1
0
mirror of https://bitbucket.org/librepilot/librepilot.git synced 2024-12-01 09:24:10 +01:00
Commit Graph

6026 Commits

Author SHA1 Message Date
Oleg Semyonov
fd9f1491ba CameraStab UI: fix spacing and add comments according to OP coding style guidelines
More info:
http://wiki.openpilot.org/display/Doc/Coding+Style
http://qt-project.org/wiki/Qt_Coding_Style
2012-08-15 10:31:19 +03:00
PT_Dreamer
d3c5fcfa7e GCS-Handle reload operation config camera stab 2012-08-15 00:13:26 +01:00
PT_Dreamer
ae962f507b GCS-Couple of bug fixes and created a new defaultRequested signal 2012-08-14 23:55:53 +01:00
PT_Dreamer
2ddb0e63d2 GCS-Couple of bug fixes and created a new defaultRequested signal 2012-08-14 23:02:08 +01:00
PT_Dreamer
0f8974178a GCS-Fixes crash if autoLoadWidgets is called after addUAVObject 2012-08-14 18:12:07 +01:00
Oleg Semyonov
64e5ff8e6d CameraStab UI: rescan widget if changed during UAVObject update 2012-08-14 17:52:44 +03:00
Oleg Semyonov
c8e90a537e CameraStab UI: use ConfigTaskWidget API to take care of other widgets 2012-08-14 17:15:53 +03:00
Oleg Semyonov
a0977b405b CameraStab UI: use ConfigTaskWidget API with objrelation dynamic property
This supports most of widgets which are directly related to UAVObjects.
2012-08-14 17:12:55 +03:00
Laura Sebesta
62041f40ad Changed variable from static to class static. 2012-08-12 23:36:27 +02:00
James Cotton
63e490a922 Update the config file so the mag scopes selection works 2012-08-12 16:12:21 -05:00
Stacey Sheldon
4d73c1e2c2 makefile: fix clobbered JTAG macro in firmware-defs.mk
JTAG rules got clobbered at some point.  This should
fix:
   make fw_coptercontrol_program
and others like it.
2012-08-11 22:28:43 -04:00
Stacey Sheldon
9f8d22961f cc makefile: enable -ffunction-sections and -fdata-sections
These compiler options place each function and each global variable
into its own ELF section in each .o file.  This, combined with the
linker option --gc-sections allows the linker to evict unused functions
and variables from the final ELF file.

On CC, the firmware flash bank is only 118784 bytes in total.

This commit reduces the .text segment from 114120 to 83536 and .data
from 572 bytes to 560 bytes.  That frees up a grand total of 30596 bytes
of flash and 12 bytes of RAM.
2012-08-11 20:13:09 -04:00
Stacey Sheldon
29ab1d8cf3 actuator: factor out settings updates from main loop 2012-08-11 19:46:00 -04:00
Stacey Sheldon
63c167c183 taskinfo: also track Event thread's stack usage and CPU utilization
The event dispatcher thread is started differently than most other
threads so it was missed in the taskinfo tracking information.  Now
it's also included.
2012-08-11 18:47:03 -04:00
PT_Dreamer
1a22eef491 GCS-Several changes according to review comments 2012-08-11 23:36:00 +01:00
PT_Dreamer
f4bd4c3120 Merge branch 'pt/MapEnhancements' of ssh://git.openpilot.org/OpenPilot into pt/MapEnhancements 2012-08-11 22:03:09 +01:00
PT_Dreamer
619b4ffa8c GCS-Prevent the map from being dragged beyond the available imagery.
Previous way seemed broken, looked ugly and like a bug.
2012-08-11 19:58:14 +01:00
Stacey Sheldon
bbe86332de hid joystick: mentioned new feature in HISTORY file 2012-08-11 11:28:44 -04:00
Stacey Sheldon
b505602c15 Merge remote-tracking branch 'origin/stac/rc-tx-joystick-emulation' into next 2012-08-11 11:21:18 -04:00
Stacey Sheldon
49c153003c hid joystick: use linearly scaled channels
Scales each channel only based on max and min calibrated values.
The neutral value is now ignored so the joystick sees a linear
range between min and max.

This is particularly useful to allow the full range of values for
throttle to be passed through to the joystick.
2012-08-10 23:36:06 -04:00
Stacey Sheldon
07fde38887 hid joystick: factor out min/max range from report items 2012-08-10 23:29:34 -04:00
James Cotton
cab661e94b AndroidGCS: Update the controller gadget to the newer style updates 2012-08-10 16:16:39 -05:00
James Cotton
9985910969 AndroidGCS TelemetryMonitor: Now that transactions are working more reliably we
can emit the connected signal at the end of fetching all the objects.
2012-08-10 16:11:20 -05:00
James Cotton
38117f6e69 AndroidGCS Telemetry: Telemetry no longer enqueues identical object
transactions multiple times so only the most recent will occur.  Since it
doesn't enqueue the data the end behavior to the UAV should be identical.

