The latest MCT in MCUX 10.3.1 seems to have an issue with UTF-8 characters. I have a Kinetis K24 project with SDK 2.5.0, and I've configured the PIT peripheral and a single channel. But when I regenerate my peripherals.c, exit, and come back, MCT thinks I need to regenerate my code again because one of the documentation comments has supposedly changed. Here's the actual code:
/***********************************************************************************************************************
* ANALOGIN_PIT initialization code
**********************************************************************************************************************/
/* clang-format off */
/* TEXT BELOW IS USED AS SETTING FOR TOOLS *************************************
instance:
- name: 'ANALOGIN_PIT'
- type: 'pit'
- mode: 'LPTMR_GENERAL'
- type_id: 'pit_a4782ba5223c8a2527ba91aeb2bc4159'
- functional_group: 'BOARD_InitPeriph_ANALOGIN'
- peripheral: 'PIT'
- config_sets:
- fsl_pit:
- enableRunInDebug: 'false'
- timingConfig:
- clockSource: 'BusInterfaceClock'
- clockSourceFreq: 'GetFreq'
- channels:
- 0:
- channelNumber: '0'
- enableChain: 'false'
- timerPeriod: '100 us'
- startTimer: 'true'
- enableInterrupt: 'false'
- interrupt:
- IRQn: 'PIT0_IRQn'
- enable_priority: 'false'
- enable_custom_name: 'false'
* BE CAREFUL MODIFYING THIS COMMENT - IT IS YAML SETTINGS FOR TOOLS **********/
/* clang-format on */
const pit_config_t ANALOGIN_PIT_config = {
.enableRunInDebug = false
};
void ANALOGIN_PIT_init(void) {
/* Initialize the PIT. */
PIT_Init(ANALOGIN_PIT_PERIPHERAL, &ANALOGIN_PIT_config);
/* Set channel 0 period to 100 µs. */
PIT_SetTimerPeriod(ANALOGIN_PIT_PERIPHERAL, kPIT_Chnl_0, ANALOGIN_PIT_0_TICKS);
/* Start channel 0. */
PIT_StartTimer(ANALOGIN_PIT_PERIPHERAL, kPIT_Chnl_0);
}See line 39 there? MCT inserted the comment "Set channel 0 period to 100 microseconds", using Greek "mu" for "micro", which is correct. The "mu" displays correctly in the Eclipse editor window and in my third-party text editor. However, MCT doesn't know how to handle this when determining whether to regenerate the code. peripherals.c is always flagged for regeneration because of that line. If I click the green "Update Project" button, then click the peripherals.c "change" link to compare what's changed, this is what I see:
It's not reading the UTF-8 "mu" correctly, so it shows two question marks in the "on disk" box, and one question mark in the "newly generated" box.
I don't know if this is an Eclipse issue with UTF-8 compatibility, or if this is purely an NXP software component, but either way, this is rather frustrating, because there's no workaround; even if I hand-edit peripherals.c to use the Latin "u" instead of Greek "mu", MCT is still going to flag peripherals.c for regeneration. Any chance you can fix this in the next release?
