* SF Audio amplifier mute port to 2.10
* added missing definition STR_SF_DISABLE_AUDIO_AMP
* exclude radios not able to switch audio amp off
* radio/src/storage/yaml/yaml_datastructs_128x64.cpp needs to distinguish between radios with or without switchable audio amps. Expample Zorro vs X9lites
* updated companion and radio language PL
* - added Companion and radio language SE
- removed type unfinished for translated languages
* added radio and companion language JP
* - reverted change to 128x64 yaml_datastructs.h
- moved exclusion of SF DISABLE_AUDIO_AMP based on hw capabilities to UI level
* confused FUNC_ with FUNCTION_
* confused FUNCTION_* with FUNC_*
* feature request #1635
- added MIN (-MAX -> -1024) as source
* fix rebase
* fix rebase
* change to use language strings for MIN/MAX instead of instance
updated CN and TX language file for MINuse -RESX for MIN instead of magi numbercode cleanup
* updated test sources
* updated JP language
* JP language: changed translation text
* updated HE language translations
* fixes#2552
- remove edit option rssiSource
- reset rssiOption to default (none) if carried over from OpenTX conversion
* deleted rssiSource functional code Companion, simulator and radio
* forgot one commented line
* fixes#2713
changes:
- use getMultiOptionTitle() instead of rfproto->getOptionStr() as getMultiOptionTitle() takes updated MPM status in account
* code cleanup: removed obsolete functions
* fix#3046
* fixed my complete misunderstanding of how the chosen protocol is updated in Multi DSM auto mode
* removed obsolete check
* update subType list only on rf protocol change
added some comments
* reconsidered implementation utilizing LV_EVENT_REFRESH to avoid sporadic loss of update
* editorial change
* fix problem where a successful MPM DSM2 auto bind would reset the receiver ID
* reset DSM2autoUpdated just in case
* - Concentrated Multi protocol definitions into a consistent single source. Actually two .h files, one for the protocol list and names, the other for subtypes and capababilites, see comments in .h files as to why
- updated available protocols reflecting 1.3.3.20 (15+ protocols)
- fixed Companion always showing f/s and channel map options in model setup disregarding the protocol capability
* Following @pfeerick's review: changed heaeders and typos
fixed Ubuntu runtime error on assert
fixed change of Multi protocol not triggering writing yaml file
* corrected copy/paste error
* fixed FrSky index juggling
* Updated Multiprotocol database to current MPM project status (ref. MPM Multiprotos.ino)
- changed protocol name FX816 to FX
- added Flysky AFHDS2A options PPM and PPWM SBUS 16ch subtypes "PWM,SB16" and "PPM,SB16
- added V761 option to select TOPRC
- added FX options "816" and "620"
- added FRSKYR9 option "-- 8ch"
* MPM protocol scan excludes some protocols which are not meant to be selected. For Companion and Simualtor (and for radios if there is no MPM connected) this was handled in different places with outdated information. Harmonized and centralized the handling of non-selectabale protocols into the single source for all companion, Simulator and all gui types.
* More cleanup:
- changed some MPM protocol names to the ones the MPM delivers with MPM protocol scan to harmonize names between CPN/Simulator/radio
- added XN2927DUMP and MOULDKG to the list of excluded MPM protocols just like the MPM does with MPM protocol scan
- added new option value limits and moved definitions to single source
- adapted CPN and radio to use common definitions of option values for all GUI types
- moved distributed MPM (as far as possible) to one coherent place under a single #if defined (MULTIMODULE)
- fixed option value not reset at change of protocol
- some editorial stuff
Note: There is still some work to be done on CPN GUI option values. DSM2 servo rates are hard coded, BAYANG telemetry options are only shown by index and not by text as in radio code, AFHDS2A servo update rate is shown as index too which is to no use for users.
I'll try to fix all this in the next session, but request to get rid of BAYANG telemetry options translations. It is much easier to define a common string list by using untranslated.h
* rebase due to fr.h
* resolve merge conflict with main fr.h
* more cleanup:
- deleted multiple definition of MM_RF_CUSTOM_SELECTED in favor of single source define
- bw radios: change to only allow defined protocols rather than a fixed 127 protocols
* Changes:
- magic sentinel number 0xfe as define
- add 10 NN protocol entries for future use of new MPM protocols not yet defined in ETX (still recommend timely ETX maintenance update if MPM releases new firmware)
* Changes by @elecpower:
- align field labels to radio
- change servo update frequency from combo box to spin box with edit only via up/down arrows to stop invalid values being entered via keyboard. Remove frequency list used to populate combo box as only used by Companion.
- add validation to module yaml decode checking for unsupported multi protocols and/or multi protocol sub types and if unsupported set module to OFF just to be safe.
* Changes:
- radio: added validation to module yaml decode checking for unsupported multi protocols and/or multi protocol sub types and if unsupported set module to OFF just to be safe.
