Code and tests to demonstrate migrating of older auto-save files, with
blacklisting support to exclude properties. Disabled pending agreement
on the required blacklisting values.
Some pieces of code such as fgMainInit() and, by cascading effect,
fgInitHome(), were careful to return a meaningful value indicating
success or error, however the main() function in src/Main/bootstrap.cxx
ignored it royally so far.
main() now returns:
- EXIT_FAILURE if fgMainInit() or fgviewerMain() throws an exception;
- whatever said function returns otherwise.
- Rename fatalMessageBox() to fatalMessageBoxWithoutExit(). This should
prevent the kind of bug that prompted this set of changes: someone
calling fatalMessageBox(), assuming the program would stop at that
point, whereas in reality it did not.
- Add new function fatalMessageBoxThenExit(). This is not vital of
course, but allows one to spare one line here and there and to apply
the DRY principle for such fatal exits.
- Replace every existing call to fatalMessageBox() with one or the other
of the two new functions. Improve formatting along the way. This
fixes a few bugs of the kind explained above.
Basically, this is because fatalMessageBox() is only safe to call from
the GUI thread, however it seems fg_terminate() can be called from any
thread (according to C++11 semantics). Additionally, fatalMessageBox()
typically requires some work to happen in the GUI thread (event loop) in
order to display something, but we can't realistically expect this while
running a terminate handler just before the program dies.
See messages around
<https://sourceforge.net/p/flightgear/mailman/message/35775803/> for a
discussion of this subject.
+ Minor header cleanup (<locale.h> replaced with <clocale>, etc.)
We probably need a warning for cross-aircraft paths, but leaving that
for a separate change since I’m worried it will warn on MP aircraft.
Maybe better checked in the Python scripts than in the app?
Compile a useful subset of FG as a shared library, and add two basic
uses of this to exercise some Flightplan / RoutePath / navaid
functions.
The test framework can/will be expanded incrementally from here, this
is just a starting point.
As part of this, add the ability to distinguish default vs explicit
airport selection via a new /sim/presets/airport-requested flag. This
enables us to more cleanly handle different combinations of startup,
especially the case where the user requests an airport but no runway
(wants auto selection), ensuring we don’t look for the default airport’s
runway (from location-presets.xml) in that case.
- Declare 'datTypeStr' and 'defaultDatFile' as public member variables
of NavDataCache ('defaultDatFile' is not *required* for this commit,
it just seems to make sense to treat both members the same way/keep
them together in the source code).
- New keys under "navigation data" in the JSON report: "fix.dat files"
and "nav.dat files".
Make a single Cmake value to expose the build type to code, and use
this to default a run-time ‘developer-mode’ property, which can be
over-ridden from the command line.
Use this to drive the different warning levels. Policies subject to
review, especially whether nightly builds should default to
developer mode or not.
Uses TTF fonts, and displays more information textually including
the application version and current aircraft.
Also rename FGRenderer::splashinit to preinit, as was suggested
a long time ago.
This change is the logical counterpart of SimGear's change from commit
79f869a7f32910197be72b21f6489fbbba02c836 that moved the following files
from simgear/misc to simgear/io/iostreams:
gzcontainerfile.cxx
gzcontainerfile.hxx
gzfstream.cxx (formerly zfstream.cxx)
gzfstream.hxx (formerly zfstream.hxx)
sgstream.cxx
sgstream.hxx
sgstream_test.cxx
Each argument creates another log file, in the directory named. Symbolic
value ‘desktop’ creates logs on the user’s desktop.
Needs corresponding SimGear commit to build
For unknown reasons this seems to alleviate the word-wrap / min-height
bug on Windows. Committing so we can test and verify this is really
the case before the next release.
Requires FGData commit: 0565eaab10a5d466cd485766b17d1870936a0a57
(which actually renames the file).
Also disables the preferences-load command since I don’t believe it
would actually be safe to reload the defaults without doing a simulator
reset (aircraft -set.xml values would be overwritten, for example)
As discussed on the devel list, only require the major+minor versions
of FG+SG+data to match by default. If we encounter a situation on
a release branch where stronger checks are needed, it’s easy to
restore.
