fly/flightgear
1
0
Fork 0
Code Activity
14233 commits 1 branch 0 tags 41 MiB
Commit graph

411 commits

Author SHA1 Message Date
James Turner
e509711075 Fix some warnings from current Apple clang 
- remove some unused vars
- remove use of std::mem_fn / std::bind2nd
 2020-03-10 17:02:08 +00:00
Erik Hofman
32d6f75305 Process the buttons even if the axes are not declared initialized 2020-03-07 09:22:22 +01:00
James Turner
219ceacc7a macOS Catalina device filtering 2020-02-24 13:09:36 +00:00
Erik Hofman
6df768fa5f Store the initial axis values and if just one of them changes declare the joystick initialized 2020-01-27 18:12:09 +01:00
Erik Hofman
65a20b5829 Read the raw values and check them against the saturation value, if one of them is larger then initialization of the joystick is still in progress. 2020-01-27 09:15:53 +01:00
Erik Hofman
2d0b44ee63 If js is valid (again), create a new instance of jsJoystick so removing an joystick from and USB port and attaching it again get the joystick in a working order. 2020-01-26 17:44:19 +01:00
Erik Hofman
d4459c8205 Fix a typo 2020-01-26 14:30:24 +01:00
Erik Hofman
3405ea2aaa Joystick axes can get initialized to extreme values, at least on Linux. Hold of working with the axis values until all values have become sane at least once 2020-01-26 14:10:07 +01:00
James Turner
0e082d5c46 Tweaking macOS HID code to avoid a Catalina crash 
Not certain what’s going on here, this is testing a theory that
deleting a HID queue which has never been started, might behave badly.
 2019-10-17 00:01:59 +01:00
Julian Smith
331ef3232f src/Input/FGEventInput.cxx: FGEventInput::AddDevice(): fix incorrect use after delete. 
Moved delete of inputDevice to after its use in a SG_LOG diagnostic.
 2019-08-31 14:19:16 +01:00
Edward d'Auvergne
34a6cb3c74 SGSubsystem classes: Registration of all subsystems. 2019-07-01 14:51:32 +02:00
Edward d'Auvergne
d259c70560 SGSubsystem classes: Addition of staticSubsystemClassId() to all subsystems. 2019-07-01 14:51:32 +02:00
Edward d'Auvergne
332c240b38 SGSubsystemGroup: Removal of the subsystem group naming. 
This is a partial reversion of 5253215065, the
changes of which are incompatible with the subsystem class IDs.
 2019-07-01 14:27:30 +02:00
Edward d'Auvergne
fb5c36fc82 SGSubsystem classes: Renaming of the subsystem ID variables and functions. 2019-07-01 14:27:30 +02:00
Edward d'Auvergne
7a6e63ddb9 SGSubsystem classes: Subsystem and subsystem group API declaration standardisation. 
This is a cleanup commit.
 2019-07-01 14:27:30 +02:00
Edward d'Auvergne
60a2444766 SGSubsystem classes: Whitespace standardisation of the declarations. 
This is a clean up commit prior to the subsystem API standardisation to simplify
the diffs.  It includes all SGSubsystem and SGSubsystemGroup derived classes.
 2019-07-01 14:27:28 +02:00
James Turner
0bea6153b0 Event input: button fire repeat 2019-06-05 17:41:07 +01:00
James Turner
5ffe96237f Fix a warning 2019-06-05 17:41:07 +01:00
Ganael Laplanche
66e6afb449 Avoid crashing when udev_device_get_parent() returns a NULL pointer 2019-05-29 12:06:07 +02:00
James Turner
bd2c524b21 HID input: more debug output 2019-05-27 16:06:27 +01:00
James Turner
955138d06e Avoid a crash on HID devices with no name. 2019-05-16 11:11:19 +02:00
Florent Rougon
9b019f6b72 FGJoystickInput: increase MAX_JOYSTICKS from 10 to 16 
Since PLIBjs uses joyGetDevCaps()[1] on Windows and this function
accepts joystick ids from 0 to 15 (-1 being special), setting
MAX_JOYSTICKS to 16 ensures that FGJoystickInput::init() can scan all
joysticks that joyGetDevCaps() is able to report about.

[1] https://docs.microsoft.com/en-us/previous-versions/dd757105(v%3Dvs.85)
 2019-04-29 12:54:03 +02:00
James Turner
9d71d09072 Enable HID-input on Windows 2019-04-27 18:54:13 +01:00
Richard Harrison
5253215065 Improvements to performance monitoring. 
* Ensure that subsystems are named.
* Nasal timers are named appropriately
 2019-02-10 18:07:07 +01:00
James Turner
07cb303e8c Fix for jumping view when right-dragging 
This happens because we were not storing the new position when the
OSG event-adapter handled the event, which includes PUI