* CPN: use sentinel define
* cpn: Sort multi protocols list (courtesy @elecpower)
* fixes#3480
changes:
- moved FrSky subtype defines to MultiProtoDefs.h
- updated bind_button_driver.cpp
* missed replacing one instance of setMultiProtocol()
* changes:
- editorial change
- added multi option to OMP protocol (ref #2713)
* rather go for STR_MULTI_RFTUNE because MPM manual talks about RF tune
* made XN297_DUMP selectable (available with MPM debug fw only)
* corrected merge mistakes
* RX stat (formerly known as "RSSI") labeling dependent on protocol (RSSI/dbm, Rqly/%, Sgnl/-) used for range check, multi rf tuning and low/critical warning settings pages
* chore: Bump libopenui
* removed define that was supposed to be for testing only
function getRxStatLabels(): wasting flash as inline, no need to be inlined
* removed inline on getRxStatLabels() - step 2
* introducing external MLink module telemetry as module PPM subtype
* implemented fix for sensor discovery display not showing all detected sensors (is in separate PR too)
implemented PPM subtypes in Companion
* add PPM subtype to model printer
finalize subtypes for PPM in Companion
* added logic to try to use HW UART with inverters turned off first (e.g. NV14), then, if not successfull use soft serial inverted (== uninverted data due to HW inverters on PCB)
changed start bit detection IRQ priority from 5 to the previous 0 (also in separate PR)
* extended telemetry timeout to 2s due to slow Mlink RSSI update rate
* changed the "NOTLM" sub-type texts to:
- "No Telem" for 128x64 (this is the max numbers of characters it can take)
- "No Telemetry" everything else including Companion
* Companion:
- lower case 1st charachter on PpmSubTypeStrings to ppmSubTypeStrings
- added telemetry hooks for ppmSubTypeStrings stringx
* simplified serial port initialization for MLink telemetry
* editorial change: corrected obsolete comment
* fixes#3509
Changes:
- change to order of include files
- remove unnecessary include
* revert deletion of include, some radios seem to use it.
* exclude code for CONFIGURABLE_MODULE_PORT for target bootloader
* restore bit start detection IRQ for soft serial
* restore bit start detection IRQ for softserial (target Taranis and Horus)
* target Taranis: EXTI9_5 is a shared interrupt for Rotary Encoder and Heartbeat for X7, X7-ACCESS und TPRO radios. Changed IRQ priority accordingly
* fix crashes where the heartbeat ISR running at interrupt priorities higher than 5 interferes with FreeRTOS calls made from within the heartbeat ISR.
* changed mixer scheduler timer initialization to allow for software generated update events to avoid freeRTOS calls from now prio 0 heartbeat ISR.
* resolved review feedback: using defines for IRQ prio settings
* following review by @raphaelcoeffic:
- new function void mixerSchedulerSoftTrigger() (to be called by heartbeat driver)
- defines in hal.h for targets horus and taranis
* following @raphaelcoeffics review
- changed all but telemetry EXIT IRQ (soft serial) definition back to their original state and added define to override EXIT IRQ (soft serial) prio
* fixes https://github.com/EdgeTX/edgetx/issues/3273
* empty commit to trigger checks
* Companion: moved yaml tag receivers correction to yaml parserCompanion and radio: updated comments
* - all telemetry sensor name definitions in one place (untranslated.h)
- all telemetry sensor name definitions follow the naming convention STR_SENSOR_<sensor>
- eliminated one redundant definition
- editorial updates to telemetry protocols code to reflect the naming convention changes
- re-introduced translations in da.h (STR_SENSOR_THROTTLE and STR_SENSOR_SPEED)
- re-introduced translation in it.h (STR_SENSOR_GPSDATETIME)
- re-introduced translation for fr.h ( STR_SENSOR_BATT)
- saved 160 Bytes of flash
* removed obsolete translation
saved 280 byte flash memory in total
* finally removed extern const char STR_MENUTEMPLATES[]; definition
adjusted da.h, fr.h, it.h local sensor name definitions to naming conventions STR_SENSOR_<senosr>
* corrected merge
removed obsolete #ifdef
* some more obsolete ones
* added TW and CN translation for TR_PERIOD and TR_INTERVAL
fixed problem in /radio/src/gui/colorlcd/fullscreen_dialog.cpp preventing builds for CZ, IT and FR languages
* fixes https://github.com/EdgeTX/edgetx/issues/2965
Changes:
- the log period range is now 0.1s to 25.5s
- upon creation of a new SD Logs SF the period is initialized to 1.0s
- text "Value" is changed to "Period" to indicate parameter represents the log period
ToDo:
- translations for "Period" if desired
rebase due to #2992
* changed Period to Interval
Added translations for DA, DE, EN, IT, PL
* added defaults for TR_INTERVAL to language files
* fixed problem with too many leading zeros for intervals less than 1s.