Essentially, adapt two places where options.xml was supposed to be found
in $FG_ROOT (in one of these, the comment was already incorrect way
before the recent change moving options.xml out of FGData).
$FG_INSTALL_PREFIX represents the FlightGear installation prefix, such
as /usr, /usr/local or /opt/FlightGear on Unix systems. Copying the
--help output and translated strings there avoids having to write to
$FG_ROOT when 'make install' (or some OS-dependent equivalent) is run
from the FlightGear build directory---that would be ugly when $FG_ROOT
points to the FGData Git repository.
In FGLocale::FGLocale(), Translations/locale.xml is loaded using
readProperties() and fatalMessageBox() (in case an error is
encountered). Note that it couldn't be loaded via fgLoadProps() in the
current state, because this function relies on guiErrorMessage() when an
error is encountered, which calls mkDialog(), which itself does
globals->get_subsystem("gui"). This last call can't be done from
FGGlobals' constructor---where the 'globals' pointer is still
NULL---hence the need for a different mechanism not relying on
FGGlobals.
For consistency, and also because it provides a better user experience[1],
load options.xml using the same method instead of with fgLoadProps().
[1] I.e., in case of an error, the user gets to see a graphical popup
window with an explanatory message before FG exits, assuming he is
either on Windows, or on Mac, or has Qt support built in FG, as
opposed to only an SG_LOG() call [because when options.xml is
loaded, guiErrorMessage() used by fgLoadProps() can't use the 'gui'
subsystem].
- Add an optional argument to flightgear::initApp(): doInitQSettings.
This argument defaults to true, preserving initApp()'s behavior in
this respect. If this argument is set to false, FGGlobals doesn't have
to be initialized.
- New function flightgear::initQSettings(), called by
flightgear::initApp() when its 'doInitQSettings' argument is true.
This allows initializing the QSettings exactly when it is needed.
- New function flightgear::checkKeyboardModifiersForSettingFGRoot().
The code it contains used to be run from initApp(), which is
undesirable because:
1) initApp() is not only called at FG initialization (fgMainInit()),
but also from QtMessageBox(), from QtFileDialog::exec() and twice
from Options::setupRoot(). However, checking the Alt and Shift
modifiers to set 'fg-root' in QSettings to the special value
"!ask" only makes sense in fgMainInit(), not in these other
places.
2) This code relies on the QSettings to be set up, and therefore on
FGGlobals. Thus, freeing initApp() of its dependency on FGGlobals
requires splitting this keyboard modifiers checking code out of
initApp().
This is likely to fix the problem preventing startup on Windows when the
username contains non-ASCII characters (cf.
<https://forum.flightgear.org/viewtopic.php?f=22&t=31320>). Thanks to
Headhunter76 for the useful report and to wkitty42 for doing the liaison
officer. ;-)
I can't actually test this, because I don't have Windows. Windows users
should report whether this works for them.
Search for the closest ground-net node near but /not/ on, the requested
runway. This works fairly well, although for some airports the selected
node is surprisingly far from the runway.
The report now looks like:
{
"meta": {
"type": "FlightGear JSON report",
"format major version": 1,
"format minor version": 0
},
...
}
When making compatible changes to the format (e.g., adding members to
JSON objects), only the minor version number should be increased.
Increase the major version number when a change is backward-incompatible
(such as the removal, renaming or semantic change of a member). Of
course, incompatible changes (like this one) should only be considered
as a last recourse.
- New methods Options::printJSONReport() and utility method
Options::OptionsPrivate::createJSONArrayFromPathList().
- Small addition to Options::processOptions(), since the --json-report
option must be processed *after* the TerraSync and download dirs, as
well as aircraft and scenery paths, have been finally set.
Apart from providing a public method giving a path to the autosave file,
the main idea of this commit is to reduce redundancy where
globals->get_fg_home() was so far used in every place where the autosave
file is needed or saved. Use an optional argument for
FGGlobals::loadUserSettings() and FGGlobals::saveUserSettings()[1],
since it should be exceptional to access an autosave file in another
location than $FG_HOME.
