired.

The documentation seems pretty good in the Installation Manual of the Air Control Display, which can be found here: https://www.air-avionics.com?wpdmpro=air-control-display-57-installationshandbuch&lang=de&wpdmdl=982

I have an OpenVario running XCSoar 7.0 preview, set-up to portrait mode, however the profile selection menu on startup initializes to landscape. After selection it switches to portrait orientation; this also happens when "Default" is selected.
The problem seems to occur irregardless of the device screen orientation setting and the XCSoar setting.

Due to the change in the newer version of IMI Erixx firmware (version 1.2 onwards, authored in 2015) it is not currently possible to download flights from the device (the request to download the list of flights fails).

The problem is caused by the buffer size in COMM_MAX_PAYLOAD_SIZE in the file src/Device/Driver/IMI/Protocol/Types.hpp, which currently has the value of 1024B.

The new correct value should be 2048.

I have been using xcsoar in a kobo which is strapped to my leg and keep having the problem of accidentally (probably with the risers or the harness) 'click' on the screen causing all sorts of misconfigurations.

I am not sure if others have this issue but I would imagine so. This since I could not find any reference to a screen lock feature. I would like to suggest to implement one.

I am thinking an option under the display menu called "lock screen". Once button is selected only the main menu button would be selectable and once clicked would show a popup in some other of the screen to confirm whether the screen should be unlocked or not. Once unlock confirmation popup is shown any other click outside the confirmation button would cancel the action and go back to main menu.

Ideas and or suggestions are very welcome.
Regards,
pq

I know this is a long shot, but:

QT is supported on all the relevant platforms:

• https://doc.qt.io/qt-5/android.html
• https://doc.qt.io/qt-5/linux.html
• https://doc.qt.io/qt-5/windows.html

One of the major complexities is the GUI Toolkit that we currently use. (IMO). Moving to a well known[tm] system would lessen this.

QT would be attractive to developers and is well understood at this point.

kflog also seems to use qt: http://www.kflog.org/2016/07/cumulus-5-29-1/

Currently when we land we send out a signal to LiveTrack24 to notify the end of the flight. It would be great if as a user I could see on the screen a list of options to notify my current status, just like LiveTrack24, FlyMaster tracker and others offer:
(from the docs, End of track packet section)

0-> "Everything OK"
1-> "Need retrieve"
2-> "Need some help, nothing broken"
3-> "Need help, maybe something broken"
4-> "HELP, SERIOUS INJURY"

I think this has a lot of value for all pilots out there for obvious reasons.

What do you think?

https://www.xcsoar.org/ returns "Error 403"

Forbidden

You don't have permission to access / on this server.

---

Apache/2.4.25 (Debian) Server at www.xcsoar.org Port 443

It seems that git.xcsoar.org is currently not or no longer available, but the android/ioio submodule is still pointing to it on both branches, which currently makes this project (and others depending on it) impossible to build.

/cc @MaxKellermann

Recently I've been experiencing problems with XCSoar freezing in mid flight. Sometimes the program just stops working and the screen stays static - eventually I get a popup to say XCSoar has stopped responding. Other times some of the data boxes lose their data for several seconds, the screen icon disappears, then they come back on again. I'm running XCSoar on a Motorola (Lenovo) Moto G4 phone and have been using this for ages like this. Recenty I paired it with XCTracer on BLE and first flights showed no issue, but now this has started. I thought it might have been a conflict with AirWhere which is also connected by BLe to the phone but I tried with AirWhere completely disabled (and a full reboot of the phone) but the problem persists.

Anyone have similar experience or can suggest a way to cure this?

Hi,

I bought a kobo aura edition 2 last week.
Installed xcsoar 6.8.11

Tapping 'Network" 'button' (before starting xcsoar on main screen) display shows: 'WIFI ON' but in fact the wifi is not working. When pressed whole line turns blacks for a couple of seconds but wifi still not working.
Second line 'WiFi' is greyed out.

Sticker says: product # N236-KU-BK-K-EP
FCC ID:NOIKBN236
IC: 8857A-KBN237

White led keeps flashing

Kobo runs the latest firmware.

Regards, Jerome