before: 00.5s
after: 0.5s
* added CZ translations for TR_PERIOD and TR_INTERVAL
changed EN TR_FUNC from "Func" to "Function"
* updated TR_FUNC for CZ, DA, IT, JP, PL
updated TR_PERIOD and TR_INTERVAL for JP
* added TR_PERIOD, TR_INTERVAL, TR_FUNC for SE
* implemented TR_SF_SWITCH and logic to change between TR_SWITCH and TR_SF_SWITCH
* 212x64 radios: single source for #define TR_LSW_HEADERS in language files
deleted redundant defines
* changed 128x64 and 212x64 SF SD Logs to same logic as colorlcd with
- minimum log interval 0.1s
- maximum log interval 25.5s
- default log interval 1.0s
TX12_Sim and X9D_Sim tested ok
* migrated functional changes due to PR #2992
* missed updating TW language file
* fix: Always `STR_SF_SWITCH` regardless of SF
Co-authored-by: Peter Feerick <peter.feerick@gmail.com>
* migrated stack diagnostics (available, free) to using built in FreeRTOS builtin function
Note: main stack must be and is treated differently as it is not under FreeRTOS control. main stack is painted by the STM32 startup routine. free main stack is calculated checking the remaining painted stack area.
* corrected my merge error
* missed a blank
* centers flight mode text, 6Pos indicator and 6Pos numbers in square
replaces ticks with numbers in 6Pos indicator
* stole @philmoz idea to bold the selected switch position to make it stand out more
* fixes issue https://github.com/EdgeTX/edgetx/issues where a change of the source value in widget settings was not written back to the model .yml file.
* fixes https://github.com/EdgeTX/edgetx/pull/2475
extended force write to .yml to all widget settings option types
* fixes issue https://github.com/EdgeTX/edgetx/issues/2465 where a change of the source value in widget settings was not written back to the model .yml file.
* https://github.com/EdgeTX/edgetx/issues/2465
extended force write to .yml to all widget settings option types
* removed unnecessary SET_DIRTY() for ModelTextEdit() as it is already handled by ModelTextEdit()
changed indent for SET_DIRTY() in lambda for Choice()
* removed unnecessary SET_DIRTY() for ModelTextEdit() as it is already handled by ModelTextEdit()
changed indent for SET_DIRTY() in lambda for Choice()
* first commit for FrSky 8s Lipo Sensor
FLVS Advanced transfers up to 8 cell voltages for Cels sensor
* correcting mistake made transferring local change to PR
* complementary extension of Companion telemetry simulator to enable simulation of 8 cells
* change to allow selection of cells 7 and 8 in calculated sensor with formula Cell
* (cpn) Add bin and yml conversions and basic semantic version helper
Co-authored-by: elecpower <neilh713@tpg.com.au>
* initial commit for Module / RX version pop up - fixes#2320
* got rid of "no information" module names. Module names are now identical to the settings internal and external RF dialog, expect PXX2 which sets name depending on hw info.
* cosmetics
* hide int/ext status when there is no status to report
* fix: Also unhide PXX2 module status if set
* feat: Move NV14 fw version to module info screen
* Added audio/haptic feedback for all hard keys excpet trim switches.
Changed logic from application decides if feedback has to be given (was inconsistent and inefficient) to key driver detects button down and triggers immediate feedback if radio settings allowed for acoustic and/or haptic feedback (radio settings, radio setup, sound set to All)
With this change all instances of the application calling for feedback were removed in favor of the centralized approach.
This first commit also introduces a dummy function for providing feedback on key presses in boot.cpp because keys.cpp is also used as the bootloader key driver but with no logic to determine if feedback is allowed. Using a define that indicates keys.cpp is compiled for use in the bootloader would make a much nicer and backwards compatible solution.
* don't want to change libopenui so silence onKeyPress (application requests feedback) in audio.cpp
* made generation of acoustic/haptic feedback conditional on BOOT define, i.e. no feedback if keys.cpp is compiled for the bootloader
* removed onKeyPress()
* fixed crash on acoustic/haptic feedback on hard keys by restoring keys.cpp and moving logic to LvglWrapper.cpp/keyboardDriverRead().
Removed call to getWindowEvent() which would just call getEvent(false) by directly calling getEvent(false). This allowed to remove getWindowEvent() entirely as it was only used by keyboardDriverRead().
* fix: Audio header include no longer needed
This addresses issue #1778 (https://github.com/EdgeTX/edgetx/issues/1778)
Select "All" in radio settings/radio setup/Sound/Mode for key and touch acoustic feedback and settings/radio setup/Haptic/Mode for key and touch haptic feedback.
"NoKey" also disables touch feedback.
The adding of UTF-8 support in 30e3ac3 / #1900 meant there was a second character present
in the symbol character strings, which was not expected here when trimming them off.
fixed issued where log steps would be missed due to 10ms timer resolution
- simulator: allow for one get_tmr10ms() tick (10ms) deviation
- radio: switched to FreeRTOS software timer at user requested period
code cleanup
log.cpp: made variable lastLogTime static
log.cpp: deleted unused variable g_logError
functions.cpp, logs.cpp, sdcard.h: renamed variable logDelay to logDelay100ms to show it is scaled to 100ms increments
* transferred respective OpenTX updates to HoTT telemetry
* integrated LUA function flushAudio from OpenTx
* sensor names moved to “untranslated.h”
reverted *.h.txt to original mainline content
removed leading "Z" in sensor name defines