Also add comments explaining how to avoid security pitfalls with
saveUserSettings() (cf. discussion around
<https://sourceforge.net/p/flightgear/mailman/message/35461636/>).
[1] Argument *added* to this method, for consistency with
FGGlobals::loadUserSettings().
Similar to the existing FGGlobals::get/set_terrasync_dir(), add
FGGlobals::get_download_dir() and FGGlobals::set_download_dir() methods,
and of course the corresponding FGGlobals::download_dir public member
variable. FGGlobals::set_download_dir() stores the realpath() of the
given directory, including into the /sim/paths/download-dir property,
which is marked as read-only just as /sim/terrasync/scenery-dir already
is.
Handle the setup of the TerraSync and download dirs all in the same
place (Options::processOptions()), since most of the work is already
done there. This allows one to get rid of fgOptTerrasyncDir() and
fgOptDownloadDir(), and to make it easier to see that
globals->set_terrasync_dir() (resp. globals->set_download_dir()) is
called on the correct SGPath, regardless of whether --terrasync-dir
(resp. --download-dir) was passed.
Always create the TerraSync and download dirs when they don't already
exist, regardless of whether --terrasync-dir or --download-dir has been
given on the command line.
Add comments explaining how to avoid security pitfalls with download and
TerraSync dirs (cf. discussion around
<https://sourceforge.net/p/flightgear/mailman/message/35461636/>).
Adjust indentation where it was too broken, hampering readbility.
1) Paths and the FG_SCENERY paths list are now printed like:
FG_ROOT=/home/flo/flightgear/src/fgdata
instead of:
FG_ROOT=Path "/home/flo/flightgear/src/fgdata"
Moreover, FG_SCENERY is now correctly printed (without 'Path ""'
wrapping, without Terrain, Objects and markers instead of the actual
scenery paths) regardless of the position of --version relatively to
--fg-scenery, --terrasync-dir, etc. Of course, the values given to
these options do influence the output of --version.
Simplify printing of FG_SCENERY via SGPath::join() and use the
correct, OS-dependent separator between paths, SGPath::pathListSep.
Write the --version output to stdout, as already done for --help;
then it can be easily piped to $PAGER and doesn't get mixed with the
output of SG_LOG (which is on stderr). This is of course backward
incompatible for programs reading the stderr output of
'fgfs --version', but has been agreed upon on FlightGear-devel (cf.
<https://sourceforge.net/p/flightgear/mailman/message/35461619/>).
2) Don't write explicit values in the OptionResult enum: they don't
matter, but giving them is error-prone in case one does a copy-paste
to add a new member to the enum and forgets to change the value.
It is not needed anymore to hardcode the number of lines of
$FG_ROOT/Airports/apt.dat.gz. The new method, which relies on
SGPath::sizeInBytes() and sg_gzifstream::approxOffset(), works as well
for the other dat files.
Add a new NavCache rebuild phase, REBUILD_READING_APT_DAT_FILES, since
the process is now made of two parts.
Rename NavDataCachePrivate::getDatFilesPaths() to
NavDataCachePrivate::findDatFiles(), because it now returns a
DatFilesGroupInfo (new struct) instead of a PathList. For the same
reason, rename NavDataCachePrivate::aptDatPaths to
NavDataCachePrivate::aptDatFilesInfo. Adapt signatures, etc.
This requires up-to-date SimGear and FGData.
Expand existing —-config option to read either property-XML files (as
it previously did) but for non-XML files, to parse them as command
line argument files.
With Simgear commit d7d59b08a2f1a77a4247ec1a89d6ff48ed73f5c7, this
allows terrasync to be initialised from files in the install data,
which avoids downloading them again.
Relocate the start-up position to a plausible hold-short position
when MP is active and a runway start is requested. This does not (yet)
use ground-net data to identify a real hold-short position.
Remove uses of .str(), .c_str() and some other methods of SGPath.
Pass SGPath directly where possible, or explicitly convert to the
appropriate 8-bit encoding.
When set on the command line, will be used for aircraft packages. When
set in the Qt launcher, will also be used for aircraft downloads at
all times.