https://sourceforge.net/p/flightgear/codetickets/2073/
 2018-10-24 22:16:51 +01:00
James Turner
951a3ea6ad Fix Windows build, missined windows.h 2018-10-24 14:28:58 +01:00
James Turner
00eec40d51 Fix JS header include path 2018-10-23 11:00:59 +01:00
James Turner
73be50b308 Event-input: axis interpolater support 
Expand the range of filtering options on axis values to include the
optional use of an interpolation table, which gives many more options
when dealing with some devices.
 2018-10-13 00:03:30 +01:00
James Turner
6589c71f62 Tweaks to HID local descriptors 2018-10-04 23:21:05 +01:00
James Turner
abbaf8f640 HID input: allow descriptors from XML 
This allows ignoring the USB-supplied descriptor in favour of our
own, which might considerably better or less broken at least.
 2018-10-04 17:43:36 +01:00
James Turner
6366dea197 HID event-input debugging 
Adding more debugEvents-enabled log output to HDI input
 2018-10-04 12:17:01 +01:00
James Turner
96cce67040 HID input: fix multi-device Nasal support 
Ensure each laoded device gets a unique Nasal module name,
otherwise they combine with each other and Stuff Breaks(TM)
 2018-10-04 12:17:01 +01:00
James Turner
8b72229ffd C++11 improvements in Mac event-input code 2018-08-19 12:34:47 +01:00
James Turner
e24134f34a HID-input: fix crash with null manufacturer name 
https://sourceforge.net/p/flightgear/codetickets/2041/
 2018-08-10 15:11:32 +01:00
Edward d'Auvergne
fff923d2f4 TestSuite: Migration of the HIDInput tests to the CppUnit framework. 
The extractBits(), signExtend(), and writeBits() functions of the anonymous
namespace in Input/FGHIDEventInput.cxx have been shifted out of the namespace
and are now exposed via the header.  This is needed as
<Input/FGHIDEventInput.cxx> cannot be imported within the CppUnit
infrastructure, as it is already built into the run_test_suite binary.
 2018-07-26 15:52:07 +02:00
James Turner
b690be1206 Fix HID input init 2018-06-25 23:39:22 +01:00
James Turner
5741cad568 HID tweaks 2018-06-25 23:39:22 +01:00
James Turner
054c4bec0f Bug-fix: for RMB-look / PUI interaction issue 
When releasing the RMB over a PUI dialog, we no longer get stuck in
drag mode due to the buttons map getting out of state. This is slightly
ugly work-around, better fix to follow hopefully (needs better OSG
integration)
 2018-05-23 14:16:48 +01:00
James Turner
8986abeb52 Fix PUI mouse for non-pass-through mouse modes 
As reported by Michael Danilov, ensure sticky mouse modes for view and
flight controls don’t send button events to PUI
 2018-05-14 14:41:06 +01:00
James Turner
102fe04e91 Import PLIB JS code into FlightGear 
- will permit bug-fixing which cannot occur for
  official PLIB.
 2018-05-03 00:02:09 +01:00
James Turner
0762af76c2 Indexed-device config files for event-input layer 
This allows multiple identical devices with event-input, using the
same syntax as for PLIB JS
 2018-04-29 21:52:56 +01:00
James Turner
2ce4bcb666 Allow distinguishing PLIB joysticks by index (count) 
This gives a simple way to have different config files for multiple
identical devices, such as the Saitek/Logitech throttle-quadrant
 2018-04-29 19:56:01 +01:00
James Turner
0cdd6c4e70 Re-init support on HID-event-input 
To ensure the device-configuration-map is re-read, added a default
ctor and moved the config-map creation to init()
 2018-04-29 11:25:51 +01:00
James Turner
921f29fbfe HID-Input disambiguation tweaks 2018-04-09 22:03:49 +01:00
Edward d'Auvergne
0f96032487 TestSuite: Shift of the globals unit test helper functions into the test suite. 
The setting of the SG log levels and developer mode has been removed as this
clashes with the output control set by the test suite main() function.
 2018-03-23 17:26:05 +01:00
Florent Rougon
21b8c7719e FGMouseInput: fix two uses of std::ostringstream 
Seeking to the beginning doesn't clear the contents. Note that we don't
use std::ostringstream::str() either, because apparently
std::ostringstream is too complex to be easily reset to a clean state,
so we just recreate an instance every time we need.

cf. https://stackoverflow.com/a/12112642/4756009 and
    https://stackoverflow.com/questions/624260/how-to-reuse-an-ostringstream/624291#comment437178_624276
 2018-01-27 11:25:48 +01:00
Florent Rougon
99c809dda0 FGMouseInput: add shutdown() and reinit() overrides 
- Modify global_mouseInput in init() and in shutdown(), because *these*
  are the places where FGMouseInput is enabled or disabled.

- reinit() does shutdown() followed by init().

Note: the commented-out block starting with "FIXME: memory leak" that is
      removed here was just an outdated comment, because SGBindingList
      is an std::vector<SGBinding_ptr>, where SGBinding_ptr is a smart
      pointer type (SGSharedPtr<SGBinding>). In other words, there was
      no leak in this place---at least, not recently.
 2018-01-27 11:25:10 +01:00
Florent Rougon
2eb583e9e1 FGMouseInput: improve initialization sequence 
Create the FGMouseInputPrivate instance in FGMouseInput::init() instead
of in FGMouseInput's constructor. This will allow straightforward
implementation of reinit() via shutdown() and init().

Also get rid of the 'initialized' bool, since bool(d) is now equivalent
(d being the std::unique_ptr<FGMouseInputPrivate> data member of
FGMouseInput).
 2018-01-27 11:25:10 +01:00
Florent Rougon
e60e519f7e FGMouseInput: refactoring: use a smart pointer for mouse_mode::buttons 2018-01-27 11:25:10 +01:00
Florent Rougon
1832ade5f2 FGMouseInput: refactoring: use a smart pointer for mouse::modes 
Using std::vector<mouse_mode> instead of std::unique_ptr<mouse_mode[]>
would have been possible of course, but a bit more awkward as
vector<>::size() returns an std::size_t but all the "adjacent" code is
based on the 'int' type.
 2018-01-27 11:25:10 +01:00