Also renamed processObjectUpdates to enqueueObjectUpdates to make it more
intuitive.
2012-08-10 15:58:52 -05:00
James Cotton
1d361a53cd AndroidGCS Telemetry: Make sure when telemetry changes settings it removes it's
listeners before adding more
2012-08-10 15:27:43 -05:00
James Cotton
de26c3297e AndroidGCS Telemetry: Get rid of a tier of redirection through annoymous
classes that was legacy from the GCS port.
2012-08-10 15:22:56 -05:00
James Cotton
6ebf4fe87c AndroidGCS UAVTalk: Pretty thorough clean up focused on how transactions are
handled at the UAVTalk level where there can only be one pending transaction on
the wire.
2012-08-10 14:18:41 -05:00
James Cotton
6b9b49734e Check there is an active telemetry before trying to disconnect it 2012-08-10 10:21:28 -05:00
James Cotton
29d5365f0a AndroidGCS: More refactoring of ObjectManagerActivity. Now the updates are
disconnected on pause and reestabilished on resume.  This only happens within
an onConnected/onDisconnected life cycle.
2012-08-10 10:04:22 -05:00
James Cotton
176922f41c AndroidGCS: Also make the map disconnect its listeners. 2012-08-10 10:04:17 -05:00
James Cotton
749d695f4d AndroidGCS: ObjectManagerActivity maintains a list of objects it is getting
updates from and disconnects them on onStop().  This should probably move to
onPause / onResume
2012-08-10 10:04:11 -05:00
James Cotton
1b8db4eeec AndroidGCS: When priority transmit queue is full this is not a reason to crash. 2012-08-10 02:18:30 -05:00
James Cotton
4bcac07c8c AndroidGCS: Because right now the object updated callbacks are not diconnected
when an activity goes away we need to check the view is valid when processing
the object callbacks.
2012-08-10 02:17:13 -05:00
James Cotton
92209f8e42 AndroidGCS bugfix: Missed a break in a case statement. 2012-08-10 02:17:13 -05:00
James Cotton
5eabb1777d AndroidGCS: Whitespace fixing for TelemetryMonitor 2012-08-10 02:17:12 -05:00
James Cotton
78469aa6a4 AndroidGCS: Use a WeakReference to stop the OPTelemetryService handle leaking. 2012-08-10 02:17:12 -05:00
James Cotton
37c4880fba AndroidGCS: Unregister the activity broadcast receiver onStop(). Create it
onStart() instead.
2012-08-10 02:15:38 -05:00
James Cotton
57b91b7763 AndroidGCS: UAVTalk no longer extends Observable since there is only one
listener anyway we use a setListener pattern.
2012-08-10 02:15:37 -05:00
James Cotton
f13f977177 Fix: Revert change to ObjectBrowser 2012-08-10 02:15:37 -05:00
James Cotton
7a42c7b10e AndroidGCS: Start supporting NAK and having uavTalk emit a transaction failed
and succeeded call.  Had to change the mask on the AndroidGCS side to match the
FC side.
2012-08-10 02:15:37 -05:00
James Cotton
1740bef847 AndroidGCS: Delete zip file of android icons 2012-08-08 14:25:52 -05:00
James Cotton
7cca5696a2 Merge remote-tracking branch 'origin/pt/boardbased_ui_limits' into next 2012-08-08 14:11:24 -05:00
Laura Sebesta
fe6ea7e0e2 Fixed getUAVPosition to use position actual data correctly.
Refactored coordinate conversion names in order to be clearer, since
two identically named functions-- one in revo, one here-- performed
very different maths.
2012-08-08 20:58:13 +02:00
Kenz Dale
f3e1e768ad Refactored variable names to be more helpful. 2012-08-08 20:57:41 +02:00
Kenz Dale
6b3cb29ea1 Fixed maximum size for title bar in map widget. 2012-08-08 20:56:29 +02:00
James Cotton
06a6e367b0 Merge remote-tracking branch 'origin/hyper/OP-661' into next 2012-08-08 13:49:48 -05:00
James Cotton
41a46258be AndroidGCS: Change Sambas to not crash in AVD for bluetooth or if there are no
devices.
2012-08-08 13:43:09 -05:00
James Cotton
dd0a5b2cf2 AndroidGCS: Make home page buttons more vertical to work better on phones.
Maybe someone could make these automatically space out?
2012-08-08 13:17:30 -05:00
James Cotton
11a08c186c AndroidGCS: Refactor the Fragments into their own package to keep the UI
"components" segregated.  Is this good practice though?  It means making a lot
of functions for the API between ObjectManagerActivity and Fragment public.
2012-08-08 13:14:38 -05:00
James Cotton
95dfc88f95 AndroidGCS: Move the PFD functionality into a fragment and then
programmatically instantiate that from the test PFD activity
2012-08-08 13:14:01 -05:00