When changing the path in the launcher, the set of aircraft catalogs
is refresh automatically. Note the default catalog may need to be
re-installed.
The --metar option has no effect unless --disable-real-weather-fetch is
also passed. This often makes users believe that --metar doesn't work.
Change the implementation for --metar so that it automatically implies
--disable-real-weather-fetch.
After removing /Models from FGData loading of shared models from
within a scenery model failed because the ResourceManager did not know
about additional scenery paths.
The problem was that the change to make FGScenery a standard subsystem caused
the particle OSG group to no longer be inserted into the scene graph. The
solution was to convert the particle group to be set up as a standard branch of
the scene graph, alongside the terrain, models, aircraft, and interior branches.
The particle system is now set up as part of the init() subsystem function call,
so it is compatible with the new subsystem design.
This follows from http://thread.gmane.org/gmane.games.flightgear.devel/78650 and
resolves the sign bug https://sourceforge.net/p/flightgear/codetickets/1778/ .
Combined with a matching change to FGData, this changes the HUD formats from the
current set of 3 (note that the text in brackets is not shown in the HUD
preferences PUI dialog, but is show here for reference):
0) Decimal degrees (37.618890N -122.375000W)
1) Degrees, minutes (37*37.133N -122*22.500W)
2) Degrees, minutes, seconds (37*37 08.0 N -122*22 30.0 W)
to (here the text in brackets will be shown in the PUI dialog):
0) DDD format (37.618890N 122.375000W)
1) DMM format (37*37.133'N 122*22.500'W)
2) DMS format (37*37'08.0"N 122*22'30.0"W)
3) Signed DDD format (37.618890 -122.375000)
4) Signed DMM format (37*37.133' -122*22.500')
5) Signed DMS format (37*37'08.0" -122*22'30.0")
6) Zero padded DDD (51.477500N 000.461389W)
7) Zero padded DMM (51*28.650'N 000*27.683'W)
8) Zero padded DMS (51*28'39.0"N 000*27'41.0"W)
9) Trinity House Navigation (51* 28'.650N 000* 27'.683W)
- Use setlocale() to ensure consistent handling of locales and
string handling irrespective of whether or not QCoreApplication
is invoked. Forces a the C locale for numerics and collation,
since many pieces of FG assume this.
- add a button to the main dialog, explaining how to adjust the
fg-root path via the GUI
- tweak the GUI flow to support explicitly changing the path even
when the default path is acceptable.
Under some conditions on my system
aircraft_paths.begin() was equal to scenery_paths.end()
This resulted in neither of them being added to read_allowed_paths
followed by failure to load Nasal scripts from the aircraft directory.
This might happen, if scenery_paths gets allocated just before aircraft_paths
in memory.
Better loop across the desired lists instead of using fancy tricks with
the iterator.
Avoid intermittent race conditions on shutdown when the OSG
pager/Db thread is running. Ensure the OSG threads are stopped before
we start tearing down any scene graph nodes.
- remove some global headers from AI headers, to avoid pollution
- change how ATC owns the ‘player’ FGAIAircraft so reset works
- ensure AIAircraft controllers are cleared on unbind for reset
This is for the FGProperties::getLongitudeString() and
FGProperties::getLatitudeString() functions. The previous algorithm was to
round the degrees up by the smallest fraction required to prevent a round up to
60 minutes or seconds, and then round down the final minutes or seconds by the
same fraction. The new algorithm is to detect if the final minute or seconds
will be rounded to 60 by the string formatting and, if so, the higher unit
(degrees or minutes) is incremented by one, and the lower unit decremented by
60.
Drop fgNormalizePath, use realpath() only
As this makes it accept relative paths, always use the returned
(absolute) version for the actual file operation to avoid check-to-use
races, or where this is not possible (NasalSGPath) explicitly reject
relative paths
Fix: do_save is a write, not a read
* Before setting /sim/aircraft-dir from the --aircraft-dir option,
canonicalize its value with SGPath::realpath() as is already done in
FGGlobals::append_aircraft_path() for the paths given with --fg-aircraft
or via the FG_AIRCRAFT environment variable.
* This fixes a bug when --aircraft-dir is used, due to the fact that
fgValidatePath() canonicalizes its 'path' argument before matching it
against the allowed patterns, and therefore will not validate paths
under the directory specified with --aircraft-dir if this directory has
been given in a non-canonical form by the user (e.g., containing at
least one symlink component).
* This fix does not lower security: the path which is canonicalized has
been explicitely given by the user. This operation is already done for
all paths specified with --fg-aircraft or via the FG_AIRCRAFT
environment variable, via Options::initPaths() which calls
FGGlobals::append_aircraft_paths().
* To reproduce the bug, create a symlink (e.g., /tmp/aircrafts) to a
directory suitable for --fg-aircraft, then run:
fgfs ... --fg-aircraft=/tmp/aircrafts \
--aircraft-dir=/tmp/aircrafts/SenecaII --aircraft=SenecaII
This will trigger many failures such as:
loadxml: reading '/tmp/aircrafts/SenecaII/Dialogs/registration.xml'
denied (unauthorized directory - authorization no longer follows
symlinks; to authorize reading additional directories, add them to
--fg-aircraft)
(from do_load_xml_to_proptree() in flightgear/src/Main/fg_commands.cxx)
I have also tested this with the ec130b4 and the 777-200ER. Same
problem, same fix.
* If one has the same aircraft in several aircraft directories,
FlightGear should not mix resources from the various aircraft
directories. For instance, if one starts FG with:
--fg-aircraft=/my/personal/dir:/path/to/fgaddon/Aircraft
and one has in /my/personal/dir/ec130 a clone of the upstream
developer repo, FlightGear should use either the upstream version from
/my/personal/dir/ec130 or the FGAddon version from
/path/to/fgaddon/Aircraft/ec130, but not some strange, untested hybrid
of both.
* This commit makes sure that when the looked-up resource starts with
Aircraft/<ac>, where <ac> is the current aircraft name [last component
of aircraftDir = fgGetString("/sim/aircraft-dir")], then
AircraftResourceProvider::resolve() doesn't search other aircraft
directories if the resource isn't found under 'aircraftDir'.
* To reproduce the bug before this commit, you may add the following
code (there is nothing specific about the SenecaII here, it's just the
aircraft I used for testing):
var file_path = resolvepath("Aircraft/SenecaII/flo-test");
if (file_path != "")
gui.popupTip("flo-test found", 2);
else
gui.popupTip("flo-test not found", 2);
in a keyboard binding for the SenecaII (for instance; you may use the
F11 binding that otherwise only prints a short message). You should
add this to the SenecaII/SenecaII-base.xml file *that will be loaded
by FlightGear*, let's say the one under /my/personal/dir in the
example above (beware of the <path-cache> in autosave_X_Y.xml). Then,
by creating or removing a file named "flo-test" in the SenecaII
subdirectory of other aircraft dirs (for instance,
/path/to/fgaddon/Aircraft in the example above), you can see that the
behavior of the loaded aircraft is influenced by the contents of
unrelated versions of the same aircraft that might be present in other
aircraft dirs (e.g., loaded /my/personal/dir/SenecaII influenced by
/path/to/fgaddon/Aircraft/SenecaII).
* Aircrafts loading resources using paths relative to the current
aircraft directory (e.g., with 'resolvepath("flo-test")') are not
affected by this kind of problem, because this scheme is handled by
CurrentAircraftDirProvider, which does not exhibit this bug.
negative latitude/longitude coordinates resulted in negative WEST/
SOUTH coordinates for the default format 0 (zero).
This should be now fixed so that
+12.3 gets formatted as 12.3N/E
-12.3 gets formatted as 12.3S/W
https://bugs.debian.org/780867
This messy approach is to minimise changes during freeze; for 3.7,
I plan to make realpath() handle non-existent files as "realpath
they would have if created now" and get rid of fgNormalizePath
This is insecure because it always (not just on Windows) converts
\ to / before .. checking. Either use the path it returns (as in
f_open()) or use an SGPath (where this conversion is already done)
Only a minor problem because the affected functions are limited to
the .sav file type
